Peer Review of Electrified Vehicle
Simulations within EPA's ALPHA Model
&EPA
United States
Environmental Protection
Agency
-------
Peer Review of Electrified Vehicle
Simulations within EPA's ALPHA Model
Assessment and Standards Division
Office of Transportation and Air Quality
U.S. Environmental Protection Agency
Prepared for EPA by
ICF International
EPA Contract 68HERC21D0016
Task Order 68HERC22F0351
This technical report does not necessarily represent final EPA decisions
or positions. It is intended to present technical analysis of issues using
data that are currently available. The purpose in the release of such
reports is to facilitate the exchange of technical information and to
inform the public of technical developments.
NOTICE
4>EPA
United States
Environmental Protection
Agency
EPA-420-R-23-004
February 2023
-------
Table of Contents
I. Introduction 1
II. Peer Review Process 2
Selecting Reviewers 2
Administering the Review and Receiving Comments 3
III. Summary of Peer Reviewer and EPA Comments 4
Comment Overview and Summary 5
IV. U.S. EPA's Responses 9
Summary Observations By EPA 10
ALPHA'S Purpose 10
Reason for ALPHA and Associated Documentation Release to the Public 11
The Goal of ALPHA Peer Review 11
Responses to Peer Reviewer's Comments by Charge Question 12
Appendix A: Comments by Reviewer (Unedited) 35
Comments by Sujit Das 35
Comments by Dr. Shawn Midlam Mohler 39
Comments by ICCT 47
Appendix B: Peer Reviews Curriculum Vitae (CV) 55
Sujit Das 55
Shawn Midlam-Mohler 63
Oscar Delgado 85
Appendix C: Conflict of Interest (COI) Forms 90
Appendix D: Notes from Mid-review Teleconference 93
Appendix E: Peer Reviewer Selection Memo 100
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
I. Introduction
As the U.S. Environmental Protection Agency (EPA) Office of Transportation and Air Quality (OTAQ)
develops its programs to control greenhouse gas (GHG) emissions from light-duty highway vehicles,
there is a need to evaluate the effectiveness of technologies likely to be used to meet these
standards. EPA has further developed the Advanced Light-Duty Powertrain and Hybrid Analysis
(ALPHA) vehicle simulation model to perform simulations for electrified vehicles. The Electrified
Vehicle Simulations module of the EPA ALPHA model uses the industry standard MathWorks
software products, MATLAB and Simulink (version 2020a). The entire model and all subsystems are
unlocked for complete transparency. A comprehensive peer review of the new ALPHA simulation
capabilities of electrified vehicles is imperative to assess the technical approach and accuracy of
the ALPHA model as EPA's regulatory data analysis tool by the regulated light-duty automotive
vehicle community.
EPA's guidelines specify that all highly significant scientific and technical work products shall
undergo independent peer review according to specific agency protocols. This process is designed
to ensure the use of the highest quality science in its predictive assessments and to assure
stakeholders that each analysis/study has been conducted in a rigorous, appropriate, and defensible
way. Therefore, EPA submitted the model for external peer review to assess whether the model has
been developed in a rigorous, appropriate, and defensible way.
The peer review was conducted from June to November 2022 in accordance with the current
version of EPA's Peer Review Handbook) At the conclusion of the review process, ICF collected all
unedited peer reviewers' comments and provided them to EPA. This technical report contains a
summary of the reviewers' comments to EPA's charge questions, along with the unedited answers
presented by each peer reviewer. The document also describes how peer reviewers were selected,
the process that ICF took to administer the peer review, and how the peer review was concluded.
Supporting documentation collected from the reviewers, including their curriculum vitae (CV) and
conflict of interest (COI) statements, is also provided.
The following materials are included in this technical report:
• Description of the Peer Review Process (Section II)
• Reviewer Responses to Charge Questions (Section III)
• EPA's Responses to Peer Reviewer Comments (Section IV)
• Unedited Comments by Reviewers (Appendix A)
• Reviewer Supporting Documentation (Appendix B and Appendix C)
• Notes from mid-review teleconference (Appendix D)
1 U.S. Environmental Protection Agency, Peer Review Handbook, 4th Edition, October 2015. Prepared for the U.S.
EPA by Members of the Peer Review Advisory Group, for EPA's Science Policy Council, EPA/100/B-15/001.
Available at http://www.epa.gOv/o , including OMB's Information
Quality Bulletin for Peer Review (Handbook, Appendix B) provisions for the conduct of peer reviews across
federal agencies.
1 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
• Peer Reviewer Selection Memo (Appendix E)
II. Peer Review Process
ICF conducted the peer review in three stages. First, ICF identified a qualified set of reviewers;
second, ICF contracted with the selected peer reviewers and conducted the review; then, ICF
collected reviewers' feedback on the model. Finally, ICF documented the peer review process, as well
as the comments and feedback from the peer reviewers in a technical memo that was submitted to
EPA. Ultimately, EPA will convey results of the peer review process to the developers of the ALPHA
model, who will respond to the comments received. The following sections provide details on these
steps.
Selecting Reviewers
To assemble the panel of three independent peer reviewers, ICF reviewed a pool of subject matter
experts both suggested by EPA OTAQ and identified by ICF through independent research. ICF first
assessed the experts' availability to perform the peer review within the timeline agreed upon with
the EPA Task Order Contracting Officer Representatives (TO COR). After that, ICF reviewed
curriculum vitae and other relevant work to select peer reviewers that represent a combined
expertise that cover, at a minimum: understanding of vehicle technology packages including battery
technology, hybrid and electric powertrains, e-motors, transmission systems (e.g., shift strategy),
and vehicle accessories as well as engine fuel consumption map, and vehicle behavior.
The initial list of peer reviewers was based on both EPA's initial recommendations and ICF's
suggestions for additional potential reviewers. Twelve candidates (five recommended by EPA and
seven identified by ICF) were considered. ICF also prioritized peer reviewers based on the relevance
of their background and experience with the topic of the report. Through an initial contact with the
selected peer reviewers, ICF assessed each potential reviewer's ability to perform the work during
the period of performance and to identify any association they have with the work that would
preclude them from being objective. ICF contacted and communicated with all candidates by e-mail.
In our outreach we identified ourselves as independent contract employees and provided initial
information on the ALPHA model, including the newly added electric vehicle model and the expected
time commitment to exercising the model. We asked the potential reviewers to assess their
availability for this study and for their hourly rate. We also collected a curriculum vitae for each peer
reviewer that expressed availability and interest in participating.
Upon completion of the initial contact, the top three peer reviewers selected for this project agreed
to participate in this peer review process. The rest of the peer reviewer candidates were either not
available, not interested (e.g., retired), or had concerns with the limited time allocated for the review
(i.e., 20 hours). The resumes for the three selected candidates were collected and shared with U.S.
EPA TO COR. Upon approval from EPA TO COR, ICF initiated the subcontracting process with the
selected peer reviewers.
2 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
The three selected peer reviewers (including the iCCT team) provide a diverse combination of
expertise in evaluating the ALPHA model. Sujit Das, with 37 years of experience in energy efficiency
research, has served as the peer reviewer of the ALPHA model back in 2016, and has published
articles related to powertrain design for advanced fuel vehicle technologies. Shawn Midlam-Mohler is
a Professor of Mechanical Engineering at Ohio State University, with expertise in engine selection,
modeling, and control development for an extended range electric vehicles as well as vehicle
simulations and powertrain optimization. Oscar Delgado and ICCT team also bring in years of
experience in modeling advanced technologies and developing tools to support global commercial
fleets in their transition to zero-emission vehicles.
Sujit Das
Shawn Midlam-Mohler Oscar Delgado
** Note that while Oscar Delgado will be our point-of-contact, ICCT has decided to review the model as a team (Oscar
Delgado, John German, Hussein Basma).
1) Sujit Das
Strategic Analysis, Inc.
300 Harp Street
Alpharetta, GA 30009, USA
suiit das2021@outlook.com
2) Shawn Midlam-Mohler
ModelTek, LLC
PO BOX3590
Columbus OH 43210
dr.smm.llc@gmail.com
3) Oscar Delgado, Ph.D.
John German
Hussein Basma, Ph.D.
The International Council on Clean Transportation
1500 K St NW Suite 650, Washington DC 20005
Oscar@theicct.org
ICF anticipated that this selected group of reviewers would provide extensive and complementary
expertise to conduct the peer review. ICF provided an overview of the final list of reviewers in the
July 1, 2022, Peer Review Selection Memo to EPA.2
Administering the Review and Receiving Comments
In conducting the peer review, ICF composed and delivered a charge letter to the three selected
peer reviewers along with the model codes and documentation, and a conflict of interest (COI) form
2 Peer Review Selection Memo for Task Order 68HERC22R0170 for Peer Review of Electrified Vehicle Simulations
within EPA's ALPHA Model, to Jeff Cherry, US EPA OTAQ, from: Sam Pournazeri, ICF.
3 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
for peer reviewers to fill out and return to ICF along with their comments. The charge letter included
EPA's charge questions to the reviewers, instructions on how to complete the review, and a timeline
of when comments were due to ICF. ICF sent these materials to each individual reviewer on
September 29, 2022. The charge questions submitted to each peer reviewer were as follows:
1) Does EPA's overall approach to the stated purpose of the model (demonstrate technology
effectiveness for various fuel economy improvement technologies) and attributes embody that
purpose?
2) What is the appropriateness and completeness of the overall model structure and its
components, such as:
i. The breadth of component models/technologies compared to the current/future light-duty
fleet
ii. The performance of each component model, including the reviewer's assessment of the
underlying equations and/or physical principles coded into that component.
iii. The input and output structures and how they interface with the model to obtain the
expected result, i.e., fuel/energy consumption and C02 over the given driving cycles.
iv. The use of default or dynamically generated values to create reasonable models from limited
data sets.
3) Does the ALPHA model use good engineering judgement to ensure robust and expeditious
program execution?
4) Does the ALPHA model generate clear, complete, and accurate output/results (C02 emissions
or fuel efficiency output file)?
5) Do you have any recommendations for specific improvements to the functioning or the quality
of the outputs of the model?
ICF then arranged and hosted a teleconference on October 6, 2022, with the selected peer reviewers
and EPA. The goal of the meeting was to introduce the peer reviewers to the EPA staff and address
early questions or concerns. The meeting included an overview of the review process, background
information on the model, and a discussion on technical and practical aspects. ICF's notes from this
meeting are included in the kick-off meeting notes that ICF shared with EPA TO COR on October 6,
2022 (Appendix D).
ICF requested that the peer reviewers provide responses to the charge questions and complete COI
form within two weeks, however ICCT as well as Dr. Shawn Midlam-Mohler requested an extension of
the deadline by one week, respectively. All peer reviewer comments and completed COI forms were
received by October 31, 2022. ICF organized all comments into tables so that the individual
comments could be easily grouped and compared for review purposes.
III. Summary of Peer Reviewer and EPA Comments
Section III presents an overview of the peer reviewers' comments received on the five charge
questions, as well as a summary of EPA's responses. This overview is followed by Section IV, which
provides EPA's detailed responses to the direct, unedited peer reviewer responses to each of the
charge questions. In Appendix A, the unedited responses by reviewer appear in a table format. In
4 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
those tables, the left column lists the EPA's charge question, and the right column provides the
reviewer's comments.
Comment Overview and Summary
The following section summarizes the peer reviewers' comments to the charge questions. The
questions have been abbreviated for easier presentation. These summaries do not rewrite the
responses or supersede the unedited comments presented in Section IV.
All three reviewers provided additional comments beyond those requested by the five prescribed
charge questions. Those are not summarized here but are presented in their entirety in Section IV.
Question 1: Does EPA's overall approach to the stated purpose of the model (demonstrate
technology effectiveness for various fuel economy improvement technologies) and
attributes embody that purpose?
The reviewers found that EPA's overall approach to the stated purpose of the model and attributes
embody the goals as outlined. Sujit Das added that considering that ALPHA model was primarily
developed to assess the GHG emissions of internal combustion engines (ICE) light-duty vehicles, it
may not be well suited to evaluate alternative pure EV technologies. Shawn Midlam-Mohler raised
that ALPHA in its current state does not offer any novel features compared to the current tool
portfolio in the public domain. If one goal for ALPHA model is to be widely accepted to quantify
technology effectiveness and C02 emissions of electric vehicle technologies, Shawn warns that
there may not be enough distinct features for ALPHA to be preferred over other models of its class.
ICCT suggests that ALPHA developers provide greater clarification on powertrain components'
characteristics and what electric motor-generator technologies they are based on. Users may be
interested in the assessment of other electric motor-generator technologies that scale differently
based on vehicle type. ICCT also raised a question on the exclusion of the fuel-cell powertrains from
the model.
EPA Summary Response: EPA supports the reviewers' finding that our overall approach to the
stated purpose of the model and attributes embodies the goals as outlined. We also agree with Sujit
Das that the earlier versions of ALPHA were focused on C02 production; however, implementing a
pure EV model (which uses the same non-powertrain sub-models within ALPHA) was a relatively
minor expansion of ALPHA's capabilities for version 3.0. In reference to Shawn's concern about the
lack of distinct features, ALPHA is explicitly not intended to compete with commercial vehicle
simulation products or supplant manufacturers' own modeling packages, and thus its user interface
features should not be expected to be at the same level as commercial simulation tools. EPA's goal
of the public release of ALPHA is to provide sufficient documentation and transparency to allow
thorough review by stakeholders of EPA rulemakings. EPA also agrees that providing greater
clarification on powertrain components would be useful to stakeholders. EPA is in the process of
documenting the electric component data used in ALPHA modeling and will be publishing the
information on a new EPA web page dedicated to electric components.
5 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Regarding clarification of powertrain characteristics\ peer reviewers were provided four electric
motor maps with detail specifications, one for use in each model (i.e., one for the BEV, one for the
PowerSph't strong hybrid, one for the P2 strong hybrid, and one for the P0 mild hybrid). Any of these
component maps can be scaled and ALPHA is also capable of accepting other motor maps as input.
Finally, the exclusion of fuel cells is not unique. The electrified configurations currently populated
within ALPHA - BEVs, PowerSph'ts, P2s and POs - represent the most popular high-volume electrified
configurations. Based on its analysis, EPA concluded that other currently lower volume
configurations, such as series parallel hybrids, PI mild hybrids, and fuel cell electric vehicles, were
likely to remain lower volume and not significantly affect EPA's characterization of the overall future
fleet
Question 2; What is the appropriateness and completeness of the overall model structure
and its components?
The reviewers agreed that the ALPHA model's structure and its components are sufficiently
appropriate and complete to achieve the stated purpose. However, each expressed that there is
room for improvement across the breadth of model components, model performance, input and
output structures, and use of limited data sets.
Sujit Das qualifies that ALPHA model's parameter files at the five major EV component levels are
accessible, provided that users are moderately experienced with MathWorks software. Although the
model as provided was unlocked for complete transparency, Sujit recommends that a separate
document with equations and their descriptions would be invaluable to users regardless of
MATLAB/Simulink aptitude. An explicit list of equations would enable a higher level of input data
validation.
Shawn Midlam-Mohler and ICCT both agreed after an extensive look at control algorithms for the BEV
and PHEV platforms that there is room for improvement. Shawn expressed concern with ALPHA'S
use of relatively simple rule-based control code and with the presence of some developmental
comments in the control code provided to the peer reviewers. Another weakness of the current
control algorithm is the relationship it has to the model's seemingly calibration-based data, which
makes it difficult to discern how well components of controls can be scaled to create other
reasonable models.
ICCT concluded the different technology and modeling choices are all reasonable, and special
attention was given to the hybrid powertrain controllers. However, ICCT questioned whether the
current control strategy for the P2-PHEV platform is representative. ICCT offered credit to the rule-
based control strategy's ability to be computationally efficient but countered that there are other
alternative control strategies that can better reduce total fuel and energy consumption. ICCT also
posed that although constant-power auxiliary modeling, as provided, is complete and appropriate,
there are recent developments in the literature that can inform ALPHA developers of how to
incorporate dynamic auxiliary operation and impacts.
6 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
EPA Summary Response: EPA agrees with the reviewers that the ALPHA model's structure and its
components are sufficiently appropriate and complete to achieve the stated purpose, and
improvements could be made to the ALPHA package. As suggested by Sujit Das, EPA will work to
incorporate a description of the underlying equations and/or physical principles coded into
components.
EPA agrees with Shawn Midlam-Mohler that ALPHA's control code is rule-based (focused on rules
observed in testing of production hybrid and battery electric vehicles) but does not view this as a
concern or deficiency. EPA models representative production (or production-ready) components
and control strategies that can, on average, be used to simulate the performance of the current fleet
to provide reasonable C02 and energy consumption estimates for technologies and strategies
manufacturers might choose to employ in future fleets. This approach grounds our vehicle models in
the actual control calibration and performance of production vehicles and avoids overlooking any
design constraints present in a vehicle manufacturer implementation of new or existing technology.
In addition, EPA agrees with Shawn that any unused experimental code should be cleaned up. EPA
plans to add more appropriate comments to the code but leave the commented-out code itself in
place as a basis for potential future development.
ICC T commented the different technology and modeling choices are all reasonable. There was
concern with the representativeness of ALPHA'S P2-PHEV strategy related to the modeled control
strategy. EPA constructed the P2 model based on the operation of a popular and well-performing
vehicle in the fleet (in this case a Hyundai Sonata) The engine control strategy in ALPHA
demonstrates a good match with test data from the vehicle operating on the regulatory cycles.
Although other more optimal control strategies may become available in the future, for the purposes
of the current rulemaking, EPA incorporated vehicle components and control strategies that already
exist within the current fleet and are representative of the performance of a broad range of vehicles.
Question 3: Does the ALPHA model use good engineering judgement to ensure robust and
expeditious program execution?
The three reviewers expressed that although the general approach ALPHA model takes is sound and
comprehensive, that some of the techniques and priorities of the model could be shifted to provide
greater functionality without much compromise. Sujit Das shared that in one instance, he
experienced a model run-time exceeding 10 minutes, without indication of progress through the
duration of the run.
Although Shawn found that ALPHA model as-is delivers fast program execution, he countered this
goal with a suggestion for developers to implement more complex control algorithms to potentially
increase model run-time. Rather than prioritizing time-savings during model execution, Shawn
proposes that ALPHA model with more complex control algorithms achieves parity with current run-
times since users will spend less time calibrating and validating data overall.
ICCT initially stated they could not determine if the ALPHA model uses good engineering judgement
without a thorough comparison between simulation results and real-world testing data at the
7 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
component and vehicle level. After discussing their comment with EPA, ICCT concluded the general
approach of using real-world testing data in the model validation process was reasonable and the
modeling run demos provided by EPA demonstrate good agreement between simulation results and
real-world data, but a more robust assessment was beyond the scope of their peer review.
EPA Summary Response: EPA agrees with reviewers that the general approach ALPHA model takes
is sound. The runtime experienced by Sujit Das was primarily due to the setup of the demo cases
sent to the reviewer, which logged every signal and recompiled between runs. In larger batch
simulations, runtimes are generally quite short for each simulation. The more complex algorithms
proposed by Shawn Midlam-Mohler might be valuable to some users, but EPA does not have the
need to implement these types of algorithms in support of current regulatory work. The current
algorithms are relatively simple in part due to the purposeful replication of the narrow vehicle
operating conditions occuring during hot start EPA regulatory cycles (e.g., room temperature, no
HVAC, warmed up operation, etc.). EPA agrees with ICCT that a robust, deep-dive assessment of
model fidelity through the comparison of simulation results and real-world testing data is a time-
consuming task, and notes that each sub-model within ALPHA was thoroughly validated against
component and system data collected from vehicle dynamometer testing by EPA's laboratory and
by others.
Question 4: Does the ALPHA model generate clear, complete, and accurate output/results
(C02 emissions, or fuel efficiency output file)?
The three reviewers found that the output results and output files are labeled appropriately and are
relatively complete. Shawn Midlam-Mohler noted that "results" are generated across log files,
console output, and figures, which should provide users with a good amount of summary and
detailed results. However, Shawn and ICCT note a few user-friendly changes that ALPHA model
developers should consider making or need to make. For example, some of the plots automatically
generated are missing axes labels, making what is to be interpreted unclear. Other user quality of life
enhancements could organize console output as .csv files so that users do not have to copy, paste,
and reformat data for their specific purpose. ICCT suggests that certain data, such as energy
consumption metrics, should be disaggregated by component so as for ALPHA model to provide
more clear and complete evaluations.
EPA Summary Response: EPA agrees with reviewers that output results and output files are labeled
appropriately and are relatively complete. EPA also agrees that some figure axes were not properly
labeled and will update the scripts in ALPHA to ensure axes contain units. For the console output,
the data are primarily intended to be used for diagnostics and are formatted for that purpose. While
console values were not included in the peer review output files, they are accessible in the
workspace and thus could manually be added to the CSV result files if desired.
The accessory energy consumption referred to by ICCT can be defined by the user either for each
component individually or as aggregated energy consumption of all accessories. For the peer review,
the aggregated energy consumption was defined as a single generic loss, whose value is derived
8 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report December 2022
from chassis dynamometer test data. Users may instead choose to define losses disaggregated by
component.
Question 5: Do you have any recommendations for specific improvements to the functioning
or the quality of the outputs of the model?
All reviewers fall in alignment with ICCT's specific recommendations. For example, reviewers
expressed that better model documentation, particularly regarding equations and modeling
assumptions, be made available in the next iteration of ALPHA model. In the additional comment
section, Shawn mentioned that the model documentation was very much focused on the
programming structure of the model and not much on the actual modeling approach. He noted that
for others to adopt the methodology developed by EPA, these details need to be readily available to
the users.
The ICCT reviewers also found through their experience with ALPHA model that one "master" input
file would expediate initial model parameterization. Additionally, reviewers expressed some concern
or curiosity regarding the core data used to develop the vehicle models. Some aspects of the ALPHA
model, such as its robustness or fidelity to alternative vehicles and emissions, will gain greater
approval and acceptance when the underlying data are made available for users to understand the
breadth of validation against real-world data.
EPA Summary Response: EPA agrees with reviewers that additional documentation would be useful
to include in the next iteration of ALPHA, especially on the modeling approach, and will work to do so.
Some additional model details have been included in previous EPA publications, which were not
included with the ALPHA documentation, but will be integrated into the ALPHA manual to be readily
available for future users.
Referring to ICCT's comments, EPA notes that ALPHA includes a batch processing script where all
inputs can be listed and multiple simulations can be performed, thereby expediating initial model
parameterization. EPA also agrees with the reviewers that explaining core data collection and its
relevance is essential for a robust, transparent modelling process. EPA is in the process of
documenting the electric component data used in ALPHA modeling and will be publishing the
information on a new EPA web page dedicated to electric components.
IV. U.S. EPA's Responses
ICF forwarded all peer reviewers' comments as a draft report package to EPA for their review. Upon
reviewing the comment, EPA provided detailed responses to peer reviewers' comments that are
listed in this section of the report. The remainder of the document contains responses from ALPHA
Model development team. EPA organized responses to the peer reviewer comments grouped by
charge letter questions, addressing each reviewer's comments.
9 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Summary Observatioi
EPA's responses included additional information regarding a) the purpose of ALPHA, b) the reasoning
behind releasing ALPHA and its associated documentation to the public, and c) the goal for this
ALPHA peer review. The intent of the additional information is to provide further clarifications and
helpful context for EPA's responses to specific peer review comments associated with public use of
the simulation tools, supporting documentation for ALPHA model structure and execution, and the
user-friendliness of the tool.
urpose
EPA uses ALPHA for regulatory purposes to simulate the CO2 emissions and energy usage of vehicles
in the current fleet and potential future fleets. EPA recognizes there are a wide variety of vehicle
components and control strategies within the fleet, both in conventional and electrified vehicles.
Rather than precisely simulate all aspects of every vehicle, ALPHA uses a limited library of vehicle
components and control strategies to reasonably represent performance across the fleet. This
approach endeavors to strike a balance between two separate goals - to precisely simulate the
performance of every vehicle while also having an easily adaptable model capable of delivering
acceptable vehicle simulation results across a vast variety of known current, and unknown future,
vehicle fleets.
To meet the simulation needs for EPA's regulation development, ALPHA was designed with a fit-for-
purpose approach that does not require precise implementation of all control strategies for each
current and future vehicle variation. Rather, EPA selects representative production (or production-
ready) components and control strategies that can, on average, be used to simulate the
performance of the current fleet to provide reasonable CChand energy consumption estimates of
the technologies and strategies manufacturers might choose to employ in a future fleet. This
approach grounds our vehicle models in the actual control calibration and performance of
production vehicles and avoids overlooking any design constraints present in a vehicle
manufacturer's implementation of new or existing technology. This approach to fleet simulation is
very similar to the approach EPA has taken in the past to model conventional vehicles for the current
fleet3 and the future fleet4.
Although ALPHA can easily incorporate a wide variety of component maps, to simplify the peer
review process only one version of each type of component was supplied to the peer reviewers.
Likewise, each model included in this peer review contained a typical control strategy and a single
calibration. These can be altered by the user; however, the strategy and calibrations provided in the
peer review models reflect those used in the operation of well-performing vehicles that are
3 Kevin Bolon, Andrew Moskalik, Kevin Newman, Aaron Hula, Anthony Neam, and Brandon Mikkelsen, "Characterization of GHG
Reduction Technologies in the Existing Fleet," SAE Technical Paper 2018-01-1268, 2018
4 Andrew Moskalik, Kevin Bolon, Kevin Newman, and Jeff Cherry, "Representing GHG Reduction Technologies in the Future Fleet
with Full Vehicle Simulation," SAE International Journal of Fuels and Lubricants, 11(4):469-482, 2018)
10 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
representative of their type of configuration (PO mild hybrid, P2 or PowerSplit strong hybrid, or
battery electric vehicle).
Reason for ALPHA and Assoc cumentation Release to the Public
Although ALPHA is primarily used by in-house EPA experts in support of regulatory development
programs, it is publicly released in conjunction with proposed rulemakings to increase transparency.
ALPHA is not intended to be used as a general-purpose vehicle simulation model when versions are
released to the public. Instead, ALPHA intentionally incorporates only features needed to estimate
C02 emissions and energy consumption over the US EPA regulatory drive cycles under laboratory
conditions as specified under the proposed rulemaking. EPA does not require, for example, the
additional fidelity or flexibility to estimate real world in-use emissions.
The goal in posting versions of ALPHA on the EPA website is to provide sufficient documentation to
allow transparent review by stakeholders of our rulemakings. ALPHA'S documentation and user
interface are intended to allow the public to observe and review the simulation inputs, modeling
assumptions & behavior, and simulation outputs as a mean of understanding how the ALPHA
simulations were conducted in support of EPA's regulatory programs. ALPHA is not intended to
produce wide-ranging, detailed vehicle simulations like those generated by commercial simulation
tools (such as Autonomie, GT-Drive and AVL Cruise). For this reason, overall documentation, user
interface features, and user support for the ALPHA tool should not be expected to be at the same
level as commercial simulation tools.
Likewise, the ALPHA interface and documentation are not created for the novice user and do require
some expertise in both vehicle modeling and MatLab usage. The ALPHA documentation and user
interface provided should allow experts in the automotive community to re-create provided
simulations using their own in-house simulation tool, a commercially available tool, or even ALPHA.
The model itself is fairly straightforward and the MatLab data Classes provide formatted
documentation for most critical components.
ai of ALPHA Pee lew
The goal for this ALPHA peer review is to examine the structure, operation, and simulation results
used by EPA to determine the effectiveness of various technologies via simulation. The examination
of this peer review is centered on ALPHA'S recently added electrified vehicle models:
• Battery electric vehicle (BEV) model
• PowerSplit hybrid vehicle model
• P2 hybrid vehicle model
• PO hybrid vehicle model
In its performance work statement to peer-reviewers, EPA requested "the reviewer's opinion of the
concepts and methodologies upon which the model relies and whether or not the model can be
expected to execute these algorithms correctly. The documentation and user interface provided
for the peer review (and planned as part of the ALPHA release in conjunction with our current
rulemaking) were primarily intended to allow observation and review of the structure, operation, and
11 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
simulation results of the supplied electrified models. Although the documentation and user interface
provided was not intended to be the focus of the peer review, EPA welcomes comments on these
aspects to help guide future development of ALPHA.
Responses to Peer Reviewer's Comments by Charge Question
In this section, comments from the peer reviewers are grouped by charge letter topic. EPA's
responses to each reviewer's comments by charge letter topic are shown below.
Question 1: Does EPA's overall approach to the stated purpose of the model (demonstrate
technology effectiveness for various fuel economy improvement technologies) and
attributes embody that purpose?
Peer Reviewer: Sujit Das
A. Four major types of electrified vehicles have been considered in ALPHA and it is appropriate
that the ALPHA simulations of two most important vehicle types (i.e., BEV and PowerSplit
HEV) in the near-term have been selected for the peer review. Of the electrified vehicles, U.S.
DOE/EIA projects new BEV sales increase faster than any other type of battery-powered
vehicles, both electric hybrid and 300-mille electric vehicles reaching at -1.2 million/year and
both BEVs and PHEVs combined would account for 13% of total LDV sales in 2050, according
to the AE02022 reference case.
EPA Response: Thank you for the comment; EPA agrees with the reviewer's assessment of
the two most important vehicle types.
B. As ALPHA model has been primarily created to evaluate the GHG emissions of ICE Light-Duty
vehicles, it is less appropriate to evaluate alternative pure EV technologies. It is also to
include other specific electric vehicle types, i.e., sedan, SUV, CUV, and pickup in the future
model updates.
EPA Response: ALPHA was primarily created to simulate vehicle behavior on regulatory
dynamometer cycles. Although the first iterations of ALPHA were focused on CO2 production,
implementing a BEV model, and tracking electrical energy usage was deemed to be a minor
expansion of ALPHA's capabilities. In fact, the non-powertrain sub-models within ALPHA are
used to simulate both conventional and electrified vehicles.
Additionally, the vehicle package included in the ALPHA model sent to peer reviewers included a
set of vehicle parameters which reflected a specific generic vehicle. These parameters can be
altered to model any other vehicle; thus, the capability to simulate the behavior other specific
electric vehicle types, such as sedans, SUVs, CUVs, or pickups, is a capability that already exists
within the ALPHA modeI.
12 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Peer Reviewer: Shawn Midland-Mohler
A. The scope of the review is focused mainly on the BEV model and the PowerSplit HEV model
with a secondary focus on the Strong P2 HEV and the Mild PO HEV model. In all of these
cases, the modeling approach is able to meet the goal of modeling energy consumption. I
have concerns about the current state of the model with some concerns about how the
control algorithms are implemented.
A main concern is regarding the likely adoption of this model or the utility for the intended
purpose given the observation that: a) this class of model is already in the market; and b)
the vehicles that this model focuses on (BEV, PowerSplit HEV, etc.) are available from many
OEMs and, thus, there performance across many different vehicle classes is well-understood.
The modeling approach used is typical to that used in industry and academia, thus, it is
appropriate. However, the approach also does not lend itself to easy adoption outside of
expert users. In general, people with sufficient expertise to modify this type of model and
yield reasonable results likely already have existing models available to them. One of the
stated goals is that ALPHA will gain wide acceptance in the light-duty vehicle automotive
community, and I do not feel that is likely to occur in the current implementation.
EPA Response: The intended purpose of ALPHA is to support EPA's regulatory actions in a
robust and transparent way. ALPHA is explicitly not intended to compete with commercial
vehicle simulation products or supplant manufacturer's own modeling packages. The goal of
the public release of ALPHA is to provide sufficient documentation to allow transparent
review by stakeholders of the rulemaking. ALPHA'S documentation and user interface are
intended to allow the public to observe and review the simulation inputs, modeling
assumptions & behavior, and simulation outputs as a means of understanding how the
ALPHA simulations were conducted in support of EPA's regulatory programs. For this reason,
the overall documentation and user support for ALPHA should not be expected to be at the
same level as commercial simulation tools (such as Autonomie, GT-Drive and A VL Cruise).
B. The model approach is very similar to that used in Autonomie which has the benefit of many
years of development. A main feature in Autonomie that distinguishes it from ALPHA is the
availability of a GUI for model creation, modification, and data analysis for users to exercises
models. The ability to scale component models via a GUI, queue up different drive cycles,
adjust control parameters, etc. seem to make it better suited for the intended purpose.
Autonomie also has a more robust library of component options as well as more robust
control algorithms.
Because of the availability of tools like Autonomie, in-house tools, commercial tools like AVL
Cruise, and even example models provided within MATLAB, it is unclear how ALPHA in its
current form will meet the objectives. To be very clear, the technical approach of ALPHA
seems to be sound. My concerns are that it is not providing a solution that is not already in
the market with more established products.
