United States
Environmental Protection
Agency
Environmental Research
Laboratory
Duluth MN 55804
Research and Development
EPA-600/S3-82-066 Oct. 1982
Project Summary
An Assessment of Wind
Characteristics and Wind
Energy Conversion Systems for
Electric Utilities: Wisconsin
Power Plant Impact Study
Carel C. DeWinkel
This study assesses the potential
contribution of wind energy conversion
systems (WECS) to the generation of
electricity by utilities in Wisconsin and
parts of adjacent states. The final report
contains a review of the literature on
wind and wind machines, an analysis
of wind characteristics in Wisconsin
and adjoining areas, and an analysis of
WECS applied to the Dairyland Power
Cooperative (DPC), a rural-based
cooperative serving parts of Wisconsin,
Minnesota, Iowa, and Illinois.
Along the shorelines of the Great
Lakes, wind energies are among the
highest in the U.S. Inland, wind
energies in many areas are suitable for
WECS; however, careful study is
required to select the most favorable
sites. The DPC, with its many small,
scattered users, is the type of utility
for which wind systems are ideally
suited. Peak winds correspond with
the utility's winter peak loads for
electric heating. An economic analysis
indicates that WECS can be econom-
ically feasible for DPC in the 1980s.
The analysis considers WECS only as
peak load fuel-savers; if their role in
contributing to base-load generating
capacity were included, the economic
feasibility would be even greater.
Combined with a controlled load of
water heaters, WECS would lend
diversity and stability to the power
system and delay or eliminate the need
for more costly additions to generating
capacity.
A detailed reliability and cost product
analysis of WECS plus direct control
of a variety of loads, as well as a
thorough wind survey, is required for
accurate evaluation of any application
of wind systems. A model for such
analysis is presented in this report.
This Project Summary was devel-
oped by EPA's Environmental Research
Laboratory. Duluth. MN. to announce
key findings of the research project
that is fully documented in a separate
report of the same title (see Project
Report ordering information at back).
Introduction
Wind machines represent an old,
prdven technology. Wind-powered
electric generators, developed near the
end of the last century, provided
electricity in many parts of the world.
Cheap, abundant fossil fuels eventually
halted widespread use of wind power.
Now, however, energy shortages and
environmental concerns, coupled with
the availability of new materials, and
advances in electronics, and control
technology are making wind energy
conversion systems (WECS) attractive
once more.
This study assesses the use of WECS
by utilities for generation of electric
power in Wisconsin and portions of
adjacent states. The final report contains
the following sections:
1. A review of the literature on wind
characteristics.
-------
2. An analysis of wind characteristics
in Wisconsin and adjoining areas.
3. Major characteristics of wind
machines on the market or under
development.
4. Power output of selected wind
machines under conditions found
in Wisconsin and parts of adjacent
states.
5. Analysis of WECS applied to the
Dairyland Power Cooperative (DPC),
a rural-based cooperative serving
parts of Wisconsin, Minnesota,
Iowa and Illinois.
6. Analysis of the potential for
combining WECS and load control
to meet DPC's peak load.
7. Appendices. Data for 12 locations
on: monthly average wind speeds
and wind power density at four
heights, average monthly capacity
factors for three wind machines,
annual average power duration
curves for three wind machines,
and cost estimates for wind ma-
chine and site-development for
large WECS.
Findings and Conclusions
Energy Use in Wisconsin
In 1976, Wisconsin's power sources
for primary energy generation were:
40% petroleum, 27% natural gas, 23%
coal, 9% nuclear, and 1% hydro;
virtually all of these sources were
imported. Any substitution of local,
energy sources would make the state
less dependent on imports and more
flexible in its energy systems.
An important characteristic of Wis-
consin is the distribution of its population
and activities. About 33% of the
population live in rural areas, and 54 of
the 72 counties have population densi-
ties lower than 40 persons/km2. The
Dairyland Power Cooperative (DPC)
typifies a utility serving small, scattered
users. About 76% of DPC's energy
requirements are for farm, rural resi-
dential, seasonal, and town residential
customers who make up 96% of the
total users. Wind energy sources are
particularly suitable for this kind of
decentralized pattern of energy use.
