January 1994
(Version 1.1)
EPANET USERS MANUAL
by
Lewis A. Rossman
Drinking Water Research Division
Risk Reduction Engineering Laboratory
Cincinnati, Ohio 45268
RISK REDUCTION ENGINEERING LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OH 45268
-------
DISCLAIMER
The information in this document has been funded wholly or in part by the U.S. Environmental
Protection Agency (EPA). It has been subjected to the Agency's peer and administrative review,
and has been approved for publication as an EPA document. Mention of trade names or
commercial products does not constitute endorsement or recommendation for use.
Although a reasonable effort has been made to assure that the results obtained are correct, the
computer programs described in this manual are experimental. Therefore the author and the
U.S. Environmental Protection Agency are not responsible and assume no liability whatsoever
for any results or any use made of the results obtained from these programs, nor for any
damages or litigation that result from the use of these programs for any purpose.
11
-------
FOREWORD
Today's rapidly developing and changing technologies and industrial products and practices
frequently carry with them the increased generation of materials that, if improperly dealt with,
can threaten both public health and the environment. The U.S. Environmental Protection
Agency (EPA) is charged by Congress with protecting the Nation's land, air, and water
resources. Under a mandate of national environmental laws, the Agency strives to formulate
and implement actions leading to a compatible balance between human activities and the ability
of natural systems to support and nurture life. These laws direct the EPA to perform research
to define our environmental problems, measure the impacts, and search for solutions.
The Risk Reduction Engineering Laboratory is responsible for planning, implementing, and
managing research, development, and demonstration programs to provide an authoritative,
defensible engineering basis in support of the policies, programs, and regulations of the EPA
with respect to drinking water, wastewater, pesticides, toxic substances, solid and hazardous
wastes, and Superfund-related activities. This publication is one of the products of that research
and provides a vital communication link between the researcher and the user community.
In order to meet regulatory requirements and customer expectations, water utilities are feeling
a growing need to understand better the movement and transformation undergone by treated
water introduced into their distribution systems. EPANET is a computerized simulation model
that helps meet this goal. It predicts the dynamic hydraulic and water quality behavior within
a drinking water distribution system operating over an extended time period. This manual
describes the operation of the program and shows how it can be used to analyze a variety of
water quality related issues in distribution systems.
E. Timothy Oppelt, Director
Risk Reduction Engineering Laboratory
111
-------
ABSTRACT
EPANET is a computer program that performs extended period simulation of hydraulic and
water quality behavior within drinking water distribution systems. It tracks the flow of water
in each pipe, the pressure at each pipe junction, the height of water in each storage tank, and
the concentration of a substance throughout a distribution system during a multi-time period
simulation. In addition to substance concentrations, water age and source tracing can also be
performed. The water quality module of EPANET is equipped to model such phenomena as
reactions within the bulk flow, reactions at the pipe wall, and mass transport between the bulk
flow and pipe wall. This manual describes how to use the EPANET program on a personal
computer under both DOS and Microsoft* Windows'". Under Windows the user is able to edit
EPANET input files, run a simulation, display the results on a color coded map of the
distribution system and generate additional tabular and graphical views of these results.
-------
TABLE OF CONTENTS
Disclaimer u
Foreword . iii
Abstract - iv
Figures • • - yii
Tables yiii
Acknowledgements • • • • • ix
1. Introduction 1
2. The Network Model.. • 3
2.1 Network Components 3
2.2 Time Patterns 11
2.3 Hydraulic Simulation Model 12
2.4 Water Quality Simulation Model 14
2.5 Reaction Rate Model 16
2.6 Water Age and Source Tracing 18
3. Installation .. • 19
3.1 System Requirements 19
3.2 Installation for Both Windows and DOS..... 19
3.3 Installation for DOS Only 20
3.4 Customizing Your Installation , 21
4. Input Data Formats 23
4.1 Data Preparation 23
4.2 Input File Organization 24
4.3 Input File Format 27
4.4 Verification File Format 55
4.5 Map File Format 56
4.6 Summary of Default Values and Units 58
-------
5. Running EPANET Under DOS 61
5.1 General Instructions 61
5.2 Contents of the Output Report 62
5.3 The EPANET4D Program 64
6. Running EPANET Under Windows 69
6.1 Overview .- 69
6.2 Launching the Program 71
6.3 Operating Procedures , 71
6.4 Summary of Menu Commands 84
7. Example Applications 87
7.1 Introduction.... 87
7.2 Example 1 - Chlorine Decay 87
7.3 Example 2 - Fluoride Tracer Analysis 94
7.4 Example 3 - Source Tracing 99
Appendix A. Files Installed by EPANET 103
Appendix B. Error and Warning Messages 105
Appendix C. The KYP2EPA Conversion Program 109
Appendix D. Troubleshooting Ill
VI
-------
FIGURES
Number Page
2.1 Node-Link Representation of a Network 3
2.2 Example of a Pump Characteristic Curve 6
2.3 Pump Curve With Extended Flow Range 7
2.4 Effect of Relative Speed (n) on Pump Curve 8
2.5 Time Pattern for Water Usage 12
4.1 Example EPANET Input Data 25
4.2 Examples of Pump Curve Input Requirements 40
4.3 Definition of TankLevels 50
4.4 Example Map File 56
5.1 Text Editor Command Summary 66
5.2 ContentsofEPANET4D.BAT 67
6.1 Example Contents of the EPANET4W Workspace 70
6.2 File Dialog Box 72
6.3 Text Editor Command Summary 73
6.4 The Browser 77
6.5 Legend Dialog Box 80
6.6 Map Options Dialog Box 80
6.7 Table Search Dialog Box 81
6.8 Graph Options Dialog Box 82
6.9 Data to Link With a Graph 82
6.10 Graph With Linked Data 83
7.1 Network for Example 1 88
7.2 Input Data for Example 1 88
7.3 Portion of Output Report for Example 1 91
7.4 Network for Example 2 94
7.5 Input Data for Example 2 96
7.6 Observed (X) and Predicted Fluoride Levels at Node 11
ofExamplei 99
7.7 Network for Example 3 100
7.8 Detail of Network Around River Source 101
7.9 Spatial Coverage of Water From Lake Source After 14 Hours 102
Vll
-------
T A B L E S
Number page
2.1 Pipe Head Loss Formulas 5
2.2 Roughness Coefficients for New Pipe 5
2.3 Loss Coefficients for Common Components 10
4.1 Summary of Default Parameter Values 59
4.2 Summary of Input Parameter Units 60
Vlll
-------
ACKNOWLEDGEMENTS
Special thanks are owed to Dr. Walter M. Grayman, Consulting Engineer, and
to Dr. Paul F. Boulos of Montgomery Watson for the many suggestions and
hours of testing they provided during the development of the EPANET program.
Without their assistance, the quality of mis product would be substantially less.
The efforts of Dr. Rolf Deininger and Kuo-Liang Lai of the University of
Michigan, Dr. Charles N. Haas of Drexel University, Eugene Mantehev of the
Seattle Water Department, and Drs. Lindell Ormsbee and Srinivas Reddy of the
University of Kentucky in beta-testing the EPANET program are also greatly
appreciated. Gayle Smalley of the North Marin Water District and Rob Tull of
Montgomery Watson supplied data for one of the example networks in Chapter
7. Dr. James Uber of the University of Cincinnati along with his students Cheryl
Bush, Mao Fang, and Ken Hickey, provided thoughtful discussion on many key
issues. The public domain text editor TE 2.61 used by EPANET was developed
by John D. Haluska.
A final note of acknowledgement goes to Dr. Robert M. Clark, Director of
RREL's Drinking Water Research Division, whose pioneering work in modeling
water quality in distribution systems and personal encouragement inspired the
development of EPANET.
IX
-------
CHAPTER 1
INTRODUCTION
EPANET is a computer program that performs extended period simulation of
hydraulic and water quality behavior within pressurized pipe networks. A
network can consist of pipes, nodes (pipe junctions), pumps, valves and storage
tanks or reservoirs. EPANET tracks the flow of water in each pipe, the pressure
at each node, the height of water in each tank, and the concentration of a.
substance throughout the network during a multi-time period simulation. In
addition to substance concentrations, water age and source tracing can also be
simulated.
EPANET is designed to be a research tool for improving our understanding of the
movement and fate of drinking water constituents within distribution systems.
The water quality module of EPANET is equipped to model such phenomena as
reactions within the bulk flow, reactions at the pipe wall, and mass transport
between the bulk flow and pipe wall. As we gain more experience and
knowledge of water quality behavior within distribution systems we intend to
update and refine EPANET to reflect this progress.
Another distinguishing feature of EPANET is its coordinated approach to
modeling network hydraulics and water quality. The program can compute a
simultaneous solution for both conditions together. Alternatively it can compute
only network hydraulics and save these results to a file, or use a previously saved
hydraulics file to drive a water quality simulation.
EPANET can be used for many different kinds of applications in distribution
system analysis. Sampling program design, hydraulic model calibration, chlorine
residual analysis, and consumer exposure assessment are some examples.
EPANET can help assess alternative management strategies for improving water
quality throughout a system. These would include:
• altering source utilization within multiple source systems,
• altering pumping and tank filling/emptying schedules ,
1
-------
• use of satellite treatment, such as re-chlorination at storage tanks,
• targeted pipe cleaning and replacement.
EPANET was coded in the C language and makes use of dynamically allocated
memory. Thus the only limits on network size are available memory. The
version of EPANET contained on the distribution disk is intended for use on
IBM-compatible personal computers running under DOS. However it is a
relatively simple task to re-compile the source code to run on other machines,
such as UNIX workstations.
The EPANET package contains two program modules. One is a network
simulator that runs under DOS, receiving its input from a file and writing its
output to another file. The user must use external programs to edit the input file
and view or print the output file. (An optional DOS shell program is provided
that interactively edits EPANET input, runs the simulator, and views or prints its
output according to selections made from a menu). The second module is a
Microsoft, Windows™ 3.x program that allows one to edit EPANET input data,
run the simulator, and graphically display its results in a variety of ways on a
map of the network. Thus there are two different ways to run EPANET — under
DOS or under Windows. We believe that for most situations the visualization
power of the Windows version is an essential aid in trying to comprehend the
results of running EPANET and recommend that this mode be used if your
computer hardware and software can support it.
Chapter 2 of this manual describes how EPANET models water distribution
systems. Chapter 3 tells you how to install the EPANET system on your
personal computer. A detailed description of the program's input data format is
provided in Chapter 4, while Chapters 5 and 6 give instructions for running
EPANET under DOS and Windows, respectively. Finally, Chapter 7 illustrates
different features of the program by running several example applications.
-------
CHAPTER 2
THE NETWORK MODEL
2.1 Network Components
EPANET views a water distribution network as a collection of links connected
together at their endpoints called nodes. Figure 2.1 illustrates a node-link
representation of a simple water network.
Reservoir
Tank
Source
Figure 2.1 Node-Link Representation of a Network
-------
As shown in the figure, links come in several varieties:
1. pipes
2. pumps
3. valves.
Besides being the junction point between connecting pipes, nodes can serve as:
1. points of water consumption (demand nodes)
2. points of water input (source nodes)
3. locations of tanks or reservoirs (storage nodes).
How EPANET models the hydraulic behavior of each of these components will
be reviewed next. For the sake of discussion we will express all flow rates in
cubic feet per second (cfs), although the program can also accept flow units in
gallons per minute (gpm), million gallons per day (mgd), or liters per second
(L/s).
Pipes
Pipes convey water from one point to another. Flow direction is from the end
at higher head (potential energy per pound of water) to that at lower head. The
head lost to friction associated with flow through a pipe can be expressed in a
general fashion as:
h = ab (1)
where hL is the head loss in feet, q is the flow in cfs, a is a resistance coefficient,
and b is a flow exponent.
EPANET can use one of three popular forms of Equation 1: the Hazen-Williams
formula, the Darcy-Weisbach formula, or the Chezy-Manning formula. The
Hazen-Williams formula is probably the most popular head loss equation for
distribution systems, the Darcy-Weisbach formula is more applicable to laminar
flow and to fluids other than water, while the Chezy-Manning formula is more
commonly used for open channel flow. Table 2.1 lists values of the resistance
coefficients and flow exponents for each formula. Note that each formula uses
-------
a different pipe roughness coefficient that must be determined empirically Table
2.2 lists general ranges of these coefficients for different types of new pipe
materials. Be aware that a pipe's roughness coefficient can change considerably
Table 2.1 Pipe Head Loss Formulas
Formula
•••
Hazen-Williams
Darcy-Weisbach
Chezy-Manning
(full pipe flow)
Resistance Coefficient (a)
'™^===^=i^ssK5i^s::s5^s:=
4.72C-1-8Sd-*-87L
0.0252 f(e,d,q)d-5L
4.66n2d-5-33L
Flow Exponent (b)
1.85
Notes: C = Hazen-Williams roughness coefficient
e = Darcy-Weisbach roughness coefficient (ft)
f = friction factor (dependent on e, d, and q)
n = Manning roughness coefficient
d = pipe diameter (ft)
L = pipe length (ft)
Table 2.2 Roughness Coefficients for New Pipe
Cast Iron
Concrete or
Concrete Lined
130-140
—' -i
120-140
Darcy-Weisbach
e, millifeet
~"
0.85
1.0-10
Manning's
n
=
0.012-0.015
0.012-0.017
Galvanized Iron
0.5
0.015-0.017
-------
Pipes can contain check valves in them that restrict flow to a specific direction.
They can also be made to open or close at pre-set times, when tank levels fall
below or above certain set-points, or when nodal pressures fall below or above
certain set-points.
Pumps;
A pump is a device that raises the hydraulic head of water. The relationship
describing the head imparted to a fluid as a function of its flow rate through the
pump is termed the pump characteristic curve. Figure 2.2 gives an example.
EPANET represents pump curves with a function of the form:
aql
(2)
400
300
"S 200
100
6
Flow, cfs
12
Figure 2.2 Example of a Pump Characteristic Curve
-------
where ha is the head gain imparted by the pump in ft, q is the flow through the
pump in cfs, h0 is the shutoff head, a is a resistance coefficient, and b is a flow
exponent. By supplying EPANET with the shutoff head h<, and two other points
on the pump curve, the program is able to estimate values for a and b.
Some pumps exhibit a different type of characteristic curve beyond their normal
flow range. Figure 2.3 shows a pump with a linear head-flow relation in its
extended flow range. In this case EPANET can describe the pump's behavior
with two equations, one for the normal flow range and another for its extended
flow range.
400
Figure 2.3 Pump Curve With Extended Flow Range
-------
Another way to represent a pump when its characteristic curve is unknown is to
assume that it adds energy to the water at a constant rate. In this case the
equation of the pump curve would be
(3)
where Hp is the pump horsepower. The latter quantity can be computed based
on an initial estimate of the flow and head at which the pump will operate. This
type of pump curve should only be used for steady-state, preliminary design
studies.
Flow through a pump is unidirectional and pumps must operate within the head
and flow limits imposed by their characteristic curves. If the system conditions
require that the pump produce more than its shutoff head, EPANET will attempt
to close the pump off and will issue a warning message. EPANET allows you
to turn pumps on or off at pre-set times, when tank levels fall below or above
certain set-points, or when nodal pressures fall below or above certain set-points.
Variable speed pumps can also be considered by specifying that their speed
setting be changed under these same types of conditions. By definition, the
original pump curve supplied to the program has a relative speed setting of 1.
If the pump speed doubles, then the relative setting would be 2; if run at half
speed, the relative setting is 0.5 and so on. Figure 2.4 illustrates how changing
a pump's speed setting affects its characteristic curve.
Flow, cfs
Figure 2.4 Effect of Relative Speed (n) on
Pump Curve
-------
Valves
Aside from the valves in pipes that are either fully opened or closed (such as
check valves), EPANET can also represent valves that control either the pressure
or flow at specific points in a network. Such valves are considered as links of
negligible length with specified upstream and downstream junction nodes. The
types of valves that can be modelled are:
1. Pressure Reducing Valves (PRVs)
2. Pressure Sustaining Valves (PSVs)
3. Pressure Breaker Valves (PBVs)
4, Flow Control Valves (FCVs)
5. Throttle Control Valves (TCVs)
PRVs limit the pressure on their downstream end to not exceed a pre-set value
when the upstream pressure is above the setting. If the upstream pressure is
below the setting, then flow through the valve is unrestricted. Should the
pressure on the downstream end exceed that on the upstream end, the valve closes
to prevent reverse flow.
PSVs try to maintain a minimum pressure on their upstream end when the
downstream pressure is below that value. If the downstream pressure is above
the setting, then flow through the valve is unrestricted. Should the downstream
pressure exceed the upstream pressure then the valve closes to prevent reverse
flow.
PBVs force a specified pressure loss to occur across the valve. Flow can be in
either direction through the valve.
FCVs limit the flow through a valve to a specified amount. The program
produces a warning message if this flow cannot be maintained without having to
add additional head at the valve.
