HomeMy WebLinkAboutBethel Wind Power Project Feasibility Study - Mar 2015 - REF Grant 7040015AEA – 7040015 TDX Power, Inc.; Coleman
Page 1 of 23
AEA Renewable Energy Fund Grant Round IV
Bethel Wind Power Project
Phase 1 Deliverable: Feasibility Study
TDX Power
March 1, 2015
This Feasibility Study summarizes the work done to date to analyze the potential for wind
power production integrated into the Bethel utility system. This summary report follows
on two earlier TDX reports, which covered:
- Current status of the utility, its infrastructure, loads, thermal loop and operations.
- Potential wind farm sites and distribution limitations
- Power plant controls assessment for wind farm integration
The two earlier reports in PowerPoint format are in the Appendix.
This report focuses the sizing of potential wind farms and the inherent issues and
limitation imposed by the site, environment and utility infrastructure.
An initial task was to define a set of criteria for the development of future wind farms that
were gated by the amount of impact to the utility infrastructure and operational protocols.
The goal was to define the maximum feasible wind farm size that fit each level or gate.
These AVEC selected criteria set the level of changes that will be required within the utility
infrastructure (power plant, controls and distribution).
The three defined levels proposed are:
o Low – No changes
o Medium – Minor changes only:
Add secondary loads
Reasonable transformer upgrades
o High – Significant system wide upgrades.
AEA – 7040015 TDX Power, Inc.; Coleman
Page 2 of 23
Bethel Electrical Load Overview:
The load on the Bethel electrical utility ranges from a low of 3 MW to a peak 7 MW, with
typical Alaskan seasonal variations. In the recent past there have been minor variations
due to weather, but no significant change in overall demand. There is some current and
future growth anticipated for the Bethel electrical demand, but that was not predicted, nor
incorporated into this study.
2014 2010 2013 AVERAGE 2014 2010 2013 AVERAGE
Max Max Max MAX Min Min Min MIN
Jan 6850 7025 6575 6817 3975 3950 3800 3908
Feb 6775 6475 6675 6642 4200 3800 3850 3950
Mar 6656 6525 6525 6569 4050 3990 3925 3988
Apr 6350 6090 6000 6147 3850 3600 3500 3650
May 5950 5550 5950 5817 3400 3200 3400 3333
June 5400 5225 5400 5342 2875 3125 2875 2958
July 5400 5275 5400 5358 2875 3150 3100 3042
Aug 5575 5425 5575 5525 3350 3300 3350 3333
Sept 5660 5725 5660 5682 3420 3300 3420 3380
Oct 5950 5975 5950 5958 3550 3600 3550 3567
Nov 6850 6300 6700 6617 3975 3900 3050 3642
Dec 6675 6800 6675 6717 3825 4050 3825 3900
Diesel dispatch in Bethel currently is operator controlled, from initial engine prep, to
starting and warm up and then to a manual synchronization and stabilizing process.
Typically the entire start sequence takes around 20 to 30 minutes and is fully ingrained
into the operational protocol. The five 900 rpm EMDs are rated at 2.2 MW each and
currently their minimum power output is limited at 1.2 MW (55% of rated). The minimum
number of diesel engines under any situation is two and the maximum is currently four
engines.
The introduction of wind energy will lower the diesel fuel consumption, however a number
of operational impacts need to be assessed, and at each of the three levels the issues
change. These issues will be identified and addressed in the analysis of each level.
An overriding issue for any wind integration into Bethel will be the capability of the EMD
governors and voltage regulators. A program to address this issue is being contemplated,
but has not been accomplished. Given the age of the equipment and the manual control
adjustment currently required, there is sufficient evidence that this could create a limiting
factor for any changes to the power plant. It is presumed that rapid ramping of the EMDs in
response to changes in wind turbine output may destabilize the control system. Currently
ramps rates are very minimal under normal operations so there is no firm evidence, but the
concern certainly exists. TDX would propose a test plan to evaluate this situation as part of
the next phase of activity.
This report focused on the dispatch protocol for the engines, as it relates to the level of
wind power being contemplated. This approach provides a solid high level understanding
AEA – 7040015 TDX Power, Inc.; Coleman
Page 3 of 23
of the basic performance parameters (fuel savings, diesel run times, wind penetration,
wind curtailment and thermal energy) using basic diesel control requirements as defined
by the current EMD engines and their controls.
It is hoped that this evaluation of levels will guide AVEC to select a configuration (level) and
rough wind farm size that fits their needs, budget and operational goals. Once this selection
is made then TDX Power can begin work on the Conceptual Design.