EPA Response: EPA agrees the modeling approach is very similar to that found in Autonomie
and appreciates the reviewer for recognizing the technical approach taken with ALPHA is
sound. The intended purpose of ALPHA is to support EPA's regulatory actions in a robust and
transparent way and explicitly not compete with available commercial products. We
recognize that Autonomie, A VL Cruise, and other packages, as commercial tools, may have a
more user-friendly interface and a larger library of components and algorithms.
13 I P 3 g' ©
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
C. Furthermore, the marketplace now has a wide variety of electrified vehicles available and
there is data associated with these vehicles in the public domain. Organizations have access
to this data from their own vehicles as well as competitor assessments. From a planning
perspective, it is not clear what a model like this is able to provide. In the area of BEVs, the
increasing offering from many OEMs gives us the ability to reliably estimate things like range
and energy consumption based on actual vehicle data.
EPA Response: ALPHA is used to support regulation development, which requires
projections of CO2 emissions and energy usage in future years as the US vehicle market
adapts to consumer demand and regulatory requirements. The simulation models in ALPHA
were developed using actual electrified vehicle data and performance. ALPHA can then be
applied to simulate vehicle configurations anticipated in future fleets but not necessarily
currently available. Because the model inputs, vehicle parameters, and outputs are
transparently provided in support of EPA's rulemaking, any of the commercial tools (or a
stakeholder's own in-house tool) can be used to re-create the simulations provided to the
public.
D. The core energy consumption of the energy storage system and traction motors which
ALPHA focuses on is quite well understood and apparent from test data that is in the public
domain via certification requirements. Aspects like HVAC load, battery cooling during fast
charging, etc. which are areas which are more challenging which can significantly impact real-
world energy usage and range are not well modeled in ALPHA or most models.
EPA Response: EPA agrees these effects could be modeled better in ALPHA and most other
models. However, the focus of ALPHA is on vehicle performance over room-temperature
regulatory cycles, where the effects from the aspects highlighted by the reviewer above are
not a factor. We believe that possibly incorporating these effects into future versions of
ALPHA maybe appropriate, to the extent that actual on-road emissions and/or energy
consumption becomes important to directly quantify in any future rulemaking program.
E. None of the above should be taken as a comment on the modeling approach or skills of the
developers. The approach seems to be typical of the class of models that others have
deployed for this purpose. The main concern is if ALPHA will serve the intended purpose in
terms of being impactful in the technical community. Given the availability of public domain
data on these vehicles, the availability of internal data on their own vehicles to OEMs, the
availability of competitor assessments, and the availability of other simulation products
capable of the same type of analysis it is not clear how widely this tool will be used.
EPA Response: EPA does not expect ALPHA to be used widely in place of other commercial
simulation packages or manufacturer's internal tools. Rather, we hope the impact is through
the acceptance of ALPHA's simulation inputs, modeling assumptions & behavior, and
simulation outputs in support of EPA's the regulatory analysis. The goal is to provide enough
information that allows direct comparison of ALPHA simulation results to similar simulations
or estimates completed by any stakeholder's internal analysis.
Peer Reviewer: ICCT
A. The proposed model looks comprehensive and of high fidelity enough to serve its purpose of
quantifying the fuel economy, energy efficiency, and C02 emissions of different powertrain
typologies under a variety of operating conditions for several technology choices.
EPA Response: Thank you for the comment; EPA agrees.
14 i P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
B. The main issue that needs clarification at this stage is the powertrain components' sizing and
scaling approach. The process seems to be technology agnostic. For example, in the case of
electric motor-generator scaling, the model appears to rely on one electric motor data-
driven model reflecting a specific motor technology. We are not sure if this is only the case
for the shared demo version of the tool and if the complete ALPHA model already includes
several components' technologies. If that is the case, please disregard this comment.
EPA Response: Thank you for your comment. EPA agrees ALPHA's sizing and scaling
approach deserves clarification and will provide more explanation of our sizing and scaling
methodology as part of the Regulatory Impact Analysis for our current rulemaking. The
version of ALPHA provided to the peer reviewers included four electric motor maps, one for
use in each model (i.e., one for the BEV, one for the PowerSph't, one for the P2, and one for the
PO). Any of these component maps can be scaled to model a component of different power.
We believe these component maps are reasonably representative of the motors used in
industry; however, ALPHA is also fully capable of accepting other motor maps as input. EPA
has used alternate component maps to simulate specific vehicles, some of which contain CBI
and cannot be publicly released. We did utilize these alternate component maps to confirm
our conclusion these four electric motor maps are reasonably representative of the motors
used in the industry.
C. While the model is clear regarding technology choices focusing on hybrid-electric and pure-
electric powertrains, it remains unclear why fuel-cell powertrains are excluded from the
model.
EPA Response: When choosing which electrified configurations to include in ALPHA, EPA
surveyed which configurations existed in the current fleet. The most popular high-volume
configurations - BEVs, PowerSph'ts, POs, andP2s - were chosen for inclusion in ALPHA. Other
currently lower volume configurations - series parallels, Pis, and fuel cell electric vehicles, for
example - have not been included but may be candidates for inclusion in future versions of
ALPHA, especially if their market penetration increases.
Question 2; What is the appropriateness and completeness of the overall model structure and
its components, such as:
2a) The breadth of component models/technologies compared to the current/future light-duty
fleet
2b) The performance of each component model, including the reviewer's assessment of the
underlying equations and/or physical principles coded into that component.
2c) The input and output structures and how they interface with the model to obtain the
expected result, i.e., fuel/energy consumption and C02 over the given driving cycles.
2d) The use of default or dynamically generated values to create reasonable models from
limited data sets.
Peer Reviewer: Suiit Das
A. ALPHA model is fairly straight forward tool only for an experienced MATLAB user for
understanding vehicle behavior, greenhouse gas emissions and the effectiveness of various
powertrain technologies of current and future vehicles by appropriately changing input
values in five major vehicle parameter files. Parameter files are organized at the level of five
major EV components, i.e.. Base (Driver and Controls), Vehicle, Electrical, Accessory, and
Transmission for running a desired EV technology. Battery and the electric machine are a
15 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
part of the Electrical component. The ALPHA model is currently limited to C02 emissions for
five different EPA driving cycles including custom driving cycles based on test fuel properties
and vehicle fuel consumption.
EPA Response: EPA agrees that fully exercising ALPHA requires some experience with
MA TLAB; the ALPHA interface and documentation are not created for the novice user.
Additionally, although earlier versions of ALPHA are limited to simulating C02 emissions from
vehicles with IC engines, with the incorporation of electrified models into ALPHA the model
can now also simulate electric energy consumption. Although the version of ALPHA provided
to peer reviewers included the city, highway and US06 regulatory drive cycles, any pre-
defined drive cycle may be incorporated into ALPHA and used in the simulation.
B. An assessment of the underlying model equations and/or physical principles couldn't be
made as they were limited to the original Simulink code without any appropriate model
documentation available including the limited peer review time. ALPHA model 0.2.0
documentation is an excellent resource for a MATLAB model user in terms of the contents of
various files, but no description of types of equations including the source and validation of
the equation parameter values used. A Data Dictionary of variables used in the model would
be useful for better understanding of a novice user.
EPA Response: EPA agrees that information on the model equations and/or physical
principles is not currently contained in the documentation provided for the peer review and
can only be found in the Simulink code. In an updated version of the ALPHA documentation,
EPA will work to incorporate a description of the underlying equations and/or physical
principles coded into components, along with additional overview information we plan to
include in our Regulatory Impact Analysis.
C. The model is completely input data driven, which need to be collected by either engine or
chassis dynamometer testing by specific vehicle system technology case. The model
application is thereby limited to the extent of validated data availability. The overall model
performance is dictated by calibration of numerous technology-specific parameters used to
determine final vehicle fuel economy and C02 emissions for various vehicle drive cycles.
EPA Response: EPA agrees ALPHA is data driven and limited by (or dependent on) data
availability. EPA believes this approach is appropriate for a model used to support regulatory
activities, as our projections are based on real world information. The majority of ALPHA
inputs are built from component and vehicle benchmarking conducted on production
vehicles, while others are derived, and quality checked using data in technical reports and
papers based on work done by others. ALPHA'S electrified models were calibrated against
specific, popular, and well-performing vehicles that contain technology packages considered
reasonably characteristic of the overall fleet. While EPA recognizes there are a wide variety of
unique components and operational strategies in the fleet, EPA has concluded that modeling
these representative examples of various configurations, and not every variation, is sufficient
for our purposes in characterizing a potential future fleet. However, it is important to note
ALPHA was constructed to easily allow incorporation of any data or control parameters made
available from stakeholders which might improve our projections.
D. The input structure is defined by five major component MATLAB files, in which the input
parameter values can be changed for the simulation of new technologies. The expected
results of fuel/energy consumption and C02 over the given driving cycles for the two vehicle
types reviewed were reasonable. For PS HEV HWFET drive cycle, C02 emissions was
16 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
estimated to be 4%-21% higher than for the UDDS cycle.
EPA Response: Thank you for the comment; EPA agrees.
E. The use of default or dynamically generated values could only be assessed by the final
summary output results. A documentation on the approach and the sources used for the
input parameter values would be useful for the model user to develop or any new
technologies.
EPA Response: The default values used in ALPHA are primarily derived from data taken
during chassis dynamometer testing. These specifically include hybrid control parameters,
accessory loads, and battery pack parameters. Those few values which are not based
directly on test data (for example, the boost converter losses in the powersplit model) are
taken from published research. EPA agrees that documentation on the approach and sources
for default or dynamically generated values within ALPHA could be improved and will work to
do so.
Peer Reviewer: Shawn Midland-Mohler
A. Overall, the model has the overall systems that one would expect for the stated goal. Given
the importance of HVAC and battery thermal management to BEV and PHEV platforms, this is
one area that is not well-developed in the model. The mechanical and electrical accessories
are divided into four sub-models, generic loss, power steering, air conditioning, and fan loss.
In the BEV model, there was no energy usage associated with the engine fan, power steering,
or air conditioning system. This could indeed be the case; however, the modeling approach
is map-based and would require this information to be specified by the user. Given that the
loads from these systems can cause significant reductions in in-use energy efficiency, higher
fidelity of these models would certainly add to the capability of the model.
EPA Response: ALPHA is used to simulate room-temperature regulatory cycles performed
on a chassis dynamometer with the HVAC system turned off. In this case, there are no losses
due to HVAC loads, although ALPHA does provide a tunable parameter to represent those
losses if the user wishes. For the remaining losses, rather than individually modeling them, the
effects of all accessory losses (including energy used for battery thermal management) are
combined into one generic accessory load, whose value is set based on vehicle test data.
However, EPA agrees higher fidelity modeling of accessory losses would enhance ALPHA'S
capability and plans to begin developing ALPHA's battery thermal management modeling
(especially for BEVs) to support potential future regulations.
B. The control models deployed in the models reviewed also poses a challenge. As with any
model of this class, a controller is necessary to manage the torque split and gear as well as
other important vehicle functions. The quality of the control algorithm can have a major
impact on the efficiency of the simulated vehicle - a great vehicle component design with a
marginal control algorithm/calibration will perform marginally. There are optimization
techniques that have been applied to this class of models to allow a more refined control to
be deployed without excessive calibration by the user.
EPA Response: EPA agrees the quality of the controller affects the efficiency of the vehicle.
The electrified models in ALPHA were calibrated against specific, popular, and well-
performing vehicles in the fleet that contain technology packages reasonably characteristic
of the fleet as a whole, and the control algorithms in the model replicate the behavior of
17 i P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
these vehicles. Although other techniques could be used, EPA chose to replicate the
behavior of manufacturer-calibrated vehicles rather than optimize their efficiency. This
approach ensures that all NVH and other reliability issues that manufacturers must consider
are accounted for in the calibrations used in our simulation.
Additionally, ALPHA is used to simulate room temperature dynamometer cycles rather than
in-use operation where much of the calibration complexity resides. EPA does not intend to
implement numerous calibration packages for each electrified vehicle, but instead plans to
use fewer calibrations that are reasonably characteristic of the performance of the whole
fleet over dynamometer regulatory cycles. However, EPA does recognize the advantage of
implementing an optimal control algorithm which perhaps dynamically tunes its behavior for
both the powertrain and drive cycles mirroring further technological advances in industry
and will consider doing so.
The control algorithm used for the Power Split vehicle was inspected which is in the
PS_control.m function. The "working" part of the code consists of less than 100 lines of code
and is what one would refer to as rule-based for the most part. There are comments
included that say"% ::What is this?" and "% ???" which I can understand as a person who has
done these things before - but also does not lend confidence in the maturity of the control
algorithm provided. Given the importance of control algorithms for predicting the efficiency
of vehicles it is critical that these be matured.
EPA Response: EPA agrees the working code in PS_controlm consists of about 100 lines of
rule-based code. Please note there are additional sources of code associated with
PowerSpUt hybrids throughout ALPHA.
To meet the narrow simulation needs for EPA's regulation development, ALPHA'S control
strategies were designed with a fit-for-purpose approach. As a result, they do not require
full and precise implementation of all possible control strategies for current and future
vehicle variations, over every operating condition. For example, ALPHA only needs to simulate
vehicle operation at room temperature, over standard regulatory drive cycles. In addition,
ALPHA needs to simulate production (orproduction-ready) components and control
strategies that can, on average, be used to replicate the C02 and energy consumption
estimates of the current fleet (which are ultimately used to forecast what technologies and
strategies manufacturers might choose to employ in a future fleet).
This control strategy approach grounds our vehicle models using specific vehicle data that
reflects production control strategies and calibrations which meet the performance of
production vehicles, thereby including the performance results due to design constraints
incorporated in a vehicle manufacturer's implementation of new or existing technologies.
EPA agrees the comments quoted are uninformative. The portion of the code referenced
contains experimental code with an alternative methodology for determining additional
emissions from hybrids during an FTP cold start and was not being used functionally. EPA
eventually opted to use a post-processing adjustment for the cold-start FTP (similar to the
one used for conventional vehicles) and ceased work on the alternative methodology but did
not remove the associated code. The comment remained from the time when the code base
was being refactored and cleaned, as a method of communicating within the coding team,
and reflects the work-in-progress state of the code. Rather than delete this section of code,
EPA will dean up the code, add comments, and leave the code as a basis for potential future
development.
18 i P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
C. Overall, this is more or less an implementation of a force balance on the vehicle. The
components are modeled via maps and the basic relationships between the components.
No errors were noted in the summation of torques/forces that acted on the vehicle inertia. A
model of this class relies on appropriate component maps and appropriate controls.
Without a more rigorous look at these with comparison data it is not possible to provide a full
assessment of this.
When inspecting the driver commands (brake and accelerator) and high-level control inputs
like gear shifts and torque commands, I did not find anything of concern. Depending on the
underlying control algorithm and driver model, there can sometimes be high-frequency
behaviors on these signals that are not representative of actual vehicle controls or driver
behaviors. This was not noted in the model outputs reviewed.
EPA Response: EPA agrees with the reviewer on these points. We have intentionally tuned
the driver model to avoid these high-frequency behaviors. Additionally the ALPHA shifting
algorithm was developed to replicate actual vehicle behavior and has not exhibited these
behaviors. (See, for example: Newman, K, Kargul, J., andBarba, D. (2015) "Development and
Testing of an Automatic Transmission Shift Schedule Algorithm for Vehicle Simulation, "SAE
Int. J. Engines 8(3):20!5, doi:10.4271/2015-01-1142.)
D. I was not able to find much documentation on the actual models used outside of the overall
vehicle mass and loss model. This made it challenging to review the modeling approach as it
needed to be interpreted from the model and input/output.
EPA Response: EPA agrees information on the modeling approach was limited in the
documentation provided for the peer review. EPA plans to share more details on the
modeling approach in an upcoming technical paper, but the content of this was not available
prior to the peer review. Additionally, EPA plans to include a description of the model
architecture, model inputs, and simulation results in comparison to validation vehicles and
similar vehicles in the fleet in its Regulatory Impact Analysis (RIA) supporting the upcoming
rulemaking. Finally, EPA plans to add some of the content from the technical paper and RIA
to the ALPHA documentation.
E. I provided very detailed comments on the output in response to the fourth Charge Question
below. In this discussion, I will focus on the input structures. In the demo files provided, the
input structure was clearly defined and using variants in the model appropriate subsystems
were enabled.
It was not clear to me if there was scaling that could be applied in the input structure - there
did not appear to be. That is one aspect that is generally quite useful to be able to slightly
adjust component sizes without having to generate new component data files.
EPA Response: Any of the component maps within ALPHA can be scaled to model a
component of different power. EPA agrees ALPHA'S sizing and scaling approach deserves
clarification and will provide more explanation of our sizing and scaling methodology as part
of the Regulatory Impact Analysis for our current rulemaking.
F. Compared to more mature projects like Autonomie, there are not many options. The
capability is there, but I did not locate any library of models or the ability to scale them.
Likewise, the control algorithms were very likely highly specific to the particular set of
19 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
components they were calibrated to work with. I was not able to locate documentation on
the nature of the controls but after inspecting the model and the input files it seemed to be
very calibration-based.
EPA Response: ALPHA is built on component and vehicle mapping conducted on specific
production vehicles. EPA recognizes there are a wide variety of components and operational
strategies in the fleet. Although EPA plans only on modeling representative examples of
various configurations and not every variation, ALPHA was constructed to allow stakeholders
to incorporate their own data if they wish.
Peer Reviewer: ICCT
A. The different technology and modeling choices are all reasonable. Special attention has been
given to the hybrid powertrain controllers, one of the main fuel economy drivers.
The P2-PHEV controller, as an example, seems to utilize a charge-depleting charge-
sustaining strategy depending on the battery state-of-charge (SoC) and the physical
constraints of the different energy conversion devices. During the charge-sustaining mode,
the algorithm seems to try to operate the engine at its minimum BSFC line as much as
possible. Assuming that this is the only modeled control strategy, the question here is about
the representativeness of this control strategy.
EPA Response: EPA constructed the P2 model based on the operation of a popular and well-
performing vehicle in the fleet (in this case a Hyundai Sonata). The engine control strategy in
ALPHA demonstrates a good match with test data from the vehicle operating on the
regulatory cycles. Although we recognize there are a wide variety of operational strategies in
the fleet, we believe over regulatory cycles this strategy would produce energy and fuel
consumption results that are reasonably representative of a wide swath of the present and
future fleet.
B. Other more complex and efficient controllers, such as the ECMS (equivalent consumption
minimizing strategy) based on Pontryagin's Maximum Principal optimization techniques, can
reduce total fuel and energy consumption, and several vehicle manufacturers have
considered it. While on the other hand, other simpler heuristics rule-based controllers can
also be implemented. In addition, the domain of hybrid vehicle controllers has been going
beyond the traditional optimization-based or rule-based methods as more advanced neural
network- and machine learning-based control strategies are under continuous development.
So, the question here is mainly about the representativeness of the adopted control strategy.
EPA Response: EPA has chosen to incorporate vehicle components and control strategies
that already exist within the current fleet and are representative of the performance of a
broad range of vehicles as part of our updates to ALPHA. A t this time, we are not
implementing hybrid control strategies that have not been brought to production. However,
for future development of ALPHA we may consider evaluating more efficient algorithms as
part of our ongoing research programs.
C. In the generated results file in MATLAB workspace, there is a mention of a "s_factor," which
seems drive cycle dependent. It is unclear why this factor is defined and how it is used in the
model. This question is raised here as the "s_factor" is a common notion in the academic
literature regarding hybrid vehicles' control strategies which is defined as a conversion factor
between fuel energy and battery energy.
20 i P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
EPA Response: The S factor is used to correct the fuel consumption to a zero-change in
battery SOC, using the method outlined in Appendix C ofSAE procedure J1711
(Recommended Practice for Measuring the Exhaust Emissions and Fuel Economy of Hybrid-
Electric Vehicles, Including Plug-in Hybrid Vehicles). In this way the effective C02 impact of
charging or discharging the battery over the length of the cycle is included in the final tally.
The S factor used is indeed cycle dependent; the values come from test data informing a
draft (not yet finalized) update to the J1711 standard. Once the J1711 standard is updated,
these values will be altered if necessary. An explanation of the definition and use of the S
factor will be added to the documentation.
D. The battery model is a resistance-capacitance (RC) equivalent circuit model that captures
the dynamic battery voltage drop as a function of the battery's main state variables, such as
SoC and temperature. RC equivalent circuit models are proven to be robust models and
widely used in battery modeling for vehicle energy assessment. The battery model also
includes a thermal sub-model where the battery temperature and different heat transfer
phenomena are estimated. The battery's internal heat generation due to the exothermic
chemical reactions is well presented. The battery heat exchange with its surroundings, either
with the ambient or the cooling system, seemed to be simplified in a single parameter: the
battery pack's total conductance. It is unclear to us what battery cooling technologies are
considered and to what extent this approach can consider different cooling techniques, such
as active/passive air or liquid cooling. While this issue might not be a game changer in hybrid
powertrains, the battery temperature is more critical on pure electric powertrains, given the
size of the battery, which would affect the battery performance and total energy
consumption. In addition, the battery efficiency seems to be quite high (96.9% on UDDS_1,
97.7% on UDDS_2, 98.7% on highway cycle, 91.8% on US06J, 97.3% on US06_2).
EPA Response: The battery pack parameters (Open Circuit Voltage - OCV, and Series
Internal Resistance - R) were tuned to match the terminal voltage and current response of
the battery pack data collected on chassis dynamometer during the drive cycle testing. This
ensured the modeled charge flowing on and out of the battery during dynamometer testing
correlated with the test data and we found the battery losses and battery efficiency also
corresponded with test data. For example, the internal resistance of the BEV battery pack -
and thus the expected losses -is very similar to published data from a VW iD.3 (see Nikolaos
Wassiliadis, Matthias Steinstrater, et ai, "Quantifying the state of the art of electric
powertrains in battery electric vehicles: Range, efficiency, and lifetime from component to
system level of the Volkswagen ID.3,"eTransportation, Volume 12, May2022,100167).
Beyond the cycle efficiency numbers, there are additional battery losses associated with
battery charging from an AC power source. As these charging losses do not occur during the
dynamometer operation, they are not included in the quoted battery efficiency number.
Instead, they are postprocessed within ALPHA using a nominal "charging efficiency" factor
which includes both losses within the charging system and losses in the battery during
charging. This "charging efficiency" factor is the ratio of DC energy used during the
dynamometer test to AC energy supplied to the vehicle, and is set to 0.90, based on average
values recorded during multicycle tests of BEVs.
E. The electric machine model is mainly data-driven, summarized in dynamic look-up tables
focusing on the maximum torque curves and energy conversion efficiency. It is unclear
whether the model differentiates between continuous and peak torque. The electric machine
losses seem to be quite high (23% on UDDS_1, 29% on UDDS_2,15% on highway cycle, 26%
on US06_1,14% on US06_2). It is not clear if these are losses during both propulsion and
21 i P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
regenerative braking modes.
EPA Response: Regarding continuous versus peak torque, the dynamic lookup tables within
ALPHA are capable of varying available torque based on a thermal model or other criteria. For
the included demo simulations, a static map was used. This is an area where further
development is planned.
The losses quoted are not the throughput losses (i.e., / - efficiency). Rather, they are the
percentage of propulsion energy that is lost within the electric drive unit (specifically for the
BEV, this would be the percentage lost with respect to the total battery energy used during
the cycle). The electric drive unit includes the emotor, power electronics, and gearing, and
the electric machine losses quoted reflect losses in all three components. The efficiency of
the EDU in the BEV model supplied to the peer reviewers tends to be in the range of 87% -
92% when operating over the EPA regulatory cycles.
EPA notes the electric machine efficiency was not reported in any output file which would be
a helpful quantity to provide ALPHA users. Therefore, EPA will add columns to the
"results.csv" file for future public releases to allow easier review of key simulation run
parameters (e.g., emotor efficiency, engine efficiency, battery efficiency, etc.) In addition,
EPA notes the demo files provided for the peer review had the output verbosity set to 1,
which produced a fairly limited "results.csv". Higher verbosity levels will automatically include
additional parameters in the results file. EPA plans to insert comments into the sample
scripts to better alert the user that additional data can be output.
F. The different transmission systems and the engine are developed with high fidelity. The gear-
shifting strategy behaved as expected for some sample runs. The transmission system
model is highly complex, which raises the concern of data availability for model calibration
for different vehicles.
EPA Response: The ALPHA shifting algorithm was developed to replicate actual vehicle
behavior. (See: Newman, K, Kargul, J., and Barba, D. (2015) "Development and Testing of an
Automatic Transmission Shift Schedule Algorithm for Vehicle Simulation, "SAEInt. J. Engines
8(3):2015, doi:10.4271/2015-01-1142.) We have used the shifting algorithm to validate multiple
conventional vehicles and have found there is a reasonably good match between ALPHA shift
results and shifting over the regulatory cycles with vehicles whose transmissions are
calibrated to emphasize fuel economy.
G. The powertrain auxiliaries, such as the air conditioning (AC) unit, fan, pump, and other electric
auxiliaries, seemed to be modeled simply as constant power demand from the battery or
torque demand from the engine. While this simplification is acceptable for most auxiliaries,
such an approach can misestimate the AC/heating unit energy consumption, which is highly
sensitive to dynamic operating conditions such as external temperature and trip duration.
Detailed vehicle cabin thermal and AC/heating models would enhance the model capabilities
to model the AC energy consumption providing a more accurate estimation of the vehicle's
fuel economy/energy efficiency. Although the user can still input different auxiliary
consumption in kW to mimic additional heating and cooling needs, quantifying such metric is
a complex process that may incur inaccuracies. Thus, modeling such behavior is quite
essential.
The academic literature has grown rich recently with modeling techniques to integrate
vehicle powertrain and vehicle cabin thermal modeling into a single platform. Some examples
of such literature are:
22 i P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Marcosa, D., Pinob, F.J, Bordonsa, C., Guerrab, J.J., "The development and validation of
a thermal model for the cabin of a vehicle" Applied Thermal Engineering,
https://doi.Org/10.1016/j.applthermaleng.2014.02.054.
Doyle, A., Muneer, T., "Energy consumption and modeling of the climate control
system in the electric vehicle". Energy Exploration & Exploitation,
https://doi.org/10.1177/01445987188064.
EPA Response: ALPHA is used to simulate room-temperature regulatory cycles performed
on a chassis dynamometer (with the HVAC system turned off). In this case, there are no
losses due to HVAC loads, although ALPHA does provide a tunable parameter to represent
that loading if the user wishes. For the remaining losses, rather than individually modeling
them, the effects of all accessory losses (including energy used for battery thermal
management) are combined into one generic accessory load, whose value is set based on
vehicle test data.
However, EPA agrees that higher fidelity modeling of these losses would enhance ALPHA's
capability to perform other types of simulations, and we are planning to develop HVAC and
thermal control modeling capability in ALPHA (especially for BEVs) to support future
laboratory research programs. We thank the reviewer for the specific literature
recommenda tions.
H. Regarding the drive cycles, it appears that the US06 drive cycle is split into two phases, with
the first phase incorporating the hard accelerations and the second phase incorporating the
high speeds. We recommend adding a combined US06 audit.
EPA Response: The US06 drive cycle is customarily divided into "city" and "highway" phases
as described by the reviewer. EPA feels there is a benefit in analyzing the two phases
independently (for both the US06 and UDDS) and thus report them as such. A combined
"EPA_ UDDS & EPA_HWFET & EPA_ US06 audit" is already included to allow a quick visual
verification of 100% conservation of energy for the entire simulation run.
I. The output file structure looks appropriate and complete. The model, however, lacks a
master input file where the user can easily visualize all inputs. While this is understandable for
a non-commercial model, a master input file would ease model validation and allow the user
to conduct parametric studies more easily, which would help provide a more solid idea of the
entire model's robustness.
EPA Response: The version of ALPHA provided to the reviewers contained only a few
electrified model simulations configured via scripts and run independently. Using different
scripts, ALPHA is capable of loading simulation parameters from a file either as a list of
individual simulations or by using the full factorial expansion capability built into ALPHA. EPA
plans to add an additional demo file to its public release of ALPHA to demonstrate this
functionality.
J. While a data-driven modeling approach is reasonable for vehicle fuel economy estimation, a
significant amount of data must be collected or provided to parametrize the model correctly.
Engine test benches and vehicle chassis dynamometers are well-developed standard
practices for data collection for conventional powertrain technologies. However, it is unclear
how data for batteries and electric machines were developed and the extent to which such
data collection methods are well developed. While this review covers the modeling approach
in general, clarifying the electric components' data collection approaches via proper
23 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
documentation is essential for the user to understand the potential limitations of the model.
EPA Response: EPA agrees that robust data collection to sufficiently represent the operation
of components is essentiaI. To model the behavior of electric machines, EPA uses emotor
test data from suppliers, contractors, and other national laboratories. These data are
collected from electric components using dynamometers, with methodologies similar to
those used in engine testing. EPA independently evaluates the data to ensure the resulting
ALPHA input maps reflect physically reasonable component performance.
Data on battery performance are developed using widely recognized battery models whose
parameters are then tuned so the current and voltage response match test data. For
example, the BEV battery model was developed using published data from a VW/D.3 (see
Nikolaos Wassiliadis, Matthias Steinstrater, etal, "Quantifying the state of the art of electric
powertrains in battery electric vehicles: Range, efficiency, and lifetime from component to
system level of the Volkswagen ID.3," eTransportation, Volume 12, May2022, 100167).
EPA also agrees that explaining data collection and properly documenting the processes for
obtaining electric component data is essential for a robust, transparent modelling process.
We are documenting the electric component data used in ALPHA modeling and will be
publishing this information on the EPA web site in conjunction with the rulemaking. Users of
ALPHA can easily substitute their own emotor efficiency maps into ALPHA if they wish,
without altering the underlying vehicle control strategy.
Question 3: Does the ALPHA model use good engineering judgement to ensure robust and
expeditious program execution?
Peer Reviewer: Suiit Das
A. The ALPHA model is a MATLAB/Simulink based full vehicle computer simulation model
capable of analyzing various vehicle types combined with different powertrain technologies.
Although both current and future advanced vehicle technologies can be explored by defining
appropriate parameters in five major EV component files but requires a fairly knowledgeable
of the specific MATLAB version (e.g., 2022a for the review version) to ensure robust and
expeditious program execution.
EPA Response: EPA agrees with the reviewer that exploring the full capabilities of ALPHA
requires a fair amount of knowledge of MA TLAB. ALPHA 3.0 was developed using MA TLAB
2020a and should be compatible with subsequent versions of MATLAB. For regulatory
purposes, we provide pre-populatedinput files so stakeholders can replicate EPA's analyses.
Independent analyses which involve explorations of various parameters can be
accomplished by stakeholders using ALPHA; however, this type of operation is best suited
for users who are skilled in MA TLAB and vehicle analysis. ALPHA does not have the extensive
user support a similar commercial package would have.
B. A specific simulation runtime is significantly high, more than 10 mins. without providing any
indication to the user progress made so far. A user-friendly front end useful for an
expeditious sensitivity analysis of key input parameters.
EPA Response: There are a few factors driving the high run time for simulations. The provided
demo cases were all set up using the REVS_log_all logging. This logs every signal in the model
and requires a significant amount of memory. Additionally, whenever ALPHA reconfigures
24 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
(changingpowertrains or component models) it must be recompiled causing the first
simulation to require additional time. Had the simulations instead swept parameters such as
vehicle mass that do not require recompilation, subsequent simulations would have run
significantly faster. In larger batch simulations, where a limited set of signals is logged,
runtimes are generally under 30 seconds for each simulation.
Regarding a progress display, there is a disconnect between MA TLAB and Simulink, where
once the Simulation is requested, feedback is not generated until the simulation is
completed. For batch simulations there is a progress display that updates after each
simulation. Considering the included demos only include one or two long running cases this
functionality was probably not visible to the user.
EPA thanks the reviewer for the suggestion of a graphical front end. Given the flexibility of
ALPHA, a user interface covering all features would probably be confusing, but a simplified
version could help new users get acclimated, and EPA will consider implementing such a
feature.
Peer Reviewer: Shawn Midland-Mohler
A. Overall, the modeling approach used seems appropriate to the technical goals. The fidelity
selected provides fast execution. This does have drawbacks as it is heavily reliant on
experimental maps as input. As outlined in a previous comment, the use of relatively simple
control algorithms rather than techniques that have some manner of optimal control is a
weakness. Adding this could result in slower execution time, however, it would likely result in
better results with less overall run time for the user as it would require less runs to calibrate
and adjust the control.