TCVs simulate a partially closed valve by adjusting the minor head loss
coefficient of the valve. A relationship between the degree to which the valve is
closed and the resulting head loss coefficient is usually available from the valve
manufacturer.
-------
Minor Losses
Minor head losses (also called local losses) can be associated with the added
turbulence that occurs at bends, junctions, meters, and valves. The importance
of such losses will depend on the layout of the pipe network and the degree of
accuracy required. EPANET allows each pipe and valve to have a minor loss
coefficient associated with it. It computes the resulting head loss from the
following formula:
h =
(4)
where K is a minor loss coefficient, q is flow rate in cfs, and d is diameter in ft.
Table 2.3 gives values of K for several different kinds of components.
Table 2.3 Loss Coefficients for Common Components
Component
Globe valve, fully open
Angle valve, fully open
Swing check valve, fully open
Gate valve, fully open
Short-radius elbow
Medium-radius elbow
Long-radius elbow
45° elbow
Closed return bend
Standard tee - flow through run
Standard tee - flow through branch
Square entrance
Exit
Loss Coefficient
10.0
5.0
2.5
0.2
0.9
0.8
0.6
0.4
2.2
0.6
1.8
0.5
1.0
10
-------
Modes
All nodes should have their elevation above sea level specified so that the
contribution to hydraulic head due to elevation can be computed. Any water
consumption or supply rates at nodes that are not storage nodes must be known
over the duration of time the network is being analyzed. Storage nodes (i.e.,
tanks and reservoirs) are special types of nodes where a free water surface exists
and the hydraulic head is simply the elevation of water above sea level. Tanks
are distinguished from reservoirs by having their water surface level change as
water flows into or out of them — reservoirs remain at a constant water level no
matter what the flow is. EPANET models the change in water level of a storage
tank with the following equation:
Ay = IA* (5)
A
where Ay = change in water level, ft
q == flow rate into (+) or out of (-) tank, cfs
A = cross-sectional area of the tank, ft2
At = time interval, sec
Thus EPANET needs to know the cross-sectional area as well as the minimum
and maximum permissible water levels for storage tanks. Reservoir-type storage
nodes are usually used to represent external water sources, such as lakes, rivers,
or well fields. Storage nodes should not have an external water consumption or
supply rate associated with them.
2.2 Time Patterns
EPANET assumes that water usage rates, external water supply rates, and
constituent source concentrations at nodes remain constant over a fixed period of
time, but these quantities can change from one time period to another. The
default time period interval is 1 hour, but this can be set at any desired value.
The value of any of these quantities in a time period equals a baseline value
multiplied by a time pattern factor for that period. Figure 2.5 illustrates a pattern
of factors that might apply to daily water demands, where each period is of 2
hours duration. Different patterns can be assigned to individual nodes or groups
of nodes.
11
-------
o
tJ
O)
a
v>
1 2 3.4 5 6 7
Period
9 10 11 12
Figure 2.5 Time Pattern for Water Usage
2.3 Hydraulic Simulation Model
The hydraulic model used by EPANET is an extended period hydraulic simulator
that solves the following set of equations for each storage node s (tank or
reservoir) in the system:
Bt A,
(6)
(7)
(8)
along with the following equations for each link (between nodes i and j) and
each node k:
12
-------
(9)
where the unknown quantities are:
ys = height of water stored at node s, ft
qa = flow into storage node s, cfs
qij = flow in link connecting nodes i and j, cfs
h; = hydraulic grade line elevation at node i (equal to
elevation head plus pressure head), ft
and the known constants are:
As = cross-sectional area of storage node s (taken as infinite for
reservoirs), ft2
Es = elevation of node s, ft
Qt = flow consumed (+) or supplied (-) at node k, cfs
f( •) = , functional relation between head loss and flow in a link
Equation (6) expresses conservation of water volume at a storage node while
Equations (7) and (10) do the same for pipe junctions. Equation (9) represents
the energy loss or gain due to flow within a link. For known initial storage node
levels ys at time zero, Equations (9) and (10) are solved for all flows qy and heads
h; using Equation (8) as a boundary condition. This step is called "hydraulically
balancing" the network, and is accomplished by using an iterative technique to
solve the nonlinear equations involved.
The method used by EPANET to solve this system of equations is known as the
"gradient algorithm"1 and has several attractive features. First, the system of
linear equations to be solved at each iteration of the algorithm is sparse,
symmetric, and positive-definite. This allows highly efficient sparse matrix
techniques to be used for their solution2. Second, the method maintains flow
continuity at all nodes after its first iteration. And third, it can readily handle
pumps arid valves without having to change the structure of the equation matrix
when the status of these components changes.
ini, E. and Pilati, S., "A gradient method for the analysis of pipe networks",
International Conference on Computer Applications for Water Supply and Distribution, Leicester
Polytechnic, UK, September 8-10, 1987.
2George, A. and Liu, J. W-H., Computer Solution of Large Sparse Positive Definite Systems,
Prentice-Hall, Inc., Englewood Cliffs, NJ, 1981.
13
-------
After a network hydraulic solution is obtained, flow into (or out of) each storage
node, q, is found from Equation (7) and used in Equation (6) to find new storage
node elevations after a time step dt. This process is then repeated for all
subsequent time steps for the remainder of the simulation period.
The normal hydraulic time step used in EPANET is 1 hour, but can be made
shorter if more accuracy is needed. Shorter time steps than normal can occur
automatically whenever pipe or pump controls are activated (e.g., a tank fills to
the level that causes a pump to shut off), or when a tank becomes either empty
or full (causing the tank outlet/inlet line to be closed).
2.4 Water Quality Simulation Model
EPANET's dynamic water quality simulator tracks the fate of a dissolved
substance flowing through the network over time. It uses the flows from the
hydraulic simulation to solve a conservation of mass equation for the substance
within each link connecting nodes i and j:
(ID
dt
where cy = concentration of substance in link i,j as a function
of distance and time (i.e., Cy = Cjj(xy,t)), mass/ft3
Xy = distance along link i,j, ft
qij = flow rate in link i,j at time t, cfs
Ay = cross-sectional area of link i,j, ft2
flfej) = rate of reaction of constituent within link i,j,
mass/ftVday
Equation (11) must be solved with a known initial condition at time zero and the
following boundary condition at the beginning of the link, i.e., at node I where
x:= = 0:
c&(0,f) = _*_ . (12)
14
-------
The summations are made over all links k,i that have flow into the head node (i)
of link i j, while L^ is the length of link k,i, Mj is the substance mass introduced
by any external source at node i, and Qsi is the source's flow rate. Observe that
the boundary condition for link i,j depends on the end node concentrations of all
links k,i that deliver flow to link ij. Thus Equations (11) and (12) form a
coupled set of differential/algebraic equations over all links in the network.
EPANET solves these equations by a numerical scheme called the Discrete
Volume Element Method (DVEM)3. Within each hydraulic time period when
flows are constant, DVEM computes a shorter water quality time step and divides
each pipe into a number of completely mixed volume segments. Within each
water quality time step, the material contained in each pipe segment is first
transferred to its adjacent downstream segment. When the adjacent segment is
a junction node, the mass and flow entering the node is added to any mass and
flow already received from other pipes. After this transport step is completed for
all pipes, the resulting mixture concentration at each junction node is computed
and released into the head end segments of pipes with flow leaving the node.
Then the mass within each pipe segment is reacted. This sequence of steps is
repeated until the time when a new hydraulic condition occurs. The network is
then re-segmented and the computations are continued.
The water quality time steps used in the method are chosen to be as large as
possible without causing any pipe's flow volume to exceed its physical volume
(i.e., have mass transported beyond the end of the pipe). Thus the water quality
time step dt^ cannot be larger than the shortest time of travel through any pipe
in the network, i.e.:
V..
dt = Min —£ for all pipes ij (13)
q
where V^ is the volume of pipe i,j and qy is its flow rate. Pumps and valves are
not included in this determination since transport through them is assumed to
occur instantaneously. Under this water quality time step, the number of volume
segments in each pipe (n;j) is:
3Rossman, L.A., Boulos, P.P., and Altman, T., "Discrete Volume-Element Method for
Network Water-Quality Models", Jour, Water Resource Planning and Management, Vol. 119,
No. 5, September/October 1993.
15
-------
(14)
where INT[x] is the largest integer less than or equal to x. EPANET limits dt,,
to be no smaller than a user-adjustable time tolerance, so that solution times and
the number of segment volumes dp not become excessive. The default value of
this tolerance is 1/10 the length of the hydraulic time step. In addition the user
can specify a maximum number of volume segments that any single pipe can be
divided into. The default value for this parameter is 100.
2.5 Reaction Rate Model
Equation (11) of EPANET's water quality model provides a mechanism for
considering the loss (or growth) of a substance by reaction as it travels through
the distribution system. Reaction can occur both within the bulk flow and with
material along the pipe wall. EPANET models both types of reactions using first
order kinetics. In general, within any given pipe, material in the bulk flow will
decrease at a rate equal to:
0(c) = -kbc - -J-
R
(15)
H
where
RH
- first-order bulk reaction rate constant, I/sec
= substance concentration in bulk flow, mass/ft3
= mass transfer coefficient between bulk flow and
pipe wall, ft/sec
= hydraulic radius of pipe (pipe radius/2), ft
cw = substance concentration at the wall, mass/ft3
The first term in this equation models bulk flow reaction, while the second term
which includes a new unknown cw, represents the rate at which material is
transported between the bulk flow and reaction sites on the pipe wall. Assuming
that the rate of reaction at the wall is first order with respect to c, and that it
proceeds at the same rate as material is transported to the wall (so that no
accumulation occurs), we can write the following mass balance for the wall
reaction:
kf(c - cw) = kwcw (16)
where k, is a wall reaction rate constant with units of ft/sec. Solving for c and
substituting into Equation (15) results in the following reaction rate expression:
16
-------
0(c) = -Kc
(17)
where K is an overall first order rate constant equal to:
K = kb +
(18)
The above discussion pertains to substance decay, with mass transfer from the
bulk flow to the pipe wall. Dropping the negative sign in front of K in Equation
(17) would model the growth of a substance, with mass transfer from the pipe
wall to the bulk flow.
To summarize, there are three coefficients used by EPANET to describe reactions
within a pipe. The pipe's bulk rate constantly and its wall rate constant kw must
be determined empirically and supplied as input to the model. The mass transfer
coefficient kf is calculated internally by EPANET using the dimensionless
Sherwood Number as follows4:
ShD
d
(19)
Sh = 0.023Re°*3Sc0333 for Re > 2300
(20)
where
Sh = 3.65 +
.Sh
Re
Sc
d
L
q
A
D
.Q668(d/L)ReSc
.04l_(d/L)ReSc]
.67
for Re < 2300
(21)
mass transfer coefficient, ft/sec
Sherwood Number
Reynolds Number (q d / A / v)
Schmidt Number (v I D)
pipe diameter, ft
pipe length, ft
flow rate, cfs
cross-sectional flow area of pipe, ft2
molecular diffusivity of substance in fluid, ftVsec
kinematic viscosity of fluid, ftVsec
4Edwards, D.K., Denny, V.E., and Mills, A.F., Transfer Processes, McGraw-Hill, New
York, NY, 1976.
17
-------
Equation (20) applies to turbulent flow where the mass transfer coefficient is
independent of the position along the pipe. For laminar flow, Equation (21)
supplies an average value of the mass transfer coefficient along the length of the
pipe.
2.6 Water Age and Source Tracing
In addition to chemical transport, EPANET can also model the changes in age of
water over time throughout a network. To accomplish this, the program
interprets the variable c in Equation (11) as the age of water and sets the reaction
term 0(c) in the equation to a constant value of 1.0. During the simulation, any
new water entering the network from reservoirs or source nodes enters with age
of zero. Water age provides a simple, non-specific measure of the overall quality
of delivered drinking water. When the model is run under constant hydraulic
conditions, the age of water at any node in the network can also be interpreted
as the time of travel to the node.
EPANET can also track over time what percent of water reaching any node in the
network had its origin at a particular node. In this case the variable c in Equation
(11) becomes the percentage of flow from the node in question and the reaction
term is set to zero. The c-value of the source node is kept at 100 percent
throughout the duration of the simulation. The source node can be any node in
the network, including storage nodes. Source tracing is a useful tool for
analyzing distribution systems drawing water from two or more different raw
water supplies. It can show to what degree water from a given source blends
with that from other sources, and how the spatial pattern of this blending changes
over time.
18
-------
CHAPTER 3
INSTALLATION
3.1 System Requirements
To install EPANET to run under both Windows and DOS requires the following:
» an IBM-compatible PC equipped with an 80286 or higher CPU
» MS-DOS or PC-DOS version 3.0 or later
« Microsoft Windows version 3.0 or later
« at least 3 megabytes of free disk space
To install EPANET for DOS only requires a PC equipped with an 8088 or higher
CPU running MS-DOS or PC-DOS version 3.0 or later. Although not required,
a math co-processor is highly recommended.
3.2 Installation For Both Windows and DOS
To install EPANET so that it will run under either Windows or DOS:
1. Place the distribution disk in a floppy disk drive (drive A or B).
2a. If Windows is not running, enter the command
WINA:SETUP
at the DOS prompt (use WIN B:SETUP instead if the distribution disk is
in drive B).
2b. If Windows is already running, select Run from the File menu of the
Program Manager, type A:SETUP in the Command Line box, and select
the OK button. (Type BrSETUP instead if the distribution disk is in drive
B).
3. If you wish to install EPANET in a directory other than C:\EPANET
enter the full path name for your directory choice in the dialog box that
19
-------
appears.
4. Select the continue button to resume the installation.
5. When the installation is completed, a README.TXT file will be
displayed in the Windows Notepad program informing you of any
modifications to the users manual and how to further customize your
installation.
After a successful installation, a new Program Group named EPANET will be
added to your Program Manager, with an EPANET icon in it.
Note: If EPANET fails to install properly, try the following:
Shut down all other Windows applications that may be running (such as
the Clock) and launch the setup program again from Program Manager.
Exit from Windows, change directories to the Windows directory
(typically c:\windows), and repeat the installation procedure.
If using Windows for Workgroups, exit from Windows, re-start Windows
in Standard mode, and repeat the installation procedure.
Appendix A lists the various files that are installed on your system. If for some
reason you need to re-install the entire EPANET system or only a portion of it,
repeat the entire installation procedure as described above. Do not try to copy
individual files off of the distribution disk since several of these files are in a
compressed format.
3.3 Installation For DOS Only
If your system is not equipped with Windows, you can install a version of
EPANET that will run only under DOS as follows:
1. Place the distribution disk in a floppy disk drive (either drive A or B).
2. Enter the following command at the DOS prompt:
A:SETUP4D
20
-------
3.
5.
(or B:SETUP4D if the disk is in drive B).
When prompted, enter the full path name of your choice for an EPANET
directory, or simply hit the Enter key to accept C:\EPANET.
Verify your choice of an EPANET directory by hitting the 'y' key.
Hitting the Escape key will cancel the installation, while any other key
will ask you to re-specify an EPANET directory.
The relevant files will be copied from the installation disk to your
EPANET directory.
Appendix A also lists those files installed for DOS operation only.
3.4 Customizing Your Installation
Running in 32-Bit Mode
EPANET's network simulator comes in both a standard version and 32-bit
version. The latter works only on PC's equipped with an 80386 or higher CPU
and when run under Windows, in Windows' 386 Enhanced mode. It runs
simulations about twice as fast and makes all of the machine's extended memory
available to the program. The standard version of the simulator can only use a
maximum of 640 kilobytes of conventional memory. Because the program size
of the 32-bit version of EPANET is about 260 kilobytes larger than the standard
version, we recommend that you use it only if you have at least this amount of
extended memory available on your machine.
If you want to run EPANET in 32-bit mode, you must modify the
AUTOEXEC.BAT file in your root directory. Using any text editor or word
processor that saves its output in ASCII format, add the following line to this file:
SETEPANET=32
(Note there are no spaces on either side of the equal sign.) After re-booting your
machine, EPANET will automatically run in 32-bit mode. To return to standard
mode, simply remove this line from your AUTOEXEC.BAT file and re-boot your
machine.
21
-------
Using a Different Editor
The Windows version of EPANET comes with a public domain text editor that
is used to edit input data files. It is a DOS program named TE.EXE that
EPANET runs within a DOS window. If you wish, you can substitute a different
editor of your choosing. It can be either a DOS or Windows program that can
accept the name of the file to edit on its command line, and is capable of editing
large files. This would eliminate such programs as the Windows Notepad which
has a file size limit of only 32 kilobytes.
To replace the editor that ships with EPANET with another DOS editor (such as
EDIT.COM that comes with MS-DOS 5.0 and later):
Use the Windows PIF Editor to change the settings in the EDITOR.PIF
file in your EPANET directory. For example, to switch to the MS-DOS
editor which resides in a directory named C:\DOS, use the following
settings:
Program Filename: C:\DOS\EDIT.COM
Window Title: EDITOR
Startup Directory:
To replace the default editor with a Windows editor: '
Use any text editor to add the following section to the file EPANET.INI
in your Windows directory (if this file doesn't exist then create it):
[EDITOR]
Program =< program name >
Caption =< window title >
where < program name> is the full path name of the editor program
(e.g., C:\EDITORS\WDSfEDIT.EXE) and < window title > is the portion
of the caption that always appears in the editor's main window (e.g.,
WINEDIT).