Performance Modeling Process
Homer was selected as the hybrid wind system performance modeling program for this
analysis. There are multiple limitations with Homer which needed to be overcome to
develop a realistic model for the hybrid system operations.
Wind Turbine Performance Adjustment
- Turbine and Wind Farm Losses:
Degrading wind energy output for real world simulations. Homer does not take into
account normal wind turbine, wind farm and distribution system losses when
predicting system performance. In our exercise, we used two methods to solve this
problem. First, we degraded the power curve for the wind turbine to account for
typical losses, which for practical purposes occur continuously. These included the
following:
Typical Turbine/Farm Losses
Tuning 2%
Dirt 2%
Array 1%
Transmission 5%
Turbulence 1%
Misc 1%
Total 12%
- Operational Reduced System Performance:
Operationally, the wind farm output is also limited by turbine and farm availability,
which tends to be longer in rural Alaska since repair time is hampered by long
distances and lack of infrastructure. Likewise the utility will have needs to Curtail
the turbine output during utility maintenance events or other situation which might
arise. In addition, Bethel is subject to icing at a significant level in the winter.
Anemometer data indicated icing sufficient to impact the anemometer system in
excess of 30% for 3 months and some impact of icing for up to 8 months a year.
These losses combined are 26% .
AEA – 7040015 TDX Power, Inc.; Coleman
Page 4 of 23
Operational Impacts
Lack of
Availability 10%
Curtailment 3%
Icing 13%
Total 26%
-
For the EWT 900 on a 75 meter tower using the wind data collected by AEA in 2006 the
wind turbine predicted outputs were adjusted for these impacts.
EWT900 - 75 Meter Tower
AEP Predictions
MWh/yr Capacity Factor
Homer uncorrected 3,095 39.3%
WaSP 3,135 39.8%
12% turbine losses 2,724 34.6%
26% operational 2,016 25.6%
The losses modeled appear to be quite conservative, in that, well run wind systems in the
lower 48 experience significantly less over all losses but cumulatively, the total losses are
in the range of AEA’s past experience for rural Alaskan community wind systems.
Homer Modeling Assumptions
Within the Homer model there are a number of factors that can be changed to adjust the
operational decisions of the model. To better simulate the Bethel power plant, two gensets
were forced on continuously, no matter what the wind output or total load was. In
addition, genset dispatch decisions included having operational reserve (diesel capacity)
for 10% load variation and a 20% wind variation within the time step. The modeled
thermal load roughly twice the current demand, with seasonal variations based upon
Bethel temperature data. One hour time steps were used for simulation.
Homer Results
The Homer results were calculated for the various size wind farms sizes with a focus on the
amount of diesel fuel saved. The wind turbine power curves were reduced by 12% at all
AEA – 7040015 TDX Power, Inc.; Coleman
Page 5 of 23
wind levels to account for the typical losses, which are essentially constant. Then, the
Homer performance results were then reduced by 26% for the Operational losses
identified. The assumption is that 26% of the time the turbine is not operational, so the
utility is essentially operating without a wind farm. The remainder of this report will
include the impacts on wind farm performance due to both the turbine/farm losses of 12%
and the Operational losses of 26%.
-
200,000
400,000
600,000
800,000
1,000,000
1,200,000
0 2 4 6 8 10 12Gallon per YearWind Farm MW
Utility Fuel Savings
Uncorrected
26% Reduction
AEA – 7040015 TDX Power, Inc.; Coleman
Page 6 of 23
Low Level Analysis: Requirements for Wind Farm Sizing
o Max amount of wind power without any changes to the power plant, controls
strategy or distribution.
o No Operational Protocol Changes
With the current power plant and controls the maximum amount of wind energy that can
enter the system under ANY condition and have no adverse effect on the power plant is 500
kW. However, this limiting condition (500 kw) exists only when total demand is between 4
and 4.5 MW and a genset dispatch decision is eminent.
Specifically the condition exists when the load is around 4 MW and operators must decide
to start or stop a third engine. If the wind exceeds 500 kW, then three engines will be
operating below their current minimum operational limit.
If two engines are running, and the load grows even moderately, and the wind turbine
output suddenly drops, then the two engines may become overloaded, until a third engine
can be brought on-line (minimum of 30 minutes). Short-term operation under this
condition would certainly not be destructive to the system but would fall outside the
current operational protocol.
The following graphic highlights this concern. The wind farm is modeled at 1 MW capacity.
The upper and lower stair steps are the minimum and maximum rated output for the diesel
genset with 1, 2, 3 or 4 engines dispatched. Each step in the max and min output lines
represents an additional operating generator.