Availability of Wind Power
At twelve airports assessed in Wis-
consin and adjacent states, average an-
nual wind speeds usually vary between
4.5 and 6 m/sec at 7 m and between 6
and 8 m/sec at 60 m. The correspond-
ing average wind power density, which
is related to the cube of the wind speed
and consequently increases rapidly as
wind speed increases, is between 100
and 200 W/m2 at 7 m and 250 and 500
W/m2 at 60 m. Coast Guard stations
along Lake Michigan and Lake Superior
record wind speeds equal to or higher
than those at the airports.
The values reported in this study are
not necessarily typical of the best wind
power locations. The importance of
local factors in determining wind power
densities is illustrated by the difference
between power densities at Rochester,
Minnesota, and La Crosse, Wisconsin.
These cities, which are only about 80
km apart, have average annual wind
power densities at 60 m of 517 W/m2
for Rochester, located on a ridge, and
152 W/m2 for La Crosse, which is
situated in a valley. The Great Lakes
also have a profound effect on the
winds. Average annual wind speeds of 6
to 8 m/sec are the norm near the Great
Lakes shoreline. These winds correspond
to wind power densities of 225 to 375
W/m2 — among the highest in the U.S.
Inland and at sheltered coastal meteo-
rological stations, however, average
wind speeds are only 3.5 to 5 m/sec.
Wind data have not been systematically
recorded at the best wind sites; therefore,
the potential for WECS in the Great
Lakes region has not been fully explored.
The variation in wind power over time
means that a given power output cannot
be guaranteed unless storage devices
are included in the system. Average
annual wind speeds at any site can be
predicted with an accuracy of 85% to
90%. Similarly, the seasonal pattern in
wind speeds remains fairly constant on
a time scale of a month or more. It is the
short term, hour-to-hour and day-to-day
fluctuations in speed that must be dealt
with.
Because of the intermittent and
variable nature of the wind, detailed
analysis of WECS and utility load
characteristics is necessary if WECS
are to be connected to a utility grid. An
array of wind machines spread over a
large area offers advantages in reliability
of power output over concentrated
clusters of wind machines. These
advantages must be weighed against
the higher operating and maintenance
(0 and M) costs of dispersed WECS.
Characteristics of Wind
Machines
The wind machines considered in this
study are designed to operate with a
constant rotational speed. The power
output of the machines is rated at a
designated wind speed, and is limited to
this value. Although this results in some
loss of potential power at higher wind
speeds, it also reduces the size and cost
of the generating system.
Associated with the rated wind speed
and rated power is the rated power
density of the machine, or the power per
unit area swept by the blades (W/m2).
Given a certain rated power capacity, a
large rotor (having a low rated power
density) will reach its rated capacity at a
relatively low wind speed. Such ma-
chines are efficient at low wind speeds
but waste part of the energy available in
high winds. Conversely, small rotors
(with high rated power densities) reach
the same rated power capacity at a rela-
tively high wind speed and are therefore
better suited to high wind regimes.
Rated power densities of about 100 to
200 W/m2 seem most suitable for the
wind regime of Wisconsin.
The power output of a wind machine
also can be presented as a curve
showing the percentage of time that
each level of power output is equaled or
exceeded. A wind machine with a low
rated power density reaches its rated
level more often than one with a high
rated power density. It is important to
select the wind machine best suited to
the wind characteristics of each site.
Costs of Wind Systems
Present costs per kW of capacity are
generally in the range of $500 to $1500,
although Southern California Edison
paid only $356/kW for a large new
WECS and General Electric installed a
1.5 MW system costing $1586/kW as
part of the Federal Wind Energy Program.