22
-------
CHAPTER 4
INPUT DATA FORMATS
4.1 Data Preparation
Prior to running EPANET, the following initial steps should be taken for the
network being studied:
1. Identify all network components and their connections. Network
components consist of pipes, pumps, valves, storage tanks and reservoirs.
The term "node" denotes a junction where network components connect
to one another. Tanks and reservoirs are also considered as nodes. The
component (pipe, pump or valve) connecting any two nodes is termed a
"link".
2. Assign unique ID numbers to all nodes. ID numbers must be between 1
and 2147483647, but need not be in any specific order nor be consecutive.
3. Assign an ID number to each link (pipe, pump, or valve). It is
permissible to use the same ID number for both a node and a link.
4. Collect information on the following system parameters:
a. diameter, length, roughness and minor loss coefficient for each
pipe,
b. characteristic operating curve for each pump,
c. diameter, minor loss coefficient and pressure or flow setting for
each control valve,
d. diameter and lower and upper water levels for each tank
e. control rules that determine how pump, valve and pipe settings
change with time, tank water levels, or nodal pressures,
f. changes in water demands for each node over the time period
being simulated
g. initial water quality at all nodes and changes in water quality over
time at source nodes.
With this information in hand, you are now ready to construct an input file to use
with EPANET.
23
-------
4.2 Input File Organization
EPANET receives its input data from a file whose contents are divided into
several different sections. Each section begins with a specific keyword in
brackets. Figure 4.1 provides an example EPANET input file. (Any text
appearing after a semicolon is a comment added to enhance readability.) The
keywords and the categories of input data they represent are:
[TITLE]
[JUNCTIONS]
[TANKS]
[PIPES]
[PUMPS]
[VALVES]
[REPORT]
[STATUS]
[CONTROLS]
[PATTERNS]
[TIMES]
[QUALITY]
[SOURCES]
[REACTIONS]
[OPTIONS]
[DEMANDS]
[ROUGHNESS]
[END]
problem title
junction node information
tank/reservoir information
pipe information
pump information
valve information
output report format
initial status of selected links
link control rules
water demand and source strength time patterns
simulation time step parameters
initial water quality in network
baseline contaminant source strength
reaction rate coefficients
miscellaneous analysis options
changes in baseline water demands
changes in pipe roughness coefficients
signals end of input file
The only mandatory sections are [JUNCTIONS], [TANKS], and [PIPES]. The
order of sections is not important, except that whenever data in a section
references a node, that node must have already been defined in the
[JUNCTIONS] or [TANKS] sections. The same holds true for any reference to
a link (pipe, pump or valve). To be safe then, you should place the [TITLE],
[JUNCTIONS], [TANKS], [PIPES], [PUMPS], and [VALVES] sections first.
Each section can contain one or more lines of data. Blank lines may appear
anywhere in the file and the semicolon (;) can be used to indicate that what
follows on the line is a comment, not data. Data items can appear in any column
of a line, but a line cannot contain more than 80 characters. Observe how in
Figure 4.1 these features were used to create a tabular appearance for the data,
complete with column headings.
24
-------
ETITLE3
SPAMEI Exanple Network 1
{JUNCTIONS}
; Elevation Demand
MO ft . gpro
* * --------
10 710 Q
11 ' 710 150
12 700 150
13 695 100
21 700 150
22 695 200
23 690 150
3t 700 100
32 710 100
ETAMKSJ
; Elev. Init. Hin.
; 10 ft Level Level
2 850 120 100
9 800
EPIPES3
; Mead Tait Length
;!D Node Node ft
10 10 11 10530
11 11 12 5280
12 12 13 5280
21 21 22 5280
22 22 23 5280
31 31 32 5280
110 2 12 200
111 11 21 5280
112 12 22 5280
113 13 23 5280
121 21 31 5280
122 22 32 5280
-"-
*
Max.
Levfet
150
Diam.
in
18
U
10
10
12
6
1S
10
12
8
8
6
"
'
f*. V '
;
" * I—
f *
- -
Diam,
ft
50.5
^ ^ -,
Rough.
Coef*.
100
100
too
1QO
100
100
100
100
100
100
100
100
Figure 4.1 Example EPANET Input Data
(Continued on Next Page)
25
-------
[PUMPS]
«"•""*.--..-»„»„...
; Head Tail Design H*tJ
;10 Mode Mode ft gpm
'9 9 10 250 1500
CONTROLS}
»-.w.-..---»^,,^». ,.---_.*,,.„»_.,.__.,.,.»,,.
; Pump 9 is on when Tank 2 Level < 110 ft
; Pump. 9 is off When Tank 2 level > 140 ft
\im 9 OPEN IF NODE 2 BELOW 110
LtHK 9 CLOSED IF HODE 2 ABOVE f40
IPATTERNSJ
; ID Multipliers
' 1 1.0 1.2 1,4 1.6 1-4 1.2
1 1.0 0.& 0.6 0.4 0.6 O.S
[QUALITY]
**""""""" ---^ .....^
; Initial
;Wodes Concen. mg/L
' 2 32 0.5
9 1.0
2 1.0
[REACTIONS]
^GLOBAL BULfc -.5 ; Sulk decay caeff.
fiLOBAL WALL -1 ; Wall decay coeff.
ITtHES] , '
DURATION 24 ; 24 hour sfmulation period
PATTERN TIHESTEP 2 ; 2 hour pattern time period
[OPTIONS]
DUALITY Chlorine ~; Chlorine analysis
MAP Ket1»map j Map coordinates file
CENDJ
Figure 4.1 Continued
26
-------
1
Keywords can appear in mixed lower and upper case. Unless specifically noted,
the default units for all data are as follows:
Length
Pressure
Flow
Concentration
feet
pounds per square inch (psi)
gallons per minute (gpm)
milligrams per liter (mg/L)
An option is available in the [OPTIONS] section to change the flow units to either
cubic feet per second (cfs), million gallons per day (mgd), or liters per second
(L/s). In the latter case, SI (metric) units would apply to all quantities, so lengths
and pressures would be expressed in meters (m). Concentration units can also be
changed in the [OPTIONS] section to any desired measure.
The diagram below illustrates which additional input sections are needed as one
moves from analyzing steady state hydraulics, to extended period hydraulics, on
up to an extended period water quality simulation. Note that some sections, such
as [OPTIONS], might appear in all three types of runs.
Steady-State
Hydraulics
[TITLE]
[JUNCTIONS]
[TANKS]
[PIPES]
[PUMPS]
[VALVES]
[REPORT]
Extended Period
Hydraulics
Extended Period
Water Quality
•> [STATUS]
[CONTROLS]
[PATTERNS]
[TIMES]
--> [QUALITY]
[SOURCES]
[REACTIONS]
[OPTIONS]
4.3 Input File Format
A detailed description of the data in each section of the input file will now be
given in alphabetical order. Each section begins on a new page. Mandatory
keywords are shown in boldface while optional items appear in parentheses.
27
-------
Section: [CONTROLS]
Purpose:
Allows pump, valve and pipe settings to change at specific times or when
specific pressures or tank water levels are reached in the network.
Formats:
LINK linkID setting AT TIME rvalue (units)
LINK linkID setting IF NODE nodelD BELOW level
LINK linkID setting IF NODE nodelD ABOVE level
Parameters:
linkID
setting
tvalue
units
nodelD
level
ID number of a pump, valve or pipe
link setting which can be:
- a pump status (either OPEN or CLOSED),
- a pump speed (relative to the speed used to define the
pump's characteristic curve in the [PUMPS] section),
- a valve setting (pressure, flow or loss coefficient)
or status (either OPEN or CLOSED)
- a pipe status (either OPEN or CLOSED)
time at which change in link setting applies
optional units on control time which can be:
SECONDS (or SEC),
MINUTES (or MEM),
. HOURS(default),
DAYS (or DAY)
ID of controlling node
control action level (either water level above tank bottom
if control node is a tank, ft (m), or pressure level if control
node is a junction node, psi (m).
Remarks:
Use one line for each control rule. A link can be subject to more than
one control rule. Control can be based on time, on water levels in tanks
(not elevations), or on pressures at junction nodes.
28
-------
Section: [CONTROLS] (continued)
The first format implements the control at the designated time. Times can
be expressed as hours: minutes or as a decimal value. In the latter case,
the default units are hours.
The second format implements the control when a specified node drops
below its action level while the third format applies the control when the
node rises above the action level.
Settings for constant-speed pumps are either OPEN (pump on) or
CLOSED (pump off). You can simulate a variable speed pump by
specifying a speed factor for its setting. When the factor is 0, the pump
is off. When it is 1, the pump operates on its original characteristic
curve. Other speed factor values shift the position of the characteristic
curve as described in Chapter 2.
Control valve settings can be either numerical values or OPEN/CLOSED.
The only allowable settings for pipes are OPEN or CLOSED. If a pipe
is closed, EPANET assumes the existence of a valve somewhere in the
line. DO NOT specify a valve in the [VALVES] section to accomplish
this purpose.
Examples:
;Open pump 23 when the water level in the tank at node 45 drops ;below
23ft and close it when the level rises above 36ft.
LINK 23 OPEN IF NODE 45 BELOW 23
LINK 23 CLOSED IF NODE 45 ABOVE 36
;Close pipe 245 at 3.2 hours into the simulation
LINK 245 CLOSED AT TIME 3.2
;Drop the speed of pump 1 to half its normal level when the pressure at
;node 10 goes above 75 psi
LINK 1 0.5 IF NODE 10 ABOVE 75
29
-------
Section: [DEMANDS]
Purpose:
Provides an alternative to the [JUNCTIONS] section for entering baseline
nodal demand flows.
Format:
MULTIPLY value
node demand (pattern)
Parameters:
value
node
demand
pattern
Remarks:
multiplier value
node ID
baseline demand flow at the node (negative if there is
external flow into the node)
optional ID of time pattern defined in [PATTERNS] section
The first format multiplies each baseline demand specified previously in
the [JUNCTIONS] section by a given amount. The second format is used
for individual nodes whose demands are to be specified.
This optional section can be used to specify nodal demands and their time
patterns, instead of both elevation and demand as in the [JUNCTIONS]
section. Any node referenced in this section must have been previously
defined in the [JUNCTIONS] section.
Unless explicitly specified either here or in the [JUNCTIONS] section, a
node's demand is zero and it is assigned time pattern 1 by default.
Storage nodes (tanks/reservoirs) should not have demands or external
inflows assigned to them.
Example:
12 245 3 ;Node 12 has a baseline demand of 245 gpm which varies
;over time according to time pattern 3
34 -450 ;Node 34 has an external baseline inflow of 450 gpm
30
-------
Section: [JUNCTIONS]
Purposes
Identifies elevations and, optionally, baseline demands and demand
patterns for all junction nodes in the system.
Formats;
id elev (demand) (pattern)
Parameters:
id
elev
demand
pattern
Remarks:
node ID
node elevation, ft (m)
optional baseline demand flow (negative for external source
flows into the node)
optional ID of time pattern defined in PATTERNS!
section
One line should appear for each junction node, excluding tanks and
reservoirs.
If you want to specify a time pattern ID for a node you must first specify
the baseline demand ahead of it.
If not specified otherwise, a node's demand is zero and it is assigned time
pattern 1 by default.
A [JUNCTIONS] section is required.
Examples:
101 124 >'Node 101 is at elevation of 124 ft.
123 245 56 -Node 123 is at elevation of 245 ft. and has a base
;demand of 56 gpm.
34 102 -245 2 ;Node 34 is at elevation of 102ft. and has a base
,-inflow to the system of 245 gpm which changes
;with time according to pattern number 2
31
-------
Section: [OPTIONS]
Purpose:
Provide values for various network properties and simulation options.
Formats:
UMTS
HEADLOSS
HYDRAULICS SAVE
HYDRAULICS USE
VERIFY
MAP
QUALITY
SPECIFIC GRAVITY
VISCOSITY
DIFFUSIVITY
TRIALS
ACCURACY
SEGMENTS
option
option
filename
filename
filename
filename
option (units)
value
value
value
value
value
value
Parameters:
option
filename
value
Remarks:
a choice from a fixed set of options
name of a file
numeric value
You need only set option values for items whose default values you wish
to change.
UNITS sets the units in which flows are expressed. Choices are:
GPM (gallons per minute - the default)
CFS (cubic feet per second)
MGD (million gallons per day)
SI (liters per second)
SI units require that metric units (length in meters, pressure in meters,
etc.) be used for all other input as well. The other flow units choices
maintain use of English units (length in feet, pressure in psi, etc.)
throughout.
32
-------
Section: [OPTIONS] (continued)
HEADLOSS selects the pipe headless formula used to calculate system
hydraulics. The available choices are:
H-W (Hazen-Williams formula - the default)
D-W (Darcy-Weisbach formula)
C-M (Chezy-Manning formula) •
Note that each of these formulas employs a different type of roughness
coefficient.
HYDRAULICS SAVE is used to identify a file in which the run's
hydraulic solution will be saved. This solution can then be used in future
runs (using the option defined below) that focus exclusively on water
quality, thereby making the calculations go quicker.
HYDRAULICS USE names a file from which a previously saved
hydraulics solution will be retrieved/thus avoiding the need to re-compute
hydraulics for the current run.
VERIFY identifies the name of a file used to verify that the network
connectivity implied by the entries in the [PIPES], [PUMPS], and
[VALVES] sections is correct. See Section 4.4 for details on the format
of this file.
MAP identifies the name of a file used to store map coordinates and labels
that will be displayed when running EPANET for Windows. See Section
4.5 for details on the format of this file.
QUALITY specifies the type of water quality analysis to make. The
choices are:
NONE (no water quality analysis - the default)
CHEMICAL (compute chemical concentration)
AGE (compute water age)
TRACE nodelD (compute the fraction of water originating from the
specified node)
As an alternative to the keyword CHEMICAL you can use the actual
name of the chemical whose concentration will be tracked (such as
CHLORINE) so that this name will appear in all output reports. In
addition, you can supply the name of the units (e.g., ug/L) in which its
concentration will be measured (the default units are mg/L). No water
quality analysis will be performed if the simulation duration is 0 hours
(i.e., steady state).
33
-------
Section: [OPTIONS]
(continued)
SPECIFIC GRAVITY is the weight per unit volume of the fluid being
modelled relative to that of water. The default value is 1.0.
VISCOSITY is the kinematic viscosity of the fluid at the temperature
condition being simulated. The units of viscosity are ft2/sec (or m2/sec for
SI units). The default value is 1. IxlO'5 ftVsec, corresponding to water at
20 degrees C. Viscosity is used only when the Darcy-Weisbach headless
formula is employed or when a pipe wall reaction mechanism is included
in the water quality analysis.
DIFFUSIVITY is the molecular diffusivity of the chemical being tracked.
It has English units of ft2/sec and SI units of m2/sec. The default value
is l.SxlQ-8 ft2/sec. This is the diffusivity of chlorine in water at 20
degrees C. Diffusivity is used only when pipe wall reactions are
considered in the water quality analysis.
TRIALS is the maximum number of iterations that EPANET should
employ when solving the network hydraulic equations at each time step of
the simulation. The default value is 40.
ACCURACY prescribes a convergence criterion for the iterative method
used to solve the network's hydraulic equations. The iterations end when
the sum of absolute flow changes in all links divided by the total flow in
all links is less than the ACCURACY value.. The default for this
parameter is 0.001. EPANET will not allow a value smaller than Ifr5 to
be used.
SEGMENTS specifies the maximum number of segments that any pipe
can be divided into during the water quality analysis. The default value
is 100. The number of pipes reaching this limit can be reported on by
asking for a status report in the [REPORT] section.
Examples:
UNITS MOD
HEADLOSS D-W
QUALITY TRACE 12
HYDRAULICS SAVE test.hyd
ACCURACY .0005
SEGMENTS 300
;MGD flow units
;Darcy-Weisbach head loss formula
;Track % of flow from node 12
;Save hydraulic results
Increase the accuracy of
;hydraulic and water quality results
34
-------
Section: PATTERNS]
Purpose:
Describes how water demands and external source concentrations change
ovftr nmf» &
over time.
Format:
pattern multl mult2 ....