EPA Response: The models within ALPHA are calibrated to replicate the performance of
specific vehicles in the fleet which have been tuned by the manufacturer. The use of input
maps and control algorithms measured/observed in the testing of representative vehicles
grounds our ALPHA control algorithms in reality. These are simple in part due to the
purposeful replication of the narrow vehicle operating conditions of hot start EPA regulatory
cycles (e.g., room temperature, no HVAC, warmed up operation, etc.). Although the models
within ALPHA could be re-tuned to increase their control strategy flexibility and scope
(which might be valuable to some users), EPA does not have the need to do so in support of
its current regulatory work.
Peer Reviewer: ICCT
A. It is not possible to provide a solid opinion about engineering judgment and model
robustness without a thorough comparison between simulation results and real-world
testing data at the component level and vehicle level. The peer reviewers asked EPA if ALPHA
has been validated against real-world results. EPA answered that each sub-model had been
thoroughly validated against data collected from vehicle dynamometer testing internally and
externally, where the tested vehicle behavior was reproduced with the model. The general
approach of the model validation process, as shared by EPA, sounds reasonable and
comprehensive. The modeling run demos provided by EPA demonstrate good agreement
between simulation results and real-world data, but a more robust assessment is beyond the
scope of our review.
EPA Response: EPA thanks the reviewer for this comment and acknowledges that a robust,
25 i P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
deep-dive assessment of model fidelity through the comparison of simulation results and
real-world testing data is a time-consuming task. We appreciate the reviewer's expertise and
observations on the importance of using real-world vehicle data to improving model
robustness.
B. One important point to mention is the impact of the simulation step size on the simulation
error. While it is understood that a smaller step size would yield a lower error at the expense
of higher computation resources, it is unclear how the simulation step size is set and what
simulation error is considered acceptable.
EPA Response: EPA agrees with the reviewer that simulation step size and resultant error is
important to robust simulation. We feel we set an appropriate step size (100 Hz) and our
energy audits (documented in the "console, txt" files) confirms that energy is being
conserved within the model during simulations. Additionally simulation step size is a
parameter which can be changed by the user if desired.
Question 4: Does the ALPHA model generate clear, complete, and accurate output/results
(C02 emissions, or fuel efficiency output file)?
Peer Reviewer: Sujit Das
A. The output Excel file is fairly simple with summary results of fuel economy (MPGe and
Whr/mile). Fuel efficiency output file is detailed and clear with both Phase and Weighted
aggregate results of energy economy, efficiency, and consumption by drive cycles. Energy
Audit report by the drive cycle is fairly detailed in terms of energy balance at the gross level
and by major EV components.
Energy consumed by Accessories has been accounted as the sum of Generic and DCDC
Converter Losses. Fuel consumed (grams and gallons) including C02 emissions for
conventional vehicles are reported under Phase Results by drive cycles.
The simulated AC usage in UDDS and HWFET drive cycle results of a 2019 Tesla Model S
Standard Range compared well (within less than 2%) with EPA certification values.
EPA Response: Thank you for the comment; EPA agrees.
Peer Reviewer: Shawn Midland-Mohler
For this charge question, the output will be considered the log file, the console output file, the results
file, and the figures that are generated from the model run.
Clear Output/Results:
A. Log File: Overall, this was reasonably well organized. See comments below on some items
that led to some confusion.
EPA Response: Thank you for the comment; EPA agrees.
B. Console Output: This was clear. Given the tabular nature of the data, it would be helpful to
output this as a .csv or use the Report Generator capability in MATLAB to give it structure as
a pdf. I can see users having to do cutting and pasting to use this data for whatever their
purpose was.
EPA Response: The console output is primarily intended to be used for diagnostic purposes
26 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
to confirm conservation of energy. The values displayed in the console output's energy audit
are accessible in the workspace and thus could be manually added in the CSV result files.
Developing scripts to add any desired columns for the energy audit will be considered for
future development.
C. Results File: This was clear, but the horizontal format is not what I would have preferred. It
seems like it would be more readable as a .csv in a spreadsheet app if it were arranged
vertically.
EPA Response: The peer reviewers were supplied with individual example runs for each
electrified technology. However, ALPHA is typically run in a batch mode with either multiple
vehicles or multiple permutations of parameter values, and the output file contains results
from all vehicle simulations. In these cases, with dozens or hundreds of vehicle simulation
outputs, the horizontal .csv format lends itself more easily to post-processing analysis and
manipulation.
D. Figures: Many of the figures did not contain proper units for some of the axes. I was able to
infer the likely units but that is clearly not good practice. Given that the figures are only
generated upon run, I would have appreciated them being saved or, at a minimum, written
into a pdf file and logged with the above files.
EPA Response: EPA agrees with the reviewer that properly labeling figure axes is desirable.
We will update the scripts in ALPHA to ensure axes contain units. For the provided sample
simulation scripts, the plot generation was done at the end of the script, outside of the batch
simulation operation. To generate and organize plots for each simulation case would require
moving the plot generation into a case post-processing script. Calling the DOR plot scripts
with the MA TLAB publish command can be used to send the resulting figures to a pdf. EPA
agrees this is a useful feature and will work on developing the postprocessing script and a
demo showcasing its usage.
Complete Output/Results:
E. Log File: The log files that I inspected did not seem to fully be populated with data or I may
have been misunderstanding the intent. In the "Configuration Keys" section there were many
entries without values, for instance the A, B, and C coefficients did not list values. I expected
these to be the values that were used to generate the results. There were many other areas
where this was the case.
EPA Response: The "Configuration Keys"section of the log files are intended to document
how the batch was configured, and not to document the individual simulations. Regarding the
ABC coefficients not having values, those configuration keys are intended to override values
that would otherwise be set via the vehicle param files. For a simple simulation, most model
parameters are set in the various component parameter files rather than in the log file, and
the blank configuration keys represent those parameters whose values default to the values
in the vehicle param files. The documentation can be revised to better clarify the use of the
configuration keys and the distinction between values set in the log file and those set in the
vehicle param files.
F. Console Output: The console output contained a good summary of overall cycle energy
flows.
EPA Response: Thank you for the comments; EPA agrees.
27 i P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
G. Results File: The results file contains an effective summary of each cycle and the overall
composite of the cycles run from an energy perspective.
EPA Response: Thank you for the comments; EPA agrees.
H. Figures: The set of figures shown was satisfactory. Obviously, a person can generate their
own. A concern is that I did not see that the data from a simulation was logged. I would
prefer to have that option to avoid having to rerun a simulation. I was able to locate the time
series output in the "model_data" structure. I assumed that this was the model parameters
based on the "data" label - instead it was the output. I think "modeLoutput" would have
been a more appropriate name or something even more descriptive.
EPA Response: ALPHA contains a couple different mechanisms for saving the simulation
data which are more thoroughly described in the documentation and configured within the
sim_batch object. If retain_output_workspace is set to "true", then simulation data is saved
in memory and is accessible with the extract_ workspace method for each simulation case.
Alternatively, data can be logged to a mat file by setting save_ output_ workspace to "true."
The "model_data"nomenclature is carried over from our model validation activities where it
is compared to a similarly organized "test_data" object. The comment is appreciated and will
be a topic for further discussion by the developers.
Accurate Output/Results:
/. Overall, the accuracy is difficult to assess because the models provided are only
representative of classes of vehicles and no direct comparison data is provided. Upon
request by the reviewers, the EPA provided simulation results that were representative for a
2019 production EV. The results cycle-based results agreed very well with the experimental
results provided. This demonstrates that it can generate valid results.
EPA Response: EPA agrees ALPHA can generate valid results and acknowledges that a
robust, deep-dive assessment of model fidelity through the comparison of simulation results
and real-world testing data is a time-consuming task.
J. Overall, the modeling approach is known to be able to predict cycle energy usage well under
nominal conditions. Additional factors like HVAC loads and cold/hot weather performance
can be challenging to model with the current fidelity of the models. I do not feel that is
something that the model is expected to be able to do at this point, so this compromise is
understood. To bring those factors into the model, additional systems need to be directly
modeled or the resulting loads on the system need to be brought in via the existing
accessory load and efficiency models. This typically requires detailed information on
components/controls and/or experimental data that is often not easily obtained outside of
OEMs.
EPA Response: ALPHA simulates room-temperature regulatory cycles performed on a
chassis dynamometer as a result, there are no losses due to HVAC loads. However, EPA
agrees higher fidelity modeling of these losses would enhance ALPHA's capability and we are
considering improving this modeling for future versions of ALPHA.
Additional Comments/Questions:
28 i P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
K. It was not clear to me if there is a warning issued if the drive trace is violated. If not, this
would need to be added because results when this occurs may not be valid.
EPA Response: EPA agrees tracking drive trace violations is necessary. ALPHA does track
drive trace violations (if any) and can report the percentage of time the simulation is outside
the drive trace envelope. In the peer review outputs, this can be found in the console. EPA
normally also indicates trace violations in the CSV results file. EPA will consider ways to
make any trace violation more obvious in the simulation output files.
L. It was not clear to me if there is a correction for SOC variation for non-plug-in hybrids.
EPA Response: ALPHA accounts for the SOC variation by using an S factor to correct the fuel
consumption to a zero-change in battery SOC. The method used is outlined in Appendix C of
SAE procedure J1711 (Recommended Practice for Measuring the Exhaust Emissions and Fuel
Economy of Hybrid-Electric Vehicles, Including Plug-in Hybrid Vehicles). EPA will update
ALPHA documentation to note the use of the S factor to account for SOC variation.
M. It was not clear to me in the structure of the model if the translational mass of the rotational
components were factored appropriately. There was some discussion of this in the manual
but without having time to really investigate this I could not be certain. I noted how the
rotational inertia was carried forward in the model to be added into the overall mass where
the vehicle inertial integrators were.
EPA Response: The physics within ALPHA are simulated via torques and rotational inertias
which are carried down through the powertrain to either a disconnection point, an open
dutch for example, or the wheels. Within the tire model the torque is translated into force
while the rotational inertia of upstream components is translated into an equivalent mass
which can be added to the static mass of the vehicle. From these values the acceleration
and speed can be computed. We agree the graphical nature of a Simulink model can make it
difficult to determine whether the physics are represented properly. A more thorough
description of these models has been included in some of our prior publications and could
be included within the user documentation to clarify the matter. Concerns regarding
conservation of energy was a primary reason the energy auditing was implemented, allowing
the changes in s tored energy to be ve tted agains t the sum of the energy consumed by each
of the component losses.
N. In reviewing the model_data variable, I noted that the modeLdata.controls part of the
structure was not populated with data. There were variables there in the EV and PS model,
but they did not contain data vectors.
EPA Response: Controls is an area where EPA has not often used the model_data variable
during development, preferring instead to use the signals in datalogcontrols which are more
powertrain specific. The model_data variable is constructed from class_ test_data which
was originally developed to provide a standard structure into which data from vehicle testing
could be loaded and is integral to the DOR functionality. ALPHA only generates a subset of
the signals class_test_data is capable of loading; thus, many variables are left unused.
Peer Reviewer: ICCT
29 IP a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
A. The ALPHA model generates automatic plots describing the physical behavior of the main
powertrain components, in addition to a console text file summarizing the energy
consumption, fuel economy, and C02 emissions. We recommend the following to make the
results file more readable:
The accessory's energy consumption doesn't show the share of each component
(fan, AC, pump, etc.). These are aggregated under one value. Although the generated
results files contain placeholders for these accessories, they show a value of zero.
EPA Response: For these variables¦, users have the option of defining energy
consumption for each component individually or defining the total energy
consumption for all accessories. For the peer review, the total energy consumption
was defined as a single generic loss, whose value is derived from chassis
dynamometer test data, without defining the breakdown into individual accessory
systems. The zero values represent placeholders for the individual quantities not
itemized.
There is no mention of the battery's thermal needs as an accessory consumption. Is
battery cooling/heating consumption considered part of the battery losses? It is
worth documenting how the model handles battery thermal needs.
EPA Response: EPA agrees that documenting how ALPHA incorporates battery
thermal management losses is preferred. The accessories are modeled in ALPHA as a
single generic loss, whose value is derived from chassis dynamometer test data.
However, EPA plans to begin developing ALPHA'S battery thermal management
modeling (especially for BEVs) to support potential future regulations.
The results file mentions "kinetic energy" as a potential energy source in addition to
fuel and stored energy. It is unclear whether this refers to brake energy recovery or
another indicator.
EPA Response: Kinetic energy mentioned in the file refers to the kinetic energy of the
vehicle, and therefore the amount of energy available for recovery when braking. EPA
will clarify this term in the documentation.
The fuel energy could be further detailed into direct flow to the driveline or energy
flow from the engine to the battery, depending on the powertrain architecture. This
can provide a clearer idea of the control strategy under different drive cycles,
operating conditions, and system boundary conditions.
EPA Response: ALPHA does track electrical energy into and out of the battery which,
like fuel energy flow, can be used to provide an idea of the control strategy. The
advantage of tracking electrical energy is that it directly coincides with a measurable
quantity in the vehicle, as opposed to fuel split which must be calculated. However,
EPA will also consider incorporating a similar fuel energy calculation.
B. In the model, it is mentioned that the final drive efficiency is already included in the electric
drive unit. This is probably why the final drive losses are always set to zero in the results
report. However, the final drive is separate from the electric motor, and combining their
efficiencies limits the model's flexibility. We recommend fixing this issue.
EPA Response: ALPHA has the option of modeling the final drive as part of the motor (in
which case the separate final drive block losses are zero) or separately in the final drive
block. For the peer review, the e-machine map used for the BEV model was based on test
30 | P 8 g 6
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
data from an entire electric drive unit (EDU), which included the gearing and the gearing
losses. However, if the user wishes to incorporate an e-machine which does not include
gearing, the gearing losses and ratio can be defined within the final drive.
Question 5: Do you have any recommendations for specific improvements to the functioning
or the quality of the outputs of the model?
Peer Reviewer: Sujit Das
A. Detailed sixty complementary graphical output files as a function of drive cycle time to the
three summary output files provided would have useful for the evaluation of the model
functioning. Unless an expert MATLAB/Simulink user, it is not intuitive to track down the
logical flow of summary final results from its initial parameter values used in underlying
equations.
EPA Response: EPA agrees the overall process from input parameters, simulation
configuration, simulation, and output processing can be confusing in part due to the effort to
make the simulation process in ALPHA highly configurable. Additional documentation
explaining the process from a higher level will be added in the future.
Peer Reviewer: Shawn Midland-Mohler
A. I provided some specific recommendations in the above section. I do not have any
additional comments to provide.
EPA Response: EPA thanks the reviewer for the detailed and thoughtful comments above.
Peer Reviewer: ICCT
A. Hybrid and plug-in hybrid vehicles' controllers (energy management strategy) are rigid and
developed based on specific strategies and algorithms. Providing the option to simulate
several hybrid control strategies as part of the batch simulation runs would be more
comprehensive and may cover a broader range of vehicles.
EPA Response: The models within ALPHA are calibrated to replicate the performance of
specific vehicles in the fleet which have been tuned by the manufacturer. Although the
models within ALPHA can be re-tuned by EPA or stakeholders, for its current regulatory
modeling EPA does not need to replicate the wide array of control algorithms in the fleet.
Please also see EPA's response to Charge Question 3, Shawn Midland-Mohler, comment A.
B. A bottom-up approach is recommended to estimate the vehicle weight. Estimating the
components' weights and aggregating these weights to calculate the total vehicle weight can
provide a more accurate estimation of the impact of different technology choices, especially
different battery, and electric motor sizes.
EPA Response: ALPHA is used to simulate vehicle behavior during chassis dynamometer
testing which is performed using the equivalent test weight (ETW) provided by the
manufacturer. EPA agrees with the reviewer that estimating the effect on weight of various
technology choices can be useful, but that analysis is beyond the scope of ALPHA.
C. A master input file can ease the execution of parametric studies and help validate the
model's behavior and the component and system levels.
31 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
EPA Response: The version of ALPHA provided to the reviewers contained only a few
electrified model simulations which were run independently. However, ALPHA also includes a
batch processing script where all inputs can be listed and multiple simulations (e.g.,
parametric studies as suggested by the reviewer) with different inputs can be simulated
together with both input and output parameters collected in a single file.
D. Improve the model documentation with explicit modeling assumptions, especially regarding
hybrid controllers.
EPA Response: EPA agrees with the reviewer that improving the documentation for ALPHA
would aid users, and specifically that an expanded description of hybrid controllers would be
beneficial. We will work to implement this in the ALPHA documentation. Please also see
EPA's response to Charge Question 2, Shawn-Midland-Mohler, comment £
E. It is essential to provide insights into the representativeness of the core data used to
develop these models. It is not entirely clear when these data were collected, how relevant
they are today, and how relevant they will be in the long term.
EPA Response: EPA agrees with the reviewer that explaining data collection and relevance,
and properly documenting the processes for obtaining electric component data is essential
for a robust, transparent modelling process. These data were collected using methods similar
to those used in engine testing which is described in detail on our website:
https://www.epa.gov/vehicle-and-fuel-emissions-testing/benchmarking-advanced-low-
emission-iight-duty-vehic/e-technology. EPA independently evaluates the data to ensure
the resulting ALPHA input maps reflected physically reasonable component performance as
described on our website: https://www.epa.gov/vehicle-and-fuel-emissions-
testing/combining-data-compiete-engine-aipha-maps. EPA is in the process of
documenting the electric component data used in ALPHA modeling and will be publishing all
the information on a new EPA web page dedicated to electric components. However, users of
ALPHA can easily substitute their own emotor efficiency maps into ALPHA if they choose
without altering the underlying vehicle control strategy.
ADDITIONAL OVERALL COMMENTS PROVIDED (NOT CHARGE QUESTION-SPECIFIC)
Peer Reviewer: Suiit Das
A. A different detailed simulation model primarily for electrified vehicles needs to be developed
with a focus on life cycle C02 emissions instead of tailpipe emissions simulated by ALPHA.
EPA Response: ALPHA is intended to simulate vehicle performance on regulatory cycles
performed on a dynamometer. A life cycle analysis for C02 would, by definition, need to
incorporate factors and assumptions beyond vehicle dynamometer behavior and is beyond
the scope of ALPHA.
B. Any comparative analysis with the similar forward-looking, full vehicle computer simulation
model such as AUTONOMIE used by U.S. Department of Energy will be useful towards the
model validation.
EPA Response: EPA has compared the ALPHA simulation of a conventional vehicle to an
Autonomie simulation of the same vehicle. We concluded that given the same inputs, not
only were the C02 emissions very similar, but the second-by-second vehicle performance
32 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
was very similar as well. EPA will consider different approaches for providing such analyses
to the public for consideration.
C. It is critical that the level of accuracy of vehicle performance results obtained from a
simplistic model such as ALPHA be frequently demonstrated and documented to meet the
stringent requirements of any Federal regulation such as CAFE in this case.
EPA Response: EPA agrees documenting ALPHA and demonstrating its accuracy is
important. In support of this approach, EPA has spent considerable time and effort producing
publicly available ALPHA documentation in publications such as the Draft Technical
Assessment Report (TAR), benchmarking data, key ALPHA input file descriptions, key ALPHA
result outputs, and technical papers and presentations. Many of these describe ALPHA
validations against detailed data captured on the dynamometer. However, an important
factor to consider is that ALPHA is EPA's tool internally developed and used to estimate
future C02 emissions and is not a regulation compliance tool
Peer Reviewer: Shawn Midland-Mohler
A. In browsing the model documentation, it was very heavy on the programming structure of
the model and very light on the actual modeling approach. For instance, the only model that
seemed to describe in any detail was the vehicle loss and inertia model. There was very little
insight provided into other plant models or the control models. For others to adopt this
methodology, these details need to be readily available to users. As it is now, a potential user
is left to interpret the intent of the model from the structure of the code. This is possible,
but it leaves lots of questions and for it to be widely adopted would need to have more
information provided.
EPA Response: EPA agrees documentation for ALPHA could always be improved and will
work to do so within our resource constraints. We thank the reviewer for the specific
suggestion of providing more detail about other plant and control models. Some additional
model details have been provided in other publications which will be integrated into the
manual. The initial purpose of the provided documentation was to establish a consistent
framework for continued ALPHA development which includes a heavy connection with the
program structure. Please also see EPA's response to Charge Question 2, Shawn-Midland-
Mohler, comment £
Peer Reviewer: ICCT
A. The review process of the ALPHA model considered both the provided MATLAB/Simulink
scripts and models and the accompanying documentation. Generally, the different battery-
electric and hybrid powertrain models are developed based on solid methodologies that
capture state-of- the-art technologies with proper modeling techniques. The modeling
approach is thorough and presents a comprehensive energy analysis of the different
powertrain physical domains, where the sub-systems' interactions are clearly defined and
well established. The core of the models mainly relies on a data-driven approach, where the
physical behavior of the main powertrain components, such as the battery, electric machine,
engines, etc., is captured using multidimensional data sets. This simplification is
understandable at the powertrain system level, as simulating the dynamic physical behavior
of every component would be beyond the scope of this model.
The structure of the model is modular, allowing a standardized modeling and simulation
environment for all powertrain technologies based on a common model core, as presented in
33 | P 3 g ©
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
the different libraries. The model includes a detailed set of controllers that dictate the
behavior of the powertrain and permit communication among its components. The model
architecture is well-organized and self-descriptive, helping the user navigate easily and
making it less prone to logical and syntax errors. Finally, the model results are communicated
through an output file summarizing the main key performance indicators at the system level,
in addition to detailed output files describing the different components' behavior.
EPA Response: Thank you for the comments; EPA agrees.
B. Nonetheless, some assumptions at the component model and control levels are not clearly
presented and require more clarification. These are discussed in more detail in question 2 in
the responses to the charge questions.
EPA Response: Controls and components for the electrified models were based on the
operation of specific popular and well-performing vehicles in the fleet (for example, the P2
model was based on a Hyundai Sonata). The engine control strategy in ALPHA demonstrates
a good match with test data from these vehicles. Although we recognize there are a wide
variety of operational strategies in the fleet, we believe our approach produces energy and
fuel consumption results that are reasonably representative of a wide swath of the present
and future fleets.
C. Finally, the provided documentation falls short of what would have been needed to develop a
high-level understanding of the model structure and its input-output framework. The lack of
proper documentation forces the user to dig deep into the code and structure of the models
to understand the model's inner workings. While it is understood that such documentation is
not meant for commercial use, a more organized summary of some critical assumptions
would have been appreciated. This also renders more transparency to the overall data-
driven modeling approach.
EPA Response: EPA agrees documentation for ALPHA could always be improved and will
continue work to do so. Descriptions of the electrified vehicle architectures modeled will be
included in our Regulatory Impact Analysis and the validation of these models will be
described in a future SAE paper. As these materials develop, we will also add more to the
ALPHA documentation to describe how the variables get passed into and out of Simulink for
the actual simulations.
34 | P 3 g ©
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Appendix A: Comments by Reviewer (Unedited)
s
CHARGE QUESTION
COMMENTS
1. Does EPA's overall approach to the stated
purpose of the model (demonstrate
technology effectiveness for various fuel
economy improvement technologies) and
attributes embody that purpose?
Four major types of electrified vehicles have been considered in ALPHA and it is
appropriate that the ALPHA simulations of two most important vehicle types (i.e., BEV
and PowerSplit HEV) in the near-term have been selected for the peer review. Of the
electrified vehicles, U.S. DOE/EIA projects new BEV sales increase faster than any other
type of battery-powered vehicles, both electric hybrid and 300-mille electric vehicles
reaching at ~ 1.2 million/year and both BEVs and PHEVs combined would account for
13% of total LDV sales in 2050, according to the AE02022 reference case.
As ALPHA model has been primarily created to evaluate the GHG emissions of ICE
Light-Duty vehicles, it is less appropriate to evaluate alternative pure EV technologies.
It is also to include other specific electric vehicle types, i.e., sedan, SUV, CUV, and
pickup in the future model updates.
2. What is the appropriateness and
completeness of the overall model structure
and its components, such as:
o The breadth of component
models/technologies compared to
the current/future light-duty fleet
o The performance of each
component model, including the
reviewer's assessment of the
underlying equations and/or physical
principles coded into that
component,
o The input and output structures and
how they interface with the model to
obtain the expected result, i.e..
ALPHA model is fairly straight forward tool only for an experienced MATLAB user for
understanding vehicle behavior, greenhouse gas emissions and the effectiveness of
various powertrain technologies of current and future vehicles by appropriately
changing input values in five major vehicle parameter files. Parameter files are
organized at the level of five major EV components, i.e.. Base (Driver and Controls),
Vehicle, Electrical, Accessory, and Transmission for running a desired EV technology.
Battery and the electric machine are a part of the Electrical component. The ALPHA
model is currently limited to CO2 emissions for five different EPA driving cycles
including custom driving cycles based on test fuel properties and vehicle fuel
consumption.
An assessment of the underlying model equations and/or physical principles couldn't
be made as they were limited to the original Simulink code without any appropriate
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
fuel/energy consumption and CO2
over the given driving cycles.
0 The use of default or dynamically
generated values to create reasonable
models from limited data sets.
model documentation available including the limited peer review time. ALPHA model
0.2.0 documentation is an excellent resource for a MATLAB model user in terms of the
contents of various files, but no description of types of equations including the source
and validation of the equation parameter values used. A Data Dictionary of variables
used in the model would be useful for better understanding of a novice user.
The model is completely input data driven, which need to be collected by either
engine or chassis dynamometer testing by specific vehicle system technology case.
The model application is thereby limited to the extent of validated data availability.
The overall model performance is dictated by calibration of numerous technology-
specific parameters used to determine final vehicle fuel economy and CO2 emissions
for various vehicle drive cycles.
The input structure is defined by five major component MATLAB files, in which the
input parameter values can be changed for the simulation of new technologies. The
expected results of fuel/energy consumption and CChover the given driving cycles for
the two vehicle types reviewed were reasonable. For PS HEV HWFET drive cycle, CO2
emissions was estimated to be 4%-21% higher than for the UDDS cycle.
The use of default or dynamically generated values could only be assessed by the final
summary output results. A documentation on the approach and the sources used for
the input parameter values would be useful for the model user to develop or any new
technologies.
3. Does the ALPHA model use good
engineering judgement to ensure robust and
expeditious program execution?
The ALPHA model is a MATLAB/Simulink based full vehicle computer simulation model
capable of analyzing various vehicle types combined with different powertrain
technologies. Although both current and future advanced vehicle technologies can be
explored by defining appropriate parameters in five major EV component files but
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
requires a fairly knowledgeable of the specific MATLAB version (e.g., 2022a for the
review version) to ensure robust and expeditious program execution.
A specific simulation runtime is significantly high, more than 10 mins. without providing
any indication to the user progress made so far. A user-friendly front end useful for an
expeditious sensitivity analysis of key input parameters.
4. Does the ALPHA model generate clear,
complete, and accurate output/results (C02
emissions or fuel efficiency output file)?
The output Excel file is fairly simple with summary results of fuel economy (MPGe and
Whr/mile). Fuel efficiency output file is detailed and clear with both Phase and
Weighted aggregate results of energy economy, efficiency, and consumption by drive
cycles. Energy Audit report by the drive cycle is fairly detailed in terms of energy
balance at the gross level and by major EV components.
Energy consumed by Accessories has been accounted as the sum of Generic and
DCDC Converter Losses. Fuel consumed (grams and gallons) including CO2 emissions
for conventional vehicles are reported under Phase Results by drive cycles.
The simulated AC usage in UDDS and HWFET drive cycle results of a 2019 Tesla Model
S Standard Range compared well (within less than 2%) with EPA certification values.
5. Do you have any recommendations for
specific improvements to the functioning or
the quality of the outputs of the model?
Detailed sixty complementary graphical output files as a function of drive cycle time to
the three summary output files provided would have useful for the evaluation of the
model functioning. Unless an expert MATLAB/Simulink user, it is not intuitive to track
down the logical flow of summary final results from its initial parameter values used in
underlying equations.
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
ADDITIONAL OVERALL COMMENTS PROVIDED (NOT CHARGE QUESTION-SPECIFIC):
A different detailed simulation model primarily for electrified vehicles needs to be developed with a focus on life cycle C02 emissions
instead of tailpipe emissions simulated by ALPHA.
Any comparative analysis with the similar forward-looking, full vehicle computer simulation model such as AUTONOMIE used by U.S.
Department of Energy will be useful towards the model validation.
It is critical that the level of accuracy of vehicle performance results obtained from a simplistic model such as ALPHA be frequently
demonstrated and documented to meet the stringent requirements of any Federal regulation such as CAFE in this case.
ADDITIONAL COMMENTS BY SPECIFIC ELECTRIFIED VEHICLE MODEL
N/A
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report December 2022
. 1 ¦ , ¦ wn Mellaril Mohler
CHARGE QUESTION
COMMENTS
1. Does EPA's overall approach to the stated
purpose of the model (demonstrate
technology effectiveness for various fuel
economy improvement technologies) and
attributes embody that purpose?
The scope of the review is focused mainly on the BEV model and the PowerSplit HEV
model with a secondary focus on the Strong P2 HEV and the Mild PO HEV model. In
all of these cases, the modeling approach is able to meet the goal of modeling
energy consumption. I have concerns about the current state of the model with
some concerns about how the control algorithms are implemented.
A main concerns is regarding the likely adoption of this model or the utility for the
intended purpose given the observation that: a) this class of model is already in the
market; and b) the vehicles that this model focuses on (BEV, PowerSplit HEV, etc.)
are available from many OEMs and, thus, there performance across many different
vehicle classes is well-understood.
The modeling approach used is typical to that used in industry and academia, thus, it
is appropriate. However, the approach also does not lend itself to easy adoption
outside of expert users. In general, people with sufficient expertise to modify this
type of model and yield reasonable results likely already have existing models
available to them. One of the stated goals is that ALPHA will gain wide acceptance in
the light-duty vehicle automotive community, and I do not feel that is likely to occur
in the current implementation.
The model approach is very similar to that used in Autonomie which has the benefit
of many years of development. A main feature in Autonomie that distinguishes it
from ALPHA is the availability of a GUI for model creation, modification, and data
analysis for users to exercises models. The ability to scale component models via a
GUI, queue up different drive cycles, adjust control parameters, etc. seem to make it
better suited for the intended purpose. Autonomie also has a more robust library of
component options as well as more robust control algorithms.
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
Because of the availability of tools like Autonomie, in-house tools, commercial tools
like AVL Cruise, and even example models provided within MATLAB, it is unclear how
ALPHA in its current form will meet the objectives. To be very clear, the technical
approach of ALPHA seems to be sound. My concerns are that it is not providing a
solution that is not already in the market with more established products.
Furthermore, the marketplace now has a wide variety of electrified vehicles available
and there is data associated with these vehicles in the public domain. Organizations
have access to this data from their own vehicles as well as competitor assessments.
From a planning perspective, it is not clear what a model like this is able to provide.
In the area of BEVs, the increasing offering from many OEMs gives us the ability to
reliably estimate things like range and energy consumption based on actual vehicle
data.
The core energy consumption of the energy storage system and traction motors
which ALPHA focuses on is quite well understood and apparent from test data that is
in the public domain via certification requirements. Aspects like HVAC load, battery
cooling during fast charging, etc. which are areas which are more challenging which
can significantly impact real-world energy usage and range are not well modeled in
ALPHA or most models.
None of the above should be taken as a comment on the modeling approach or skills
of the developers. The approach seems to be typical of the class of models that
others have deployed for this purpose. The main concern is if ALPHA will serve the
intended purpose in terms of being impactful in the technical community. Given the
availability of public domain data on these vehicles, the availability of internal data
on their own vehicles to OEMs, the availability of competitor assessments, and the
availability of other simulation products capable of the same type of analysis it is not
clear how widely this tool will be used.
40
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
2. What is the appropriateness and
completeness of the overall model structure
and its components, such as:
o The breadth of component
models/technologies compared to
the current/future light-duty fleet
o The performance of each
component model, including the
reviewer's assessment of the
underlying equations and/or physical
principles coded into that
component,
o The input and output structures and
how they interface with the model to
obtain the expected result, i.e.,
fuel/energy consumption and C02
over the given driving cycles,
o The use of default or dynamically
generated values to create
reasonable models from limited data
sets.
The breadth of component models/technologies compared to the current/future
light-duty fleet:
Overall, the model has the overall systems that one would expect for the stated goal.
Given the importance of HVAC and battery thermal management to BEV and PHEV
platforms, this is one area that is not well-developed in the model. The mechanical
and electrical accessories are divided into four submodels, generic loss, power
steering, air conditioning, and fan loss. In the BEV model, there was no energy usage
associated with the engine fan, power steering, or air conditioning system. This
could indeed be the case; however, the modeling approach is map-based and would
require this information to be specified by the user. Given that the loads from these
systems can cause significant reductions in in-use energy efficiency, higher fidelity
of these models would certainly add to the capability of the model.