Annual costs for O and M are
between 0.5% and 3% of the investment,
depending on the size and other
characteristics of the system Other
costs are taxes and insurance, which
vary geographically.
Although uncertainties exist in both
capital and O and M costs for WECS,
these uncertainties are much smaller
than those for coal or nuclear power.
Furthermore, most uncertainties in
costs of WECS involve how much
present costs can be decreased through
experience and mass production, while
uncertainties for coal and nuclear
power involve expenses for health,
regulation, and environmental protection
as well as rising fuel prices and
generally tend to increase costs rather
than decrease them.
-------
Analysis of WECS for the
DPC
The major components oftheanalysis
were as follows:
1. Evaluation of siting and wind data.
2. Choice of type of WECS.
3. Selection of different penetration
levels of WECS, based on load and
energy forecasts.
4. Calculation of new monthly dura-
tion curves, based on projected
hourly WECS output and loads.
5. Calculation of production cost for
proposed plan plus WECS.
6. Calculation of savings in production
costs (compare proposed plan with
and without WECS).
7. Calculation of break-even capital
cost of WECS (considering lifetime
system costs for 0 and M, taxes,
and insurance).
8. Comparison of break-even capital
cost of WECS with estimated
capital cost: is WECS economically
competitive?
The three WECS variables were:
wind machines with rated power
density of either 90 W/m2 or 220
W/m2; WECS capacity of from 0 to 400
MW; and 0 and M costs for WECS from
1% to 3% of capital costs (a 5% level
might apply to dispersed small scale
systems). In addition, fuel saving
resulting from the WECS was calculated
for annual increases in fuel prices
ranging between 5% and 12%.
With 100 MW capacity and the range
of other variables indicated, the break-
even cost of WECS was found to be be-
tween $550 and $1300/kW for wind
machines rated 220 W/m2, and be-
tween $800 and $1800 for those rated
90 W/m2. Therefore, WECS will fail to
be economically competitive only if fuel
prices increase less than anticipated
and if cost estimates for the wind ma-
chines of $700 to $1000/kW are too
optimistic.
The analysis considered only the
ability of WECS to save peak-load fuels.
It did not take into account contributions
of WECS to base-load capacity. Because
the most expensive peak-load fuels
would be displaced first by WECS, the
greatest economic advantages appar-
ently accrue from the first 100 MW of
WECS capacity installed. The break-
even cost is highest for 100 MW
capacity and lower at the 400 MW
level. The break-even cost is highest for
100 MW capacity and lower at the 400
MW level. The break-even cost would be
higher and the decrease in break-even
cost with additional WECS capacity
would be less if the ability of WECS to
substitute for between 10% and 15% of
base load fuels were taken into account.
Furthermore, the wind data employed in
the analysis are for airport locations,
which are not the best WECS sites. For
all these reasons, the estimated benefits
of WECS are conservative.
WECS and Load Control
Load control techniques and energy
storage systems can be used by a utility
to flatten its load curve and increase
average load factors. There are two
major load control strategies:
1. To drop or defer certain loads
during times of high marginal
operating costs or high probability
of system failure.
2. To shift loads on a regular basis to
times when they can be provided
for most economically.
Certain loads in the industrial and large
commercial sectors are candidates for
the first strategy, while many loads in
the residential and small commercial
sectors can be shifted on a daily basis
according to the second.
A growing proportion of DPC's load is
for residential and commercial low
quality energy tasks such as space
conditioning, refrigeration, and water
heating. Growth in this load makes load
control and dispersed thermal energy
storage systems not only practical but
perhaps necessary. The large and
growing space heating component of
DPC's load also contributes to the
strong positive correlation between the
utility's load and the availability of wind
power, and enhances the economic
benefits and technical feasibility of
WECS. Peak winds and peak demands
both occur in winter and early spring.
Several other utilities serving Wisconsin
have, or predict, similar winter peaking
loads.