Parameters:
pattern
multl,
mult2,
etc.
pattern ID number
multipliers applied to baseline values in consecutive time
periods
Remarks:
Use as many lines as it takes to describe all time patterns. If a given time
^ — * Sta* «* ™ **'
A time pattern consists of a collection of multipliers that are applied to a
^ T^l S°UrCe concentration «ver a sequence of consecutive
periods. Within a time period, the demand (or source concentration)
^ t0 the muWPlier ^es the baseline vSue
length is -1 hour' but «** can be
There is no limit on the number of patterns or multipliers per pattern
However system memory is conserved if you number the patterns in
SS^rfT8 wltY' If ^ duration of a pattem is less th^ the ^
duration of the simulation, the pattern will repeat itself. For example
Where
35
-------
Section: [PATTERNS] (continued)
Nodes are assigned to specific demand time patterns in the [JUNCTIONS]
or [DEMANDS] sections, and to source concentration time patterns in the
[SOURCES] section. Unless assigned to a specific pattern in one of these
sections, the default pattern associated with a node's demand is pattern 1.
For contaminant sources, the default is to have no pattern, meaning there
will be no time variation in source strength.
By default, all multipliers for all time periods in all patterns are 1. Thus
if you do not define any time patterns nor make any assignment of nodes
to time patterns, demands and source concentrations will never change
from their baseline levels during the course of the simulation.
Examples:
1 1.1 1.3 1.5
1 1.1 .87
2 0 150 0 0 0
;Pattern 1 extends over 5 time periods with
;multipliers ranging from .87 to 1,5
; Pattern 2 is used to describe afire flow condition
;that exists in time period 2 at a node that
;otherwise would not have any demand
36
-------
Section: [PIPES]
Purpose:
Provides a description of each pipe in the network.
Formats:
id nodel node2 length diam rcoeff (Icoeff) (CV)
Parameters:
id link ID
nodel ID of beginning node
node2 ID of ending node
length pipe length, ft (m)
diam pipe diameter, inches (mm)
rcoeff roughness coefficient, unitless for Hazen-Williams or
Chezy-Manning headloss formulas, millifeet (mm) for
Darcy-Weisbach formula
Icoeff optional minor loss coefficient (0 if not specified)
C V used if the pipe contains a check valve.
Remarks:
One line should appear for each pipe.
A headloss formula can be specified in the [OPTIONS] section. The
default is the Hazen-Williams formula. Note that the units of a Darcy-
Weisbach roughness coefficient are millifeet (or mm for SI units).
A [PIPES] section is required.
Examples:
12 34 45 1200 8 120 ;Pipe with Hazen-Williams roughness
Coefficient
123 34 65 2500 10 .8 5 ;Pipe with Darcy-Weisbach roughness and
;minor loss coefficient
56 12 7 1200 8 120 CV ;Pipe with check valve preventing flow
from Node 7 to 12
37
-------
Section: [PUMPS]
Flirpose:
Describes each pump in the network and its characteristic curve.
Formats:
id nodel node2 hp
id nodel node2 hi ql
id nodel node2 hO hi ql h2 q2
id nodel node2 hO hi ql h2 q2 q3
Parameters:
id
nodel
node2
hp
hO
hl,ql
h2, q2
q3
Remarks:
link ID
ID of node on inlet side of pump
ID of node on discharge side of pump
pump power rating, hp (kw)
shutoff head, ft (m)
design head, ft (m), and design flow
head, ft (m) and flow at upper end of normal operating
flow range
maximum flow in extended flow range
One line of either format should appear for each pump.
The first format is used for pumps where the characteristic curve is
unknown and a constant power output is assumed.
The second format is used for a "standard" pump curve with no extended
flow range, where the cutoff head is 133% of the design head and the
maximum flow is twice the design flow.
The third format describes a pump curve with no extended flow range.
38
-------
Section: [PUMPS] (continued)
The last format describes a characteristic curve with an extended flow
range. If q3 equals q2, then the extended flow range begins at q2 and
follows the slope of the pump curve at this point until zero head is
reached. If q3 is greater than q2, then the extended curve is a straight line
between these two flows.
Graphical examples of the input requirements for the various types of
pump curves are shown in Figure 4.2 on the following page.
Unless modified in the [STATUS] or [CONTROLS] sections, all pumps
are assumed to be operating throughout the simulation.
The program automatically prevents reverse flow through a pump, and
issues warning messages when a pump is operating out of its normal
range.
Variable speed pumps can be modelled by establishing speed settings in
the [STATUS] and [CONTROLS] sections. See descriptions of these
sections for details.
Examples:
101 1 2
102
193
200 ; Constant horsepower pump
3 4 50 1200 ; Standard pump curve
5 6 100 75 1000 50 1200 1500
; Custom curve with extended range
39
-------
Standard Curve
No Extended
Flow Range
h1,q1
Flow
Flow
Custom Curva,
Standard Extended
Flow Range
h1,q1
k H2,q2,q2
Custom Curve,
Custom Extended
Flow Range
q3
Flow
Flow
Figure 4.2 Examples of Pump Curve Input Requirements
40
-------
Section: [QUALITY]
Purpose:
Establishes the initial water quality level at network nodes at the start of
a simulation.
Formats:
nodel (node2) quality
Parameters:
nodel,
node2 node ID's
quality initial quality, (concentration for chemical constituents,
hours for water age, or percent for source tracing)
Remarks:
Use as many lines as needed to specify the initial water quality throughout
the network.
Each line can either specify water quality at a single node or for a range
of nodes.
The units of water quality depend on the type of analysis to be performed:
concentration is used for chemical propagation, hours are for water age,
and percent is used for source flow tracing.
Remember to change the entries in this section when you change the type
of water quality analysis being performed.
If not set in this section, a nodes's initial water quality is assumed to be
0.
Examples:
1 23 1 ;Nodes 1 to 23 have initial quality ofl.
10.1 ;Node 10's initial quality is changed to .1.
41
-------
Section: [REACTIONS]
Purpose:
Specifies reaction rate coefficients.
Formats:
GLOBAL BULK bulkcoeff
GLOBAL WALL wallcoeff
BULK pipel (pipe2) bulkcoeff
WALL pipel (pipe2) wallcoeff
TANK node! (node2) bulkcoeff
Parameters:
bulkcoeff
wallcoeff
pipel,
pipe2
nodel,
node2
bulk rate coefficient, days'1
wall rate coefficient, ft/day (m/day)
pipe ID's
tank ID's
Remarks:
Default reaction rate coefficients are 0 for all pipes and tanks.
GLOBAL sets a single coefficient for either bulk reactions or pipe wall
reactions that applies throughout the network.
BULK and WALL establish reaction coefficients for individual pipes or
groups of pipes in the network. These will override any global
coefficient. J &
TANK establishes a reaction coefficient for individual tanks or groups of
tanks. This will override any global bulk coefficient.
42
-------
Section: [REACTIONS] (continued)
Note that the bulk reaction coefficient has units of days"1 while the wall
reaction coefficient has units of ft/day (or m/day). One way to compare
the relative magnitude of these two types of coefficients is to divide the
wall coefficient by the hydraulic radius of the pipe (i.e., 1/2 the pipe
radius). The resulting quantity will have the same units as the bulk
coefficient, days"1.
Remember to use negative signs on all reaction coefficients that
represent substance decay.
Examples:
GLOBAL BULK-.1
BULK 23 52 -.5
TANK 102 .05
GLOBAL BULK -.5
GLOBAL WALL -1.5
WALL 32 -2.0
;Example 1:
;Global bulk decay throughout network of .1/day,
;modified to .5/day for pipes 23 to 52,
;and a re-growth rate of .051 day in Tank 102
;There is no wall reaction.
•Example 2:
;Global bulk decay throughout network of .5/day.
;Global wall decay of 1.5 ft/day.
;Pipe 32 has watt decay of 2 ft/day.
43
-------
Section: [REPORT]
Purpose:
Describes the contents of the output report.
Formats:
FILE
STATUS
PAGESIZE
NODES
NODES
LINKS
LINKS
variable BELOW
variable ABOVE
filename
option
lines
nodel (node2)
NONE
linkl (Iink2)
NONE
value
value
Parameters:
filename
option
lines
nodel,
node.2
linkl,
Iink2
variable
name of report file
YES, FULL or NO
number of lines per page in output report
node ID numbers
value
link ID numbers
output variable name:
DEMAND
ELEVATION
GRADE
PRESSURE
QUALITY
DIAMETER
FLOW
VELOCITY
HEADLOSS
numerical value
(in flow units)
(in ft or m)
(in ft or m)
(in psi or m)
(in concentration units,
hours, or %)
(in inches or mm)
(in flow units)
(in ft/sec or m/sec)
(in ft/kilo-ft or m/km)
44
-------
Section: [REPORT] (continued)
Remarks:
K If T* ° a teXt ffle t0 Which the outPut reP°* will be
Normally this option would only be used when running under
Windows since under DOS, the output report file name is given on the
command line. Under Windows, the default is not to sa4 the outpS
report to a tile.
STATUS is used to include or exclude system status reports from being
generated every time hydraulic conditions change during the simulation
£fn±fn f "^ thCSe rep0rtS' A FULL status reP°rt *****
information for each solution trial of the, hydraulic equations and is
normally used only for de-bugging hydraulically unbalanced networks.
LINK lines Specify which nodes ^ ^ks should be
in the output report. Use as many lines as it takes to include all
of the items you want reported. The default is to report on all nodes and
links Using NODE NONE or LINK NONE eliminates any node or Ikk
output, respectively. J
The last format tells EPANET to report on those nodes and links whose
computed variables meet a certain criterion. Multiple criteria are
P^SIJ^ OR'S> F°r examPle' PRESSURE BELOW 30
PRESSURE ABOVE 100, and QUALITY BELOW 1 will report ail
nodes that meet one or more of these conditions. This will be in addition
to any nodes or links specified in the NODE and LINK lines.
Examples:
NODE 12
NODE 22 36
PRESSURE BELOW 30
LINK NONE
.-Report on Nodes 12 and
;22 through 36, as well as
;nodes with pressure below 30 psi.
;No links are reported on.
45
-------
Section: [ROUGHNESS]
Purpose:
Provides an alternative to the [PIPES] section for easily adjusting
roughness coefficients for groups of pipes.
Formats:
ADD value
MULTIPLY value
pipel (pipe2) roughcoeff
Parameters:
value
pipel,
pipe2
adjustment value
pipe ID's
roughcoeff roughness coefficient, unitless for Hazen-Williams or
Chezy-Manning headloss formulas, millifeet (mm) for
Darcy-Weisbach formula
Remarks:
The first format adds a given amount to each pipe's roughness coefficient.
Use a negative value to reduce roughnesses.
The second format multiplies each pipe's roughness coefficient by a given
amount.
The third format is used repeatedly for each pipe or group of pipes whose
roughness coefficients will be set to a specific value.
This section allows you to easily modify roughness coefficients without
having to change them individually in the [PIPES] section. This can
prove especially useful during calibration runs.
Example:
ADD -5
1252 110
;Subtract 5 from the roughness coefficient of each pipe.
;Set roughness coefficient of pipes 12 through 52 to 110.
46
-------
Section: [SOURCES]
Purpose:
Assigns baseline concentrations to nodes that serve as sources of chemical
constituents into the network.
Format:
node concen (pattern)
Parameters:
node source node ID
concen baseline concentration of constituent entering the node as
an external source
pattern optional ID of time pattern defined in [PATTERNS] section
Remarks:
Use one line for each node that serves as a constituent source.
The data in this section pertain only to water quality analyses for
chemicals, not to water age or source tracing.
For junction nodes, if there is no external source inflow (negative
demand) assigned to the node, then the quality at the node always equals
the source quality. Use this feature to simulate satellite treatment such as
chlorine booster stations.
The optional time pattern ID identifies which time pattern defined in the
[PATTERNS] section will be used to vary the source strength about its
baseline level over time. If omitted, there is no variation in source
strength over time.
Example:
102 100 3 ;Node 102 has a baseline source strength of 100 mg/L
;which varies over time according to pattern 3
47
-------
Section: [STATUS]
Purpose:
Establishes initial setting of selected links at the start of the simulation.
Format:
linkl (Iink2) setting
Parameters:
linkl, Iink2 link ID's
setting link setting at the start of the simulation, which can be:
- a pump status (either OPEN or CLOSED),
- a pump speed (relative to the speed used to define the
pump's characteristic curve in the [PUMPS] section),
- a valve setting (pressure, flow or loss coefficient)
or status (either OPEN or CLOSED)
- a pipe status (either OPEN or CLOSED)
Remarks:
Use one line for each link or range of links whose initial setting you wish
to specify.
Normally all pipes are open, all pumps are on (with a speed setting of 1),
and all valves are at their original pressure, flow or loss coefficient
settings at the start of the simulation. To change a link's status at some
future, point during the simulation, you would use a rule in the
[CONTROLS] section.
You cannot set the status of a pipe containing a check valve.
Example:
24 CLOSED
1 4 1.2
;Close off pipe 24 at the start of the simulation
;Begin the simulation with pumps 1 through 4
;operating at 1.2 times their normal speed
48
-------
Section: [TANKS]
Purpose:
Describes each storage tank or reservoir in the network.
Formats:
node elev (initlevel minlevel maxlevel diam (minvol))
Parameters:
node
elev
initlevel
minlevel
maxlevel
diam
minvol
node ID
bottom elevation of tank where water level is zero, ft (m)
initial water level above tank bottom, ft (m)
lowest allowable water level, ft (m)
highest allowable water level, ft (m)
tank diameter, ft (m)
volume of water below minimum level, ft3 (m3).
Remarks:
One line should appear for each tank or reservoir.
For reservoirs you need only enter the node ID and elevation. By
definition, the water surface elevation in a reservoir remains fixed while
for tanks it varies as flow enters or leaves.
Water surface elevation in tanks equals the bottom elevation plus water
level. Tanks are assumed to be cylindrical between their minimum and
maximum levels. Non-cylindrical bottom sections can be accommodated
by supplying the volume of the section as the last parameter on the line
(minvol). See Fig. 4.3 for a description of tank levels. For non-circular
tanks, use a diameter equal to 1.128 times the square root of the area.
A [TANKS] section is required.
Examples:
101 423 15 10 20 45 ;45 ft diameter tank with variable water
;surface level
102 390 .-Reservoir with fixed water surface
49
-------
Minimum
Level -
Minimum
Volume
Maximum
TT- Level
Elevation
Datum
Figure 4.3 Definition of Tank Levels
50
-------
Section: [TIMES]
Purpose:
Describes various time step parameters used in the simulation.
Formats:
DURATION value
HYDRAULIC TIMESTEP value
QUALITY TIMESTEP value
MINIMUM TRAVELTIME value
PATTERN TIMESTEP value
REPORT TIMESTEP value
REPORT START value
(units)
(units)
(units)
(units)
(units)
(units)
(units)
Parameters:
value
units
a time value
optional time units which can be:
SECONDS (or SEC)
' MINUTES (or MIN)
HOURS (default)
DAYS (or DAY)
Remarks:
You need only specify those time parameters that will differ from their
default values.
Use of time units is optional. The default units are hours.
DURATION sets the length of the entire simulation (both hydraulic and
water quality). The default value is 0 hours, which implies that a steady
state run will be made. No water quality analysis will be performed for
steady state runs.
51
-------
Section: [TIMES] (continued)
HYDRAULIC TIMESTEP determines how often a new hydraulic state
of the network is computed. The default time step is 1 hour.
QUALITY TIMESTEP fixes the time step that will be used to track
water quality changes through the network. If not supplied, the program
uses an internally computed time step based on the smallest time of travel
through any pipe in the network. If you ask for a status report in the
[REPORT] section, a listing of the quality time steps used throughout the
simulation will appear in EPANET's output report.
MINIMUM TRAVELTIME establishes the smallest time of travel
through a pipe recognized by EPANET. Travel times smaller than this
number are set equal to it. (Travel times through pumps and valves are
instantaneous and are not affected by this limit.) The default minimum
travel time is 1/10 the hydraulic time step.
PATTERN TIMESTEP determines the length of a time pattern period
(i.e., the period of time over which water demands and constituent source
strengths remain constant). EPANET will adjust your HYDRAULIC
TIMESTEP to be no greater than your designated PATTERN TIMESTEP.
The default time pattern period is 1 hour.
REPORT TIMESTEP determines the interval of time between which
network conditions are reported on. If need be, EPANET will reduce
your HYDRAULIC TIMESTEP so that it is no greater than the REPORT
TIMESTEP. The default value is 1 hour.
REPORT START specifies at what time into the simulation results should
begin to be reported. The default value is 0.
Examples:
DURATION 120
QUALITY TIMESTEP 2 MIN
PATTERN TIMESTEP 2
REPORT TIMESTEP 2 HOURS
REPORT START 2 DAYS
;120 hour (5 day) simulation period
;2 minute water quality time step
;2 hour interval between changes in
;demands
;Reports are generated every 2
,-hours beginning on day 2
52
-------
Section; [TITLE]
Purpose:
Attaches a descriptive title to the problem being analyzed.
Format:
less characters
Remarks:
The [TITLE] section is optional.
of
53
-------
Section: [VALVES]
Purpose:
Describes each control valve in the network.