The control models deployed in the models reviewed also poses a challenge. As with
any model of this class, a controller is necessary to manage the torque split and gear
as well as other important vehicle functions. The quality of the control algorithm can
have a major impact on the efficiency of the simulated vehicle - a great vehicle
component design with a marginal control algorithm/calibration will perform
marginally. There are optimization techniques that have been applied to this class of
models to allow a more refined control to be deployed without excessive calibration
by the user.
The control algorithm used for the Power Split vehicle was inspected which is in the
PS_control.m function. The "working" part of the code consists of less than 100 lines
of code and is what one would refer to as rule-based for the most part. There are
comments included that say"% ::What is this?" and "% ???" which 1 can understand
as a person who has done these things before - but also does not lend confidence in
the maturity of the control algorithm provided. Given the importance of control
41
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
algorithms for predicting the efficiency of vehicles it is critical that these be
matured.
The performance of each component model, including the reviewer's
assessment of the underlying equations and/or physical principles coded into
that component.
Overall, this is more or less an implementation of a force balance on the vehicle. The
components are modeled via maps and the basic relationships between the
components. No errors were noted in the summation of torques/forces that acted
on the vehicle inertia. A model of this class relies on appropriate component maps
and appropriate controls. Without a more rigorous look at these with comparison
data it is not possible to provide a full assessment of this.
When inspecting the driver commands (brake and accelerator) and high-level
control inputs like gear shifts and torque commands, I did not find anything of
concern. Depending on the underlying control algorithm and driver model, there can
sometimes be high-frequency behaviors on these signals that are not representative
of actual vehicle controls or driver behaviors. This was not noted in the model
outputs reviewed.
I was not able to find much documentation on the actual models used outside of the
overall vehicle mass and loss model. This made it challenging to review the modeling
approach as it needed to be interpreted from the model and input/output.
The input and output structures and how they interface with the model to obtain
the expected result, i.e.. fuel/energy consumption and C02 over the given driving
cycles.
I provided very detailed comments on the output in response to the fourth Charge
Question below. In this discussion, I will focus on the input structures. In the demo
42
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
files provided, the input structure was clearly defined and using variants in the model
appropriate subsystems were enabled.
It was not clear to me if there was scaling that could be applied in the input structure
- there did not appear to be. That is one aspect that is generally quite useful to be
able to slightly adjust component sizes without having to generate new component
data files.
The use of default or dynamically generated values to create reasonable models
from limited data sets.
Compared to more mature projects like Autonomie, there are not many options. The
capability is there, but 1 did not locate any library of models or the ability to scale
them. Likewise, the control algorithms were very likely highly specific to the
particular set of components they were calibrated to work with. 1 was not able to
locate documentation on the nature of the controls but after inspecting the model
and the input files it seemed to be very calibration-based.
3. Does the ALPHA model use good
engineering judgement to ensure robust
and expeditious program execution?
Overall, the modeling approach used seems appropriate to the technical goals. The
fidelity selected provides fast execution. This does have drawbacks as it is heavily
reliant on experimental maps as input. As outlined in a previous comment, the use of
relatively simple control algorithms rather than techniques that have some manner
of optimal control is a weakness. Adding this could result in slower execution time,
however, it would likely result in better results with less overall run time for the user
as it would require less runs to calibrate and adjust the control.
4. Does the ALPHA model generate clear,
complete, and accurate output/results
(CO2 emissions or fuel efficiency output
file)?
For this charge question, the output will be considered the log file, the console
output file, the results file, and the figures that are generated from the model run.
43
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
Clear Output/Results:
Log File: Overall, this was reasonably well organized. See comments below on some
items that led to some confusion.
Console Output: This was clear. Given the tabular nature of the data, it would be
helpful to output this as a .csv or use the Report Generator capability in MATLAB to
give it structure as a pdf. 1 can see users having to do cutting and pasting to use this
data for whatever their purpose was.
Results File: This was clear, but the horizontal format is not what 1 would have
preferred. It seems like it would be more readable as a .csv in a spreadsheet app if it
were arranged vertically.
Figures: Many of the figures did not contain proper units for some of the axes. 1 was
able to infer the likely units but that is clearly not good practice. Given that the
figures are only generated upon run, 1 would have appreciated them being saved or,
at a minimum, written into a pdf file and logged with the above files.
Complete Output/Results:
Log File: The log files that 1 inspected did not seem to fully be populated with data or
1 may have been misunderstanding the intent. In the "Configuration Keys" section
there were many entries without values, for instance the A, B, and C coefficients did
not list values. 1 expected these to be the values that were used to generate the
results. There were many other areas where this was the case.
Console Output: The console output contained a good summary of overall cycle
energy flows.
Results File: The results file contains an effective summary of each cycle and the
overall composite of the cycles run from an energy perspective.
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
Figures: The set of figures shown was satisfactory. Obviously, a person can generate
their own. A concern is that 1 did not see that the data from a simulation was logged.
1 would prefer to have that option to avoid having to rerun a simulation. 1 was able to
locate the time series output in the "modeLdata" structure. 1 assumed that this was
the model parameters based on the "data" label - instead it was the output. 1 think
"model_output" would have been a more appropriate name or something even more
descriptive.
Accurate Output/Results:
Overall, the accuracy is difficult to assess because the models provided are only
representative of classes of vehicles and no direct comparison data is provided.
Upon request by the reviewers, the EPA provided simulation results that were
representative for a 2019 production EV. The results cycle-based results agreed
very well with the experimental results provided. This demonstrates that it can
generate valid results.
Overall, the modeling approach is known to be able to predict cycle energy usage
well under nominal conditions. Additional factors like HVAC loads and cold/hot
weather performance can be challenging to model with the current fidelity of the
models. 1 do not feel that is something that the model is expected to be able to do
at this point, so this compromise is understood. To bring those factors into the
model, additional systems need to be directly modeled or the resulting loads on the
system need to be brought in via the existing accessory load and efficiency models.
This typically requires detailed information on components/controls and/or
experimental data that is often not easily obtained outside of OEMs.
Additional Comments/Questions:
45
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
It was not clear to me if there is a warning issued if the drive trace is violated. If not,
this would need to be added because results when this occurs may not be valid.
It was not clear to me if there is a correction for SOC variation for non-plug-in
hybrids.
It was not clear to me in the structure of the model if the translational mass of the
rotational components were factored appropriately. There was some discussion of
this in the manual but without having time to really investigate this 1 could not be
certain. 1 noted how the rotational inertia was carried forward in the model to be
added into the overall mass where the vehicle inertial integrators were.
In reviewing the model_data variable, 1 noted that the modeLdata.controls part of
the structure was not populated with data. There were variables there in the EV and
PS model, but they did not contain data vectors.
5. Do you have any recommendations for
specific improvements to the
functioning or the quality of the outputs
of the model?
1 provided some specific recommendations in the above section. 1 do not have any
additional comments to provide.
ADDITIONAL OVERALL COMMENTS PROVIDED (NOT CHARGE QUESTION-SPECIFIC):
In browsing the model documentation, it was very heavy on the programming structure of the model and very light on the actual
modeling approach. For instance, the only model that seemed to describe in any detail was the vehicle loss and inertia model. There was
very little insight provided into other plant models or the control models. For others to adopt this methodology, these details need to be
readily available to users. As it is now, a potential user is left to interpret the intent of the model from the structure of the code. This is
possible, but it leaves lots of questions and for it to be widely adopted would need to have more information provided.
ADDITIONAL COMMENTS BY SPECIFIC ELECTRIFIED VEHICLE MODEL:
There are no additional comments beyond what has already been provided.
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report December 2022
Comments by ICCT
CHARGE QUESTION
COMMENTS
1. Does EPA's overall approach to the stated
purpose of the model (demonstrate
technology effectiveness for various fuel
economy improvement technologies) and
attributes embody that purpose?
The proposed model looks comprehensive and of high fidelity enough to serve its
purpose of quantifying the fuel economy, energy efficiency, and C02 emissions of
different powertrain typologies under a variety of operating conditions for several
technology choices.
The main issue that needs clarification at this stage is the powertrain
components' sizing and scaling approach. The process seems to be technology
agnostic. For example, in the case of electric motor-generator scaling, the model
appears to rely on one electric motor data-driven model reflecting a specific
motor technology. We are not sure if this is only the case for the shared demo
version of the tool and if the complete ALPHA model already includes several
components' technologies. If that is the case, please disregard this comment.
While the model is clear regarding technology choices focusing on hybrid-electric
and pure- electric powertrains, it remains unclear why fuel-cell powertrains are
excluded from the model.
2. What is the appropriateness and completeness
of the overall model structure and its
components, such as:
o The breadth of component
models/technologies compared to the
current/future light-duty fleet
o The performance of each component
model, including the reviewer's
assessment of the underlying equations
and/or physical principles coded into
that component,
o The input and output structures and
how they interface with the model to
The breadth of component models/technologies compared to the
current/future light-duty fleet.
The different technology and modeling choices are all reasonable. Special
attention has been given to the hybrid powertrain controllers, one of the main fuel
economy drivers.
The P2-PHEV controller, as an example, seems to utilize a charge-depleting
charge-sustaining strategy depending on the battery state-of-charge (SoC) and
the physical constraints of the different energy conversion devices. During the
charge-sustaining mode, the algorithm seems to try to operate the engine at its
minimum BSFC line as much as possible. Assuming that this is the only modeled
47
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
obtain the expected result, i.e.,
fuel/energy consumption and C02 over
the given driving cycles,
o The use of default or dynamically
generated values to create reasonable
models from limited data sets.
control strategy, the question here is about the representativeness of this control
strategy.
Other more complex and efficient controllers, such as the ECMS (equivalent
consumption minimizing strategy) based on Pontryagin's Maximum Principal
optimization techniques, can reduce total fuel and energy consumption, and
several vehicle manufacturers have considered it. While on the other hand, other
simpler heuristics rule-based controllers can also be implemented. In addition,
the domain of hybrid vehicle controllers has been going beyond the traditional
optimization-based or rule-based methods as more advanced neural network-
and machine learning-based control strategies are under continuous
development. So, the question here is mainly about the representativeness of the
adopted control strategy.
In the generated results file in MATLAB workspace, there is a mention of a
"s_factor," which seems drive cycle dependent. It is unclear why this factor is
defined and how it is used in the model. This question is raised here as the
"s_factor" is a common notion in the academic literature regarding hybrid
vehicles' control strategies which is defined as a conversion factor between fuel
energy and battery energy.
The performance of each component model, including the reviewer's
assessment of the underlying equations and/or physical principles coded into
that component.
The battery model is a resistance-capacitance (RC) equivalent circuit model that
captures the dynamic battery voltage drop as a function of the battery's main
state variables, such as SoC and temperature. RC equivalent circuit models are
proven to be robust models and widely used in battery modeling for vehicle
energy assessment. The battery model also includes a thermal sub-model where
48
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
the battery temperature and different heat transfer phenomena are estimated.
The battery's internal heat generation due to the exothermic chemical reactions is
well presented. The battery heat exchange with its surroundings, either with the
ambient or the cooling system, seemed to be simplified in a single parameter: the
battery pack's total conductance. It is unclear to us what battery cooling
technologies are considered and to what extent this approach can consider
different cooling techniques, such as active/passive air or liquid cooling. While this
issue might not be a game changer in hybrid powertrains, the battery
temperature is more critical on pure electric powertrains, given the size of the
battery, which would affect the battery performance and total energy
consumption. In addition, the battery efficiency seems to be quite high (96.9% on
UDDSJ, 97.7% on UDDS_2, 98.7% on highway cycle, 91.8% on US06J, 97.3% on
US06_2).
The electric machine model is mainly data-driven, summarized in dynamic look-
up tables focusing on the maximum torque curves and energy conversion
efficiency. It is unclear whether the model differentiates between continuous and
peak torque. The electric machine losses seem to be quite high (23% on UDDS_1,
29% on UDDS_2,15% on highway cycle, 26% on US06_1,14% on US06_2). It is not
clear if these are losses during both propulsion and regenerative braking modes.
The different transmission systems and the engine are developed with high
fidelity. The gear- shifting strategy behaved as expected for some sample runs.
The transmission system model is highly complex, which raises the concern of
data availability for model calibration for different vehicles.
The powertrain auxiliaries, such as the air conditioning (AC) unit, fan, pump, and
other electric auxiliaries, seemed to be modeled simply as constant power
demand from the battery or torque demand from the engine. While this
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
simplification is acceptable for most auxiliaries, such an approach can
misestimate the AC/heating unit energy consumption, which is highly sensitive to
dynamic operating conditions such as external temperature and trip duration.
Detailed vehicle cabin thermal and AC/heating models would enhance the model
capabilities to model the AC energy consumption providing a more accurate
estimation of the vehicle's fuel economy/energy efficiency. Although the user can
still input different auxiliary consumption in kW to mimic additional heating and
cooling needs, quantifying such metric is a complex process that may incur
inaccuracies. Thus, modeling such behavior is quite essential.
The academic literature has grown rich recently with modeling techniques to
integrate vehicle powertrain and vehicle cabin thermal modeling into a single
platform. Some examples of such literature are:
Marcosa, D., Pinob, F.J, Bordonsa, C., Guerrab, J.J., "The development and
validation of a thermal model for the cabin of a vehicle" Applied Thermal
Engineering. https://doi.org,/10.101G/i.appIthermaIens,.2014.02.054.
Doyle, A., Muneer, T., "Energy consumption and modeling of the climate
control system in the electric vehicle". Energy Exploration & Exploitation,
https://doi.ore/10.1177/01445987188084.
Regarding the drive cycles, it appears that the US06 drive cycle is split into two
phases, with the first phase incorporating the hard accelerations and the second
phase incorporating the high speeds. We recommend adding a combined US06
audit.
The input and output structures and how they interface with the model to
obtain the expected result, i.e.. fuel/energy consumption and C02 over the
given driving cycles.
50
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
The output file structure looks appropriate and complete. The model, however,
lacks a master input file where the user can easily visualize all inputs. While this is
understandable for a non-commercial model, a master input file would ease
model validation and allow the user to conduct parametric studies more easily,
which would help provide a more solid idea of the entire model's robustness.
The use of default or dynamically generated values to create reasonable
models from limited data sets.
While a data-driven modeling approach is reasonable for vehicle fuel economy
estimation, a significant amount of data must be collected or provided to
parametrize the model correctly. Engine test benches and vehicle chassis
dynamometers are well-developed standard practices for data collection for
conventional powertrain technologies. However, it is unclear how data for
batteries and electric machines were developed and the extent to which such
data collection methods are well developed. While this review covers the
modeling approach in general, clarifying the electric components' data collection
approaches via proper documentation is essential for the user to understand the
potential limitations of the model.
3. Does the ALPHA model use good
engineering judgement to ensure robust
and expeditious program execution?
It is not possible to provide a solid opinion about engineering judgment and model
robustness without a thorough comparison between simulation results and real-
world testing data at the component level and vehicle level. The peer reviewers
asked EPA if ALPHA has been validated against real-world results. EPA answered
that each sub-model had been thoroughly validated against data collected from
vehicle dynamometer testing internally and externally, where the tested vehicle
behavior was reproduced with the model. The general approach of the model
validation process, as shared by EPA, sounds reasonable and comprehensive. The
modeling run demos provided by EPA demonstrate good agreement between
51
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
simulation results and real-world data, but a more robust assessment is beyond
the scope of our review.
One important point to mention is the impact of the simulation step size on the
simulation error. While it is understood that a smaller step size would yield a lower
error at the expense of higher computation resources, it is unclear how the
simulation step size is set and what simulation error is considered acceptable.
4. Does the ALPHA model generate clear,
complete, and accurate output/results
(CO2 emissions or fuel efficiency output
file)?
The ALPHA model generates automatic plots describing the physical behavior of
the main powertrain components, in addition to a console text file summarizing
the energy consumption, fuel economy, and C02 emissions. We recommend the
following to make the results file more readable:
The accessory's energy consumption doesn't show the share of each
component (fan, AC, pump, etc.). These are aggregated under one value.
Although the generated results files contain placeholders for these
accessories, they show a value of zero.
There is no mention of the battery's thermal needs as an accessory
consumption. Is battery cooling/heating consumption considered part of
the battery losses? It is worth documenting how the model handles
battery thermal needs.
The results file mentions "kinetic energy" as a potential energy source in
addition to fuel and stored energy. It is unclear whether this refers to
brake energy recovery or another indicator.
The fuel energy could be further detailed into direct flow to the driveline
or energy flow from the engine to the battery, depending on the
powertrain architecture. This can provide a clearer idea of the control
strategy under different drive cycles, operating conditions, and system
boundary conditions.
52
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
CHARGE QUESTION
COMMENTS
In the model, it is mentioned that the final drive efficiency is already
included in the electric drive unit. This is probably why the final drive
losses are always set to zero in the results report. However, the final drive
is separate from the electric motor, and combining their efficiencies limits
the model's flexibility. We recommend fixing this issue.
5. Do you have any recommendations for
specific improvements to the functioning
or the quality of the outputs of the model?
Hybrid and plug-in hybrid vehicles' controllers (energy management
strategy) are rigid and developed based on specific strategies and
algorithms. Providing the option to simulate several hybrid control
strategies as part of the batch simulation runs would be more
comprehensive and may cover a broader range of vehicles.
A bottom-up approach is recommended to estimate the vehicle weight.
Estimating the components' weights and aggregating these weights to
calculate the total vehicle weight can provide a more accurate estimation
of the impact of different technology choices, especially different battery,
and electric motor sizes.
A master input file can ease the execution of parametric studies and help
validate the model's behavior and the component and system levels.
Improve the model documentation with explicit modeling assumptions,
especially regarding hybrid controllers.
It is essential to provide insights into the representativeness of the core
data used to develop these models. It is not entirely clear when these
data were collected, how relevant they are today, and how relevant they
will be in the long term.
53
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
ADDITIONAL OVERALL COMMENTS PROVIDED (NOT CHARGE QUESTION-SPECIFIC):
The review process of the ALPHA model considered both the provided MATLAB/Simulink scripts and models and the accompanying
documentation. Generally, the different battery-electric and hybrid powertrain models are developed based on solid methodologies that
capture state-of- the-art technologies with proper modeling techniques. The modeling approach is thorough and presents a
comprehensive energy analysis of the different powertrain physical domains, where the sub-systems' interactions are clearly defined and
well established. The core of the models mainly relies on a data-driven approach, where the physical behavior of the main powertrain
components, such as the battery, electric machine, engines, etc., is captured using multidimensional data sets. This simplification is
understandable at the powertrain system level, as simulating the dynamic physical behavior of every component would be beyond the
scope of this model.
The structure of the model is modular, allowing a standardized modeling and simulation environment for all powertrain technologies
based on a common model core, as presented in the different libraries. The model includes a detailed set of controllers that dictate the
behavior of the powertrain and permit communication among its components. The model architecture is well-organized and self-
descriptive, helping the user navigate easily and making it less prone to logical and syntax errors. Finally, the model results are
communicated through an output file summarizing the main key performance indicators at the system level, in addition to detailed
output files describing the different components' behavior.
Nonetheless, some assumptions at the component model and control levels are not clearly presented and require more clarification.
These are discussed in more detail in question 2 in the responses to the charge questions.
Finally, the provided documentation falls short of what would have been needed to develop a high-level understanding of the model
structure and its input-output framework. The lack of proper documentation forces the user to dig deep into the code and structure of
the models to understand the model's inner workings. While it is understood that such documentation is not meant for commercial use, a
more organized summary of some critical assumptions would have been appreciated. This also renders more transparency to the overall
data-driven modeling approach.
ADDITIONAL COMMENTS BY SPECIFIC ELECTRIFIED VEHICLE MODEL:
N/A
54
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Appendix B: Peer Reviews Curriculum Vitae (CV)
VITA
SI ,111 DAS
5oh I larp Mrecl
Alpir.tiett,*. (.» \
OlSJLCliNL
18651789-0299
Email: ujit_das202 l(goutlookxom
ICC
in research, analysis, and knowledge dissemination on energy, water, anil material
efficiency in the manufacturing sector by developing unci deployment, and promoting
hitpjclful pravtseev n*oK ask) technologies
I Hanked 2 •" in v l.tvs v ith Honors..
vmn i ssio\ \i » \i't Rii m i
2022 ~ Present Pfinc.p^l Enginter, y.f ategic Aria'vvi Inc. Ar'mgton. VA
Manuf JCtufing cost jnijlyy.* of iOtid oxide electrolyse ;¦ eel; jnd fuel cell storage nutenali
1«»S4 - W2 R c-c.tici« i Ik. I iicf i:. I>i'i i-si<>;) u.iK Rijfv ! ,i'\ 1 tiik Rs^IiX, IN
l'»'»2 2MM5 RtVD Suit Member, I ttciu'\ S>;•. i-u'ii < Ktk Rabe N;Hi«»n tl ! ah t Kik KiJik. 1 \
2#05 2W22 sr Rcscurih Stall'McwNk, I oeiin aod 1 un^twUtttna Vscikc l*.\i*:on.«
Radee Yitsvn.il L.dvrafor., hcv ember i'^-i parent
A dc:iV4Wt.i«cd J-.""ii_ term R«kl) team Scadct ,n the nuriuf.k.nmn^ civic) itV.vicrhA :c>cardi A.
"~upportcd b\ a •>* idv rat'iuc <•»! I' \ tedcial juictK.o tm»t.*Hh h\ \4«u*u> I'M '1 I'l Rl.
.~jTkc* and I **.wl Incrin i, jnd induct icv loihr*«cvt«m«nw, and Istc c\Je cncrtn ami
cm »>nd m.imil.K".t.nn.' "-up^N vham n. writu.tm»n. impleinc;Uiiihi«> c.irlv M.IJ.V
\ I- i Rl ettkv supported kvi;ru'N*.ic> ti» driw Atnc?tc»m eMmpctimcnc^ una ati»«t
in ciKi''i\. w ate:, and materia] elUucikx ,n the n\a!ii;Jaer.innt! -.cv",RM. prolan) iiunagcf RaiI) ..wtn tth> .ic/t>s> num di\iMon> in \,u.\>Un directorates hi
a 'Aide r.inec »>i tcchiv •<*«($ aic.iv. c g . bi'htue^ht matciuiK mditdititi avivaiiccd i\hcr> ;md
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Su|»t Das Page -2- Apr.l 2022
viiajpusik'h niiitiuiactiiniiii, ach iinvcd clem enerin miirjiitjitiifiiw, \ chicle maixifaeniriny. vs.dei
sypph diiJ dcimnd. digital and elci*ri1ication and encri>\ mfnistructures-
The Stratcuiv Advanced Mjmifji.ti.inng project ^nppc)n*_-J by the DOL Advanced
Manufuctuntu> < HTke Ix'iiil; led curreiiih L,rc« niofe than l\\ tee *,o a lc\cl of in FY2n
rcMillinu in >e\efal new st.;tf hire. < tshci new miluiiWN ha\e K-oti the multive-ai
tevhncwnomie and life cycle aruty-is of coal lu produib by 1 K'ssil h.ietL'y off.ee
SocrnI teehnoeevwnik and !ifc cycle ciiere,} models e>f ;xh .meed materials f e.g . carbon tlhcr
and it> composites arc the % >tiK models u-«ed widely h\ -ij.rk.ic. DOI KERF. offices. coal 'o
products tor the DOL 1-ussii htK-ry\ office i. urxi clean cnery* nuiniifjcfLirjii.' tech no lot as ic >j..
Additive imnuiaclunriu jnd Roll '.i> koli nianiif«u.!tinnj| were developed. tJccisiiirMnLikiM tuok
lur scleral resource mirko such as petroleum arid nuclear materials were developed in the pact
inciudine '.he recent ntppK wham competith ene^s analysis • »!"clear, encrjn nk-nufaelurinL!
leehitoicj.'ieN le.y . wide haiidujp materials .aid senitconeluekv manufoctLiriiiL1t. and integrated
enertn and de^ili.natmn >%stem desiust,
L nl!,'iKirdied «,>th several .nditstrv partner •> in the techno economic analysis oi miiteriaU
tcchiiok'jiics iur identifying cotismerc. tali/anon opportunities ir. uvrldw ide nutlets Published
icsearch in iclerreJ jowruk and presented in inieriiulicisal cunfererke» Sc\ era I energy project--
ui developing countries such is BangLnksh and India were led tn she past and recent tn\ ited
Ulk» on the lite e\c le analyse uf hghnv eight material- were m I joiada. C htry>. and Europe. The
folio winy Ikt prtn ides highlights of mult i disciplinary and a wide rmit'e of n-unulactuniw enertn
ettkiciicv research and anaivsjs pro;ec:> m the jjcjs v">[ encrt"., water. auid material etiieiene\
led:
¦ Strategic jainlvsH ol «cmicctrniuctor. ^iTurt. carbon fiber composites and w ate* use .n she
.tdv .viiced nianufaetuitnu tor ihe EliRh AM() fiilkc
¦ rcihtiK etimunnc and life c>vlecncrg> jnalysi1- ul vva!-U<'prwJi>c:*. e.ij.. carbon filvr
sonipositc.> and kiildmt: msiib'.U'n Jiwserials !l>r the 1 X >1. I osstl fnerii's (.Hike
• Integrated energ1. and desalination ^y^tem design and industrial nvder use for tlie IXJF
» L ead author ol Teciinnloi'y Asses-,men; >>f Thermoelectric .mil Wide B.uid (iup materials
ibr the eauoiiiL' L'.S Dept. id' i.ncryy t^Kiadrenmai Fechnoio;j\ Re, tew
» Fethno eeoitomic and life cycle an.ilysi,% ofcarhun f.ber ce>n-ipi.wi!i> fur the VVind Lncry %
Techmi^iji.es ot'liee
« Life t yde Fnerey and Em tmmncnu) Asses^inent of Aluminusn -InteE^n c \"ehic[e
Design
• Lite i ycle Lnertty and l:tn trotiirsemjl A^ess.nent of \!ul(< Matermi Lmhtw eight
\"ehicie<
» Supply <. ruin VtnnuJaeiurini: ( ompetr.iccrK^* \naivsis of .'\ddui\e \tanuf;tc!t.rini:,
I u.'hon i iber. arid \\ >de Band (lap Ma'.criaK for L'.S. Depi ».>f Fneruy
¦ Next ^ene.'jtioii inalcriiiSs wish cnet*ay e:ritssions reducttun pmensul jn the L'.S .rtdu.itrv
for DiH. Vdvaneed SSnnuI'dcHmiii; ilfik'e
» System-H \naly>b for the I ".S. Dept ol fiierijy Fuel f ell Teehnoku>ie-> t Mtlce
56 1 P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
kipt Ddi -3 Aoni 2022
• pruco-. trunk'Sme »i icnijvr.mirc iitd »cK «.», -.lam in the
^0'4Uit kV> p»»»cr «:mev i»«r |H »| t u.-i *. cli I'i%hmik-vics IVusnam
• ! ilc ivi!c nioUchiiL' t>i .liicrn.cnc h»hmc;L>hi em-aic M rhe l>< )1
PiiipuKvn \!;itc:i,i:-- Pivj.'.iiit
» Market pniciHi.il ,nul i:ii-.i-»nik*!iIr,* .'^..w.riem .if e'h.mol ,t:u! h\dto^en .i* aiiurume
tucA
• «'>•-! ipmiehr.i.' ,«kJ litciv.lc jin.ih jd% ,«k cJ* emv L'-. an J la btv\ c«?ht jn,»wrwK
t-.tr lit M- < Mtkc t-t \ c'lule "1 C(.hnckicic>
• Mj'.eml I',SNC^nu,nt> uij'.ed K' Pjiii.ei'-litp fur \ Vev (leneratinH ut
Vehkie-. i rcedem ( tn'tx'r.tm e Re--euuh il uv.ium*. , \ f< i
• ot'K'fK-'A.ih!^ tcvhi:c|i'!iic-. m .iirvl li.vu'l.nW-di
• Mil(-.i'liti.T. .tr«K-.-.
• I'Cini5>-nit ni ,«ih.sikc.I p.l;i'H)ti;nu;u'a h; iIimm stj^c t'voiiomiv J.irn.tt'e--
• f he cumr.t;iu! iinpik.ttnTA o! pm .n -he puvw: i:i indu
» M.ulvL't ,t>vcN~n;cii(> <•: ciu'i^v elhucrn U'el.n«>!o»;io. ma'i Nt.k rctti(jcr,iioi\ in Isidu
• in^\.vik«ii .Hid \l,«me;ntr,ee nft'.M. v, he vie i \ehkle-- in India
\-se->:r.cii; »•; <.tMn. \r* lU'lhi, India. October W2-
June I'WJ,
IKnelitpcd j ,
iLt tin 11 Ri ihc i'.w m:h) • n, Ind;»> .uul |
cnerey and economic nw-tiefersH actmtiec
Kist ;trt h Knrr»\ and I mnomte Nn^ihw Nvction. Oak Ridge National
i ;if>iiratm\. V-pUfiilu'f OHI-Dtciinhtr r)X4
l)>,>vur»K*i;ed .ind ev.ikused l-'A, IX >1 maiiiUsineJ uniipiiter1. I'mdel'., I c . Ikvuiv.
Benefit Lik-iua < mhi-- MnJel and (he l-Y;r<<'cutii Ait>v.i>iiin MmleS I)e\eloped .< v.-mpviser
"••ifsn.nc "Hlf H I 1" t.ii I I .iI'jjUoii M.nlet u i nc-u1.
LIST OF PUBLICATIONS
B(H)K ( H \Pll RS M RU Slit ft
"hce'. elir,^ e.nd !iie ^tut iijiitw eji.'hl ciia k'> " \lk>r\t it M iteriuK, 1 K.'-.i^ii
utui M ,,n;iii>Hinr,L' i.'i 1 i-Jr.weii'l:i \ e'.i^lcv 2c (n ( Nev ter hu iI xn'ikonicu 2it2
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
buiit Das Page-4- Apnt 2022
I book chapter* published in "Aduinied CumpuMlc Kki'.en.'.h ;ur Auiomuti\i Aoplicalmn.-
Mruc'.urul Inkyntv -md Cru.-limnr'.hincs>." I diled b\ Ahmed ttmarakhs, I niv u!
Sundcri.md. I'fv and published by Wile\ A, Sons i Aut< "13)
duplet Low Lust t'drKui 1 J\*e tnj'Aulijimilive Applications- (f';iet 5 Low Gkf. t arhim Fibre
DcvcltipaicnU: C hunter 1 "?¦ Low Cos! t urKm librc tic Au;oim-u\e Appjeuiuw i fait 2'
Appkeaiie>n->. EVrterratnce and Reduction \1udd-,i
"Rex". cling ,ind Lite C\Je Ivucn fur Liglmvciuhr, Vehicles..'' -\ Buuk Chapter in MateiuK
and Vljr.uLicsun.nt- lit Liuhmciulu Vehicle^. edited In i' k. Matliek. U\xidhe.ul
Pivb];>hinu Limited, pp jtw 33n. 201U
(..(eri.il t !r,o in \uioinnb:les " A Bonk i hnptcr in I'ncvcWcdia of rncrav publs-died h\
Eisner Inc., Vol. 3. pp #5** 2inh.
' Pki>',ic \\ as:e> MaiiuL'eaiicnt. < ontrol. Rewcltnu. a.iJ Disposal. 1 No) es Dnta t orporation, NJ
u muine.itimi Tedtuaiogy Etect»ai« to Cememd
lulu net O;' ThmiJ. IX'* iixV to.rnal cf 3>j'>ttiriifc.f<- fnergy., Grids an#Networks,.New. 202&
btlp-%:.'['d.cti.n[»/ k>. HM.2irtH>5i-i i lenv
el al.l
"I lie l'.JK'it'\ Kmlpt tul el \utnnmts\ v I'.iceti^me Seuviix" SuiLtUhiht-.' \)a!i'rna\ jtui /nwMi 'i-w-
Vohime 2^. Seplcniivi 2'C'i, U>l'lL'5. htlnv Ai itl1 Id Hi Hi i.-tinn.il.2u2ihelfl11'¦M i Aini^lnniu el
iO
"t lp!ni',;/L'd I'.jibuii 1 sbei ir« \\ irni 1 uibuie ttUiic DeMUii." S.Wi >2n|l<-1-II Sdhiiu
Suuu;mI LjlvfjUMu:*. \i'':iv|iie;e*ui.. N.M (At, (»flh l.mu» et j|.'<
" J ceiiiii."%.eumiiiHe AikiUms i»l < md Pitek I i. 'i»ni 1-itKi Nbi'Uhuluiiiiij.'" p:e-entii!>nii juiic I'aifcn 2lM''
Ltsnl'eie'tive. lie 'J m Lc\injrtun. on Juh 14-lV. 2IM>» 11>.». S.i
' (i'kikil Cji l".i'ii ;¦ il-et ("iiiiip^Mte*. Si psik C Iimhi i,"i>il^peli1i <¦ ene~-» AimK m^." "] eihs'iieul Rc|jnit. I lal.