An hour-by-hour analysis of DPC's
load and WECS power output with a
range of assumptions showed that
WECS plus load control may meet peak
loads at least as reliably as conventional
peak load generators. A WECS capacity
of 1.0 to 1.3 times the peak load range,
coupled with a controlled load of water
heaters equal to 25% of the WECS
capacity, will meet the peak load
between 93% and 96% of the time while
saving 94% to 99% of peak load fuels
and 10% to 15% of base load fuels. The
WECS alone could meet the peak load
more than 60% of the time; load control
would be used only about 35% to 40% qf
the time that the peak load range is
reached.
Break-even costs for WECS increase
by at least 10% to 15% when load
control is combined with the system.
The combination of WECS and load
control may also delay or eliminate the
need to add new conventional generating
capacity. Therefore, according to this
preliminary analysis for Wisconsin,
WECS plus load control appears to be a
suitable alternative — economically
sound and socially and environmentally
acceptable — to additions of coal or
nuclear capacity. Detailed analysis is
warranted and recommended.
Benefits of WECS
Economic benefits are only one type
of benefit to be obtained from generation
of electricity by WECS. The final report
discusses various social benefits, such
as creation of skilled local jobs. In
addition, to the extent that wind power
replaces power from coal and uranium,
the problems resulting from these
technologies will be diminished. WECS
can be installed without a long lead time
(about 2 years, as compared to as much
as 12 years for a nuclear facility).
Additions of WECS can be made on an
incremental basis and adapted to local
conditions. WECS increases the diversity
of energy resources and the stability
and resiliency of energy systems. All
these features are valuable in times of
uncertainty.
Recommendations
1. Wind data presently available are
insufficient for an accurate as-
sessment of WECS. Initiate a wind
energy survey to identify locations
that have good wind availability
and are close to utility load
centers.
2. Participate in Department of Energy
demonstration and test programs
of both small and large WECS for
use in all sectors.
3. Initiate state-funded demonstra-
tion/test programs of WECS to
complement the federal program.
4. Create additional incentives such
as property tax exemptions and
low interest loans for WECS
owners, including utilities.
5. Include detailed evaluation of
WECS in the assessment of ad-
vance plans for utilities. These
analyses should address issues
such as a change in the mix of
conventional generators (fewer
base-load units); possible reduction
ft U.S. GOVERNMENT PRINTING OFFICE: 1882 -559-017/0845
-------
in conventional generating capac-
ity; the value of shorter planning
lead times, maintenance planning
of small units, and diversity in
energy resources; and the poten-
tial of load control.
6. In evaluating possible WECS,
consider the benefits of increased
local control and diversity and the
creation of skilled local jobs.
7. Address the problem of the sale of
electricity by non-utilities back to
the utility or to third parties.
8. Assess the nature of the utilities'
load in terms of the quality of
energy needed by consumers. An
increasing part of the load is for
tasks such as space heating/cool-
ing and water heating that can use
energy sources of lower quality
than electricity. Compare the
advantages of having some low
quality load that could be met by
thermal storage systems, thereby
flattening daily load curves, with
using the second law of thermo-
dynamics to set priorities among
the consumer activities served.
9. Question whether consumers
have the right to electric utility
service without any restrictions on
the type of system installed.
Consider mandatory controlled
storage heating systems for space
and water to reduce growth in the
peak demand for electricity.
Carel C. DeWinkelis with the Institute for Environmental Studies, University of
Wisconsin, Madison, Wl 53706.
Gary E. Glass is the EPA Project Officer (see below).
The complete report, entitled "An Assessment of Wind Characteristics and Wind
Energy Conversion Systems for Electric Utilities: Wisconsin Power Plant
Impact Study," (Order No. PB 82-258 971; Cost: $15.00, subject to change)
will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Environmental Research Laboratory
U.S. Environmental Protection Agency
6201 Congdon Blvd.
Duluth, MN 55804
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
PS 0000329
U S ENVIR PROTECTION AGENCY
REGION 5 LIBRARY
230 S DEARBORN STREET
CHICAGO 1L 60604
------- |