Format:
id nodel node2 diam type setting (losscoeff)
Parameters:
id link ID
nodel ID of node on upstream side of valve
node2 ID of node on downstream side of valve
diam valve diameter, inches (mm)
type type of valve:
PRV for pressure reducing valve,
PSV for pressure sustaining valve,
PBV for pressure breaker valve,
FCV for flow control valve,
TCV for throttle control valve,
setting pressure setting for PRVs, PSVs and PBVs in psi (m), flow
setting for FCVs, or loss coefficient for TCVs
losscoeff minor loss coefficient for fully opened valve (0 if not
specified)
Remarks:
One line should appear for each control valve. Check valves are
identified in the [PIPES] section, not here.
Control valves should not be connected directly to tanks or reservoirs.
Note that pressure settings for valves are pressures and not total head (or
hydraulic grade line elevation).
Examples:
301 12 34 8 PRV 75 ;Valve 301 is an 8" PRV that keeps the
,-pressure at Node 34 below 75 psi.
54
-------
4.4 Verification File Format
The verification file provides an optional means of checking that the links
specified in the input file connect the correct pair of nodes together. If you
choose to use a verification file, its name should appear in the [OPTIONS]
section of the input file using a line that reads:
VERIFY filename
where filename is the name of the verification file.
The verification file consists of a single section whose format is:
Section: [VERIFICATION]
Purpose:
Verifies network node/link connectivity.
Format:
node linkl link! ...
Parameters:
node node ID
linkl
Iink2, etc. ID's of all links connected to node.
Remarks:
Include one line for each node in the network.
The file can contain blank lines and it treats the semicolon (;) as a signal that the
rest of the line is a comment. EPANET will report an error message if any
inconsistencies between the connections established in the input file and those in
the verification file are found. A warning message will be issued if all nodes are
not listed in the verification file.
55
-------
4.5 Map File Format
The Windows version of EPANET allows you to view simulation results on a
map of the network being analyzed. To use this feature you must first create a
file that contains the X-Y coordinates of the network nodes included in the map.
The name of this file should appear in the [OPTIONS] section of the EPANET
input file using a line that reads:
MAP filename
where filename is the name of the map file. This will cause the map to be
displayed automatically after a successful simulation. If you do not specify a map
in an EPANET input file or you wish to change to another map, you can use a
menu option in the Windows version of EPANET to specify the name of the map
file to be displayed (see Section 6.3).
The contents of the map file are divided into the following sections:
[COORDINATES] specifies nodal X-Y coordinates
[LABELS] specifies location and text of labels
[END] signals end of map file
The file can contain blank lines and it treats the semicolon (;) as a signal that the
rest of the line is a comment. An example map file is shown in Figure 4.4. The
following pages describe the formats of the [COORDINATES] and [LABELS]
sections.
[COORDINATES]
;Node
r
2
9
to
11
fLA&ELSI
;X* coord
0
1E
55
IEND]
X.- coord
50
1«
20
30
V-coord
70
63
90
y-coord
90
70
70
70
Label
"Source"
"Pump"
"Tank"
Figure 4.4 Example Map File
56
-------
Section: [COORDINATES]
Purpose:
Assigns map coordinates to network nodes.
Format:
node Xcoord Ycoord
Parameters:
node
Xcoord
Ycoord
node ID
horizontal coordinate
vertical coordinate
Remarks:
Include one line for each node displayed on the map.
The coordinates represent the distance from the node to an arbitrary origin
at the lower left of the map. Any convenient units of measure for this
distance can be used.
There is no requirement that all nodes be included in the map, and their
locations need not be to actual scale (such as when trying to depict a
cluster of nodes that are close to one another).
57
-------
Section: [LABELS]
Purpose; .
Assigns map coordinates to text labels.
Format:
Xcoord Ycoord "label" (node)
Parameters:
Xcoord horizontal coordinate
Ycoord vertical coordinate
"label" label text (in double quotes)
node ID of optional anchor node
Remarks:
Include one line for each label on the map.
The coordinates refer to the upper left corner of the label.
The optional anchor node anchors the label to the node when the map is
re-scaled during zoom-in operations.
4.6 Summary of Default Values and Units
The only mandatory sections required in an EPANET input file are the
[JUNCTIONS], [TANKS], and [PIPES] sections. Table 4.1 summarizes the
default values used by EPANET in lieu of specific instructions supplied in other
input sections. Table 4.2 lists the units in which the various input parameters
must be expressed. The "flow units" in this table can correspond to either gallons
per minute (the default), cubic feet per second, million gallons per day, or liter
per second, depending what option is specified for UNITS in the [OPTIONS]
section. SI (metric) units apply only if SI is designated for UNITS.
58
-------
Table 4.1 Summary of Default Parameter Values
Parameter
Nodal Demands
Initial Quality
Source Quality
Time Pattern Multipliers
Demand Time Patterns
Source Time Patterns
Time Pattern Time Step
Simulation Duration
Hydraulic Time Step
Headloss Formula
Flow Units
Hydraulic Trial Limit
Hydraulic Accuracy
Specific Gravity
Viscosity
Difrusivity
Water Quality Analysis
Reaction Rates
Pipe Segment Limit
Minimum Travel Time
Reporting Start Time
Reporting Time Step
Lines per Report Page
Nodes Reported On
Links Reported On
Status Reports Generated
Default Value
0 gpm
0
0
1 for all patterns
Pattern 1 for all nodes
No Pattern
1 hour
0 hours
(steady state)
1 hour
Hazen-Williams
GPM
40
.001
1.0
I.lxl0-sft2/sec
l.SxlO'8 fWsec
None
0
100 per pipe
Hydraulic Step / 10
0 hours
1 hour
55
All
All
No
Relevant Section
[JUNCTIONS]
[DEMANDS]
[QUALITY]
[SOURCES]
[PATTERNS]
[JUNCTIONS]
[DEMANDS]
[SOURCES]
[TIMES]
[TIMES]
[TIMES]
[OPTIONS]
[OPTIONS]
[OPTIONS]
[OPTIONS]
[OPTIONS]
[OPTIONS]
[OPTIONS]
[OPTIONS]
[REACTIONS]
[OPTIONS]
[TIMES]
[TIMES]
[TIMES]
[REPORT]
[REPORT]
[REPORT]
[REPORT]
59
-------
Table 4.2 Summary of Input Parameter Units
Parameter
Junction Elevation
Junction Demand
Tank Bottom Elevation
Tank Levels
Tank Diameter
Tank Minimum Volume
Junction/Tank Quality
Chemical
Age
Source Trace
Pipe length
Pipe Diameter
Pipe Roughness
Hazen- Williams
Darcy-Weisbach
Chezy-Manning
Minor Loss Coefficient
Pump Power Rating
Pump Head
Pump Flow
Pump Speed Setting
Valve Diameter
Valve Pressure Setting
Valve Flow Setting
Bulk Reaction Coeff.
Wall Reaction Coeff.
Specific Gravity
Viscosity
Diffusivity
English Units
feet
flow units
feet
feet above bottom
feet
cubic feet
milligrams/liter
(or user supplied)
hours
percent
feet
inches
none
millifeet
none
none
horsepower
feet
flow units
none
inches
pounds per square inch
flow units
days"1
feet/day
none
square feet per second
square feet per second
SI (Metric) Units
meters
flow units
meters
meters above bottom
meters
cubic meters
same
as
English
meters
millimeters
none
millimeters
none
none
kilowatts
meters
flow units
none
millimeters
meters
flow units
days"1
meters/day
none
square meters per second
square meters per second
60
-------
CHAPTER 5
RUNNING EPANET UNDER DOS
5.1 General Instructions
Step 1. Create an Input File
To run EPANET under DOS you first create an input file that describes the
network being simulated. This file must be created with an editor or word
processor that saves its text in ASCII format. The format of the input file has
been described in Chapter 4. You might also want to prepare a verification file,
following the format described in Chapter 4, that will verify the nodal
connections contained in the input file.
Step 2. Invoke the EPANET Simulator
Assuming that your current directory is the same one in which the EPANET files
were installed, the command issued from the DOS prompt for running an
EPANET simulation is:
EPANET inpfile rptfile
where inpfile is the name of the input file and rptfile is the name of a file that
will contain the output report with the simulation results. Note that if either of
these files resides in a different directory than the EPANET files, then their,
names must include their full directory paths.
You might prefer to keep your network data files in different directories than
your EPANET program files, especially if you will be working on many different
network analyses. To avoid the need to supply full path names for your date files
you can add the EPANET directory to the path variable that DOS maintains, and
then launch the program directly from any directory you choose. Consult your
DOS manual for instructions on setting up your path.
61
-------
Step 3. View the Output Report
To view the output report over the video display you can employ a file viewer
program capable of handling large files. Such a program, BROWSE.COM, has
been installed in your EPANET directory. To use it, simply enter the command
BROWSE rptfile
where rptfile is the name of your EPANET output report file. Then use the
arrow keys to move through the file one line at a time, or use the Page Up and
Page Down keys to move one page at a time. You exit the viewer by pressing
the Escape key. In addition, you can be print the output report by issuing the
DOS PRINT command as follows:
PRINT rptfile
5.2 Contents of the Output Report
EPANET's output report consists of four different types of tables. If the STATUS
option was set to YES or FULL in the [REPORT] section of the input file, a
SYSTEM HYDRAULIC STATUS table will be produced at the start of every
hydraulic time step or whenever one of the following events occurs first:
1. a filling tank reaches its maximum level or an emptying tank
reaches its minimum level,
2. the control time for a pump, valve or pipe is reached or the
control level of a tank is reached,
3. a reporting period occurs.
If a full status report was requested, each System Hydraulic Status table will be
preceded by a listing of the accuracy achieved at each iteration of the method
used to hydraulically balance the network and will show which links are changing
status (e.g., opening and closing) during these iterations. Normally this level of
detail is needed only to debug systems that fail to converge hydraulically.
The SYSTEM HYDRAULIC STATUS table itself displays the following items:
1. whether the system is hydraulically balanced or not, the number of
iterations required and the accuracy achieved,
62
-------
2. the total water demand exerted by the system
3. the water surface elevation of each storage node and whether it is
filling or emptying
4. the status (OPEN or CLOSED) of each pump and valve
5. those pipes which have been closed off.
If a water quality analysis is being made, a second type of status table the
SYSTEM WATER QUALITY STATUS table will appear next. For each
hydraulic time step, this table displays the water quality time step and the number
of pipes that required more than the maximum allowable number of segments.
The third type of table presents NODE RESULTS at each reporting period of the
simulation. Unless changed" in the [TIMES] section of the input file, the
reporting interval is 1 hour. For each node that has been flagged for reporting
(the default is to report on all nodes) the following information is produced:
1. Node ID
2.
Elevation
3. Demand flow (negative values denote source flow)
4. Hydraulic grade (elevation -f pressure head)
5. Pressure
6.
Chemical concentration, age of water or percent of flow from a
given source, depending on the type of water quality analysis being
made (only for non-steady state runs).
For tank nodes, a positive value for demand flow means that water flows into the
tank, while a negative value means that water flows out of the tank. The
[REPORT] section of the input file allows you to limit output to only those nodes
below or above specific limits on the above variables. This helps reduce the
amount of output produced and focus in on critical points in the network.
The final type of table presents LINK RESULTS at the end of each reporting
period. Each link flagged for output reporting (the default is to report on all
links) has the following items listed:
63
-------
1. Link ID
2. Head and tail nodes
3. Diameter
4. Flow rate (negative if from tail node to head node)
5. Velocity
6. Headloss per 1000 ft (or m) of pipe
For valves the headless reported is the actual change in head across the valve,
while for pumps, it is the negative of the head supplied by the pump. Pumps also
have their power consumption displayed. As with nodes, the link listing can be
limited to those links whose output variables meet criteria specified in the
[REPORT] section of the input file.
5.3 The EPANET4D Program
EPANET4D is a DOS batch program that lets you interactively edit an EPANET
input file, run the file through EPANET, and then view or print the results all
from within a single program. EPANET4D comes supplied with its own file
editor and viewer, but you can substitute your own choices for these if you like.
You run EPANET4D by issuing the following command from the DOS prompt:
EPANET4D inpfile rptfile
where inpfile and rptfile are the names of an input and report file. If you are
running EPANET4D from the EPANET directory, then make sure that the full
path names are included with the input and report file names if they reside in a
different directory. Alternatively, you can launch the program from any directory
you choose as long as the EPANET directory is included in your DOS path.
64
-------
After the program loads itself, it displays a menu with the following choices:
EPANET MENU
—
1 - Edit input file
2 - Run EPANET
3 - View report file
4 - Print report file
5 - Quit
=====
and prompts you to enter the number of the choice you wish to select.
press the F4 key. After exiting, you are retured to the
mu scren
Reminder: When exiting the editor, you must respond with Y (for yes) when
af in A°U Wish t0 Save the chanSes y°u made t° the file if you
want EPANET to recognize these changes. Y
Choice 2 executes the EPANET simulator. After execution, you are prompted
to hit any key which then returns you to the menu screen. PromP'ed
be ™ H y°U toubrowse throuSh I* EPANET report file after a run has
been made. You use the keyboard arrow keys to move up or down through the
file a line at a time. The Page Up and Page Down keys will move you ! page
at a time. You exit the file browser by pressing the Escape (ESC) key
Choice 4 prints the contents ,of the report file by calling on the DOS PRINT
command. The menu screen re-appears afterwards.
Choice 5 exits the EPANET4D program and returns you to DOS.
65
-------
TEXT EDITOR 2.6 COMMAND SUMMARY
TE IPathJlFileNamel J c- Ctrl s- Shft a- Alt
File
c-KD, c-KO. F4 Save ffle and quit editor
c-KE, F2 save and/or toad another file
Cursor Movement
e-S, left Char left
c-D, Rt Char right
c-A, c-Left Prey word
c*F, C'Rt Next Word
c-E, Up Prey line
c-X, Dr> Next Une
c-W Scroll 153
e-Z. Scroll down
c-R, PgUp Up 23 lines
c-C, PgDn Dn 23 lines
c-Q4 To specif fed line
c-QP Dn page I en lines
c-QS, Hone Line begin
c-QD, End Line end
t-QEj c-Home Screen top
c-QX, c-End Screen bottom
c-QR, c-PgUp File start
C-QC, c-PgDn File end
c-QB To block start
c-QK To block, end
Tab Next word, prey line
s-Tab Prev word, prev line
'c*Kn Set Ifne mark n=0-3
c-Qn To line mark n=0-3
Insert/Delete
c-V, Ins Insert/Replace
Entr Split/Insert line
c-N, F9 Insert line
c-Y, F10 Delete line
c-H, BkSp Delete left char
c-S, Del Del char/ join line
c-T Del next word
c-QY Del to end line
Block
c-KB, F7 Mark blocfc start
c-KK, F8 Mark block end
c-KK Hide/display block
c-KI Block right 1 char
c-KU Block left 1 char
e-KC Copy block
c-KY Delete block
c-KV Move block
c-KR Read block from disk
c-KW Write block to disk
Miscellaneous
F1 Display sumnary of Text Editor comnands
c-QF Find phrase (1-31 chars} in file or blocfc
c-QA Find/replace phrase (f-31 chars) in file or block
c-KP, F5 Print file or block to LPT1. LPT2, or LPT3
a-Xa-Ya-Z ASCII code XYZ = 32-255 on keypad
c-P Then a-Xa-Y on keypad for ASCII XY = 1-31
c-KS, F3 Temp return to DOS. Back te TEJ EXIT
c-QM Set left/right margfnSt page length
c-B Forjnat paragraph to left/right margins
Figure 5.1 Text Editor Command Summary
66
-------
If you don't care for the file editor and viewer that comes with EPANET4D you
can substitute your own choices for these programs. The programs you use must
be able to accept the name of an input file on the command line and the viewer
must be capable of handling large files, since the output from a long term
simulation of a large network can be quite considerable in size.
Figure 5.2 lists the contents of the EPANET4D.BAT file. If you would like to
teXl I?**;. replace the word "te-exe" that appears below the
me editor ********** Une with ^ name of the new ^^
*******r ?*viewer' rePlace the word "browse.com" below the
********* L^f* fite vie^er ********** iine with the name of the new file
^oLr^°te lf thCSe Pr°Srams resi
-------
^
:edit
ran
TO ********* Uuncf)
te.exe %i
goto start
irun
cts
if »%epanet%»==«<32» goto run32
epanetf^.exe %1 %2 S3
pause
goto start
:run32
set DOS4G=qufet
echo Running 3Z-bit version of EPANET..
dos4gw epanet32.exe ft 5K %3
set pos4G=r
pause ' ^
goto start
cts
if Hot exist %2 goto err3
rew ^ ,
********* Launch file viewer *********
goto start
: printout
cts'
if not exist 52 goto err3
print %2
goto start
echo Correct syntax is; EPAMET4D inpfile rptfile
echo where inpfile is the name of an input fie
echo rPtfUe fS ^ name Qf a «porf file.
goto end '
:err2
°r fnput 8nd report f{tes-
created
goto end
:err3 '
Icho
pause
goto start
squit
cts
Figure 5.2 Continued
68
-------
CHAPTER 6
RUNNING EPANET UNDER WINDOWS
6.1 Overview
The Windows version of EPANET, referred to as EPANET4W, provides a
graphical user interface for running network simulations and viewing their results
It allows you to edit EPANET input files, execute EPANET's network simulator'
and view EPANET's output in a variety of formats that include:
• color-coded network maps
• time series plots
• tabular reports.