Rul sjt' I.jIsmikU-r>. I uk R'iJl'c. I N I Dun. S. al ji.f
l.ilc L vck" lu»pd«ub oj \ulufiK»U^ Si^hitfhtnu- S'< \U
l.tiM. i i2(!< 2 H .,,¦¦,>/ h »/,u III 1 '' •.•.tismul.y Ct /1 >.'< 11 :is >u d u e Ut I. >
"V'flneie l.!L'lil«ei,i!lil;He t licit", I V lm(wei% in l.'S l.iL'hl-Jkih X'ehiek ! Ieel."<:u,-Utiiu'ble \1:i'.ei uls ;««:
58 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Sujrt Das Page -5- April 2022
T-ciiiniloi-^cv \ '.p| S. >ul\. p n >2>'fKm
'li'iiui'. jLi-mi :s I riL'iruxim;.' 1'jjiiu.v t'jikm H'vi A. I nimHi-sk-M' muted psc.saiiiiiiHi. ji liiv .Vh jislvJ
OvMt'ii & MjiiuLteUtim;.' tu'ileienee, helJ us i Lid.uid. Oil en M.ir 24
"Ol M MrulcjjiCN on LiglnWeitiht McUiL>" im itcd pre.-enmtion At the Advyucd I.iL'ihfy cit-'hl
N'chicLe-. and Matenak 2Hl't Forum, held nn I • 14 i tel.'! 6. Heriin. < ienmny
"Life t vele AiijIvms o('Auli';T)oti\c Llcclr«iK<.." paper pre.-cnced ,it :1k At Lt A XIV
amfcrcixv. held un Sep*. 2? Zk>.' 16 in «.IwviicNtnn. St
'(.irhon I ihc I timpt'sites in i Ik'h Vohcrte < i round Trjrtspnrtj.tmn C >>m petition Between
Altenmth eC inipcr prepared tbr piesentutinn and publication lit !'' OA M ,-V /k U1 on
Hi! /i /.'i /-J f(i Diiunth'. I-:.
"Lite t u'le Fuele1. ami l-:n in>r.mctitjl .Wcv.iia'iiI .¦>! \li„;mm:i»-lrileilMve \ ehielc IXMjjr.." H \F Pjjtet
No. MM 1 ' \\ „it ^iiiLk. f A I Api 14K puhli-luh in the S 11 M > ;ik Rulgt* Nalional
Laboratory -- Aluminum Offers Snwllt'Ki Tut.il Cn btm Fool print Amuji" Cmfipetn?"
M.iten.ils" - DRIVE ALUMINUM press i dense on- Api *> {h'emAnwk puHishal ,m
Apnt i ii, J014 • http. .'/nnnv.'jnee»LUt rein, t ix.%, ont/tww%/I0!>J3TV ntummtm-whttks-
"(jlnkil l ji(x.n Fiber C umpoMtes Supply i. ham I uniuelitn eness Ai'UiK^i^. d:.ifi report
tor ?hc Doh S'rir.euie I'ktnumii Assessment Ufilte. May 2H i i
"Mc?ji Mow, Recycling. Fncruy and Fnviro»micnt' '\i. t is. Ms and Ii". Paper presented a* the
TMS Annual Meeting on Feb. IJj, 14
' t'osi c»t t h\nersliip and Well t»'»\ heels C t >il L'u\ihk' I Vn''iipUm.ittle I'ailnsav in tlu 1 uiiNf-i'il b>s.etiH "" Sir. iled p.riei pie^enl-tien J!
it. MRS-H. \S1I'i| ' lull mi \k\ umed MjIl-iuja.'JiripjaA t linu. Vpl ZZ-JS. 2l'!.l
¦I'kil.tumi t n.Hij' Xfeui> I PC i.\l1 ti'i I i;.'l:i-ilun \ elueles". F .icl C ell Feeliiwli'-'ie- Phvuun ReeeM e. 11 S
nt F:K,rus. Sen; Ziti i.
"M.mutlieiyi ma Pi'Mees> \iodeline ol HJil 4l»l k\\A- tMiishmeij ile.it ur«d Pnuei tor Stjliuiwiy
t ue 1 f. eil Systems." Papei nu, t-.SFueli eilZii JZ 41 !S3. /'r>K>>!
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Sujft Das Pege-6- April 2022
Sewed is one of the eatpert KViewets far the fialloiwite torn «;««: U.S. DOT/U.S. EPA reports
-N'irrs\ fee! economy role making draft report "Mass Reduction I* Lifsht-Duty Vehicles, ft*
MihvI \ cj!- 2«ll "-25" i2n|ni
¦t PA \vh,ifK\J I .t'hf-lluit Pi'.\«train Hybrid Analysis (ALPHA) M«M*"|2Cl«!&J
"Vehicle. Mass Reduction and Cast Analysis - Heavy Dm? Pickup Truck and Liisht CoininefciaJ
Vans" report for EPA {2016)
"Costs of Medium- and Homy-Duty Vehicle Fuel Efficiency and Emissions Reduction
Technologies," Tetn Tech. Ik/NMTSA. Feb. '15.
•\i*« Ifcdudi.r. .inJ i 'iist Aiiiifsis _ Ueht-Dutv Pickup Tracks Model Years. 2020-2025" FEV,
!nv. t 5* \. Jutv.' M
RhJuvIhh-. ini Lijilit-PnH X tiitdi-i lut \lmicl \ 2mi~.2'I2\" l.l) \i; Tin- r cmw.
\V.ls.lil»iJ|n|i I'liunjih Refill, \|«I 2n!2
* ^ -i-lll -DlfU I ivIhimIi-^'V ^ i •-! V Ilk iH Pih i! Vt i#. fk¥ "Rtiftf i ?um
'A
"L:L'hh\ciiihtii!B OppnrtutiHic. in thefiMxd Auluim'tiv c Industry." in\ .led prcsentaliun ul ihc
2ih 1 Inlcrnjltiinjl Auk^rm'U'.c Lijihiwoiiilit Material-, Development I muni, held m
1 hiiiigqiiw. China. un Mur 24 25.'I 1.1 \[-o al the 12 " II MRS Inteniattun.'l (. onk-rcucc
>">n Ad'. anted Manuals. held id fjnnt}d»\ ( inna on V*pt. 22 2$, 2« M iS
' 11aportaixe ^f flc<«iurn:e Viability .WesMnent nl'Auhmome L mhnxuyht Mutetiub" niviti'd
pr^ciiution jt liic ird Annual Advanced Liglnwoigh! Materia h lo? \ cfln.lt> (.unicreiiec
hdd *>n Auj;. I M 2. IP. Detroit. Ml.
"Annlv-*^ niTucl F ;lunnl Trampan.iliem Acti\ lis and fitscnlinl Ri-arihuinn C eii^trninK'
Tronspurtjtinn Rf-caah Record. Jt'lima! r4 liic Tr;yiM">oft,i'.ion Research tk'jfd. Nn.
2 K».S. T rmiNfHirt.ition Research lioani of the \utJOJial Acadcmic-v Wj-Jimuhm. LK , 2I>1(S,
pp I3<« i 45
"Reducing it. 2H1H. pp. V*. V)
""MifdJiii;.' IVunil:- l 'ii j Mjur.c-iUii: Dili." \uin'ip«'.i\l liilcnijilii'ii.-.i. Am. •>. 2*Ml', pp. '4 .'K
80 1 P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Supt Das Page-?- April 2022
Hik'i v ic.i .iitidc 'H Slcv en A.-hltA.
"Anuh -i* ui i ihJ t.lhuii»i| rum>]Mil.i'io:: :l\ jr.i.; iVKHtul Pisnihutum l
11im-.pint.iti-ui Ki>tasd'i Rtvi>iil. Ji.iL'injI nl Uk. I'luimvnuKu't'. Kt^caidi Burnt Nt-. 21(v\.
Iiilii-pois.ilsoii lio-.t-im. b Rujuj re the NjImiijI A,.LRk-nm>. ft IX". 2(11!*. pp 1 'b-145.
"Low -< \nbot: I Ut'l SutaiiKl N!aUi> Jiij jr.ui .Ik' issue*." Hit'im iVlivi. -. > > I, jh. \<>.}. .Uts 2n|il,
j,|i 5Xi>-5'f!
*"lni|H'i Uiiii.o ..it f cunumic \ 'ijhihu A^t."-.Njii,jisC <»i" AuU»ilioli\ e Liu ji<» Mali'i uN," mi :L*u
,->ii.>v'ntdt!.i]! i'.l the Anmuil At^untv-tl t.i;jbc*.«.ci-'h; l>.i \ diu'lo." hchl i:; lXiri»il. MI
uii Any, 11-12. 2ti]f>
Chih>'ii.ili'tc l.ik I U'i; As^cvMiiciii ui \1.:usk.»Hun brunt I mi I'ltrt.v" SAt }'i|Vi Nu 211tO-(l 1 ->)2~5.
ie!v i>";Aiilnimili\c U jncm.Uk, P \
"Pt Jiniirs Mauiii'sium Pim'talmii !4n Awti'mutiv,. \tin*i^dliniis." Kuim.il ¦ ii'Mclalv \ >«l. Hb Nu. i1
2tJiM,pp, 5i«5K
*"A Systems Approach» Life Cycle TruekCast EsUmtUoii." HAl Pa|vr Nu 2Wi~hI ^5;i2. Nulkis >>1
Aulamotive Eagtiieera,. WanemMe, PA.
'Ai-.ti-iiu-ti'.i,' LmlitHciuhliiiy MalamK IkT.Jd I wiU.uuui," HRXL 1 M-2tl'lfi 545, < K4 llhiiv MuUhmiI
l.isbi u.itot •,. t kik RrIuc, '1 X. tv.n . 2'"IIIf-,
"l.ijjlstAtfisjIu I jppuriuiiiuc:- :w I-ucl t ell \ v-litclcx" S \1. P-jim Nu. 2INi5-M| .ihiu?. Stuclv rf
li.tkin'.utix t I.iiuir.ivi h. \\ dncialjJe. 1'A.
"A (.'< 'itipai i-.m c Aim-mikiM .if Ail*mj1itv Powciluims jr.ii BuC\ -in-V. hilc MultiuK tut AtltulK^'U
1 Vdikks." SAt. Parvi \l..2i'KI4-il!-IH?.S, S^uch tit A>.,kuiiuti\i: l-iiyniLVis.
\\ aitL'ikh'.k-. PA.
"HaA SI• IJamcn" 1!x \ ubilU', ul'R^v*dmy Plji>li<. s m IL-iliuis AtijiitUvlxs." ftuikniL" l\i|v; : 5,
Ptiv_nam on ^oliti W a^Cc IVlicV. SvL-.m! rl i uic-lrx jiiJ f n\ noiiincnl.tl Sutdit.v \*ak I. nu cim!\ .
Nl"a Hdvctt. t. I scjik-nilvt hM'i
' IX'la-niinatints Aiuj\iI» »>!'! iicsyv 1 i1ih:i\Wi>i; SijiiJjuV S.)i>l; liuiliw. Tiiu-,su. ORNL-^XJ".
i')ai Ric.vt LiJuitak'i. < Uk RiJt.'i., IN. JuiI
\\\ AKDS & PttCtKHSSIONAl. AtTIVI I IK>
RbjiUiJ sviJinii ll.c i«"p 2"-. >i| Aur U sciu.li>!!> u< ilu- hiicms iCM.'diJi tittvipimv l>>i 2 I'i.liic i.mlilw tiL'lu liuimulnniN Hit 1 iMnuiriuv I i ,11' i I AJi jhcl'J MatciuiK fui <'jic
tt olcm Rt^oU1 l*nivi.'j%!tv Sayd f.'i.inlir:Uiiiy Piuli.^^kiiuj «MyUu'.>a VVRl'i piLiiiiain -¦ 2IHH
Rcci(Vi:iil i >t llic 2nP I. ,S -1 *» i j.c I 11! [f ui stvii I jkji h'* itn) IK'^iluuli. ui Ik-Myn (lu! L'(1l'C min.1
.,ii"lr.u I. S. IX'piniiueiit ol kiium Pi4k:\ jikl MsltitV:- Aiuiv^i.s OMkv
61 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Su|it Dai Page -8- April 2022
Recipient >15 I t<-tc>l R, McI'jj!j;hJ AajiJ
1 k;cc-\e,ii l2Ulrv.2nb)| ietni .t* ¦ McmlsetAt LiU^c ot'SAl. 1 ii;jnieciiii(j Meecmii-.
American I enter tor Lite C.Xde ANsessrnem v t nmmittee member |2I)I.S rre-.cn;
Awuiiice.' J(ld4 jiiUiiui ul MeUiS l.k'sl Pj(vi Ii\ (lie MiucijI. Mei.tiv and XLilcnuk Snuel't I 1 MS|
l. hiiii Strict) nl lutMiuoti'il' S'.iii.'iricciirx i\S-j SuvtLiMkibk S'liwjr.iui Pc, eknuncHt I (Mii.iuUec
.'Jaii. Dee CflUi
Man hit .>1 S ijn-.ii'iitulii'n Re>,i*-iicli Hn.-.n! I '3 RBH tiitls_'c> s2lv
Tr.ir,-.pttrUi)>ni r.cuinnii.-e-.
\Jieuulnc ['luniiiumUini 1 Uek and I eelmtik>L*i5>
limlei! Spcjkv-i <.'!!• lilt' Ltle s uic VwomiivIH k»1 MjK'H.'.Is f«y Lkinii;.- I mvci\il» i*|" 1 eehiKihnn. ( ii.iu
I ¦.'•itleieitee bic-.in.ni I itejinA-Tj. ii>i SAI _i.t S »*. leJuc'ior. an J cost UiuivH:- ol hehc-iiuh
car and pickup ttuck. Jt.ii medium- ;y»tj ho.t v-i'jiv vchu L-.
Pie; Rexti'Aci it*i ncickci .u.H I iiuir.cc nil't.\uiik;I ei I
I've! Reviv»ci Sm Sevcul hint;.-'* and H*,» mmiiii'tititl Refuted .''mriiai,!-
621P a g e
-------
Curriculum Vitae
Shawn W, Mdlam-Mohler
Current Appointments:
Professor of Practice - Primary Appointment
Ohio State University Department of Mechanical and Aerospace Engineering, Columbus, OH
Director
Ohio State University Simulation Innovation and Modeling Center, Columbus, OH
Fellow
Ohio State University Center for Automotive Research, Columbus, OH
8 2019 to present
7/2017 to present
8/2012 to present
Education:
Ph.D.
M.S.
B.S.
Mechanical Engineering 6/2005
The Ohio State University Columbus, OH
Dissertation Title: "Modeling, Control, and Diagnosis of a Diesel Lean NOx Trap Catalyst"
Mechanical Engineering
The Ohio State University Columbus, OH
Thesis Title: "A Novel Fuel-Operated Heater for Automotive Thermal Management"
Mechanical Engineering
Wright State University
Siimma cum Laude, 4.0 GPA
Dayton, OH
3 iiol
6/1999
Academic Experience:
Associate Professor of Practice
Ohio State University Department of Mechanical and Aerospace Engineering, Columbus, OH
Associate Director
Ohio State University Simulation Innovation and Modeling Center, Columbus, OH
Assistant Professor of Practice
Ohio State University Department of Mechanical and Aerospace Engineering, Columbus, OH
Research Scientist
Ohio State University Center for Automotive Research, Columbus, OH
Senior Research Associate
Ohio State University Center for Automotive Research, Columbus, OH
Research Associate II
Ohio State University Center for Automotive Research, Columbus, OH
9/2015 - 8/2019
1/2014 - 6/2017
8/2012 - 8/2015
10/2008 - 7/2012
11/2005-9/2008
2/2004- 10/2005
Professional Licenses:
Professional Engineer
State of Ohio
License 75703
Inactive
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
I'sfuvs Miiii.ytiimii htiUfr.Mtsil License 1622962
(Vpicct Muriit'i-mem In-ailuti' Inactive
Awards?
&sHLSasMii .A'-ha,,'! Vjf.ti..1 1 K
• ftwawi u-aw J;V* "AK lsbuHv jJ> imd *lftl'u:n« .vmnhutKwv. i.«v.wK pivwii'S jihI Mfwgtbcmn? ilv
llofklu-t i,SI" !';!!lner:hip
'.XfcWyttUi I'jcuto .Uu !.Jii'/.L'kUi > mrciMt\ ' '.¦Ife.a.Mt I'.iiuiiiet'iiiiu -1'' '•
» he viiisi In mo-,i rrmudfnl (Vioiln ,nh i»« >>l ••tuWiu orv/iitvrmy a ml: m the • \'lk"rv
¦Vnpln'J Aiil-'ni'-mr l'n.-'tr[,vtmii I Vivmnrm ol I'n'iin i. u|->
• iVc-L'iuciJ t.* iitim",s ..ornniniiiori . \ . ,ipp!ie>l au'omoin t- Tin?.* t>.' <.'aie!> .>n>l
education
¦ im-tl,tnUmu IV.-hrvl-w Team 'JVcj- -"I-
• i'iCM.-mtfO U- j k.iin ol < iSt'-i 'Ak l.icultv jud :i>cjfch out! Kv:iu« vt itwii c\tc»M\ c f\iri(tu;.inn.-- ilm iiiu
t>vh«ol •w ftwv.ml m i ihio
•'"'in
• be u'nk-d l<- ihc- |unior K<.v< 'Ak Jaasllv uJv .r uho 1v\! pronvMe", ih»* l\iu!s i b| ihc I w* \\k MuduilcomiX'UUui
1 KAtfllM* AM? MOIORIM;
f>»ui»ntl Sitiiirn! Xdustul:
1 ' !^k' WH I In1 •»!»{«> 'mh klrmmh'j ,s,'hi
4 Vmi"l \'i AuMu-inows <'! vund Winkles ~r0'^
jm i J, .m4 i-norvx
\l;in;n2cnii*nl m ) kbnd hk\m^ Vohuk';> u itii« VnsivJv-i-«lK*n iM I hiec-W ,iy < \it:uysi 1 Hnnrnics Jt0* *
3 2n\~' 2'C' rh»,'hp • I he < >}iio St.tw ( ritvcr • «u
4 Zu>4 ? Mnte f nivc-'^iv
^ I "i\'i!:-un f-'ejuv. \ he ' !UiUo I nu cr- ny
^ 2'«i> hex'rtt .Ulttva K.t;;n]fa,'i<.hi i he ' M,iiv i
? \ iYo.M'ni f',n« 1 Vl^nuf' I he < >hio St,)?,' i n'vcr-.nv
s' jUi i-Ve.se?il \ius >ii pMil I'hc ! «hu-Stak* I t-!su;['rsi\
Dmionil Students Mcnlorcd & J'uiiclid8
1 ! ik' ( 'h;^ Slitc t m\ erstS;
2 j.mon ''.•lover I ho ^htn S?me ^ nivemtv
; Isslh't! i'uv* « >|miv S)ais* S mumi\
1 Pm-r w u I'V.^rssor kr;>rta\3 I >r was no\ pff'nui»'il ^ ^inliK of sludenL-;
64 IP a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Mast* rs Student (Ashisor / Lead CVAthNw)
1 M rang \nnl>r,v,ot Vanabilm ami Jniecn.m t>plmn/ation el at Vmpression Ignition Engine, 2009.
2 R, S M.umgami. imctx-l Hmmiwi Intetpolaiion Havd Hdpomt Mciaration Methods for Closed-Loop
f "omhsiion ('omrol ot\n iJ >1 l-naine 2>*>'
* '* M Hoops I ii^ertainP. \nal> st. lit«Vntwl Inpa'so! I'•!«* I l-ngm«% 2010.
4 R. is i VoIe>. l-.ngme Selection. \louelmi:. ami1 Vmtvl 1 H i'e/iiiu. Modeling and > "omrol oi an l-UvinvulK - I leated' "atuhM 2011.
<> K V l.teiett At! impfot, ed \SoUvl-IWJ \fctl*<«k>»¥V !o1' 'akNuUotiofaB Alternative Fueled Engine,
7 J \i I i.ivi>., 1 Ir'ncI l.iig;nc kxpenmental 1 VMgn and Aduimed An.nv.js 11\Unique.. 2< I i I
N i iitpta ' Iwraesen/anen of htgine and I r>tnsmr«Mon i uht leant-. fur ''kvtri<, llvhnd »tid I lug-in Hybrid
whk-io-; 2m 2
•' Mi kucia. reedd'oniaid An-1 ucl Ratioi 'o««iol Jim mu I lam.ient1 tpcnition of an Alternative Fueled
I ngme. 2i > 13
H. \ Hvde 1 nneiopmeni ofn I'mei-on <'onlro; S\m
13 N \ I uvcHaiisH AiuH mn oi Ktswix e Svxn Scnx'i.-. !u 1 'n-lxwi J i >ugr»stics of Automotive
r,iri:cul:rte Filters 2rgnni.-ic;«k Model 1 vised Ni'.pension > 'nlihranon tor Hvhn.l \ ehiele Ride and Handling Recovery,
2014.
!<• I1 k Ihlhtiom. kii-k! 1 >uU katm.il ua\ hsiume i "har.kteis/utiein, 2i'!4
17 1 Mukhertee i' mtrol ^trau\s:v _*til ^
J1 I' ,llgde I ook \lieiii! Mamiecni<'nl Sit.ikyk"; for Hxhtki \ chicks S
2; J \\ jid. Modeling and Simulating a IViiuimancc II Mm id k.lectiK \ cIikIc, 2h« ?
21 S 1 'ictntho, Svstom SalVto I >cv, lopmen! of i IVrt'o-rmarii e ''IlkA" I hrouuh a Model-! Wd System-!
1'ngtneeniM \ppK«aoh 2")o
24 Khamia, l'ulk\ elude Model i H.'\elopmeiitol a Ihbud l.k'CSia \kli:ek And I knekipraeiit v>l"a i \ifitiok.
I estinu l-'r.miewi rk ^ulo
2> I. A ('..rdireile Automiitiiij the ^ukR\tn c XtiulyMs of knock during Hot k'nuine otutu.. 2tioti with MAI I. \l.
OMtmi?.iUon koiitiiuN. 2ulii
27 M ,1 Yalslo I V\ elopmenl of a H\ hrid \ ehicle . kmtrol Ss stem 2id'<
2-S tikw-tei, I ieiogn and Wdidatioii o! an A.live 5teieo \'isK.n >'vster.! loi the ; »M 1 ,eo< "Ak 3, 2i'l 7.
2'' Modnk. Mivleling and ' "octroi of an Automated Manual I raiwinvdon f.'-r Kv>; * \k » \'ehiele 2nl 7
PS kibalama I K'siuii and Iniplenientaiion ol a kelled \ltenialot N'.aitet Sjt.tom tor die 1 'SI !u.oi" \R 3
\ek.ek 2H17
M 11 1 Ushop \kxle'4 !.i%ed SuvpenMon i >ptimi",ition of the t ihto. ^!ale I ooMAk" i \ ehiele 2>ilS
C S Iiahk, S% Hteini, and Sufeh (•ngincetmg Wlvole:., 2'd"
M \ MaiiJokhot I V\ ef'pinent ol kiedtetsx e i' ssf ihne i >ae^l i ue! Inse^toi \ kde! hnpioeed In-
ev lint lei i "omhiritKiii i *hai;n teii'.mo.n
34 h 'i i leppei. Aiiile 1'ioieet Management SwteinN kuiiHieeimg ol an A\ liueaoi hotoUpe, 2i'lS
\ Ihoma*: 'v'odehtvi and i'erfortn.wve \naiva. of ;t In Speed Vsiloroaiie I raT.mv;Mon for X m-the-1 oop
Simuliiuoii 2i d s
3(> lluid. ! K.',siun of keeoiifigwahle Intel 101 foi Autonomou-. \ ehiele kiototspe. 2nl^,
?7 K 11 kavathia, t neertamly Annlvasofan k'n-eme lest i'ell 2uIS
651P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
3S
S K
Sahu. XKnlel-Suppoited lie.a- Hu\ Honour 1 V\efopment, 2olx
V1
r k
1 i;tmhhn;L
iV-wertram'ptum/.iUon of an Meetrie \ ehiele. 2ol S*
4o
U Stoddari. <• "ompuier Vision Techniques lot \uiomouvo Peiception System-; 2oH
41
S
2'0i)
P 1 mike The >\no St.M?e ' nivei^ih
4:
2m is
202" >
M I'atil "I he < ^t;ile I rm ers.n
41
2i J1 S.
2i t2t'
K laiu.ihtra The < >hio Stale i nneimlv
44
| S'
2s CO
1. 1 onjinuie The \ State \ ruv eisitv
45
2^1 X
2«0t'
\\ Satta 1 he ! ^hfv% State 1 "m\ er*st>
2upi
2i0u
j KatIT Vl-'ram The Ohi^ Stale 1 'm\ et-atv
4 ?
2019
2^ 20
S ' h>el 1 he i ihto State \ nivetsits
4K
:mi^
2< O3
A \aiassmhan 1 he > >hu- State f ns\ ei>Hv
4i]
2» 0*>
K ^lena. 1 he Ohio State i asn iiMts
N'i
20 |u
2* *21
11 k'angaiajan i he « jhJo State \ nneisiH
>1
-HIS
2« 121
,lit> I he * shio State 1 am'eiMh
\ ¦*
2oi<-
2* *21
1'J Kyln 1 lie Slate i 'ins et.-.m
Ji0»>
- 2'0l
\'ikhyat K.dta The < ^iuo State 1 ns\ er^ity
<4
-•Oil
2i CI
Kanna Sisnds'aranKin Wnkateshu'aia The ' State I 'i\\\ e?Ml\
2u2o
2* *21
!ne P.ernal '\ he i >hu> Stale itn ei>;u>
}'!)!
present
Xh i Pndtiman 1 he * Siarr tuversih
2o2u
prosem
Ymeente k 'apuo-Ruj/
2t>2«'
present
UaL'e So\ey I he nhu« State 1 mveism
2*0o
present
i'ohn Knight j he t )hio State I 'niv erMK
20:1
present
Ron Smuh The ^ Suite \ ]m\ er^ny
2*01
present
Shaumva Jha I he ^ >hu< State 1 niveisity
<0
ji 02
present
\hhijeel Killol i he (ihu> ^'tate \ nn ersity
2*02
present
Kanti Russell The * >h'.o Stale ! nt\ef-al_v
!t4
.(02
pttsent
Katun Suidi 1 he 1 Muo State 1 amcf^K
MS Muilenl- \Uutm td ,unl t umlril:'
! hne SnvJei I he > 'hi.> Suto t m\eisiu
! Atlum \\k.7 _'»»> I Iv < thi>< Suite t'ima-uts
iVuitnc) Cobtirn .""i»¦ Hie < Sute I niver\iU
4. Kciinv l-'ollen, 2UU7. The i >hu> Slue1 rm ero> iu"utti!ill >m7 I he1 ,*hni St-ate1 m\er,ti\
Visiting Scholars Supers itt'd:
1 he tolk'wiitt imlw ijist's«onilucied research at Ohio Stale for periods of 3 - 12 while completing MS/ThD programs
a; other inshlulionx
t Smnmc Hc.-ruhvoiu >« >7_ I lie •, fbio Suite 1 inn itmu
2 Patrick Retvehi 2t«»S The t 'hio State I hmersuv
3 Amliea 2i»W. 1 he t >hu> 8tak i imersily
4 Ulailxnt \\ e>l,mv, I he Uhu> ht.itc I mveiMtv
> t'-cinharj tit nnin, Huln I he C >hm Slate I eiMH
i> Amci Mi!line;' 2"! I, I he 1 >li!M Si:rtc ! imcrxih
7 i Serous Kjhilamy 2«)I4 lite ' thie> Sute I hmoi
K All tea iunttt! 2nl4. the > Hno Suite! niveiMH
'' I I'm K;ge/i. :n)4, fix < )hto Shite 1 mu-tsiH
)H Mtiidu (luetcutni The < >hio S!;>le t'rmerMtv
11 ViiKYtrie ' 'upito, ritl'K 1 lie Olm> Stuie I ratei\n>
I iuk'ii
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
i
2* «if»
M>»7
khisee Bh.ul 1 he v *hm St «e > nn eisih,
2
:nii7
.kn.huii Suftfilce 1 ho< "hio State I nnerstn
3
2«K)S
John i.utr The (%e .Mate ( rwersity
4
2* IMS
kitrtruj S\ /eJ 1 he ('hiu Suite I nn e Stale i im etsitv
E>
2oun
2UU'f
\l i uxlfrev ! he ¦ ho state ; mver-uiv
lti
2011"
-010
Jer.mlei lan 1 he < Omt State ' imersit}
i;
\'|i N
:ini»
kilsit Miieaale\ 1 he ohi,< state ( nn e> -ah
i:
:o,!»f
:nid
Altx itvlia kei! 1 he i >hio State 1 niveisti\
n
2* H >0
iiv.ii
\ntiiew \muM I he > 'hk* State I 'rnwoaty
14
2i K N
:i>io
R\ail Ineiett lite >h
h
2K*"
John ! iivh 1 he >hia Stale t nn etsih
In
2<>U
:t:;;
Sal ah ladxv m 1 he i >hi I mlervusxl S he > 'hn> State i rmei ,!t\
IV,
201!
.'ei i m 1 Aitesh 1 he t >hio Si.4< I 'nn ei mH
ji
2oi2
2n;»
1 um ! ',un\ n i he i )hw M.i'e t nivetMtA
:i
2m 12
:.i; *
.liixin \\ aid The t >1 trv State nn ersit}
22
2012
2iiio State I ntveistty
:4
2ol2
:un
VnUre« SjX-iue! 1 he ' >hu- State I nn eiM'A
2^
2ol"?
2n!4
I'.ryan Sih ermail iht ' >iih- State S tm eisiH
2*>
2ol1
;n!4
\S.I "t iitskn 1 he i 'hit. State I tin er.it>
:?
2o M
:<>:4
• hiut.n Kttshpuiat I he' lno State 1 n:\ etiaK
2o 12
: State, t nix-eisit\
yy
2oIo
:in?
Shuhim Vang 1 he t >h;o State i .iiu'Imu
2o| /
:oix
I'-nanu Anle-raif Ut» ' *Hhi State ! m\ er:.n\
M
2«t|S
;m«i
'atv|iiehne kai 1-1 vt t.un 1 he < 'hi- > State ! ija ust'e
*2
?>>|s
:<>i"
XIinmii \ielUn ! he < '•hio State i tuveeah
2>>|X
2> 11 v
ktistina Ktmahita 1 l,e ' tino litale t rm etsah
34
2M1S
3H"
Philhp i !alke 1 he t ifuo State i am eistt\
2o]0
2< Cii
llie Beutal 1 he t >\\u- State t tmetsin
_5 !
2iC 7
21 CI
2* 01
k;>nii Kussel! 1 lie t line State 1 ittveiMH
5K
:n:i
:«e:
James h'tiJeis I'he t thie State i an ersiH
Research Stuff Supervision:
I*i has-^vmsed the mg teebnieuUtatY i *ik thou euitenl tetniitml j>t>s(tioi! is listed
1
2oN -
2nN
I'unii IulpuL' Ko-^Mrch .Viemr.t 1 he \ m.iUm nwsr^w
2
2o! ^
2017
Uu'ub kc/aeian I-Vn!-1 Kki^rul Ki*>.ear^hcr ! he v)hio Stale I nn crsUs
>
2is^
2« CI
\'.vm\\ \utwoil KoscS|k\".ilut 1 he <¦ -ItK' Stale i ns\eiMtv
4
2o|S
?0lo
shen^ 1 Ro^vrch Nctcntjsi 1 Iv ^ MiU* { nn^rsiiv
2ol ^
20)'1
/ hui\ u \\ \mj KtM:uv.h .Vk'ntisl UieOhioSudeS mverMt\
JltlN
2021
l\a)U i \intuiut i Sej^vM Kes^.Hrl, \s.i kvsCAuh s^ienu-A ! he t 'hir State I nivesMtv
!o
2"1N
- 2020
Juki l;(i\lrUe i h »',? 1 Kvloni! ReseaKfuM 1 lie ^ 'h;o Stale i nn ei^;tt\
li
2m)h
KodiUh* \u/vi ( hinerr-e/, Ke^eaKh \:.vo^iale 2 f-.n^iiuv! 11k ' ^Ir.o Slate ' ni\ ersity.