This chapter assumes that you have a basic understanding of how to work with
Windows applications, and are familiar with the different parts of a window (such
as- the menu bar, the control-menu box, and the minimize and maximize buttons)
and with mouse operations.
Figure 6.1 shows how the EPANET4W workspace might appear at some point
during a run. It gives examples of the different types of windows that the
program can generate. These include:
1.
2.
3.
4.
5.
a window listing the network's input data
a Browser window that controls what aspect of the network
simulation results should be viewed
a window with a map of the network
windows that display output results in tabular form
windows that display time series graphs of output results.
The figure also shows the program's menu bar across the top of the workspace
Using the windows version of EPANET involves making a repeated sequence of
choices from this menu to select, edit and run an EPANET input file (or select
a previously saved output file), and view the results in the form of maps, graphs
and tables. The following paragraphs explain these various procedures '
69
-------
£dlt Bun Report firaph Map Window Help
[TITLE]
EPANET Example Network
[JUNCTIONS]
10 710
11 710
12 700
1865.06.1 .2.35.1 1.8.
14.00 11233.57
10.00 £ 129.41
2.57 • 2.89
Chlorine for Node 22. mq/L
Mj
0.78 i. .0,47
0.081
0-46 i ""7 0-3
0.96: 0.3S
1J7J 2.6J
0.54 ! _ 0.19
b"l'9 j "' " ~ 0.0
goif n"?
0.60
0.55
0.50
0.45
Figure 6.1 Example Contents of the EPANET4W Workspace
Note: EPANET4W allows you to keep multiple windows loaded in its
workspace at one time. To conserve system resources, it is recommended
that you close windows when they are no longer needed by double-
clicking the control-menu box in their upper left corner.
70
-------
6.2 Launching the Program
EPANET4W can be started either from DOS or from within Windows. To start
it from DOS, issue the command:
WIN patf?\EPANET4W
vsheiepath is the full path name of your EPANET directory (e.g., C:\EPANET).
There are two different ways to start up EPANET4W if you are already in the
Windows environment:
Method 1. Double click the EPANET icon in the EPANET program group
within Program Manager.
Method 2. From the Program Manager, select File from the main menu, and
then select the'Run option under it. When prompted for the
command line to use, enter jpa*MEPANET4W where path is the
full path name of your EPANET directory (e.g., C:\EPANET).
6.3 Operating Procedures
Getting Help
EPANET4W comes with a complete Help facility that functions the same as other
Windows Help systems. You access it by selecting an option from the Help
menu or by pressing the Fl key. From the Help menu you can choose to view
the system's table of contents, go directly to the section dealing with input data
formats, or search for help on a specific topic.
Opening an Input File
After EPANET4W begins, your first operation should be to open an input file (or
previously saved output file) for processing. To open an input file, select Open
Input from the File menu. A file dialog box will appear as shown in Figure 6.2.
You can change drives, directories, or file name patterns within this window (the
default file name pattern for input files is *.INP). To specify a new file that does
not currently exist, type its name into the File Name box. Press the Enter key
or click the OK button to load your selected file. Click the Cancel button (or
press the Escape key) to cancel the operation. After a file has been selected, its
name will appear in the title bar of EPANET4W's main window.
71
-------
File Name:
*.inp
netl.inp
netZ.inp
netS.inp
w
'&&
m.
Directories:
c:\epanet
List Files of Jype:
Drives:
Figure 6.2 File Dialog Box
Viewing the Input File
Editing the Input File
72
-------
TEXT EDITOR 2.6 COMMAND SUMMARY
TE tPathHFUeName] j c- Ctrl s- Shft a- Alt
FHe
c-KD, c-KQ, F4 Save f He and quit editor
c-KE, F2 save and/or load another f ite
Cursor Movement
c-S, Left Char left
C*&, Rt Char right
c-A, c-Left Prev Word
c-f, e-Rt Next word
e-E, Up Prev line
c-X, Dn Next Ifne
c-W Scrotl up
c-2 Scroll down
c-R, PgUp Up "23 tines
c-C, PgDn Dn 23 tines
c-Q4 To specified line
c-QP Dn page ten tines
c-QS, Home tine begin
c-QD, End Lihfr end
c-QEf fc*Hoaie Screen top
c-QXt c-End Screen bottom
c-QR, c-PaUp File start
c-QC, c-PgDn Fil« end
c-QB TO block start
c-Qfc T& block end
Tab Next word, prev tine
s-Tab Prev word, prev tine
c-Kn Set { ine mark n=0-3
c-Qn TO line mark n=0-3
Insert/Delete
c-V, Ins tnsert/Replace
Entr Split/Insert tine
(HI, F9 Insert line
c-r, F10 Delete tine
e-K, SkSp Delete left char
c-G, Det Del char/join tine
c-T Dei next word
c-QY Oet to end line
Block
c-KB, F7 Mark block start
c-KKr F8 Mark block end
c-lCH Hide/display block
c-KI Block right 1 char
c-KU Stock left 1 char
c-KC copy btock
c-KY Detete btock
c-Ktf Move btock
c-KR Read block from disk
c-Ktf Write block to disk
Miscellaneous
F1 Display sunmary of Text Editor comnands
c-QF ' Find phrase (1-31 chars) in file or block
c-QA Find/replace phrase C1-31 chars) m fite or block.
c-KP, F5 Print fite or block to LPT1r LPT2, or LPT3
a-Xa-Ya-2 ASCII code XY2 = 32-255 on keypad.
c-P Then a-Xa-Y on keypad for ASCII XY =* 1-31
c-KS, F3 Temp return to DOS, Back to T€i EXIT
c-QM Set left/right margins, page l«ngth
C-B Format paragraph to left/right tnargfns
Figure 6.3 Text Editor Command Summary
73
-------
Hints: You can substitute a different DOS or Windows file editor
program for the one supplied with EPANET. Consult Chapter 3
for instructions on how to do this.
If you are using a DOS file editor, then pressing the ALT-ENTER
key combination will toggle the editor between appearing in a
window and in full-screen mode.
Editing Other Files
At times you might find it necessary to edit other files, such as the map and
verification files. There are two ways to do this. One is to invoke the
EPANET4W editor for your input file as described earlier, and then from within
the editor switch to the file you want to edit. Hitting the F2 key in the editor
supplied with EPANET will allow you to do this. Be sure to save your file after
making any changes to it. A second approach is to switch back to Program
Manager and launch any editor you normally use as a new task to edit your
choice of files. Then switch back to EPANET4W.
Running the Input File
Use the Run menu to request that the current input file be processed by the
EPANET's network simulator. Selecting Windowed will run the simulator in a
window so that you can observe its progress. If you select Minimized the
simulator will appear as an icon until it is done processing. Select Cancel to
cancel the request.
If the run was unsuccessful, EPANET4W will display the error messages
generated by the simulator in a report window explaining what went wrong. At
this point you could invoke the editor once again to fix your mistakes, leaving the
report window displayed on the screen so that you can see where in the input file
the errors occurred.
If the run was successful, EPANET4W will display its Browser window and will
also draw a map of the network in its Map window, providing that the name of
a map file was included in the input data. Instructions on using the Browser
window to view different aspects of the output are given below. If there were
any warning messages generated, an Output Summary for the run will also be
displayed in a report window which will scroll down to the line containing the
first warning message. See Viewing an Output Summary below for details.
74
-------
Opening an Output File
As an alternative to opening an input file for editing and analyzing with
EPANET, you can choose to view results generated from a prior session of
EPANET4W that were saved to a file. To do this, select Open Output from the
File menu. A file dialog box will appear from which you can select the output
file to view. After selecting the file, its name will appear in the title bar of
EPANET4W's main window. EPANET4W will then display its Browser window
and, if a map file is associated with the output file, will draw a map of the
network. The program will not be able to display any status reports or warning
messages that were generated when this network was first simulated.
Note: The output file referred to here is not the same as the report file produced
when EPANET is run under DOS as described in Chapter 5. It is
possible to run EPANET under DOS and have it generate an EPANET4W
output file by adding a third file name to the command line as follows:
EPANET inpfile rptfile outfile
where inpfile is the name of input data file, rptfile is the name of the
report file that can be viewed or printed, and outfile is the name of the
output file that can be accessed under EPANET4W. The EPANET
simulator tends to execute about twice, as fast when run directly under
DOS (and not in a DOS session under Windows). Although it might be
more efficient to generate simulation results in this way, you loose the
advantage of being able to interactively make changes to the network and
quickly visualize their impact.
Opening a Map File
If no map file was specified in the current input data file (or output results file),
or the specified map was incorrect, or you simply want to switch to a different
map of the network, you can select Open Map from the File menu. This will
Open up a file dialog box from which a new map file can be selected.
After viewing the map you might find that it needs to be modified. For example,
text labels might need to be re-located. You can use the editor to make these
changes to the map file as described earlier under the Editing Other Files, and
then use the Open Map option of the File menu to load the updated map.'
75
-------
Saving the Current Output
Select Save Output from the File menu to save the results obtained from the
latest EPANET simulation to a file. This allows you to come back to
EPANET4W at a later time to view the results of the current simulation without
having to re-run it all over again. Enter the name under which you want to save
the current output results in the File Name field of the file dialog box that
appears. The suggested file extension to use for output results is ".out". You
can use the drive and directory boxes to select new choices for the file's path.
Viewing an Output Summary
A summary report of the current output results can be generated by selecting
Output Summary from the Report menu. At a minimum, the report will
summarize the nature of the network being analyzed (e.g., number of nodes and
links, head loss equation used, type of water quality analysis performed, etc.).
In addition, if the results were generated from running an input file, the report
will include any warning messages produced by EPANET (such as pumps
operating out of range) and, if requested in the [REPORT] section of the input
file, a status report on each tank, pump, and valve throughout the duration of the
simulation. This type of information will not be listed if you have selected an
output file for viewing. The output summary window can be moved, resized, and
closed when no longer needed.
Using the Browser
The Browser window (Figure 6.4) allows you to control how the EPANET output
is viewed This window is used to:
a) select a node view variable from among:
i. None
ii. Demand
iii. Elevation
iv. Hydraulic Grade
v. Pressure
vi. Water Quality (non-steady state runs only)
76
-------
2.28 ft/sec J
b) select a link view variable from among:
i. None
ii. Diameter
iii. Flow
iv. Velocity
v. Headless
vi. Average Water Quality (non-steady state
runs only)
c) view the value of the node and link variables for
specific nodes and links
d) select a time period for viewing
Functions (a), (b), and (c) are accomplished by making
selections from the dropdown list boxes in the Browser's
Nodes and Links panels. The upper boxes in each panel
contain the list of node and link ID's, respectively. The
lower boxes contain node and link view variable choices.
The list boxes are activated by clicking on their arrows.
A list of options will appear from which you click on the
desired choice. Whenever a choice on any of these
items is made, the following events will occur:
Figure 6.4 The Browser
a) The values of the view variables displayed in the Browser will be
updated.
b) The network map will be color-coded corresponding to the current
selections for node and link view variables. Nodes are represented
as colored circles (tanks/reservoirs are squares) and links as lines
between nodes. A legend indicating what value ranges the colors
represent can be displayed by selecting Legend from the Map
menu.
c) After a specific node or link is selected from the Browser, its
location on the map is highlighted.
Whenever you click the mouse on a node or on the midpoint of a link displayed
in the map, that node or link is highlighted and the value of its current view
variable is displayed in the Browser.
77
-------
The Links panel contains a button labelled Info. Clicking on this button will
overwrite the panel with a listing of the input information for the currently
selected link. For pipes, this information includes diameter, length, roughness
coefficient, minor loss coefficient, and reaction rate coefficients. For pumps it
includes the equation of the pump curve. The listing also contains the current
status of the link (e.g., open or closed). When you are done viewing this
information, simply click anywhere in the listing to restore the Links panel to its
original state.
To change the current time period being viewed, you move the slider bar in the
Browser's Time panel. To move forward one time period, click the right arrow
on the slider bar. To move backwards, click the left arrow. Whenever the time
is changed, the values in the Node and Link panels are updated and the map is
re-colored to reflect changed conditions within the network. You can simulate
a time animation of the map by keeping the mouse button depressed over one of
the slider bar arrows. If a steady state run was made, no slider bar will appear
in the Time panel.
As with any other window generated by EPANET4W, you can move the Browser
to another location by moving the mouse into its tide bar, pressing the left button
while moving the window to a new location, and then releasing the button. You
can also minimize the Browser window to an icon by clicking its Minimize button
in the upper right of the window. To restore the window, simply double-click
on the icon. You cannot close the Browser window.
Viewing the Map
The appearance of the network map can be modified by making choices from the
Map menu. The choices and their resulting actions are as follows:
a) Zoom In - allows you to magnify a portion of the map by
defining a zoom window. The area bounded by the zoom window
will be drawn to fill the entire Map window. To define the zoom
window, move the mouse within the Map window to where you
want a corner of the zoom window to begin, then click the left
mouse button. Next move the mouse until the outline box
displayed encompasses the area you wish to zoom in on. Then
click the left mouse button once again. Clicking the right mouse
button cancels the zoom operation.
b) Zoom Out - restores the map to a state that existed prior to the
last zoom in
c) Redraw ~ redraws the map at its original scale.
78
-------
d) Display Legend - toggles the display of a node or link legend on
and off. The legend can be dragged to a different location on the
map by moving the mouse with the left button held down.
Double-clicking the left mouse button in the legend will remove it
from the map. Clicking the right mouse button within the legend
allows you to modify the legend as described next.
e) Modify Legend — displays a dialog box for your choice of node
or link legend that lets you define what numerical ranges
correspond to a particular color on the map for the current node
or link variable. It also lets you change the colors used for the
legend (see Figure 6.5). The Default Values button restores the
ranges to their internally computed values while the Default Colors
button does the same for colors. Modifying the legend can give
you a quick picture of where in the network a certain condition
holds. For example, the legend shown in Figure 6.5 will display
all nodes where the pressure is below 20 psi in the first color.
f) Options - displays a dialog box for map labelling and style
choices (see Figure 6.6). You can elect to label nodes and links
with their iD's and values of the current view variable, change the
size of nodes and thickness of links, have flow direction arrows
drawn on the links, have pump and valve symbols drawn on the
links, change the map's background color, or change the way in
which objects on the map are highlighted. (Point highlighting
accents the individual object while area highlighting highlights a
region around the object.)
When the mouse is positioned over the map, the following keyboard-mouse
combinations can be used as shortcuts to implement operations on the Map menu:
Shift-Left Mouse Button
Ctrl-Left Mouse Button
Alt-Left Mouse Button
Alt-Right Mouse Button
— > Zoom In
— > Zoom Out
— > Redraw
— > Options
In addition to the options offered by the Map menu, the horizontal and vertical
scrollbars on the Map window can be used to pan across the map either horizon-
tally or vertically. The Map window can also be moved, minimized, and resized,
but it can not be closed.
79
-------
wede Pressure;
Figure 6.5 Legend Dialog Box
Generating Tables
"Nodes
Display:
DlD's
D Values
KlText
Size:
O Small
® Large
'Background'
<§> Black
O Gray
O White
F Links
Display:
DlD's
D Values
d Arrows
LJ Symbols
Style:
Thick
'Highlighting'
O Point
rea
Figure 6.6 Map Options Dialog Box
The Report menu can be used to display two types of tables of output results.
Selecting Current Time from the Report menu produces a table showing the
values of all variables at the current time period for either all nodes or all links
in the network. Selecting Time Series from the same menu creates a table
containing the values of all variables for all time periods for either the current
node or current link. Recall that the Browser window establishes the current
time, current node, and current link. Multiple tables can be displayed in
EPANET4W's workspace at one time. These windows can be moved, resized,
and, when no longer needed, closed. •
80
-------
Searching Tables
You can search for entries in a table that meet a specified criterion by selecting
Search from the Report menu. The table you wish to search must be the
currently active window in the EPANET4W workspace (i.e., have its title bar
highlighted - see Switching Between Windows below). A dialog box like the one
in Figure 6.7 will appear where you can define your search criterion. After you
specify your criterion, EPANET4W will report on the number of items that meet
it and will ask if you want to show only those items in the table or not. If you
answer yes, then any subsequent searches on the table will be made from among
its remaining entries. To restore the table to its original full contents, select
Restore from the Report menu.
Diameter
Flow
Velocity
Figure 6.7 Table Search Dialog Box
Generating Graphs
To produce a graph showing how the value of the current view variable for the
current node in the Browser window changes over time, select Current Node
from the Graph menu. Select Current Link to produce a similar plot for the
current link. If the current link is a pump, selecting Pump Curve from the
Graph menu will produce a plot cvf the pump's characteristic operating curve.