67 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
i:
2o2"
n
JiOl
lYrvnl
14
:u:i
¦ lV.*'ienl
Rnsuui i'VnhU'tlpous I V > St.tic i rm vr>ih
I 'cnnis Rs: scinch ^plvuIss! The 1 »hnv I nnerm\'
riidcr^niduulc and (inidusiie C oursev
Ur MiJlum-.N J^hicr •; toachirui Mcuv.; .*n aut^rmMive lechnw.\ii oKvuvcn .jmJ cupstouo J.v.ikir, Hi-; mivmlutpv
mk*;.n ekv.boriu- -;tnJi*nt cvaUjaiion sv.oh' K>f ' < v^ill k;inn;_'" r. 4 7 out .>!' u ^ jvmi stsita
IYi'miiI Ofltrcd
< 'mine Nunihrr mid llllc
\prme jut?
Ml 7Vi internal i v.mHi-.ih'iii 1 nuino M.xkiuv:
\\ wi 2mi"
Ml: I'M \iHuntoiiu' kabsiii't.an
Sj-iiiw
Ml*. 7*ii IrsH-im! iVmhusiion l;rrt>iiH* Minklmy
\\ msei In|t>
Ml* i-HI \uK»iHtlne Kmcrlnim 1 alhK.fWiv
VYmior 2< 111
k hv-t i >1 Muilkh'H'ip'inutx i'apy-tunc 1
'A inlei 2< 11 1
Ml- ii? MAh.inu.il I'iij;tn(.\THi^ I Vsign 1
Wsmei 2: ill
Xth.iol Aulomom c 1 Y>u<.iti:un 1 ah* Hat a v
Spring 2dl 1
F\< Ik Mull:uKcipNnarv 1 "apKlone 2
Spnnjf 2i:l 1
vn; tft Xkvhimicul I'njmeoimi; l iCMgn 2
\\ IliVl "'II 2
H\ il2
Ml1. n.
Nil-, i>' 1 Auu>im4rt s! 1 'owi'rirain 1 abonilan
i',lH2nH
\U'. -I'-'iC ' M \k't Kmh'iil 1' rgiiK-cnnj" < "apsti tic 1
! ;ilS 2"l,'
\'U: 4h'4 Ar^lieJ Muiuucincirt anil Svncni bnumwi in^ 1 u'ik*>
Spiint! 2i!|4
Ml- 4M*Cu2 Motliuiucifl 1 'apsl.ior 2
Spring 2(>I4
Ml >1 ,"uil^in...ti\'c 1 \m cr'.ruin ! uln -tutors
Spnnc. ?i M 4
Ml' H"4 ,\ppli..\i M-in,i^..-nH'iit ami Kv^m h'lcmeeinw 2 ''i'ili-4 ¦
I'rtll 2iH-4
Ml-. 4!-'h2 i'2 hivjiikvi mt;
Kail 2<.>14
Ml: >1"'4 Applied 1'h>|ia) Misiragt'mcni unci S>Hem hiKmcvuiiu 1
.'>P!inp 2JM
Mb. '>>'! Aulonvluc 1 \«wcrir.iin 1-aiiumkiH
Spur® 2<>l *
Ml- ^I'M 1 iaril«Mii>in-!lu*'lA\>p f.ir < 'vnlwl stem tVvdopiTKrii il'iKHi
spring 2i>l"*
Ml" 4««Cn2 NHvhaniral "i-ncmivrmg' 'ap^uinc 2,
Fall 2s'» 1 ii
\iri "mm Xpphcd I'rofca M.in.ipomt'nl .tnJ Sv-nem Prtiimctfrms
Full 2>)l!;.
\tl' 4,M2ii2 Itfchamcal i-'niimcenrv rapsione t
Spnng 201 ?
\fl* 4:.l!l2 l'2 Mtvliankal Imamcenni!'.'ap^ionc 2
hill 2<>I?
Ml; 4«0mc t
Sfiinu 2'<1 ?
Ml. 55.M Auiomutnc IVmciUaiii l.«K">rak>P
'^pmiL> 2i M 7
Ml1 4'ni2'i2 Ivk-chiirik-al 1 nynwcntiy ' "apviuni; 2
AiFumn 2iM ?
MI-.»"1II:'\0 4-fiO i'l \|l: i '.(pstoiw 1 '^.icri <1 x'-tuJent 1 icsian i ".impctitions
Spiutt: 2iMX
MI-> ":ll"\t.! ^'1 \nkinii) 1 M>-1 "apsloiio 1 V-,wn ill SiuJt'iil 1 v.wn <. \>m|X*tlSkm:.
Spring' 2(>lX
Ml- \yli.nii>u\».M-\'V,'i-niaiii l.alH'rak'H
Vlltimii 2"1 H
Ml- 4"!Ill ',11; 1 "apsloac 1 VM^ll S
\utun<.'mprtiti-in'-
'rniiu 2i>l"
Ml 4;HCt!2 Mi- ' 1 Ml Hsiukni 1 kwi^n ' "i'lnpaliliuu-.
\uiumn 2ni'.>
Ml. -1' 1111 \IKt\tpst.ilk' FlrSfJll 1
-Uiiun'n 2itl,'»
\-ll:. 4*,h2 ill Mi1 < 'ap-iicne t v-:i;jn it Sst.uk'nt riftfign * \-miwiuiim':
SpimL" ?i C.11
Mlv 'llk's'i' 4«!i2 i'2 Ml' » 2i|^!oiU' 1 V'M'.in ill Smckin 1 i."-'ign i 'pmpetilk'ii:
881P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Spring VII" f*\.HcHmm S.ahs.r;Uun
Auatriii ;iCi' \It- 4um>i \lh ntpMi'tk' I I
\uw:hh Ml; 4'''e.iut! ill Student I 'CMgn1 V-mpcUisom
Spnnji ."'CI MI.' "-^1 Aiiinmoinc I'o\i airaitt I ,ilvrjf>r\
Autumn ?iCi Ml: \i!- ' '.ipitoiv Design I
\mnmn Mh -BO HI \0- t -»psKW T'c-Mgii !I I >c.iun • "omivtmurr;
* iiiriruluni - InU'iiial"
.ML'.'JIL Vi' imv.l \uHiro..iU">- sk-im \»tah-k 2019
i'hi., x>«;sl>" i tt'\'cr-M\ 'Vlumhu- t >11
• J n»'\ Mis -i*! MH.I'C it. truli.,U autonomous vehicle sensing anil electrified powertraiiis
• Reused <. 50% of previous msteriftl
• Secured donations of tmmmtml ft* the lab Crow Fiat Chrysler unci Honda
. MbJlteMijjfi l s
< >hi«-> r-'SBic I nnvrsln. i Vlumhus < ill
» I )e* ekipeJ :i JcJicuk'J 4"'111 ivursc |m sU-kk'nis (.'iiga^eii m (he Mutlem I Vst^ti t "unipditkm ( ,i)vU"t el celk-.ijiue s oui.yc in.aerial
* 'hiti Sutlc I nn i'fmU \>lumN«s. > '11
• \ >?<< doped new course bused on \i \\- f-'Af' icodkuf* ^>n \,i\ue lor -.-ur undt'ryiaauau-\\
• * \hu so MaUcnb in a .v, swtp vim.ncv? in a and piouM nwiuucniuii K'U-ptos inc. mjciuiio
Mi: >i:
' duo Si.nc I. m\ «mc\. i Vlumix;:., < >11
• [ vi chp.'.l mnv ..¦¦..ui'ii'. >rt IiiI u> itimH.-'nvtil cmi''nlt.il. -A'-'.k'rn ntiUi'lmy riiut Vs
• l*!l'->il \v Is tiiiiik'J * 1:1 :t >'>«itivym <¦ ytnW licit: I'lu" Mnih'.suk •
All ! JC "HI
* '-hio Stall' I invert) t Viumbns. *
• u!apltsd L'Mslmk: S !.\l' and LW5U !u \wni unh sliKkm d-j^iim ^.inuvnlum Icvw.
• 1 \*Volv'pO»i lo i'.Hh! \Uk- a'ln as, laH\ ,itni h'u o tu'Vvl UaCJ luiul?.
\H t<*} ' VinhjKtH'n 1-nuHtv Mi'vlriiui' ^t5!1"
s 'hiv IMnn- i iii\ '"oluoihu'- s •! i
• koiis-sclo; -cd c-uiNv v. flh lw*'' - uuh • A ic-us ieau^ ni.ftci \h\ tc-lal r -J
•!SNign?:u'rslh
• 1 m cK*pc*u ivnteni Hi.u unii-cJ ,>!uth"nts ihrimch bmlainL' .in enure cncmc moJe! m JevoK'-poJ
ptoiCvt that uxrU mUa>frs->UuiJ.irJ ..smu?ahvn p,ivk;tu-:
MLZilLbliODUU S^hlA ^H1 't
¦ 'in-' I ii«vt'f:;i!s < Vlumhu, « >11
• Ko pu^ nk
Hfdu-^Sn -rekn ant c\p».no?k'*.! !ri! pi-^'lain
69 IP a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
ttajBJ£ea&siifflJ^ 2015
Ohio Slate University, Columbus, OH
* Developed a 6 hour seminar from cm IC engine control
• Supported by the Department of Energy
Ohio State University, Columbus, OH
• Developed a 6 hour seminar from on IC engines from a systems perspective
• Supported by the Department oi'Bnergy
MHam&MsMm 2014
Ohio Slate University, Columbus, OH
• Developed a 6 hour seminar from on modeling of internal combustion engines
• Supported by fee Department of Energy
Matlab tor l.Ma Analysis and Calibration Seminar 2013
Ohio State University, Columbus, OH
• Developed e 10 hour seminar on the use of Matlab for debt analysis and calibration
• Developed for the CAR Distance Education program
Simii Techniques for Automotive Control Development
Ohio State University, Columbus, OH
• Developed a 10 hour seminar on the we of software-iii-ilie-kxjp and hardware-in-thc-IiMp icchmqucs for
control code validation and verification
• Developed for the CAR Distance Education program
Alternative Fuck Seminar 2013
Ohio State TJntversity, Columbus, OH
• Developed a 10 hour seminar on automotive alternative fuels
• Developed for the CAR Distance Education program
Model-Based Control of Hvhrid Electric Vehicles 2012
Ohio Slate University, Columbus, OH
• Developed a 6 hour semimi Isi-m en nu-del-bsseilcorttiol of hvhrid vehicles
• Supported by the Deportment of 1 nergv
INTERNALSERVICE
11? to present
Ohio State University, Columbus, OH
• Responsible lor al Center leadership activities
• Supervise 3 business and 12 technical staff
• Grew center to S6.5M in research annual expenditures
• Supports research of more than 70 faculty, 30 graduate students, and 60 undergraduates
Simulation Innovation and Modeling Center, Associate Director 2014 to 2017
Ohio Slate University, Columbus, OH
• Responsible for day-to-day operation of the center
• Co-responsible for strategic leadership of the center
70 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
• Responsible for all hiring and staff performance
• Responsible for status reports to College and Honda
Bttsbit-" Si :it I' s» [n r\ Won:
1 li ^ 2i il 7 \k'\ixhufH t'uwt tut Managei. The ohm Slate i 'tmeisih
^ ,Ni|.. I V%em l.tiv In\lohi!m' Xtu'.e 1 iiveisiiy.
< *"<>!? lYcwnl Hcnihci V\et ule 1 Hicctor I he t>hm Si in-! m\erMly
4 :u!7 1'iesent \mlv; I'.mtemal I'tour.ini Klimatvr !'he ' ^hto State 1 mvcr-ulv
s 'il!'1 heseut •'simiHe Wi'iK'i Iv.miiin l'u>j;!a!! V.M.vml I ho (>hto Kt.m* I Tin i isity,
Inlcnl.il fto.ud < iinuiiilUc li»ul\< mini
Simulation Innovation and Modeling Center Steering Board 2014 to present
Ohio State University, Columbus, OH
• Work with other faculty to advance the mission of the SIMCenter
Center for Automotive Research Faculty Advisory Board 2014 to present
Ohio State University, Columbus, OH
• Work with other faculty to advance the mission of the Center for Automotive Research
MAEOmtuatt Admisstons Contntjlteg 3a; to
Ohio State University, Columbus, OH
• Review's graduate student applications and recommends acceptance to the Department and consideration
for Department and University fellowships
Student Organization Advising;
EcoCAR Mobility Challenge Hybrid / Autonomous Vehicle Team 8/2018 - present
Ohio State University, Columbus, OH
• Save as lend co-adviser of a 40 manlier (~S0% undergraduate) student design project team competing in
U.S. Department of Energy sponsored vehicle competition
• The.l«n« won the competition in the iir-tn-ar. (long with multiple honors amongst the various award
categories
\-kk ' \\K 3 1 Khikl Hearts. \ elm K- le.mi S „M14 ^
* >luo State 1 imeiMlv < Vlumhus, \ >11
• Ser\ e ,ts lead co-ad\ i\ei oi a 4U iremK-r t ^ » under giaduitle i Huden: design piojeet team competing in
i S ) >ep.uimc-nt of Stkn^v sponvied \ Jvcie competition
• i he team has won the eompetition m ;\kh of ihe four competition year* along with multiple honors
amongst the various ;mu?d ^aleeones
l-co- WK2 IhnndhleeUk Vehicle Icam 7/201 i -6/2014
1 %<,> ^Uie I iiiv ojmiv,' 'olumhuv < 51 i
* Sen ed us I cud a'-adviser oi a 4i>memhei» unde^iaduuteisiudent design, project team competing
m 1 S ! VpuiinienU'f j nety\ ^porwied \ chu le v.i«m|vtitu»n
* i lu team I unshed 2,al plu^e m the fu.a eo,u ot ^ompeiita>i,. .V* place in the second year, and 1*1 place the
imal \ea? o* the competition
®2«« - 6/20ti
Ohio State University, Columbus, OH
* Sewed m lead co-adviser of a 40 member ( 80% undergraduate) student design project team competing
m U.S. Department of finer©' sponsored vehicle competition
71 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
« i w ,ni I"'. >n »lv {bice yc.ir. c>l\?,>rnfvntK n ami »\-n r-tsTt.-tvu. event .'r.v.mU
' 'lnil!.'[ii!c-\ H\Ii'k1 lik-cuiv \chtdc K'mii X J*»
' tluo M:itc I iw i VJumhus •
• i Y-;Uvim.\I pnmfirih ur; in 1 J". E 'cpaiinicnt .n I nuigy -:pi>nsiuim» u! slit U'ur yoat uMnpfiilion limji 31' 3"'S" >' >!:l pluvtH*'1 -I"' -I'1' tnW »*''
refwtiretv
Oncrsity AftKitirs
lit UhIMih VSi (tiler I- ¦,•!'< 'Al\ Ic.im li.u I" n< miK l« -i -.up^nt i >1 dn vi hi eiytm-cttiHi ! h c mcniouvt tliuv
SK1 V 11 'KM'.i «iiivl iunUt'J tVH> thcM i>n m .kkt.tU' sWvun tiics ivnducU'J then 'jwi.lu.iW '.Unites .H * v'-H
KM i'HN M.ShKYU'K
vr\ ict to < •overnnttnt asukk-v
h.lW \ ehivic U'.i.wn X'ii-Ucl Kg\ iwnv'l 2i11' >
I '«j !vc\ knw-i
* > ''TiUuctoJ .i In>jr fta'r tii tow v I ,t icn< iiivUe! lor hrtuic \ elmlc l«xhrH'li>try ti: et! m muknii
lie*. 4cu;:i*.>nn So* (ulurc IUcl-Ci.s.i?Kiftl\ iCCubtiuJlN
¦JU:iS taM bM.CTfl'
I V.'i k«.n ic'.vct
* I Vutluck'd .1 J1 reel ic\ w\v •« a siuUic-. ut luluu- • cIik k- kxiinokuv in making p' >!icv Ji'cisioii;. lor
I'unttv liict-i'i oivum rmiliiti. ¦!>.
\ui.im4nv hvliiK^-K. lu'ts H KV' <>"wi :«!N
!'Vc; Kr\ un\ c?
* > "orultK'.teJ a pcof ic\ io\\ f >t\i i -I luUn Ji.vt-.iotN i >?
tuinic lurl->H;Mimnn f'^UMUK'ns
IJoa-d Member 301344
• Work with Department sf"Energy staff, Argoiwe N«tw»»»l Late Sail General Mown; stall, ant! tour usher
tieoCAS faculty a^riicm to unptwe the sfucletil dc v,-iTience for the BcoCAR I*owt am
r-l the IJ-<-arJ >'!'! 'mvtor;
* f-.Icctc»t f Htvcb>r-. -h*i Vi!.n ! iu4U 1 'hi^. 4 n^n-rrofn cl'^mcrKan^oti-itiAn uk*i •
^ iiich n pan of the * S f 'qv.nmeni ,>i i-net^n »'ie.u* ^ pr^gr.im
* Sem**! us S,v?,'Uu\ *ukI nici ^nhc l'\»vunv» ! \i'n«utKv U>t ih-/ fjpunu'-.iijun
L11L
Few: Reviewer 2012
• 'miJiM.-J ,i |v«i iv\ *•« >-t''rii«h ,.f future light-duty vehicle technology used in making policy
tk\ thi'.h;uh\ ,Vlv>vir-1
vC.f.i 2011-12
72 1 P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
• \\ "fL H'tth t wpswfmttH >M' twrcy -.l.it'f Vg.w«* Si»n.w»! I .ih-. iwff. >-»nvral 4a»Y t>w .<(*•!
l;.,o<' \1< iiKuH\ iiiiviHirs u- 'inprfve Itv siuUcnl JeMun expci iowc for the f'Ci« ' \K privnm
hlM i ml'H Hit\ \ clmlc ke\ u",u'r
IV.'I lv"\ UAU'l ' 1 1
• < \»kl(irk\l ,i - Vi Isru; fi-ei to a-w >>! ,i iiuh >>i lulun lr:liMlut;. tvtiiclrf. b>i tho I • S t-J' \ um\I fiii
t'Mi.liny kiitiu' luol ivw ,ir>J yrivnhimy fnov iun f>-au!.rti^ns
'¦'j'WiJ'M \KhJo1
> Vvr i It-hi! s vci »:s •¦>( .t lic,i> v ->kit\ Uu. I m-ulcl >ln ¦r'.ipal h\ t!v I S H -\ n-;c«! i> fKit'iiutt in ul tisK'rmil iiomNi "i^ii
« imfm-nce and St-ssiim Organi/aitim:
'• 1'itl'1 :-ic«.suni • 'ii.iu 2019
• 1 'ig.itii-.ej ,i.w.cii ku mcxk'lmsi 'u.J .mu'Satn-n itMi
h ''' »'iM .3M '• «V1«iw 20IS
• Htitkctl wnl«' 1 i,iii i iv n< • .>hh> a|,H.' I w\ ef-;t;\
journal /1 ttnfrrreri- PuhlitalMin Rcucwer-
1 »r Mnlbm-> Mikr in ,i i--vrwct |.>i tin* !* .umc puhln dh'tr. >\nU'n-ikv-
1 lcl<-inl Mihto! \ <-hkb- i v>.iyn
,? InU'mntKnuil Jouinu1 ul Ivkscismiik
.< S»>uol* <->1 AuMiiOUn < l-.ntHWti'-\WilU 1 V.iyK;'.*-
1 \mci ton :hvicI\ <.•! \kdiunic.il hiKiiiv.1!-. - I .•t'niiinn. M Mvim. jnU 1 \ulml • v'nk'Kikt.
> huuliik- til hlcclra'fti ,mJ i'lccln-utcs ¦ snicrtc.tn 1 \>nl!i'ls ' 'i-nicicn^,
i' Iti.-llluU'm) hlcclr^.tl ,mJ l-jeilti-fii'. • i-ngiiKViN i' unlonnKV sin I >t\i\Km mid ' \-r,lu'l
i liwiituk- oi' Ukcu tc.ii anJ hlcciu-aic-- latemccn 1'iiiiKn.si.inN i.-it AuhmiuCk •!
N lii;4iuilc v I Llccliit.tl mtJ l.kcttv.ih.". - I v«, (V>vw: SyMCiii-
Hrtiltsvhuial Societj Nicniln'iship:
• ASMK
• If"!*]'
» s w
M IIOI IKS! I It"
I >r Midira!i-\!tiliii'r ihihhIuii!". ,hi ;u:tn <. sc v-irch <>mgiTifn rind fsr- Hccn iik>k' ifiiui i In nul!n*i m it'scucH
luixiinj; H? hit;; iji *' • uliUnns jihJ nit h-indc\ol 1 \ i;t !h h>'iai = ' <.i|;itninh.iikl ,m h iiKk", ul M * i;i
->'t 11 S ! (>• Has tnore than I' M i-cei-tov nnvi\l lU'nlcivnoi' puhlic-it'-iru ii-itrri.il pnh'u'iil'uT, urn I mutul I S p.!leni,
lu UhSir.in i it- In-. Ivoi l!l oi < >< I'I ^|n'in nnnli'C'. \ i.i
-iXl'i'I COfliT.UllUl
Kf»»rth lirniitsmiiwitt*
fir Midlam-MoKler was PI or eo-PJ. on the following research iwojecfir
Start Duration Sj»i»»«f ¦ Project Till*
73 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
(Years)
1
^ *
3 :m(»7
A 8
:»^7
1 H
IVnnecc- \uu>rmiivs.!
1 e
UrtoniutiVi-
(jenond Nk.kv^ > 'crp
<¦ t ?i it •'/
i ^
V.sr i-ivrt'v 1 i\h \ jf*>
1 ^
Y?meco Nutom^ivr
4 2*s> 'K
4 t>
< Jummin^ hk'
O -x
\ n
: Vp'ijiltMoiU
i nc?>j\ t k-fK'J'a)
Motors
i
^ i)
r \r i'Ih- n
' 'cnvjiijursi
4 s! )K>
: 3
l-'mv.ir-.'rn i?v
4 2' » 4
.3 4
< 'ummins. fnr
4 ;j«ty
^
Slonci idgc
'/
K3
(IciikcS' 'oip
5 -«>i^
3.H
1."hn.'^lcr iJioup LU."
1 h
i \ \k InUu^u !;il
< 'oriso'liufn
K !
\ u
v.^ncrk^^
">n
11 j
] 'opartntcnr of
l-'i>ertJLH f kMK'ial
\ k'ki! '¦
H >1 ^
_ ts
' 'U 1 runs ^
l'!ivironnn-?ii
2
1 \\t< hkiu,HU nil
^ 'vn^rtuitn
i :»>i >
M %
1 K,1 {4
\DK*i iv-a;.
1 \ S
Ameiican !;iectrk
2 t5
< 'hrrslor < rroup LI,''
2o\4
2 0
Ik'Eldt kJct
Xmoncn.t
J1 H -4
J 11
[loiivb kS: \ 5
A messes
" .0! J
i N
< ',\K' hiduhtt Lii
(''>n^'ihw:p
i •>
<\\[< huhisli in^
f \'r^<-r}Khi>
•' >M-i
¦ 11
«"hn'-dtT > iivup U.t'
)!'':!!Ucuiirte Film kVjjetiefuliM's n nh r.xlcm;il Burner
keuuctau>!cnoi:-iiu>11 lot \< H l-Ieniedwtk'.ii,
nay Hurne! Iy.mokivs (mj i 'ontivl 1 wdvpnu-m
1 k'vel'Wirnt urn I Imp.omt-niii'.Kinot'fi \ldhoJtj)og>
i'l'Wvvs ijrnl fv.*'s k> "n>«enruin
VksJds that l-.ir.il'k* a Mqth-ln < 'ouiml
I-ko1 'AK 5 Uh .mtcvi (oduiiilogy \ d'.ido«\mipt?iit(uii
1 kvt SniJi).-. unit I'tHti-toimei Molding t-1 I'HIAv
I ,(rtjl'i!l > i.vi I icnvc.J • "¦ l-nd i Vi k' Analyst.-.
• .'iimnnns!'|I H hnyiro.- \ ji whim XlruMSK-rm-nts
N<>'•! Sensor ;md Smsoi Ii,\!uiy
• Vmhusiion i lumix'i < Valine hv.iluatH-n
AJ\ .iiK'vJ 1 tfcliin>k'u) IVweilum-, kn
1 'ni-vlmc iutjiim; HiniculuU' Muttci 1 'ontiul
Autr>r*\i!eil Vdnck f I .miil* f .vMV-1, \>st Soivvims
Fiujt-lti kkctnc WliK-k- lUHi An:itv ws, Insight:. and Rqwti-.
\l<\k*!-Bn%\l i 'piiraiffl1)i"n find t Vtnti\M Mahixlokvjv for 'he
i 'cswn oi Clm sUv's \exi <¦ kwrution Knvenrum < Vntrvl
kname Sutuhiht* SmiiiliitK-n, \k\lcbrH? and <*• >iihfl
Modcl-Hused l-ngifu:('ulihrutinn Itvhniquus
nil s ':ip.ihilino-' 1 'cvdopment
Klexikk' l-'fhllHi' U' '( \ Y\ i'k'pHH'Ht
XlixId-t'-nwU !*,iriii-ni,iu-1 iltor itml I k-nlio-I
74 1 P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
ilO
1-!
I ?
'17
>17
'1 ?
S
K
9/2019
6/2020
8/2020
9/2020"
t 1
I s
7S
.5
.75
3
3
i k'pailnR'ry r,s
< ivtici.il
I larlcv-! uv U>on
X KM* >1 l-M'l!'-
Uenewl s.K¦ii-i»
IMiti.t (!.1 H&V-
\nu'nc!i<
HoiiAi IOC I.'
!C,!S
'vh^Olllei
! XA !nc
I liirk*y-< mvkIm'H
] u.fd,i '
.vmcnciis
'ii. \pNuiuc
k-fc! •
America*,
Honda Ri-i °
Str.erieas
1 Kind;! k&l •
M'iei >' U-ncn.l
Mouv,
I 'r'tsJii kiST"
\menui;-;
I Kflvk) R.X! i
Americas
NHTSA-VRTC
I'A'.oi '-\k > XJiHnced I'ivhnol » Y'sitsnllcj
S:numc 'ahbuttfon I .-.in# iugcnvai luHe,-.
\h*\ HiifM' I 'Vmvpi Vehicle kvV:l)
1! > and *!> \k\k*M-n-icd Liv^me I Vm^h FcchnKju«v:
1H) 'i u-.'i I >c\ dopnicnt tu fmprow. VmhiihU^n uoddnu.1
I 'c? ?> h m;*p.vv
'1 r.iilct Sv.'av Win do ,uk! ^
Fuel \ir i >\ tiling U-:- Modeling »ukh Vnipen;*iWt>n lor I'M
l< 1 hncmes
Ir'httk. S\ sUmu 1m -\! ,o< ' \K Mohiluv * 'h.iUenuv
Humun-* Vmnc Moines of ,\f )M< Yehiolr I .'in o vuhImy
> A o| ' if X ' V'nihiNh-.>ri
Evaluation of Open Soiree Stancliwla for tmmmmm Velticte
Tesl Case Creation
Air Foree Reseacb Com'ergeuce Criteria Diwetopnient «xi AppMoition iff SIM
I,ah! Perfuco Seemms
Depirtifte* of Simulation-Driven Design Oplimi aiion at*! Aijlomalion
limfgf; l«,CoKlwooiW«jeletJ KwrnJieaters.
DepMtme* of ; NEXT CAR 2 PrtigBiin
Encngy
Journal Publication* - Publhlinl;
Indwuhials wlwi were *l»ia:d
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
0 /.hunswMi /iiu Shawn Midlam-kMilei Mait-al;,, ','anuvii ] levt'iopinem >>: Hivw- baaed tlim -Hits tankiu
rrU"( mold for rt*akMm<* dotribuk-d u-mpcraUiT*" pK'tliitu*n -ii.um pft'fvr ortboy-Mial dtxonii'oMMon mo
col location" hvenunmiaU'W'iu! oi i'p^nh' Zulo
4 -long. Qiuimuij .~>lwvu Midlnm-Mohlct i'lmtmiHiot.; Sein. \'!tnvn Mfrano. ««l tiioiyo Ki.yr.m "i'l*\
Ahareinji ''onliol ! 'oWiJainu Irirotorroc! 1 ilV and I'Aivniiicril.il \ dkUiori ot .i ^ kYA ! iNrihrtion
Ttim4<-n\ct ' Siihut iTnd !I'"FF l'\tnMcih>n\ r"(it >•! a IXnaimo! •timeline\kidcl lot ;i I'.n tlM Vw
ri!1 \." sai; im 1 -Ut f.xu-i j.h i"4:7< \'ii4-> >14
kuia^opalan S M Mniiam-Xlohk*! Yuri ovioh. J' I'HkiiA K I' • iua.vrauv Y < "i , X' \kn ct I \ t orlTut
dCMiin atkl oiliNiith.ii lodliclion rtfolhiulo'oyv lot \i-lf conttoi in whiic ajwiai"- i'onltol ktiyimetma
I'tacttcc 77 4?-^> "'<114
Hwce, k llyjt,'. A MaUum Alohk-t N , Ri.vom i! V ,h J ;\| r el al 1 .vuyn .if a I-' itailei Saiv.
I'M' v ioi the kcof' \k 7 1 .S-U- InkithiU'i'hil J^uuim" .'j i ;\i\ ,mj I ttfrn<.uiv'\ 7i*j I M ' I <44
at-i in 4:71 :nf:-
>! 1 !«nw 'j Mtdlam-Vohlct X Marmio \ Ki.wni u !V\ t"iui'j.jiik* Imp,ml loihe I 'taiiibftluih Irariak-tmei
] Uc Hi 7 a > M :
M k* ( Mff5^ S MidlamAlohtai V Mafano i» K's* '-'taliwti* al AuakMs t.>r HlkV Vastus! 1 Lvl "itudt."
ktkinaliomtl Joiiitml oi' Whiclo I vs'^n Vol Ot \o-. 7 ' 4 ?'Ih:' iltvl -wJi
Hhci) on Monte 'mmitiMH-H ' 9«iui <>/ !\-fatk no 2. 4 Cul .^i yi<
|n !5u4 IJ\n n4?4XK :>>t;
II I • ' 'otilev. IJ W/.Ai. .S \iull;nii'\!t'hk-i. i> Ki//oni "Motkl |-n>;iiK* i onlt<>! lAnciupmcnt arnlliaitlwar* -
m-the-l oop Teslini: lie the* I-.co'' \K \ik anceil \ fhick' i">nnpeisiiun", :¦> \l; Inlematir-reil Journal .'n
U.I 4 N,, l.pr K.>m.(7ii7. .Mi |
i: Mfva. ,l;«on \. 1 utVovkh Sk-plirn Mullar.i-^Sr.hkT Sha'an 1 \ ,\ktlcH'>aM\l {'siijnaloi lot <"'\hnJa ''pivilV
\ii-to4"uel k,ii!o<\wfiMn>" J: >1 R" M ' * SV«.l fMS \ li: kl-MI:\"l \\r» O i\VK< 1 -
1 k WS.V "I l< :V:' >!¦ I HI* Vit.ll. Vol !« HM ^ .>1)1*1! I
I "> »/oha S \lidlasti-Mofilcr. \ N lasano ' ¦ Ri'/fiu An %kjnuiv, Mm U-v i luiO -\|!| =ioath ka l. :cacraliou'
\ olnt.lo Hi i\ <• < 'yckV SAk liiH-inmiooai k>uii liiVi-l'M^ ^'>11
14 "'I (a S MhllamA'ohlrt I I'i-h A Soliin.ni, "All hI<-|-! I f-aul> « >|i an»l !--• ^1.it!• >n toi a ( vsH
I ..':ai v> 'K \ rar Uaatuu-ii! 'U Ntani A\hHk4 lat^oiof uw haotKC onibci
1 *¦ M 1 'ursoui S Mulhm Molilci *i ¦ Mi.-.vimai' <> ki'/om ""-knii N aim1 Vo-iclim! ,n«l \nakmsol'Ik'»'( I ms> I
kni'liw..« nil i-xk*iiial Mi\liir.-konii.ilKin \''\|t k'urnak'i i '\i> itru S\ ^ McaMiicnu'iH amli 'ooiml \ > i|
r> Kh.'oiii I 'm-otoii. alanJ Kpnim, ittal ins i-Jt^atioiion 1 >u-.o!
11' '< "I i'nitiuMiun «¦ lili I ^k'li'al ''.kxltu.- IVapa'-tnoii Ini.'rnaliaiial journal -4 \Vlnt la I %naiinc^ \ oluiiu' 44,
\oai-: :»m7
IV S >' I.re. 5 l'i«ntM iun» 4
of Matcru!" Sctonco '•.¦ivt-mlvr ^n.n
t'rt r Knh'Httl ConfH'tm-t Fapewi
\|l ol ilia I'olk'as, ing CMt-li'i'i not' j>ul-lk a"ooit-> maliiitaj lac-anti'lions h\ I w MiJaniiA foiilt! or t-ne t.l iha oo-iulhoi'>
I lain idual. w l:o u cu' ad\ isial nseiik-ietl s»|vn (h I >t M.Jlam-Muhka aia Jenolts! \ t.l tsr.-n luflhkijhtiiig
1 Smgh H N iitllaal»\loiil.'i S _V- I ulpuse, I' ( oCI I '• i>iu;l slua; ka-.t'd > ulual It^taii' M Mil'ah' ol \l a\N
ligitrilhirK a.i^* 'iiiklic; S VI; I .v.hnu'.i! 1 'ajvr
: Hnhfit. V lnl|n;ii; I* Ji Vtidlani-Mohiv'i ^ 'iCI, inline kohusi \lf" ' hasad kmoigi'ticj Muiii'menm:
Mem k.! AultHii..mau> \'eliitiv'a \!\.r l'ic|u'rt M.\i\
< I uljMilc i' Midlam-Mohk'f. S KarDmaivhi A . ft Jin. "l OiOli \ Smut,warn K\>l l.^r \ inisjl Validation
atitl WniicaUuii ol \dvano,tvl 1 *>oor XaaMaikt; 1 irohtaoHl i'upet
4 I'atil. M . I a h-at'oi \ Msdl.tHoMohlei N A i A C! i I mh n>^ \Mlia"aahoa 'a-sleia 11"4 S» oaai lo
1V\ i-lopmom I Wex< 'uanioti atid Soi'iu.it.- nt-ih.'-t..v.p Impytnomniaa; \k, I t\.hn'. at I'ap.'i
76 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
wrth Connected and
Shawn MWtatn-Mohter,
¦ '¦ ¦¦"¦!¦¦ ¦ ,• ¦ ¦¦¦¦..¦>, "MocW-BasccI Design of a Hybrid Powertrst
: "" i : r.i. ¦ •• w Fuel Economy Improvements". SAE 20204)1-1458.
Stiifclfirt, .E, :C!»Mb, SL and Midtetn-MohJer. S, "System Engineering of an Aclfntweci Driver Assistance
System," SAE Technical Paper 2019-01-0876. 2019. {Student Presented si Conference]
f»*, s., IfSwttrt, M> Kmrnm T. .
7.
S
?».
Id,
it.
12
13.
14.
15.
10.
, Andre# iMrcia, Shawn MiJlani-MoMer, and Cnoigio Rizitotti, Plmtt Modeling and i
Urifk'atian far s Plug-in Hybrid Ekelm i-thtde m the Ik-oCAH 2 Competition. "Ho 20154) 1-12
Technical Paper, 3015, (Student Presented at: Conference)
17, Hegde, BfcumtkMiaif, Shawn Mklfaa-Mohler, and Pwiil J, Tulpule. "Itierawl Model of Fuse Dylswiies for
Simuliiiion Under Tntermfuent DC Fwifs.* In ASUE 2015 Dynamic Systems mid Control Confirmee, pp.
VlB2T34AC*)8-VW)2"i"MA.Ml»B. American Society of Mechwictl Mngmccis. 3)15. [ htudoni ftt
Conference]
1 g. Rtmwi, 0., Mjiliara-MohlBr, S.» ¥*4 M.;.Ttt4<>, if.J. il'eil-B-wkivl mafyxis mtlmeasurement of
tmrgy w$ ami greenhouse gas ami erifmiu missions in a Plug-in Hybrid fishiek; "ifc EeoCM 3 erne stut^'.
2013.) Student. Presented at Conference 1
19. !«»>*•,. A, Ystsi®, ll,» Villi,. M, ftpniwil, SI,, Cr«ll«>.;l... -t Miclhitt-Mohler, S, W„ "Ref'inement
ofaPiimllel-ScriesPHBV Cor Year3ortheEcoCAR2Competitien" SAE Tecbnioall,apcr2lH4-,01-29tS, 201.4
| Student Presented at Conference J
20. Alley, Robert lease, Patrick W»tsh, Nicole lambiase, Bran Bciioy, Kristea tie I,a Rom, nougteWelsBn, Shawn
Midkai-Moliler, J«iy K«fc and Bran f tbien, "ESS 'Design Process 0»ew»w and Key Outeotnes oi" Y«ir Two
Plugging in to tl* future." SAE Technical Wiper, 2014.
Sfuiwti iVltcllam-MoWer, and. Qtofgio liizzoni, "De¥«l(|inient of a Dynatatc Drivelm# Model for
' Paper. 2014, (Student to-sented at Conference]
l,,.0®«i»r A„ Qrgi»iw«aik, M„ Midtam-Mohter, S, W. St Eiami, G,,
for the BecCAR 2 Cotiipelilion" ,SAB Technical, ftiper 21)13-111-2491,
2013. f Student Presented at CeafercnceJ
22. Clfetig, Q., S, Midhitt.MnHer, 1"!. Serm. V. Afara.no. ami Ct. S izaro "PEV charjiiria control for a p»rk mg M based
on queuing theory" In Amriam Control Cmfiumee iACO, 2013. pp. 1124-1129 LEHlI 2013. fStadent
24. Me*#n Ibsoh,Stephen Yurktwich, and Shawn Mtillam-MoMer, "Air-to-Fuel Rniio Switciwng fwtjuency Control
for CiiMftlBw Hngines.*" In Coord Systems T*eimolm\ {BEE Tmmetian «t 21:636.48, 2013, [Stwlcnt
lsrcsent«!d at Coiiiariiix]
SAIi
77 1 P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
. and Shawn MUIaiii-M»liIer, "A Physicattv-Bsscd. Lumped-Parameter Model of an. Electricall}'-
- " SAE Technical Pupa; 2012. (Student I
Shawn Midlam-Mohter. Gtwgjo Risseoni, !
et at, "Design of a Parallel-Series PUBV for the BcoCAR ;
Presented at C
Competition," " SAE Technical
Shawn Midlam-Mohler. ami Giorgio fttzzMU, "Rapid Vehicle Architecture Selection With
I %e of Autonomic." In ASME 2012 Sih Annual Dynamic I
J012 nth Motim ami mmt,rn Cafitmc*. U9-28.
' " : at Conference]
hawn Midbm-Mohler, Vmemm Miwimi, ami CJiorgio Risaiw. -Study of PEV Charging on
: DistributionTransformer Life." In Smart Otid IEEE Transactions on, 3:404 41 3012.
29. fliiftg, «J, 3. Micllam-Mohfcr, V. Mm and&. Know, "Distribution of PEV Charging Resources to BuSmm
• 1 .»fe. and Customer Satisfaction" to likctnc Vthielt Conftmtet limv}, 3012 l£KK International
[Student Presented « Conference!
PEV
Internal
ingines; Feasibility Study mi ait Organic Rai&ine Cycle With Application io the Ohio State BcvCM
PflllV " In mm 20J3 Interna! Combustion Engine Division Fall Technical Cmtfermtx. 608*13, American
Society of Mechanical Engineers, 3)12.
32. S. Midlsm-MeWer. V. Marano, G. Kizzoai, "Optimal Control of PEV Charging Bated m Rimtenlw!
Base Load Prediction", ASMS Dynamic Systems awl Control Conference (DSCC), 2D J1 (Student Presented al
Conference)
33. I ||e#sr, S, Midkm-MoMor, S. Ywkovich, "fa-cylinder Oxygen. Concentration Estimation for Diesel Engines
Via Transport Delay", American Control Conference, 201 i. [Student Presented at Conference)
34. E. follen. M. Unora, & Midlam-Mohter, Y Quezennec, 0. Rizzoni, B. Tee, 0. Matthews, "A High fidelity
Uimped-tferam eter fingti* Model for f'owertisin Control Design and Validation.", ASMS Dynamic Systems and
Control Conference, Cambridge MA, United States, 2011. [Student Resented at Conference)
35 fffmlif, IMlife -Vem. Shawn Midlain-MoHet, and Giorgio Rurnmi "Model Based Engine Control
Development and Hwdware-in-the-Loop Testing for the JieoCAR Advanced Vehicle Competition." SAE
Technical Paper, 2011. [Student Presented at Conference)
36. liottg, CJiuBjIng, Shawn Midhm-MMtJer, Ywmm Marano. anil Giorgio Rizami. "An iterative Markov Cham
Approach for Generating Vehicle Driving Cycles." SAI: Technical Paper. 201.1.
37. cSongS. MMlam-Mohlcr, V. Marano, and G. HtzactM, "PEV Charging Impact on Residential Distribution
Transformer Life." in Emrgyiick 2011IEEE, 1-6, IEEE, 2011. [Student Presented at Conference]
38. SWfciQ., Tulpule, V Marano, S. Midlaro-MoMer, and G. Rkhwb. "The Role of ITS m PUBV Vm&mmm
Improvement." \n American Central Conference (MX), MIL 2119-24.1BEE, 2011,
3ft Msiraito, V„ P, Tulpule, Q/Ooog, A„Alirtiii«, 8. Mullatit-Mohfcr, and G, Rixmrn. "VeWcIc filcetriitcation;
liupliCBtiom cm OertBtitjeitt twtl Disirtiiultofi Network," In Bmririmt Machines mtl Systems (JCEMSl 20}}
Menmtkwml Confermm en 1 «, HIRE. 2011
40. jMlitfec,. IfSsrt, Shawn MtS-7tt2. American Society of Mechanical lUigineers. 2010
43. filming, Shawn MidliBB-MoMet. Viuecnzo Mitral*), Giorgio lltzsjiti, md Tann (luezwwe. "Statistical
Anslvm of Phev Fleet Data,'* In FeMch Pmmr ami Pwpitesw Ctmftmm-t (JVC), 2019 IEEE, 1 -6, IEEE,
201 o:
44. Meyer, Jason. Stephen Yurfewick, and Shawn Miclhiti-Mohler "Architectures for Phase Variation Compensation
in AFR, Control.™ In .Autos Cm/ml Conftnmx (dCCh 20111 1447-52. IEEE. 3)10.
78 1 P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
4" fv Mjiiifigitnii' S MklfomAMil^i \f h,wg b .1 ol a' 1i 'i 1-ngmv 0Aj|< s )S>" < NcpH=nihcr '<<°
4'. i Mcwi S kMjiigopnlar 5» MidlamAlohkr » kkvcnncc, .A Yurk-.n ich 'Apphcaliou ut'nn 1 - (k'omclrv
I'Vist'i) I vlii% PVdiclti'n Mi\kl tonn sntcnrd *VmhuMK-n Pnunv.\ " \^MP PS*A"?hkr. K ManngmU I • 'Kv»«-i M < \in- a a "> 'pumal I VrlArmanrc \«'< Scpievnher
4K S Mtdiam Alohkr, P M;u.*no S f-w mo. if ' A Ki/auh TUkY f-kvt Haia i ktikvih^ and \nah
JHkk VH* "«iv. Janw
4° 1 Headings A \Va,.hmi!!^ri s \ t lalnm -N fobicr i } Utlwhk I k,nmv!;\.i'ru: io\wr < kni-faiion for HvPpd-
I• Jx'v!i w Wk.-k \u\5U-nv Pov.rt Pt.v 'M\b In? • Vm^two ^ri v>m;ui Smirtwesaml MuU'imk \ 12,H> \A
n ?.r*iQ
I Ikadm^ *' i W^hmyu-i? S Mid'am Ak >likn and 1 !kirntair> !h;_ih vmpciautk nmlu ~us?h^i-
P^nu'Tw! tliv-nnucl^s Iru' aa\diaf\ p-aua uivf Pu\ ceding M v\| A Ms Mi ^ \S". ip, mxi*' < Mnkscn-.t' <*te> nnJ Ints I i hi,, tit :M •-kin-' < >\r>nu[, * ' \ i \ \V'A If
* 1 M S Xlullan? Ufhloi, • f Kt/rt'iH I* Sicuni-; h lU^ina \1 hajp^SkK1 \ Santibhon SUuh olanrK^
{•Jtv-Mnc -\\'\ loi a M.-ru"- ihhnU I IcUtso \chsv,k' '-in Mun^ait ^nicnahnuai Sunp-,isiunt uh AiiU-caoJfi o an u'liiia-r. ^ ,,-t
*0 K kopHii^M. .\ 1V//15H H IV.'huv,' k <' \vics i' ! ulpuft Ki/.-oin V < luc.vunoc S Ihih ur kcM^h. \ ('thwiun-i k fnult k ^ "onlful • M ^nk'J MoJiMmji ot a
I hrcv \\'a> '."auk "t ' 'ouplcvi ^v ith < f\\ ucf '^cn^or- ASM!: I a sU-i> and « \f 'ontVuTkv. Air?
srk•? Ml ?JMK
¦»-) I Ik'iiihnes S U WaKbm^ian and I P Hci«*m.iiN MjuI: l'empc«;Uuu- ihct«hvk\iiic
Ausdin?) k'Mvcri ml i-xr Au!'M«>>i|vt' AppI'c.iU- "is \S\if ! \-nk'Tcncc • >n ^n^srl Maknalv Aviapt>» SiuKlnn*.
and iniclhpenl '•^ssoms Paper'kin
k \\ Amclt k kopiukj-a. >' t \,huin \\ Shafciha-!iciet k P-a\a? < i K\.ww \ nuc/r-uia-t >'
Mid}anj-\l"nk{ ' loancr 1 '>od t nii^ \l^tcf-P-avJ he^n'u .snJ \Jv iikcd \lK!lu.atmcnt. S \f Mufs.i'l-
I'SiVS NiftX Ink'ir-aiivfus! < Vnyros, ! 'oiran Ml \pn!
>>-¦< ]•-. i »udcL H \U..'snta??v wcu^fv? Z-l > .n»K'?
^'' M < ani'V.t N Midlam ^ l^hk'f ' inc.'rpno>. \ '-'ohman tj-Hlai}np rw A
AlVrtr^atincnt M --umx " s \f' U"'| "i^7 • Vn^?vn<«. « <-*,pn It'-slv ^'pkr,ike^ ?i 'i>7
>S M Taih^a I' <'1iirirn ! i'tr*\£jii s slhlkni-\k4iiti \ Um'vnnr«- m Ki.k'^ni !'.\|'hmmn-ntai < 'hatac^'ij/amut
^1 s !»\cd-Modc H( ** 1 f •[ <-AunhuMu'ti or? a * 'ontm^n kail kn-s.d litiino ^''' lnu-u; iJk'n.s' * ViiMcmv or.
P'n^mos Ik? \ulomotiili' > "apn hak
>,J M b j1ua?a M >-k>ry S Mnllai)i-\k?hkM 'i < kj«v vmuw ki/vuru ?'AjV!in!C!5!;d ^ "huiaivn.'uikMi
^i' \h\.\l M' ie H' '1 I t '«,»r"I^K'n"a <.-ii .<' ¦ -muu-n Karl i-namc vihmU!' dioS.\h!'V'^?i k'nU"r-.N«vi-
i 'apn Hal) S^pU?ndvr !u\i?
M 1 'anovu. M Y\vi\ \ »v* \'^«ii'a'n •M^hlo! H Kk.'.ohi and!' kliiam I }\nni]tu> amh \muv\ o\
1 >x ;«nd }k < '1 i \'mhusiion in a rnultt-v \ hs^lrr ! 'jcm-I >'?yu\- i'ap^r 44 sukmllcJ lu :-lb ll'W* * ^kmptisiun) on
¦\d\ ar\cvs m Xulomonvt. ^ Antral Paiar^ F'-urie'; >A'a^capo '" \ Au^sixt 2?SU7
• H vl Nnitctiiitg usvgcn Scfj-a>i
Pt'b:i\ h-j ikicU'l \hau^l • - ts iik'd'« a k- Papet Pvli" I''iM-ia'H "*• kluciiio I! ki.^fmivs
2i n
'0 N Midl.unAU.hk't ind 'i * ?u.\*vnrvv. " \ kaitpcrak»i"'k:wd i -.\hnupc tw iomp>r xl
I can *\ hap 1 'alakst N Xk'aNUU'nu-als " F'ivc at I'.ilk '! M;* f\ip?^. I, 'kvas^bi W OhwHi -ai-d b 1 "hja: i.
P'.pt'isHHTiial \ahdaiton k-r < Vnif^k ;ficra^d M<\klinu o! MfHHa Ahn«K*r ih'i 1 I knt^rif." Vwt
IMi-i'P he. i'apoi IMP' 1-. 5 MP; < 'hscapv I! '\in oftthcf
"4 \ AMtkf -,r /uu ^ :
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
• / /V-u S Mkilam Ak>hk-t k \mranuda! '« Swkiamamam 'i lioiwuio Suivv<» o! f
H\ virv^-f ' K-nrrnlrt'.n \ hMhinJ* k
'*> >' Viuli.unAMiK'J and \ muca-ihk^. Kct.A*»u*Mt!^n *\wltof tor w hyjSKs-^vucik'iation lean N'!\ has>
storp." Xmcncvtn f 'v^nlii 11 Vnh'rcjvv • «\ > fmn«'.tpoli*. M\. innU'd pipci. lunc
tollman I 1 hoi S 'hlci Y »jikventiot ^ R\/Axnt Modeling and f>uigno-.n':s v>i* "w >k MWr-
'! teal merit Suicins " N.\k kupoi >nuw>*a:?us ¦>«:<> miomatH-nai < Viiuu-ns I uiumi Ml '\pni
% Mklban Ak>kk*? and \ tiucv.-nna "I V^icn,\KmHmw: an.l \ uhdanon i.'l a Hamv IscU'Imkm kn 1.1". ! I:\Uin il
I a-hiv, kt'L'twr^tu'ri " n I'-iper i*»»7 loik* < W Ink'tnaiionhh Ivu.mI \|| \pni
'** S Midk:i!i-MohkM. and s, i lik'.vnniv Akkkhn^ ,a m j\5n^l r!,>\\ [ V«vl !'up ^ .icm kuv wt
IM! I 'k »*\ JNpor I \ II < 'j, ?n, J-..NJ . Mid{am MohSer, '< 1 Hv.VHiVi' ;ux!p» ku'A>n» \ ~ Vnttukwuehted
\U-.m \ .,h»r Mode! >?l H»'' 'I i kn^iuvs n < k\k'mal <.h\ttu«4 K-mialion I'rvc •«! IM:,('k INprt
j \ j! i • _kn /) t 'i hifiJo ! i
71 \ Ik>hman P l;u'l;rtun k Mkihmi MoMm ^ * kkwitt w* mUu Kl-v^m ! o!;» \\ \iicaio
Sv si cm |« 'f- ?tfi ItucTnjhon.ii • -.'r^<-rciKo ^ -fs t-rfiMivs i^f b>k"s, ' 'apn Hah Scpu*mb<-' *' *s
\f ' ":un.>\ 'a i \ i,if/ato)hi N ! 1 iliis'.'hi \uJluos-N, \ wLK'/sfnni't atk!11 k'tz/wHi ' A \loan-\ra!uf' \k'Ji'i
• >( a hitK> < Hiuuh' u ish kxtoinal \{!\!Uh* i-nimaisofi.' V'\- 7ih InUTinJMini
! '.^nhMonrt' on u>i < ,?pn lia'v S.*pi.;:nh-r
7^ M i anovii k >ki?cm S vfu!Inm-\MhkT ^ >1 uo xnncc and s \ in Pri'vf-s u» Ih'- 'f 1 ^cst-i I'jvzino. " \: ki.'/w. ' Sufc?\isorv «'^nu._>i i\>r vJf \\
kt-ducis^n of an HkVu 'th a \ StseJA k>dc I ^ \ 1 >, "?r;l 1 'r.L'tn-j \nk,fK un 1 Vnifl > \m!cicm.v f'onhind.' )k
June 2mi5,
>''' M i'amna A I^umKmkl K kksfv.u\ s \]i.lf.tm-\t»-.hlor \ - ku'/cnncv' anU >} Ri/zoii!. "\k\k'l an-,1
kxpcfimonis >>f ~> 1 ucl ! i? "i 'I ' v^mki-a?<>n n nh kruemu5 x.1?\turo I'-waiion ^ih SiuiUTari InknMh<\ uuh i'Motnai \k\UMo k'oj»:!<;iucu Wi\ u>i i Oom4\'^v/. Proc l-'iSi! \ '-M \\V?kl 1 NnwriN-
haru !• -na Spam. \iu ^ uM
77 ^ ' iiKVt*n:iL\ . tv \lu-v4iiv.in'!v.,ii,r k\l \Vsp.i'acn. \
'hk^/enjitv I- C 'fiK-iiiaciu'r a;uf u ! hr ^!'C Mw S'.at-* lnixersii\ i'mun-FiurK i he
! kk'kl 1" h hi( «> ii-\\: kiU^nakonal ' .> ;n. ! -VUv-js ^ \\ N tu<, 't
V ¦, iuiwnrKT v. MkHamAj.vhhn \\ 1 Hk'fi4 .md >.k M'.'pka \ _-,,laa^e Vp^rouch k> ) ac's«H kmis^k n
Manapcnii'iU in Du^vl W\h h\ l»kv>ni- \ ehk-IrNJ '''iiv 2!,|C II- V Maorttu: Ikikvli'fiiL Spain. .kiK ?>'f0
Patents:
in mkiikon k- lin- loHowiiie p-aionk- i -j \hilhmi -Mix!]ki !ia^ a nnmiv; ^1 it ado scouis nt u:v in sndiMiy uruund >. ai'iio5
aigi'f Ilfjni:- o! pt-H's^isi'!, ilkllHtfy i'liv'll »ls >t K< pa(i*fi' (ik'M' a1- it ilk'"- il'i'M.'! pii«i''1loH iXH'aii SO •'! the
diilk'ulh :n enlarcenu'm
I S Mkllnni-M-^ik'i. I \k-wi S u'La ui\ \" saijun * \M?iku-:kO!i r.inih-koi kit MjU'inal«'smhusikm ^ninn.;'
! IS I 'aUMil li ^V^k'Nj issUCv! \l.»\ -n}
k>M;uopalaii Nij S\ Kcnfk"-lh !' ! Hidd S^ir-an ki-t k>tcprx:n ^ iji N!um n W \iidi.i?^-\k4;k'» > itm i)
(."uo/onnoi ahd > Iran 15n "I \v\ csal a!i Hic1 u^ulak^ kM (n5cmal > ^'illiu:-n OM^iuen " i S ka!«'n!
S ^ M , ?N ¦* h '^K'r T'1
f '• h-s«Vt. S M'dlain-Mohk'f L i 'lidck S '> n.h ^ i l»iCAsM?iCs,' ! tiH i-onir^.i %\s!c!n and i%acth»Mj f>
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
5. .1 XSeyet S Mklbtv-Mottle; k 1 \tdek >' s utkot icb s oie.'eniiec ! Vtiv y iilibiatton syatems ;ti>d methods"
1 S Patent H,2oS,xsS. .m anted September i I. 2>'I2
6. S kajastolnlart J Mevet S Midlam-Mohlet K ! >udek, S Yurkovteh Y > Httvenrec i "onlv.l s\stems and
met hints usiny geometry hused e\hnu-.t nmiity fmxiel" I S katem X.224 ss7 awarded luh 17 20)2
7. f Meu't S \lklI,io!-slohk'r K I 'udek. S \urkoueh \ (uie/enney, "fuel control stsu-m and method tor
trnpryived resporvie to feedback from an exhaust s\ stem" 1'S f'ateni X. 1 S(\ Vns auardedMas 2'\2tt!2
8. I \K*\. S Midiam 'viohlei K I HkK'I S "i hi kov ion, *) uao.vmnv IVI,i\ oompuisution st, stems ami
methods' I'SS'atentX tl\IS? a« auied 2 12 2< < 12
9. V Hti K. IXidek S Midlam-Mohler Y t Hie 'ennec S YurkyVtch I Wt^'ms S\stem and method lot
defcprimmj a cam.shalt position m a uihahk\ai\e tnmoi.: engine .1 - Patent S :mi> 271 awatded I I ? 2>112
10. s Iau. K 1 >tidek S k uacoplan S ^ uikov kh \ llti \ i-ue'enikv S Mkltam-Mohler i >tt-line eai illation
ot um\ etsal Hacking ait met utkoxteli *i tin, t i uistn.i s\ stem loi interna! combustion enpmes 2 \ :¦> I'atcai ' "17 2i1,1 ,rx adod •> 1 2m|I
12. X Mkllam Vohlei I'. Masteison tut «"omntllm;.' V >\ Hitts-sioits iltitrt" Rest-tits i>l fhbtid and
i '• >m cnlkikal Whales. t >• I'atent 7 _S ] 4li5 .maided j Mi""'
I i X \ 1 .klljuit-\kns aj'k'i 1'nel i'(it-> iff M* enK " I'S 1'atenl
7.t>s|AI4 ;mank>! ^ i>i i>o
intellectual Property Royalties:
1 he tolkny trig In c te\earih pr^'Jikts mvoh eil io>alroent< to i 'SI tot cNeluw e rights to II' Irum the tescaieh
prokLt None ol these projvet- lesulkk! «! patents instead, the company iwes the if as a trade secret
I Model-Kused optimization and sontrol methodology lor the design ol'< 'hr\ slei s ne.-.t generatior! pow eitnim
eomiol systems, 12 (U 2t' 13 • 11 3'1 2"i.\ < V-I'I
2, Model-Basesi i iasolitie l*aitieulate I iltei System i 'esii>n. > Vntioi. \nd I >uu»oMsi!7 ill 2!i|4 - H7 j s 2>tl'>,
Sole PI
3. Engine Starlahilit} StnntSaiion, Modelmi: And i "onttol o'« o| 2>'i|4 - "2 2S 2n| ? Sole 1-1
4. Engine Calibtsuiviii I'smgiiigctwajubks, ll', li,2l'l4 • in 1 s 2'>it\ I vad '.'I
5, Model-Based Heat Has Sensor Development, C©/01/2u] 7 - 12 »! 2»IS I end 11
Applied R&D Projects .lutlged In Jurird C#B»petittons:
The following table contains awards earned by? the HcoCAR team which Dr. K.4kttam«MoMer advises. r;ntr\ m>o the
competition is competitive artd incltided maity well-regrsdcd engineering schools.. Awards $m decided h\ eishe;
quanlitatite e¥8luation of vehicle perfonntnee or via qtialitative assessment by panels of-6 to -!2 expert* from
industry, the Department of Bncigy, and Aigonne National Lab.
EcoCAR Mobility Challenge 2020
Competition Sjaorisors: U.S. Depurtinent. ofEiiergy, Geneistl Motors, artel The Mathworks
1. 1 * Place Subsystem Design Report
2. Best Final Teclmieal Report
3. Best Human-Machme Interface nrtd User Interface Video
4. Best faxeeutical Plan
5. He*« Impart Report
6. Women in ^'1 HA! \ward t»i I earn Member kttsttna Kuu.ilvtttt
7 \o!e I tue tot\n id-1" thete was no o\ etui! n mnet in this i omjxHilion vear
[''eo'." \R Nk'hililv I'hallenne 2019
< 'oitipesttion .spvinsns I S I icparttiienl of .Energy, General Motors, and The Mathworks
K I st i'fke < 'x eiall
" !s| place I aa'et Market t'lesentation
||" r't plaei i 'otrttor. and S\stems Modelitiu & Simulation t'tesentalion
II 2nd place < "oonevted and Automated \ ehic'e SssWnv. ktcse.tiatiott
12 >id pike I'lopulaion System 11110!*,ratior 1'iesentation
81 1 P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
!-» H<-\i (-)lu! 1hiik'il
2nd rimv in ' Vjm>vs.\l ml Vehicle ] VhvciahU^
i > 2nd (I'-iue \SV fNiellenoe in < Vr.neou.vl and \uk'm.iu\l \ ehicltA
I" I II I'lw SSI i )i\ t |M<\ 1)1 I'TWill'Vimi'
ikiil'iti.!! iauslflia?
:n!.'
' 'or,i|VtiH,»n .Spniv.rt;i I S i VpititnHiiK'I t-rios-.toil' n'tlt'rul Mw.iis
t 7 IM lift.1' i\cmII
IS 1 ;.;nI I mi'.sioi)-.. I'ven! IVtS.>rm.m>c
I1' \S>- 1 InmnulK'i) kfi'-caKh I'.ijic.-.
2>< i <>!
.'I I >.( 1 R!' IVi' \\ IkvI ell IVtutieum I'ni'igj I
22 Km ;\ci ihi ku.iJ l.ven!
2} Us JJJP.M '11 1'iiiheJJi'U Sikccw, A\\;t;tt
2-* ' Kama! \K'U'r.s HAraen i;i I nuiiMVimu lv><4 }-¦ A.uutl
^ 'VjM' I 'A t'fMh 11* Nymc.'nm* Av*-I'u
2! - I'll \ tilklt I kt|Mt
2 I h.t : \ mtlfMi .\|i|X*,ll
2 k l-.: !miv>*alKH) I'n-sonl-iiH'n
2" 2nd (Vntrol ami SM.s hesenutu-n
>, 2nd \!cch«tiiCi>S I'ics.'tu.itK'n
31 2ndM;it!nu>rk > UhJcIiiij Vw.tid
.52. JtJ ADAS I'"n:M!!)tiu«ai
-l,! •'
•' Vmpclilion Sponsors 11 I Vp-irlr.icnl ol hnvt^x ,uu! .jcncr.U Wmv
> * 1 Iveiaih Vm|vtitint> K! Ilnu
W - tvcmlh'rc.H-ci M.in.tv'cnii'nt i -J I'];ky
^ \ oliK li" 1 vsijin Kep>"fl ! ">i
ii' i Mice! StaiUi h.tPbKt'
*? Ima-s alien ^ !|t'v,i<1.i!soi5 >|J l'!,uv
i,s t Vi«um>-r \pjvii| l.u*nl 2niJl'im'
(>/ |.i 'j, lYrAvni.'iUni',. IM I '111' >
-*i^ N lecb.'iiiK al Hiru nuiii'ii 2nd I'Lkt.
li l-.i-r-t • I R\ i ;H-rgs VintimpSh'.n :•:» i'ia.'c
J"! LvSTii ' I !-'H W I \V < 1-mHvK.ns, Kt flail'
41 l-XI-i 'I t A \\ l"i\ I. 11m Ithishhvi:* U I'I.vl-
4-i KV:I:'" I K\ U r,\ I'M . is) t'Uv
< ¦.Miipotitiivi SpoiKcrr l1^ I vparmvtmM",inJ i"r,T„*r,)i '>l>'.ivT>
-l"> i ¦>. erall > \»npv*iition Kl i'laie
4'. < !>'eriill f'riiK'fl Managemi-nl 2nd f'laco
•>? i'..niipj'iiiul
-iX 1'.-limit.". rie'vmnli>>ti Is) ("lu.v
-i'' 1 !i"ctr»A'nM 'ii-NcnhtiHtR I -a Pl;k e
I'in.ii I'l't'hnu-al K'i"p.irl, 'ru! ''I:«v
•1 \!i'ihtiirn"iii I'lcvntiiioiv 2ihI lJ!in\
'2 \ cliK-ti- I Vnvm bl I'l.io-
Vehicle I h-;kti I1 Flu.','
^ WW lilt kevien t-,l ''I.kv
Iv.a"1 \i< ^ .'.iwik'UiuMi 2015
> 'ompelHio.i SpnmoiN 1 S I •cpaiimoiit ol I'Ui'il") aiiil ii,*iu'iui
2016
821P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
!sl H;kv C twiall
Best Kmal Sin!.chakki Stains '"tesenuition
^7 Host WmUr UVrUhi'p Innwatkin t»>pie ke\ lew
v"i Best limi!V;itH>n I'tescntatior
He-,l \Vnlifl Ssms IVsentntt.»n
r>0 IV ;l I rade Show hescmution
it! SMS he'ienlntkifi and I >emt>mlraihtn
<0 Bom t'omtimn Market Re;,esu> h ke|V!t
• ?rvl Hace 1'ifiCet Management
(4 dST \t T, l-mlvdded Sueee \uaid "Ykl I lace
2014
iVmp<'tilit>n ,S]»itiisi-is l S I ft ami1 ;eneta!