Multiple graphs can be displayed within the EPANET4W workspace, and their
windows can be moved, resized, and closed when no longer needed.
81
-------
Customizing Graphs
Select Options from the Graph menu to customize a graph whose window is
currently active (i.e., has its title bar highlighted - see Switching Between
Windows below). A dialog box will appear as shown in Figure 6.8. The lower
portion of the box lets you customize the appearance of the Y-Axis (the vertical
axis) of the graph. If the Auto Scale box is checked, EPANET4W will choose
its own axis scaling for you. The Data File box at the top allows you to link a
data file containing observed data to the graph. You can either type in the name
of the file you wish to use or click on the button to the right of the box to bring
up a file dialog box. Each line in the file linked to the graph should contain an
X-value (time) and a Y-value to be plotted on the graph. Figures 6.9 and 6.10
display the contents of a typical time series data file and how a graph would look
after these data are linked to it. This feature should prove especially useful for
model calibration.
Data File
C:\EPANET\NET2-N19.DAT
-Y-Axis
Title
Fluoride for Node 19, mg/L
Minimum
Maximum
0.07
1.04
Tic Interval
I Auto Scale
Figure 6.8 Graph Options Dialog
0,25
2.75
5.70
8.60
12.OO
13,25
14.73
17.77
2O. 52
23,53
27,17
29.87
33.92
35.67
38,48
42,08
44.68
47.50
51,17
53.45
1.04
1,04
1.08
1.00
0.81
0.9S
1.O2
1.01
0.87
0.28
0.98
0.85
0.12
0.17
0.64
0.79
0.87
0.16
0.56
0.70
Figure 6.9
Data to Link
With a Graph
82
-------
Figure 6.10 Graph With Linked Data
Copying to the Clipboard
If the currently active window is a map, table, report, or graph, selecting Copy
To CUpboard from the Edit menu will copy its contents to the Windows
Clipboard. The copied data or image can then be pasted into other Windows
applications. When copying from a table you can first select the range of data
to copy. Clicking on the upper-left grayed "cell in the table will select the entire
table. Clicking on any column's top label will select the entire column. Data
copied from tables are stored as text in the Clipboard, while map and graph
images are stored as bitmaps.
Printing
Select Print from the File menu to print the contents of any active window
(except the Browser) in the EPANET4W workspace to your Windows default
printer. If the active window is a graph or the map, then a dialog box will
appear in which you can specify a title for the plot and the page margins to use
Selecting Printer Setup from the File menu will open up a dialog box from
which you can select a different printer and alter certain printer settings, such as
printing in portrait or landscape mode.
83
-------
Switching Between Windows
The Window menu.offers a convenient means for switching between the windows
currently displayed in the EPANET4W workspace. This menuSto^S
all open windows in the workspace. The currently active window will ha?e a
check mark next to it on this list and will appear on top of any ote widows k
^workspace with its title bar highlighted. To make a different window *
smpy click on its name in the Window menu. Another way to make a
visible window active is to click anywhere within it. i
Exiting EPANET4W
Select Exit from the File menu to exit EPANET4W.
6.4 Summary of Menu Commands
File
Open Input
Open Output
Open Map
Save Output
Print
Printer Setup
Exit
Opens an input data file for processing
Opens a previously saved output file for processing
Opens a map file and displays its contents
Saves the current output results to a file
Prints the contents of the currently active window
Selects a printer and its features
Exits the program
Edit
Copy .to Clipboard Copies the contents of the currently active window
to the Windows clipboard
Input Data Edits the current input data file
Run
Windowed
Minimized
Cancel
Runs the EPANET simulator in a window
Runs the EPANET simulator as an icon
Cancels the run request
84
-------
Report
Input Data
Output Summary
Current Time
Time Series
Search
Restore
Displays listing of input data
Displays listing of output summary
Displays table of all node or link results for current
time period
Displays table of results for the current node or link
for all time periods
Searches currently active table for entries meeting
a specified criterion
Restores currently active table to its original state
Graph
Map
Current Node
Current Link
Pump Curve
Options
Zoom In
Zoom Out
Redraw
Display Legend
Modify Legend
Options
Window
Window List
Displays time series graph for current view variable
at the current node
Displays time series graph for current view variable
at the current link
Displays pump curve if the current link is a pump
Links observed data (in a file) to the current graph
and customizes its Y-axis scaling
Zooms in on a selected area of the map display
Restores map to state prior to last zoom in
Redraws map at its original scale
Toggles display of node or link legend on and off
Modifies colors and scale ranges for displaying
node or link values on the map
Changes appearance of the map (e.g., adds labels,
adds flow direction arrows, changes node/link size,
changes background color)
Activates selected window
Help
Contents Displays help system's table of contents
Input Formats Provides help on input data formats
Search for Help On Provides help on a specific topic
About EPANET Displays EPANET version number
85
-------
86
-------
CHAPTER 7
EXAMPLE APPLICATIONS
7.1 Introduction
This chapter presents three example applications of the EPANET program. The
data sets for each example are provided on the EPANET distribution disk and
should already reside in your EPANET directory if you followed the installation
instructions in Chapter 3. The file names of the data sets are:
NET1.INP (Input data for Example 1)
NET1.MAP (Map coordinates for Example 1)
NET2.INP (input data for Example 2)
NET2.MAP (Map coordinates for Example 2)
NET2-N11.DAT (Node 11 sampling results for Example 2)
NET2-N19.DAT (Node 19 sampling results for Example 2)
NET2-N34.DAT (Node 34 sampling results for Example 2)
NET3.INP (Input data for Example 3)
NET3.MAP (Map coordinates for Example 3)
7.2 Example 1 - Chlorine Decay
Figure 7.1 depicts a small distribution system that will be used to illustrate how
EPANET can model chlorine decay. Pump 9 takes water from a reservoir at
Node 9 and feeds it into a system containing a storage tank at Node 2 The
operation of the pump is controlled by the level in Tank 2. A 24 hour simulation
of chlorine transport will be made assuming a first order decay of chlorine in the
bulk flow occurs with a rate constant of -0.5/day and a first order wall reaction
occurs with a rate constant of -1 ft/day. The input data for this problem appears
in Figure 7.2 (comments have been added throughout the data set to enhance its
readability - use of such comments is purely optional).
87
-------
Tank 2
Reservoir 9
13
Pump 9
113
23
Figure 7.1 Network for Example 1
[TITLE!
EPAtlET Example NetWork 1
i: JUNCTIONS}
; Elevation
; It> ft
10 710
11 710
12 700
13 695
21 700
22 695
23 690
31 , 700
32 710
CTANKSJ
i ,, Elev, iniu
J TO ft Levet
2 850 120
* 9 800
Demand
gpm
Q
150
150
100
150
200
150
100
100
""•"--••-•-•"•-—-.--.-
Mln, Hax.
Level Level
- 100 ISO
'
Dfatn.
ft
50,5
Figure 7.2 Input Data for Example 1
(Continued on Next Page)
88
-------
[PIPES!
Head
Node
Tai I
Node
Length
ft
Diam.
in
Rough.
Coeff.
10
11
12
21
22
31
110
111
112
113
121
122
10
'11
12
Z1
22
31
2
11
12
13
21
22
11
12
13
22
23
32
12
21
22
23
31
32
10530
5280
5280
5280
5280
5280
200
5280
5280
5280
5280
' 5280
18
14
10
10
12
6
18
10
12
8
8
"6
100
100
100
100
100
too
100
100
100
100
100
100
[PUMPS! ' " ' ,
•- J >
; Mead Tail Design B-6
I ID Node (lode ft gpra
*+rt.* + r~^. + ~S, .,-».*-*»-»*.-••„*.-.
*9 9 10 250 1500
[CONTROLS] _ .
'u»K 9 OPEH IF NODE 2 BELOW 110
tlNK: 9 CLOSED tF NODE 2 ABOVE 140
[PATTERNS^ _,_ _
•"ID Huftiplters
' 1 1.0 1.2 1,4 1,6 1.4 1.2
1 1.0 0,8 0.6 0.4 0.6 0,8
[QUALITY! ,
; ' trvitfal
jNodes Concen. rag/L
' 2 32 0.5
9 1.0
21.0
[REACTIONS!
'GLOBAL BULK -.5 ; BUU decay
GLOBAL.HALL -1 ; Walt decay coeff.
[TIHES3 ' / '
' 'DURATION 24 j 24 hour simulation period
PATTERN TIMESTEP 2 ; 2 hour pattefn time period
[OPTIONS! ' " "% " ' '"*'*"'"
'QUALITY Chlorine ; Chtoripe analyst?
Wftp . ^^etl.rijap ; Hap coordinates
[EU&J
Figure 7.2 Continued
89
-------
These data also appear in the file NET1.INP in your EPANET directory. You
can run the program on this data several different ways:
a) vfrom DOS, issue the command:
EPANET NET1.INP NET1.RPT
and use a file viewer to view the contents of NET1.RPT or have
it printed,
b) from DOS, issue the command:
EPANET4D NET1.INP NET1.RPT
and select to run EPANET from the menu and then view or print
the output file NET1.RPT,
c) if you've installed EPANET for Windows, then launch it, open the
NET1.INP file, run the file through EPANET, and then look at
the results using EPANET4W's various view options.
A portion of the output file, NET1.RPT, produced by methods (a) and (b) above
is shown in Figure 7.3.
90
-------
Page 1
, _Thu Jul 29 09:48:12 1993
E P A B E T
.Hydraulic and Water Quality
Analysis for Pipe Networks
, Version 1.1
*
*
*
*
EPANET Example Network 1
Input Data Fjte ..,,„.,.,,,.„,,.,, netl.inp
'Verification Pile ..... , ...........
-Hydraulics File ..... ........ 1..V/. '
Map file. ......... .,.„..* ......... Ketl.roap
Hunter of Pipes ...,„,,..,.*„.„...*, 12
Number of Nodes .,,...,,..«..,„„..« }f
Number of Tank* »,..,,..«..".»...,.. 2
Nwnber of Pumps »,...1,., .,,,...,.,. 1
Kuenber of Valves „„„,., „,..,., ..•*,„. 0
(Jeadloss formula .................. Razen:Williams
Hydraulic rimestep .- ............... 1JJO lirs
Kydraul ic Accuracy * . . , . k « , ...... , . 0.001000
Maximum Trials .» ...... „...,,., ____ 40
Quality Analysis ...x.... ....... ,,. Chlorine
- MininxJtn Travel Time ..«,..»...„...« 6.00 min
Maximum Segments per Pipe ,..,..,,. 100
Specific firavity ...,,,....,...,.',. 1.00
Kinematic Viscosity ........... ---- 1.10e-005 sq ft/sec
Chemical Diffusivity .„ ........... 1.30e-008 sq ft/sec
•.Total Duration ..... » .............. 24.00 hrs
Reporting Criteria:
All Nodes
AlC links '
. £
' V f. r
Node Results at 0:00 hrs;
- Elev. Demand firade Pressure Chlorine
Node ft spm ft ^ psi rag/L
10 710.00 0.00 1004.50 127.61 0.50
" 11 710.00 150.00 985.31 119.29 0.50
12 700,00 150.00 970.07 117,02 0.50
13 695,00 , 100.00 968.86 118,66 0.50
21 700,00 150.00 971,55 117,66 0,50
22 695.00 200,00 969,07 118.75 0,50
23 690.00 150.00 968.63 120.73 0.50
31 700.00 T00,00 967.35 115,84 0.50
32 710.00 106.00 , 965.63 110,77 0.50
2 850,00 765.06 970,00 52.00 1.00 Tank
9 800,00 -1865,06 800.00 0,00 1,00 Reservoir
Figure 7.3 Portion of Output Report for Example 1
(Panel 1 of 3)
91
-------
Page 2
Link Results at O.-OO hrs:
EPANET Example Network 1
Link
10
11
12
21
22
- -' 31
110
111
112
*13
121
T22
9
Start
Mode
10
11
12
21
22
31
2
11
12
13
21
22
9
End
Node
11
12
13
22
23
32
12
21
22
23
32
10
Diameter
in
18.00
14.00
10.00
to. oo
12.00
6.00
18,00
10,00
12.00
8.00
8.00
6,00
Flow
gpm
1865,06
1233.57
129.41
190,71
120.59
40,77
-765,06
481.48
189.11
29.41
140.77
59.23
1865,06
Velocity
fps
2.35
2.57
0.53
0.78
0,34
0.46
0.96
1.97
0.54
0,19
0.90
0,67
96 hp
Headloss
/1000ft
1,82
2.89
0.23
0.47
0.08
0,33
0.35
2.61
0.19
0.04
0.79
0.65
-204.50 pump
Made Results at 1:0 hrs: , ', , ™
Node
10
<. 11
12
13
21
22
23
31
32
2
9
Elev.
710.00
710.00
700.00
695.00
700.00
695.00
690,00
700.00
710.00
850,00
800.00
Link Results at 1:
Link
10
11 '
12
21
22
W
111
Start
Node
10
11
12
21
22
~ 31
2
11
Demand
spm
0,00
150.00
150.00
100.00
150.00
200.00
150,00
100.00
100.00
747.57
-1847.57
00 hrs:
Grade
ft
1006.92
988.05
973.13
971.91
974.49
972.10
971.66
970.32
968.63
973.06
800.00
End Diameter
Mode in
11
12
13
22
23
32
12
21
Pressure
128.65
120,48
118.35
119.98
118.94
120.07
122.04
117.13
112.06
, 53.32
0.00
Chlorine
1.00
0.45
0.44
0.44
0.43
0,44
0.45
0.41
0.40
0,97
1.00
y
Tank
Reservoir
--- -
Flow Velocity Headloss
spm fps /1000ft
18,00 1847.49
14.00 1219.82
10.00 130.19
10.00 187.26
12.00 119.81
6.00 40.42
18.00 -747.49
10.00 477.68
2.33
2.54
0,53
0.76
0,34
0.46
0.94
1,95 J
1.79
2,83
0.23
0,45
0.08
0.32
0.34
, 2,57
Figure 7.3 Panel 2 of 3
92
-------
Page 3
Link Results at tsOO tirs; Continued)
EPANET Exanple Network 1
Link
112
113
121
,122
Start
Node
12
13^
, 22
End't
Node
22
23
7" ^i
10
liameter
in
12,00
8.00
8.00
6.00
Flow \
gpm
192.14
30,19
140.42
59,58
1847.49
Velocity
fps
0.55
0.19
0,90
0,68
97 hp
Keadloss
/1000ft
0,20
0,05
0.79
0.66
-206,92
Pupp
tfode Results at 2:00 fcrs:
Node
10
11
12
13
21
22
23
31
32
2
9
Elev.
ft
710.00
710,00
700.00
695.00
700.00
695,00
690,00
700.00
710.00
850,00
800.00
Demand
gpm
0,00
180,00
180.00
120,00
180.00
240,00
180.00
, 120,00
120.00
516,44
-1836,44
Grade
ft
1008.43^
989,77
976.09
974.02
975,41
973.81
973.33
969.96
968.13
976.06
800.00
Pressure
129,31
121,22
119.63
120. 9Q
119*34
120.81
122.77
116.98
111.85 '
54.62
0.00
Chlorine
>
1.00
0.87
15.81
0.37
- 0,76
o.sa
0,40
0.34
0,31
0.94
1.00
Tank
Reservoir
Link Results at 2:00 firs:
Start
• Link Node
10
11
12
21
22
31
110
111
112
113
121
122
9
10
11
12
21
22
31
2
11
12
13
21
22
9
End Diameter
Mode m
" 11
12
13
' 22
23
3a
12
21
22
23
31
32
', 10
18.00
14.00
10,00
10.00
12.00
6.00
18.00
10.00
12.00
8.00
8.00
6.00
Flow Velocity
gpm fps
1836,44
1163.77
173.00
' 150.^47
127.00
42.20
:516.44""
492.67
294,33
53,00
162,20
"77*80
1836,44
S -1
2.32
'2.43
0.71
0.61
0.36
0.48
0,65"
2,'D1
0.83
0.34
1.04
' 0^.88
97 hp"
Head toss
/1000ft
1.77
2.59
0.39
0.30
0.09
0.35
0.17
2.72"
0,43
0.13
1.03
1.08
-208,43 Pump
Figure 7.3 Panel 3 of 3
93
-------
7.3 Example 2 - Fluoride Tracer Analysis
Our second example shows how EPANET can be used in conjunction with tracer
studies to calibrate distribution system hydraulics. Figure 7.4 displays a portion
of an actual distribution system which was subjected to a fluoride tracer test.
Fluoride addition at the treatment plant feeding the network was turned off, and
periodic fluoride measurements were taken at several points in the network for
55 hours thereafter. Three of these sampling points are Nodes 11, 19, and 34.
Although it was possible to compute hourly total demand variation in the network
over this time period from pump station records and recorded tank elevations, no
direct measurements were made of either nodal demands or flow velocities. The
EPANET model was used to adjust values of individual nodal baseline demands
so that predicted fluoride measurements came as close as possible to observed
values.