(>!' Nl t'l.HV > n ci.ill
c/ l.uwest IVtroleum kiiem) kse
i-s 1 est < 'iitcila hmi.'-Mons
(•"> IVm I'lniil I eUmiutl Kcfvit
7n IkM M iti< i'uriMinier ."uxepUbiliu
71 Iie.il i "frrtifis I'rcsentututn
7; HiM I Jeeti icat I't cst'iitatKHi
1 icst Ptogtev. kv|vrts>
74 I-.r \S HI KMeltfiitv V.MirJ. I si Pl.it.o
7s dST,V.*l- fmheddcd Su<.i.ei.,s \w ard I si 1'laee
?(> MathWfiks Modeling Aw .lid 2nd 1'lace
Pcf^Ak^AVmpffltMl 2013
i Vm|vutn>ji S|\tnscis 11 S I VpattitterJt ft li'iciuv and < Icncial Motors
7' Vd Hare overall
7K fkst!>ma! I eclmit. a! stepfit
?! Host I lettilea! I'leser.taiifit
St) Best I'repress kepftis
K| Women in )¦ ngmeermi! A v. aid
N2 Math\\fil.s Modclme \uaid 2nd Place
i itK'ji Sponsors l S I k-pailinent fikneisn and > lanei il \lot»rs
S3 ;,ttd tltkv Ou-iall
K4 Ivxi \\ inlet Workshop Controller I 111 Haluutton 2nd I lace
N* He\t !'io|eet Initiation \ppnn al iV'sematien, 2nd Place
S>- Best; Vntiols I•resenfauoti, oid Dace
S7 1 'esl Final t'ontroller 1111, H uluataon lM 1'lat.e
Sfi Be-,! Trade Show V\ aluation 2nd I lace
M'1 dSl'Vi'k hmhedded Success \waids 1-4 I'lace
mi MathUoiks Modeling \\und Kt I'luee
"1 lUimen m iht \\ inner •;»Vole 1 tmnibiti> and Ucneral Motors
0 2nd iluce i iuuII
'M (-esl i 'ontio's kn\entaluni
"4 Kre>w ale linn \ aiion Award
,;5 I lie Math Wot ks MoJeliifj Asaid. 2nd ilkc
•;o dSlW *1: l-anhedded Suece-a. A«aid, 2iw Place
20! 2
83 1 P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
97. B\ >S< "if 1 nvoistlv if I-wiihv)ir>t» Vi >h< Kl P1,i. e
' 'AH ; \impoiilion
2010
Competition Sponsors t' S Department of Fnerav and General Motors
"N 'Hii I'i.K-e i ,n t'sat!
1-'! 1111,1 valuation Hum;, 1st 11'u.e
!on 1 >ymimie Consuinot Acceptability Kt l",>ee
IDl 1*'icckk'J Success .\w;u«l 1st S'Uv
UM I H)'A<,"!1 I in l-ngnuviinc \uuk1. *td 11,ue
h't Women mthe \\ inner M'mk1 i-ouml.itton W^men in 1 .tmmcei ing Awtrcl
Competition Spomors I' S 1 H'p.it tmont ofFnetuv ami i U'tuth! \U«t>rs
II if', 1st I'I.'hv (iverull
1 i>7 Best \\ .iliei! 1 \sign Report
li'S (HI, hvalumion H\ent 1st 1'hk.e
im> Controls l-\ent iVseirtatton, Is! I:I;kv
HO Best 1 rialc Slum IVpln) tmJ hoseiituiioii
III Post "1 echnital Success .Story
i 12 Freeseale Silicon on the Move Award Vu Place
111 <1SI> \CH l-4iiIwUeJ Sikvosx \w:iid Is! Mace
114 HOSCH IVorsiiy in hrtgmeeinig Waul 1st Place
'AN i Vmnetilipn
2009
84 1 P a g e
-------
^feuropass
Curriculum Vitae
Oscar Delgado Neira
personal information Oscar Francisco Delgado Neira
pi
' V
9 1500 K St NW, Suite 650, Washington, DC, 20005, USA
V. 1-202-407-8341 Q 304-288-1398
® Oscar@theicct.ora
@ http://www.theicct. org/staff/oscar-delgado
Sex Male | Date of birth 13/09/19791 Nationality Colombian
JOB APPLIED FOR
POSITION
PREFERRED JOB
STUDIES APPLIED FOR
WORK EXPERIENCE
From 2020 - to present
Zero-emission Fleets Center Manager and Latin America Lead
• Co-lead of the Zero-Emission Bus Rapid-deployment Accelerator (ZEBRA) in Latin America
• Coordinate the Fleets Center, which concentrates ICCT's expertise, data, and tools to
support Global commercial fleets in their transition to zero-emission vehicles.
• Support policymakers on the development of heavy-duty vehicle efficiency and emissions
standards.
Business or sector Research, non-profit
From 2016-to 2020
Senior Researcher
International Council on Clean Transportation, Vtoshington, DC. vww.theicct.org
• Project manager and/or principal investigator in heavy-duty vehicle efficiency technology
related projects in the U.S, EU, China, India, and Latin America.
• Management of green freight and technology verification projects.
• Development of ICCT's vehicle simulation modelling capabilities.
Business or sector Research, non-profit.
Researcher
International Council on Clean Transportation, Wishington, DC. vww.theicct.org
• Assessment of the potential for energy savings and GHG emissions reductions from heavy-
duty vehicle efficiency technologies through modelling and simulation.
From 2013 - to 2015 * Analysis of the cost-effectiveness of heavy-duty vehicle efficiency technologies
• Assessment of regulatory options for the second phase of US GHG heavy duty regulations.
• Analysis of potential scenarios on methane emissions and policy recommendations for
heavy-duty natural gas vehicles
• Green freight and technology verification programs
Business or sector Research, non-profit.
© European Union, 2002-20131 http://europass.cedefop.europa.eu
Page 1/5
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
m ¥ti3<5
Vteiting Inductor
! . I ( 1»* ' i I '!, i - ( ''!((. ! _iv i11 I i 11 ' " ,
* , • i"< " i iu *i i ' 'i • • l r .i~> i \"ri"-1 i'H ' ,¦ ^ * ">!*' >
p.i-f,ir(h A".- rtnnt
! * if i > " v > ¦ *, i > ¦ ¦ i. i - m »*¦•, <-t
» j u t. -i t ^ , '• > lv> ^ jr iii 'it* ¦ - I < * i* \ ' T»'t> f
: ci t-i . ik ¦ i 1 'v V ^ <* .5 r* \ 1 -•<: '¦ j' i . '.•« *
. f>.i "u. I'J, t.' I.I. U \ ^'u v \ i a U pd! »Lt' t\v>„> !"*
isfe of ctiassrs dynanToroeter ffsessufeFct&fis and driving cycle statistical pa farriers
jcted a research program to mode! fue! consumption and emissions using neural
y» t t f T'i'- m f j, i nu - r 1 j.i ' ¦»,: ,i" 1 1 '! Tmh i .t- " ^ f •< .
1 v,,r ->* "1*1 -1 (*<• •!" s"i, 1111 ,1fl~ -t * ^ m 'i >"i", 11 ^ 11 * i
2007 - '2012 Ph.D. Mechanical Ett§ineertii0
* i , i - ->,ii 1 u " > 5 •/«*' 'J* - tr i «1 i > i ' 1
.< , • .Jj V r< M<-Uvuih, ,il rrkji'k
UnVersidad tie las Andes, Bogota. Colombia
» r uv )>w i_, i." 11 ? • .i»v h i't j,
< ' D5 MocKint^i Ef,yirx>.Mng
,. 'i,,» m t • i-> <. i', ,• , -1
* i,», if f.
PERSONAL SKILLS -
88 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
-f" •*,.- vA*?|7>.|n
>!¦>., H>D "• < •! •• <¦ ¦ . U,l' il.,'"' I 1
' «. v! \lMi h ,'V. ,» • ;• 'jiJ JH, i(,i: i!t ! ' » " \ \ >J V MK-, >1
> >»; ¦„ t v ,t n!
i) fi'i-1 os'i,it" 'k r , •/. "<
, ~,i , ,» r.Vi, . ¦!'» !y<-», *!,.«'*•? I*-} r |,1 ^ "hvl , >vt
^ M ---• ' M - -,1> M . Tj
• .•.n-tp.i, ,, . r„- ^4. v r- , r,t
VECIQ. ami GEM
* "iv , ; % 'i >!»• ii** , a 11 < J ',r V- *
» V r 1 \'E ¦ « • - . , n , < o t '
¦* >«. [V> "< s -v - , Vs vr ,\j m1 t~, i, 111 w
• , i ^ 5 ^ 4 'ih , k *- (i 1' „< n _ • »
~ I* I t 11 1 ' ,, in v i ! n , ,(1 t|,( ¦ »• .r - , . ~ \
• Strong background and excellent record of teaching in thermal sciences, fluid mechanics.
and vehicle emissions
— ¦¦•"»
• Hands-on expenence in emissions test ma and vehicle efficiency monitoring
• « .ft -<~t — • * »1. . . > v,< \ »,«. u-mm, chassis tymmamemt, and portable
« > " • < *| J C ¦ >V» '" r *. i v * jfv ' <
87 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
r, • ' . ' '< ( • i r> ' 11' " -"'f "'H h* 1111 jf t '
" >• j! ~ r f ¦> 51 ~ f? 'H v if f <¦ > ' jji i 1 "r ;i"'i 11«-ii' , * •'» . "< •>" <"*, , 'i" hj **'! i~1 '"i ^ 1
\ >\ r>".hvif ii-1>s -j'* ¦ ij' • ir,\* ¦>. '--i,- -< •<-> „ r v-v i! svr,
• '• i j w,- {«, « * • -" \ -! " T . ,,W > *W .{'"itv r't r » 1 11 '» 1 1
r-'r I'tWfrr it C *< i"~ j-v'i-. « rT'W"
' 8-1 *v,"A 1 >5 <1 r i 5 ; ' .K .« I t » i. r ' 1 ;x ' ¦<" v v • ® * t ' 1 • . i|«- > lif„i I >>) M i , : I Vi' -PU i
\ * ' 1 "I1 0 1 1 U' 1 ,r 1 n O' I
h* 1 " - ' - - ir •>,>.j i / , i l > t a.U •««% iJ >r . Hn >w i ~' ' *r s' •
< . H .t "'t. *-J> M >1 V' >> ' V'i i j ' ) !¦ n 11 - i tt i » 1 4 r'J's"" t.f •, i| ii. » >mS< 1 L -k ' • 1 ¦ 1
! ransnorauon
Ur\'h ">i i . ' ,--<»•!)( t s„ q i r> {,' r "im. < ¦> (,f n" 1 ^ ^ i» " ' > k' » < < ' ' . ! '
•'--T- - -1 ri ,-r*i ;m~4- >,r, f,\r v'v>'V'vvv ' ' •> ' U1-
' - M(|. I ,J vV< * -1 "i > S- M . ' , | '« l »«> V^-." ~ ,! '1 ¦> /js. r - * '
fre^ht system in Chtna ' internationa! Council on Clean TranspottatJon
r> i M i?." i-i v i > i1! i U- i ii1 (•! '* ' ^ iT i i ^ "x ' I >r . t,„ 1 \ s.il, ? ; " 11, . * J-> 15 *'
European Union international Council on Clean 1 rartspoftation
?I - I i '»lf . >J * ^ K \ ' ' ! . i, -pi' "l" / »'K' n > v I 1' s' ^ ,1 - '!'I I
I' * H ! i ' H i -* I ! ' i j'l I I ,'i ' I I ' «' 1 1- 1 "I r 1 l 1 I ' -V -I I' l' t" '
l ' t i" » ' i . "¦ < 's ¦ 1 m *.• «r 4 i »»«',• "i "• *ir ,m.« .-i 4 (« i'
European Uni
on interratiO;
arg M, and D
rial Councti on Clean *
*' ' il (l »M'
tM , tlu ' " " i 1 . 1, 1 !• i, 1 V 1 !' ' ' " OP' " <>' ' '
in Indiatliter1
nafiofiai Coure
Lit on C^ean Transport
Wltiil.!'1;
i - '
.1 , ¦!.{,«
elqado 0. (2018) "Cc
trnpliancepathwaystnil>eUS Piiase2hoavy^ulyv«hicte&9kxm^
Iff
andl on Clears Iramp
o i< ,'ni . v >'
T ransfxtftattof
icxitipes, F.. <
and Muocrtef, R (201
' -ft - , fi-*- '"i,i M, ' 1 i!s •' " s '» , 1 ft i1 1 1 i '
Baseiirie and
potential for th
e2(K20-2030 time fra
t}.n - ii,'t-- < f ,? f o> - p-,r< vi >,
< *1!1
nd U. H (201"
, r in - < *i - i'
1 ' t . % !¦ it "J, V > <( > ,• '<- *" < !•» 5 i
f prr ¦" ""Iis
;¦ t> "ii
"'H r^">'^-h'^n
/-< ik f
fmtmm, G.,i
, ¦; i «t. - « 1 kl -
3,, &wicis, D , Oeigado 0 and Rodikiiiez, F (20l7)."Est»iiatifKiths
^
. 'V'/.i i\.
tentiai of v3f ¦» '..1 '« •! Vl . ' i ' .Ui.il' '
• ? -m i/ - :u' t •,
r«. >*,< i 'K1^ ,, r^tii .\J, -,i «>
incfe:sSAET^
*ctmseat P&per
i J" "1 :4
Rotiitqt&z f,
Muncrief. R.
L'teigado 0. Baidino,
C (/?017) ' Mai"ket penetration of iueI efficiency teohnofogics for heavy
>V, ,'t ^
'I l*' '"si 1 k
! « 'K", ,s - "
tetes, and Chinan' Interrtationaf Councif on Clean Transportation
" 1 M ' 1 "t
A Siarpe
, B, Delgado, 0. Ban
divadekar, A , and Garg M (2017) "Impravdd heavy-cJuty vehrcl© fuel
efioency in lr
id is. Benefrts t
»ss and envirofifmn
f'(t ,ri j.-m
posiKla.P, B
§lgsdo„ Q.: an-
d Oerman, d |331 B) P
\ c ' ^ f f- t -' M i ik •, i.i
!j),M .Kv-«
¦ i ?, r/m ;
« ' 111 « t{ . *' >' 'l 1 \"rS<» t '¦ <„.!./ MW ,
duty trucks tri
major nwfe&tss
. around ttn? wfkJ,'" Gi
.'I << ) . ! "v , s*. II" ' ^ ||\. I-
a-.w
1 China fuel consumpt
',i ij„-. i >f. i»'. vWj , ii ' A' i l > '...h" v.'
on Clean Trar
isporVation
Sharp© B m
i Delgado O.
{2018} Engine Ww«
i,>j\ , i kwo, ^ i <• • \ In , i ••;,»»< i • in i!», ,»«« 1
on Clean Trar
i iP..! ,'"l i'(
-".vi' ¦ »n«
i Delgado,. 0 s
[2016) "Enginasandlj
lr,r . >uli
Sharps. B an
: p' -t'1*
, /" - i-A -;*f -
4 ii," T!,^
1 '<* ' C "*9* / n >1
Jounc!! on
, -in; h •
"11
¦J |
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December
CutneuluTi viae
:• Lubsey, N, Myneoef. R. Sharpe, B. and Detgado. O. {2015) U S. efficient . .m ;i
MY 2018-202? hewy-duty vehicles, engines, and 'tellers" Inter national Council on Clean Transportation
US EPA (2015) Peer reviewed the greenhouse gas emission madef(GI=K' «« <• r j"",
fi- rjv.ewi
•• Kodjak. D, Shatpe. B„ ami Pefratto, O (2015) "Evolution of heavy-duty >
Miftgafci and Aefaptabon Strategist* Global Change,
• Franco, V anil Delgacto, 0 tmn)'nmsY^ui^ veiicM mmmSor, A comparison of USandEU tools.'
international Council on Clean Transportation
. Master D,Uiteey, N .and Pelgad®, 0, tang-haul
tractor traitefs sti the 2020-2030 tirnefenTe.'* frttefiisfional Council on Clean Tisnspoffsferi,
Datgado, O awl Lulsey, N. {231 5) "Advanced tracks -toiler efficiency tashnrfaw potential m this 2020.2030 timeframe"
in"j-« 'nr.-;! rur C J-An I
•• Thwuvengadam, A . Thirwengadam. P, Ptadhan, 8, Beach. M. Carder, 0., and Deigacfe, O (2314) Heavy-duty vehcie
oiesei engine emc>eiK.y evaluation anu energy auari vw,tst vnginta unsversrcy
.¦ amp* R• Deiptfo, O'., and Munerief, R. OBI# Xempwa*# assessment of Heavy-duty vehicle regwtato»y sfestgn
options tor US g»8enfiotis» gas and efficiency regulation" 1 rtemaionat Council on Clean Transportation.
... Sharp®, B, 0etgado, O.. mi U*m, N, <2014) 'Megmtiwj traitors into HDV wgolaton: Benefit**** analysts.''
Internafcnat Council on Qm Transportation
Delia*, 0. and Ufcey N {2014) "The US SuperTowk program Expediting development of aiiranoed heavy-duty
efficiency fechricsloBBs" International Council en Clean Transportation
v' 'V."W 4 f j, w w «- iwiWjtvw * ' "f '&>»s w w *#) cm-i* >¦
• Kappanna, R, Besch, Ivt., Tftiruwsngadam. A, Deiaaclo, O. rt at, 3013 tSmetftouaigas eiwssorisc# MY 2010
advanced heavy dn% dieset engine measures! civet a MWKXKttwnM trip of USA" SAE Teehmcsl Pa|»r 3313-24-0170
Cfct^?icto, O.F, C3a*. N N.. Thompson, 6 J„ J012 'Hftwy-sJiily trittfiiel sonsumpfcin prediction lasetj on flriving eyct#
prep*®" international Journal of Syst»»ble Transportation, 6, pp 1-24,
• • Dslgiitfo, OF.. Clartt, N.N., Thompson, G J.. 2011 "Method for ftansiation of rhm ta* eonwmptSon witl NC« emissions
between DM heavy--duty wehiete routes - Prtwfegs of tt* ASMS 3011 Intermit Comtastfen Bn9m OMskxt m
e ¦ « ..i i. ¦¦¦ j » a /, *
\emmmi t.orffefems, worgantoMi, wv
. Dt|}#!Jo, OF,, Ci»fc, N.N., Thotnpson, G.J., 2011 "frtxfctiog ttmst bm Imi on it* bass of cfete
properties." Journal *.jf lie Air & mm Mawgemert ksmmatioix S1, pp. 443-452
Delgado, 0,P„ Mwiina, JA, 2003 "ExtusiiSn d* pwflM espumados de nwteot Blastca (Bttuman nwlcima of wood
polymer foams)."" Ra'ista de (ngiatierta (Bogota. Cotorota) 18, pp ®63.
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report December 2022
Appendix C: Conflict of Interest (COI) Forms
\
'ICF
ORGANIZATIONAL CONFLICT OF INTEREST CERTIFICATE
Customer: U.S. Environmental Protection Agency
Contractor; ICF Incorporated, LLC, 9300 Lee Highway, Fairfax, VA 22031
Prime Contract: Task Order 68HERC22F0351 under Contract 68HERC21D0016
Subject Report: Peer Review of Electrified Vehicle Simulations within EPA's ALPHA
Model
Subcontractor/Peer Reviewer: Sujit Das, Strategic Analysis, Inc.
In accordance with EPAAR 1552.209-70 through 1552.209-73, Subcontractor/Consultant certifies to the
best of its knowledge and belief, that.
X No actual or potential conflict of interest exists.
An actual or potential conflict of interest exists, See attached fuli disclosure.
Subcontractor/Consultant certifies that its personnel, who perform work on this contract, have been
informed of their obligations to report personal and organizational conflict of interest to Contractor and
Subcontractor/Consultant recognizes its continuing obligation to identify and report any actual or
potential organizational conflicts of interest arising during performance under referenced contract.
Subcontractor/Consultant
10/24/22
Date
90 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
\\/_
'ICF
ORGANIZATIONAL CONFLICT OF INTEREST CERTIFICATE
Customer: U .5. Environmental Protection Agency
Contractor: ICF Incorporated, LLC, 9300 Lee Highway, Fairfax, VA 22031
Prime Contract: Task Order 6SHERC22F0351 under Contract 6SHERC21DO016
Subject Report: Peer Review of Electrified Vehicle Simulations within EPA's ALPHA
Model
Subcontractor/Peer Reviewer: Shawn Midlam-Mohler - ModeJTek, LLC
In accordance with EPAAfi 1552-209 70 through 1552-209-73, Subcontractor/Consultant certifies to the
best of its knowledge and belief, that:
X No actual or potential conflict of interest exists.
An actual or potential conflict of Interest exists. See attached full disclosure.
Subcontractor/Consultant certifies that its personnel, who perform work on this contract, have been
informed of their obligations to report personal and organizational conflict of interest to Contractor and
Subcontractor/Consultant recognizes its continuing obligation to identify and report any actual or
potential organizational conflicts of interest arising during performance under referenced contract.
ILTKJLtiL.
Su bcontractor/Consu Itant
J!ft Hi
Date
91 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
'ICF
ORGANIZATIONAL CONFLICT OF INTEREST CERTIFICATE
Customer: U.S. Environmental Protection Agency
Contractor: ICF Incorporated, LLC, 9300 Lee Highway, Fairfax, VA 22031
Prime Contract: Task Order 68HERC22F0351 under Contract 68HERC21DQ016
Subject Report: Peer Rewew of Electrified Vehicle Simulations within EPA's ALPHA
Model
Subcontractor/Peer Reviewer: International Council on Clean Transportation
In accordance with EPAAR 1552.209-70 through 1552.209-73, Subcontractor/Consultant certifies to the
best of its knowledge and belief, that:
No actual or potential conflict of interest exists.
An actual or potential conflict of interest exists, See attached full disclosure.
Subcontractor/Consultant certifies that its personnel, who perform work on this contract, have been
Informed of their obligations to report personal and organizational conflict of interest to Contractor and
Subcontractor/Consultant recognizes its continuing obligation to identify and report any actual or
potential organizational conflicts of interest arising during performance under referenced contract.
Oscar Delgado, PhD
ICCT
Subcontractor/Consultant
10/28/2022
Date
92 IP a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Appendix D: Notes from Mid-review Teleconference
To: Jeff Cherry, TO COR, U.S. EPA
From: Sam Pournazeri, Project Manager, ICF
Date: October 6, 2022
Peer Reviewers' Kick-off meeting for Task Order 68HERC22F0351 under Contract
Re: 68HERC21D0016 for Peer Review of Electrified Vehicle Simulations within EPA's ALPHA
Model
Meetir tion
• Date: Thursday, October 6, 2022
• Location: Virtual using Microsoft Teams
Meetir ;ipants:
• Jeff Cherry, EPA, TO COR
• Brian Olson, EPA, Alternative TO COR
• Sam Pournazeri, ICF, Project Manager for the peer review
• Ramon Molina Garcia, ICF, Support staff for the peer review
• Oscar Delgado, Ph.D., International Council of Clean Transportation
• Shawn Midlam-Mohler, Ph.D., Professor, Ohio State University
• Sujit Das, Principal Engineer, Strategic Analysis Inc.
Meetir nutes:
• Sam Pournazeri from ICF kicked off the meeting with a brief introduction
• Ramon Molina Garcia from ICF provided a slide presentation (appendix A) walking the peer
reviewers through the process, charge questions, and timeline
• ICF, EPA staff, and peer reviewer panel introduced themselves.
• Jeff Cherry from EPA provided an overview of the model and gave some guidance to peer
reviewers on the purpose of charge questions and that EPA is looking for peer reviewers to
look at different aspects of the model.
• Peer reviewers asked a couple of questions regarding the areas where they should focus on.
• Sujit Das asked whether there is any other technical documentation aside from the one
provided that shed light on the logics behind the model simulations where he can look and
comment. He was concerned that the existing documentation mainly describes the codes
and not the logic
• Jeff responded that there are several links provided under the Readme file that might be
helpful and that the PDF documentation included in the zip folder should have sufficient
information for the peer reviewers to comment on.
93 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
• Shawn Midlam-Mohler asked if there are more information available on some
verification/validation data associated with the model and sub-models that are included in
demo runs.
• Jeff recommended Shawn to send his questions through an email to ICF and for EPA to
respond to them.
• Oscar Delgado from ICCT also asked whether EPA has validated the ALPHA tool results
against real-world data or other commercially available vehicle simulation tools?
• Jeff recommended Oscar to send his questions through an email to ICF and for EPA to
respond to them.
• Ramon continued the presentation by going over the materials provided, expected
deliverables, and the timeline.
• At the end Sam mentioned that ICCT may need some extension on the peer review and Jeff
confirmed that it is fine. Later after the meeting, Oscar mentioned that they will make sure to
get their feedback back to ICF by October 28
• Both Shawn and Oscar submitted their questions through email, and Ramon forwarded those
to EPA.
Next Steps:
• ICF will compile the reports from peer reviewers and submit those "as is" to EPA by
November 2.
• ICF will prepare a technical process memo that describes how peer reviewers were selected,
the process that ICF took to administer the peer review, and how the peer review was
concluded. As part of this memo, ICF also include the unedited peer review comments and
responses into a tabular format, with two columns as described above so that the individual
comments may be easily grouped and compared for review purposes.
• Upon receiving the peer review reports from all reviewers, ICF will also start drafting the final
report and deliver it to EPA two weeks after receipt of peer reviewers' comments.
94 IP a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model
ICF Slide Presentation
- Final Report
December 2022
--Introductions
^EPA: Background of the Peer Review
^ ICF: Peer Review Overview
- Charge Questions
- Materials to Review and Submit
- Schedule
Questions/Comments
-> Agenda
95 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Introductions
"her
96 IP a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
> Background of the
Peer Review
8 "*icr
r The Peer Review will evaluate the accuracy and completeness
of the new ALPHA electric vehicle model
o BEV model concepts and methodologies
o Expert feedback on technical aspects of ALPHA
oSpecific recommendations to improve quality of outputs
Technical Background and Information
97 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report December 2022
Peer Review Overview
'ic*
Does EPA's overall approach to the stated purpose of the model (demonstrate technology effectiveness for various fuel
economy improvement technologies) and attributes embody that purpose?
What is the appropriateness and completeness of the overall model structure and its components, such as:
• The breadth of component models/technologies compared to the current/future light-duty fleet
• The performance of each component model, including the reviewer's assessment of the underlying equations and/or physical principles coded into that
component.
• The input and output structures and how they interface with the model to obtain the expected result (i.e., fuel/energy consumption and C02 over the
given driving cycles).
• The use of default or dynamically generated values to create reasonable models from limited data sets.
Does the ALPHA model use good engineering judgement to ensure robust and expeditious program execution?
Does the ALPHA model generate clear, complete, and accurate output/results (C02 emissions or fuel efficiency output
file)?
Do you have any recommendations for specific improvements to the functioning or the quality of the outputs of the
model?
Charge Questions
98 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
-> Materials to Review and Submit
September 29, 2022:
Charge letter and
draft distributed to
peer reviewers
October 6, 2022:
Conference call with
EPA, ICF, and
reviewers
October 20, 2022:
Comment/review
due via email to
Sam Pournazeri
Sani.Pournatefl@icf.com
Schedule
99 IP a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Appendix E: Peer Reviewer Selection Memo
To: Jeff Cherry, TO COR, U.S. EPA
From: Sam Pournazeri, Project Manager, ICF
Date: July 1, 2022
Re: Task Order 68HERC22F0351 - Peer Reviewer Selection
Under Task Order 68HERC22R0170, ICF is coordinating an independent peer review of the Electrified
Vehicle Simulations within EPA's ALPHA Model on behalf of the U.S. Environmental Protection Agency
Office of Transportation and Air Quality (EPA OTAQ).
To assemble the panel of three independent peer reviewers, ICF reviewed a pool of subject matter
experts both suggested by EPA OTAQ and identified by ICF through independent research. ICF first
assessed the experts' availability to perform the peer review within the timeline agreed upon with
the EPA Contracting Officer Representatives (COR). After that, ICF reviewed curriculum vitae and
other relevant work to select peer reviewers that represent a combined expertise that cover, at a
minimum: understanding of vehicle technology packages including battery technology, hybrid and
electric powertrains, e-motors, transmission systems (e.g., shift strategy), and vehicle accessories as
well as engine fuel consumption map, and vehicle behavior.
While all candidates were highly qualified to act as peer reviewers, ICF sought to select candidates
that can bring diverse and complementary perspective to the peer review process. ICF also
evaluated actual or apparent conflicts of interest that would preclude an independent review, in
accordance with the EPA Peer Review Handbook Sections 3.4.5 and 3.4.6. To the best of ICF's
knowledge, no conflicts of interest were found for the proposed peer reviewers in our preliminary
research but will finalize the COI evaluation as part of the contracting process. This peer review
selection memorandum presents ICF's initial selection of three proposed reviewers.
Upon the selection of the peer reviewers, ICF shared the qualifications and resume for each
proposed peer reviewer with EPA to discuss the strengths that each peer reviewer will bring into this
project. Upon discussion with TO COR, ICF finalized the list of peer reviewers.
iewer Selection Process
ICF first compiled a set of suggested peer reviewers for the report. This list was based on both EPA's
initial recommendations and ICF's suggestions for additional potential reviewers, twelve candidates
(five selected by EPA and seven identified by ICF) were considered. ICF also prioritized peer
reviewers based on the relevance of their background and experience with the topic of the report.
Through an initial contact with the selected peer reviewers, ICF assessed each potential reviewer's
ability to perform the work during the period of performance and to identify any association they
have with the work that would preclude them from being objective. ICF contacted and
communicated with all candidates by e-mail.
In our outreach we identified ourselves as independent contract employees and provided initial
information on ALPHA model, including the newly added electric vehicle model and the expected
100 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
time commitment to exercising the model. We asked the potential reviewers to assess their
availability for this study and for their hourly rate. We also collected a curriculum vitae for each peer
reviewer that expressed availability and interest in participating.
List of Peer Reviewers
Wallace R. Wade
Chief Engineer and Technical
Fellow, Powertrain Systems
Technology and Processes
Center for Automotive Research
Patrick Hammett
Lecturer, College of
Engineering, University of
Michigan
Sujit Das
Principal Engineer at
Strategic Analysis, Inc.
Francisco Posada
Project Lead for Engineering
Associates, South America
Aymeric Rosseau
Interim Director of Center for
Decarbonization Solutions
Deployment for Argonne
National Laboratory
John M. German
Senior Fellow for International
Council on Clean Transportation
Hussein Basma
Heavy-Duty Vehicles
Associate Researcher for
International Council on
Clean Transportation
Shawn Midlam-Mohler
Professor of Practice and Director
of Ohio State University Simulation
Innovation and Modeling Center
Dan Meszler
Principal Researcher at MES
Oscar Delgado
Latin American Lead / Fleets
Center Manager for
International Council on
Clean Transportation
Anup Bandivadekar
Environment Program Officer
at William and Flora Hewlett
Foundation
Linda M. Miller
Ex. Manufacturing Director, Powertrain
Operations, Ford Motor Co.
101 I P a g e
-------
Peer Review of Electrified Vehicle Simulations within EPA's ALPHA Model - Final Report
December 2022
Final List of Peer Reviewers
Upon completion of the initial contact, the top three peer reviewers selected for this project agreed
to participate in this peer review process. The rest of the peer reviewer candidates were either not
available, not interested (e.g., retired), or had concerns with the limited time allocated for the review
(i.e., 20 hours). The resumes for the three selected candidates were collected and shared with U.S.
EPA TO COR. Upon approval from U.S. EPA TO COR, ICF initiated the subcontracting process with the
selected peer reviewers. Below is the final list of the peer reviewers that will serve on this task order.
The three selected peer reviewers (including the ICCT team) provide a diverse combination of
expertise in evaluating the ALPHA model. Sujit Das, with 37 years of experience in energy efficiency
research, has served as the peer reviewer of the ALPHA model back in 2016, and has published
articles related to powertrain design for advanced fuel vehicle technologies. Shawn Midlam-Mohler is
a Professor of Mechanical Engineering at Ohio State University, with expertise in engine selection,
modeling, and control development for an extended range electric vehicles as well as vehicle
simulations and powertrain optimization. Oscar Delgado and ICCT team also brings in years of
experience in modeling advanced technologies and developing tools to support Global commercial
fleets in their transition to zero-emission vehicles.
Sujit Das
Principal Engineer at
Strategic Analysis, Inc.
Shawn Midlam-Mohler Oscar Delgado
Professor of Practice and Director Latin American Lead / Fleets
of Ohio State University Simulation Center Manager for
Innovation and Modeling Center International Council on
Clean Transportation^
** Note that while Oscar Delgado will be our point-of-contact, ICCT has decided to review the model as a team (Oscar
Delgado, John German, Hussein Basma).
102 IP a g e
------- |