Tank
Figure 7.4 Network for Example 2
94
-------
Figure 7.5 displays the final calibrated input data set for this example, which is
contained in the file NET2.INP in your EPANET directory. Note that in the
[PATTERNS] section of the input, three time patterns are defined. Pattern 1
pertains to the nodal demand flows (all nodes are assigned to time pattern 1 by
default). Pattern 2 is for the flow entering the network from the pump station at
Node 1. (The pump station is represented as a junction node with external source
flow rather than a true pump.) Pattern 3 defines how the fluoride concentration
level entering Node 1 died off after it was shut off at the treatment plant. Also
note that in the [REPORT] section we only ask for output at three nodes (where
fluoride samples were collected) and that there is no [REACTION] section
because fluoride is a conservative substance.
The NET2.INP data can be run either with EPANET under DOS or under
Windows. Three text files have been supplied that contain the observed fluoride
values at nodes 11, 19 and 34. They are named NET2-N11.DAT, NET2-
N19.DAT, and NET2-N34.DAT, respectively. Each line in the files contains a
time value and measured fluoride concentration value. If you run the Windows
version of EPANET, you can use its time series graph feature to compare the
predicted and observed fluoride values at each of the three sampling locations.
(See the instructions on page 82.) As an example, Figure 7.6 shows the results
obtained at Node 11.
95
-------
CTITLEJ
EPANET Example Network 2
[JUNCTIONS!
; 10
1
z
3
4
5
6
7
S
9
10
*1
12
«
14
15
16
17
'18
19
20
21
22
23
24
25
27
28
29
30
31
32
33
34
35
36
Etev.
ft.
50
too
60
' 60
100
125
160
110
180
130
185
2W
'210
200
190
150
180
100
150 ,
170
150
200
230
190
230
130
110
110
130
190
110
180
190
110
110
Demand Demand
apro Pattern
-694.4 2
8
K
8
8
5
4
9
14
$
34.78
16
2
2
2
20
20
20
5
19
16
10
8
11
6
8
0
7
3
17
17
1.5
1.5
0
1
ETANKSJ
t*
to
M{n- Ma*~
ft. Level tevel Level
ft.
26 235
EQUALITY!
56,7
50
70 5C*
; First Last Fluoride
; Node Node jng/t
* 36 1.0**'
Figure 7.5 Input Data for Example 2
(Panel 1 of 3)
96
-------
(SOURCES!-
t
*
i.
-, ID
r.
[PIPES!
t
* ,
*
i
2
3
4
5
6
'7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24 ~
25
26
27
28
' 29
30
31
32
34
35
36
37
, 38
39
s 40
41
Fluoride Source
mg/L
1.0
Head
Node
1
2
2
3
4
5
6
7
7
8
9
11
12
13
14
13
15
16
17
18
16
14
20
21
20
24
15
23
25
25
31
27
29
22
33
32
29
35
28
28
Pattern
3
'tall
Node
2
5
3
4
5~
6
7
8
9
• 10
11
12
13
X4 .
15
16
17
17
18
32
19
20
21
22
22
23
24
25
26
31
27
29
28
33
34
19
35
30
35
36
Length
ft.
2400
800
1300
1200
1000
1200
2700
1200
400
1000
700
1900
600
400
300
1500
1500
600
700
350
1400
1100
1300
1300
1300
600
250
300
200
600
400
400
700
1000
400
500
500 ,
1000'
700
300
Diam.
in.
12
12
8
8
12
12
'12
12
12
8
12
12
12
12
12
8
8
8
12
12
8
12
8
8
8
12
12
12
12
12
8
8
8
8
8
8
8
8
8
8
Rough.
Coeff.
100
100
100
100
100
100 '
' 100
140
100
140
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Figure 7.5 Panel 2 of 3
97
-------
[PATTERNS]
;0etnand pattern
t 1.26 1,04 .97 .97 .89 1.19 1.28
1 .67 .67 1.34 2.46 .97 ,92 .68
1 1.43 .61 .31 .78 .37 .67 1.26 1.56
t 1.19 1.26 .6 1.1 1.03 .73
1 .88 t.06 .99 1.72 t.12 1.34 1.12
1 .97 1.04 1.15 .91 .61 .68 .46
t .51 .74 1,12 1.34 1.26 .97 .82
1 t.37 1.03 .8t .88 .81 .81
;Pump flow pattern
2 .96 .96 .96 .96 .96 .96 ,62 0 0 0 0 0 .8 l' 1
2 f 1 .15 0 0 0 0 0 & .55 .92 .92 .92 .92 .9
2 .9 .45 0 0 0 0 0 .7 1 I 1 1 .2 0 0
2 fl 0 0 0 .74 .92 .92 .92 .92 .92
;Ftuonde source pattern
3 .98 t,&2 1.05 .99 .64 ,46 ,35 .35
3 ,35 ,35 .35 .35 .17 .17 .13 .13 .13 .15
3 .15 .15 .15 .15 .15 .15 .15 .12 .1 .08
3 ,1f .09 .09 ,08 .08 ,08, .08 .08 .08
3 .09 .07 .07 .09 .09 .09\09 .09 .09
3 .09 .09 ,09 .08 .35 ,72 .82 .92 1
[REPORT]
node It " """ ' %
node 19
node 34
Ifnks none
tTIMESl
DURATION 55
[OPTIONS!
QUALITY Fluoride
MAP NET2.MAP
EENDJ
Figure 7.5 Panel 3 of 3
98
-------
Figure 7.6 Observed (X) and Predicted Fluoride Levels at Node 11 Of
Example 2
7.4 Example 3 - Source Tracing
This third example is one that uses EPANET to determine the coverage achieved
by one particular raw water source in a two-source distribution system. Figure
7.7 is a map of the network being studied. It represents only a portion of a
larger system into which it feeds. There are two raw water sources - one that
is used continuously from high quality river water and another used for a portion
of the day that comes from lower quality lake water. In order to design an
effective sampling program we would like to determine how far into the
distribution system the lake water source penetrates under average water demand
conditions.
99
-------
LAKE
Figure 7.7 Network for Example 3
The input data for this example is contained in the file NET3.INP and because
of its size, will not be reproduced here. Some noteworthy features of this data
set are:
1. Node 4, the river source, feeds the system continuously through
either Pump 335 or Pipe 330. (See Figure 7.8 for a blow-up map
of this portion of the system.) The pump is controlled by the
water level in Tank 1. When the pump is on, Pipe 330 is valved
off. When the pump is off the pipe is opened. The statements in
the [CONTROLS] section that accomplish this are:
[CONTROLS]
LINK 335 OPEN IF NODE 1 BELOW 17.1
LINK 335 CLOSED IF NODE 1 ABOVE 19.1
LINK 330 CLOSED IF NODE 1 BELOW 17.1
LINK 330 OPEN IF NODE 1 ABOVE 19.1
100
-------
RESERVOIR 4
PUMP 335
Figure 7.8 Detail of Network Around River Source
(Not to Scale)
Pump 10, which supplies water from the lake source (Node 5),
operates only during hours 1 to 15 of the simulation (9:00 am to
11:00 pm for this example).
Connections to other portions of the system not contained in the
network occur at Nodes 15, 35, 123, and 203. The baseline
demands for these nodes are set to 1.0 and the actual hourly
demands appear in time patterns 3, 4, 2, and 5, respectively, as
defined in the [PATTERNS] section.
The [OPTIONS] section indicates that our water quality simulation
will trace the percentage of flow originating from the lake source,
Node 5, over a 24-hour period.
The [REPORT] section turns off any tabular output for links.
101
-------
1
Advisory: If you run this data set as is with EPANET under DOS, the output
report produced will run some 57 pages.
Figure 7.9, generated by the EPANET4W program, shows the extent to which
lake water propagates through the network after 14 hours of simulation. After
this, when the lake water is turned off, its coverage begins to recede. Note that
it would be very difficult using the tabular output data from EPANET itself to
visualize the spatial reach of the lake water over time.
LAKE
Figure 7.9 Spatial Coverage of Water From Lake Source After 14 Hours
102
-------
APPENDIX A
FILES INSTALLED BY EPANET
D°S
the
in your
EPANET16.EXE
EPANET32.EXE
DOS4GW.EXE
EPANET4W.EXE
EPANET.HLP
EPANET.BAT
EPANET4D.BAT
SOLVER.BAT
SOLVER.PIF
EDITOR.PIF
TEMPLATE
TE.EXE
BROWSE. COM
GETDGT.COM
KYP2EPA.EXE
NET1.INP
NET1.MAP
NET2.INP
NET2.MAP
NET2-N11.DAT
NET2-N19.DAT
NET2-N34.DAT
NET3.INP
NET3.MAP
README.TXT
standard (16-bit) version of the network simulator
32-bit version of the network simulator
DOS extender used with EPANET32.EXE
EPANET program for Windows
Windows Help file for EPANET
runs EPANET simulator from DOS
menu program for running EPANET from DOS
runs EPANET simulator from EPANET4W
Program Information File for SOLVER.BAT
Program Information File for EPANET4W editor
template file for input data
text editor program
file viewer program
utility program used by EPANET4D.BAT
translates KYPIPE input to EPANET input
input data for example network 1
map coordinates for example network 1
input data for example network 2
map coordinates for example network 2
fluoride data from node 11 of network 2
fluoride data from node 19 of network 2
fluoride data from node 34 of network 2
input data for example network 3 .
map coordinates for example network 3
latest information about EPANET
103
-------
In addition, the following dynamic link library files will be installed in your
Windows \SYSTEM directory unless a more recent version already exists:
COMMDLG.DLL
VBRUN200.DLL
VER.DLL
CMDIALOG.VBX
GRID.VBX
TEXTVIEW.VBX
(If you are are running a shared version of Windows over a network, then these
files will be placed in your local Windows directory.)
An installation for DOS only will copy the following files to your EPANET
directory:
EPANET16.EXE
DOS4GW.EXE
EPANET4D.BAT
TE.EXE
GETDGT.COM
NET1.INP
NET2.INP
NET2-N11.DAT
NET2-N34.DAT
NETS.MAP
EPANET32.EXE
EPANET.BAT
TEMPLATE
BROWSE.COM
KYP2EPA.EXE
NET1.MAP
NET2.MAP
NET2-N19.DAT
NET3.INP
README.TXT
104
-------
APPENDIX B
ERROR AND WARNING MESSAGES
Error Number Description
Error in trying to open a temporary scratch file used by
EPANET to save hydraulics results. May be caused by disk
being full.
Error in trying to open the file containing hydraulics results
from a previous run. Most likely caused by an incorrect
file name used in the HYDRAULICS USED line of the
[OPTIONS] section in the input data file.
The hydraulics file specified in the [OPTIONS] section of
the input data does not appear to match the network
described in the input.
12
j. . . -. _ —
described in the input
13 Cannot read data from the hydraulics file. Most likely
caused by specifying a file that does not contain hydraulics
data.
14 Cannot write results to the output report file. Could be due
to an illegal file name or the disk being full.
15 Error in trying to open the network verification file. Most
likely caused by an incorrect file name in the
VERIFICATION line of the [OPTIONS] section of the
input file.
105
-------
Error Number Description
101 Not enough memory to store network data.
102 Not enough memory for hydraulic analysis.
103 Not enough memory for water quality analysis.
104 Not enough memory to write the output report.
200 One or more errors were detected in the input file.
201 Format error in input line.
202 Node defined in input line was previously defined.
203 Link defined in input line was previously defined.
204 Input line refers to an undefined node.
205 Input line refers to an undefined link.
206 Incorrect pump curve data (e.g., the pump curve is not
concave or does not have decreasing head with increasing
flow).
207 Input line refers to an undefined time pattern index.
208 Input line contains an illegal numerical value.
209 Input line contains illegal pipe data (e.g., values less than
or equal to zero).
210 Tank levels are mis-specified (e.g., a minimum level was
greater than the maximum level).
211 There are no tanks or reservoirs in the network.
212 One end of a control valve is a tank or reservoir.
106
-------
Error Number Description
213 An unconnected node was detected. The node ID is
displayed.
214 The verification file indicates a different set of links
connecting to a node than are found in the input file.
215 There are not enough nodes in the network.
216 An illegal status was set for a check valve (e.g., OPEN or
CLOSED).
217 The reporting start time is greater than the simulation
duration.
218 The input line contains more than 80 characters.
300 The network hydraulic equations become ill-conditioned
and cannot be solved. The program indicates which node
is causing the problem and prints a status table for the
network at this point in time.
400 The water quality transport equations could not be solved.
107
-------
Warning Message
Pump cannot deliver head.
Pump cannot deliver flow.
Flow control valve cannot
deliver flow.
System unbalanced1.
Suggested Action
Use a pump with a larger shutoff head.
Use a pump with a larger flow capacity.
Reduce the flow setting on the valve or
provide additional head at the valve.
Use the STATUS FULL command in the
[REPORT] section of the input to identify any links
whose status keeps switching back and forth
between iterations of the network hydraulic
equations.
Situations that produce this condition might include inconsistent pressure control
levels and closed-off tanks or links that isolate a portion of the network from any source.
Computational results produced from an unbalanced network are not physically
meaningful. If the condition persists over several time periods, it could result in an ill-
conditioned set of hydraulic equations that cannot be solved.
108
-------
THE KYP2EPA CONVERSION PROGRAM
KYP2EPA is a program that takes a Kentucky Pipes fKYPIPF^ inrmt A * *
converts it into an EPANET input data set The KYPTPP J? P ^ S6t —
KYP2EPA kypfile epafile
where kypffle is the name of an existing KYPIPE
the EPANET file to be produced.
o A ^ • ,
d 6pafile 1S the name of
to
was
the
Aided
109 .
-------
110
-------
APPENDIX
TROUBLESHOOTING
EPANET fails to install correctly under Windows.
Try one or more of the following:
1. Shut down all other Windows applications that may be running (such as
the Clock) and launch the setup program again from Program Manager.
2. Exit from Windows, change directories to the Windows directory
(typically c:\windows), and repeat the installation procedure.
3. If using Windows for Workgroups, exit from Windows, re-start Windows
in Standard mode, and repeat the installation procedure.
When running under Windows, a message appears saying that the
Editor or the Simulator cannot be found.
Make sure that the files EDITOR.PIF and SOLVER.PIF reside in your EPANET
directory. If you have replaced the default editor that comes with EPANET then
make sure that you have modified the EDITOR.PIF file correctly (if using a DOS
editor) or have modified the EPANET.INI file correctly (if using a Windows
editor). See Section 3.4 for instructions on modifying these files.
The EPANET simulator returns with one of the Not Enough
Memory error messages.
Try one or more of the following:
1.
Use the 32-bit version of the simulator if you have an 80386 or higher
CPU with extended memory on your PC. See Section 3.4 for instructions
on how to run EPANET in 32-bit mode.
Ill
-------
2. If the error message occurs when running under Windows, try running the
same input file under the DOS version of EPANET. By adding a third
file name to command line that launches EPANET under DOS, you can
view the results of the run at a later time under Windows. After starting
up EPANET for Windows again, select that file to load under the Open
Output option of the File menu.
3. If the error message refers to a water quality analysis (error number 103),
then make sure that your QUALITY TIMESTEP parameter in the
[OPTIONS] section of the input file is not too small and that your
SEGMENTS parameter is not too large.
An EPANET run produces a System Unbalanced warning
message.
This condition typically occurs when a pump or valve keeps switching its status
back and forth between successive iterations of solving the hydraulic equations
for the network thus causing the system to fail to converge to a solution. Possible
causes for this are a pair of pressure controls that turn a pump on and off whose
pressure settings are too close together, or a collection of pressure regulating
valves whose pressure settings influence the status of one another. You can
specify the STATUS FULL option in the [REPORT] section of the input to
identify those links that might be behaving in this manner and then modify their
pressure settings to avoid this situation. Another option would be to use a
slightly larger value for the ACCURACY parameter in the [OPTIONS] section
of the input and see if this solves the problem.
An EPANET run produces obviously incorrect results (e.g., large
negative pressures) or stops running with Error Message 300
(Cannot Solve Network Hydraulic Equations).
First check your input data to make sure that:
1. all parameters are in their correct units (e.g., inches (or millimeters) for
link diameters, millifeet (or millimeters) for Darcy-Weisbach roughness
coefficients, psi (or meters) for valve pressure settings),
2. parameters are entered in the right order (e.g., pipe length precedes pipe
diameter, the node ID on the suction side precedes the node ID on the
discharge side for pumps),
112
-------
3. tank levels are specified as height above the elevation of the tank bottom
(and not as total elevation).
Then check your output for the following conditions that will cause a portion of
the network to become isolated from any source of supply:
1. pumps that are shut down because they cannot deliver the required head
(try using a pump with a larger shutoff head),
2. valves that close to prevent reverse flow (either modify the pressure
setting for the valve or provide more head on its upstream side),
3. tanks that reach their minimum water level (either reduce the minimum
level or have supply pumps online before this condition is reached).
Note that it is possible for several of these conditions to exist at the same time.
113
------- |