-1
0
1
2
3
4
5
6
7
8
9
10
0 2 4 6 8 10MW of DemandMW of Supply
"w 1 MW Wind"
DEMAND
Maximum
Minimumminimum loadArea of Concern
AEA – 7040015 TDX Power, Inc.; Coleman
Page 7 of 23
When looking at these curves, these 1 and 2 MW lines represent the turbines rated output
which only occurs at high winds. The winds in Bethel will reach turbine rated power 10%
of the year. The Bethel load currently is below 4.5 MW about 20% of the time, mostly at
night, so this concern is real and it would happen almost any night of the year. Grossly this
means there is a relatively small amount of time during the year when the turbine power
control must be implemented and a relatively small amount of curtailed energy.
Options:
- Pick a wind turbine model with controllable adjustable power set point. This would
allow a controller to automatically dial back the wind turbine output to avoid under
loading the diesel engines. This type of control would allow for a much higher rated
wind farm, whose practical limits might be the financial metrics associated with lack
of revenue due to excessive curtailment of wind power.
- The EWT-900 and the GE 1.7 MW turbine allow for controllable power adjustment
while the turbines are operational.
The following graphic shows the areas of concern (diesel dispatch) for wind farms of 1 and
2 MW capacities. With a 2 MW wind farm the area is expanded up to 7 MW.
From just the dispatch analysis, summarized above, it appears that a 2 MW wind farm with
only wind turbine controlled power output control is doable. It appears that this technique
would be deployed across a wide range of load conditions, whenever the winds are high.
Wind curtailment would occur most when the winds are high and the load is low.
-2
0
2
4
6
8
10
0 2 4 6 8 10
DEMAND
Maximum
Minimum
w/ 1 MW Wind
w/ 2 MW Wind
Areas of Concern
AEA – 7040015 TDX Power, Inc.; Coleman
Page 8 of 23
To analyze this situation will configured the Homer model with two diesel engines forced
on, so that there would always be a solid diesel base load, and then look at the amount of
“excess” energy that would be kicked off to the secondary load. This should provide a
reasonable assessment of the amount of energy that would need to be curtailed, since both
control techniques are geared to address the same problem. So in our case instead of
shunting the excess energy to a secondary load, it was simply curtailed by dialing back the
output of the wind turbines.
The following graph summarizes the curtailed energy that would be lost as the wind farm
size increases. The modeling shown is for the EWT 900 turbine where from 0 to 5 units
have been deployed (0 to 4.5 MW)
With wind farms 2.7 MW the total curtailment is near zero. As the wind farm grows the
percentage of wind energy curtailed increases. At 4.5 MW the curtailed wind energy is
close to the annual output of a single turbine. The curtailment of this significant amount of
energy reduces the economic viability of this large a wind farm, if curtailment is the only
control technique.
The following graphs provide comparison of the curtailment activity for 2.7 MW and 3.6
MW wind farms. The amount of total curtailment (kW) and timing for curtailment as
predicted by Homer are shown to illustrate the impact. These graphs provide relative
information because they don’t include the 26% Operational Impacts, since those losses
cannot be modeled within the current version of Homer.
-
2,000
4,000
6,000
8,000
10,000
0 1 2 3 4 5MWH per YearWind Farm MW
Useful vs Excess Wind Energy
Useful Wind MWH
Wind Heat MWh
AEA – 7040015 TDX Power, Inc.; Coleman
Page 9 of 23
Therefore the following graphs are valid but should be dialed back across the board by
26% to account for the wind farms non-operational periods.
Homer Modeling for 2.7 and 3.6 MW Wind Farm for Curtailed Wind Energy
2.7 MW Wind Farm
3.6 MW Wind Farm
The mean level of curtailment with the 2.7 MW wind farm is about 25 kW with peaks of 1.2
MW. With a 3.6 MW wind farm the mean increases to 100 kW with a peak of 2 MW.
Viewing the same information as a Duration Curve, direct from Homer without the 26%
Operational Losses. These graphs illustrate that active power control for the wind turbines
is a viable technique without incurring excessive wind energy losses.
AEA – 7040015 TDX Power, Inc.; Coleman
Page 10 of 23
2.7 MW Wind Farm
3.6 MW Wind Farm
The ultimate value of integrating the wind farm with the utility is the reduction if diesel
fuel burned by the power plant. The predicted fuel saving in the operational scenario being
considered is presented in the following graphic for a range of wind farm sizes. This graph
takes into account the 12% wind turbines losses plus the 26% Operational losses.
AEA – 7040015 TDX Power, Inc.; Coleman
Page 11 of 23
It is interesting to note there is no difference in system wide performance between
curtailing wind energy by controlling the output of each turbine and using secondary dump
loads. Both control strategies perform the same basic function. Essentially the same MWh
per year would be accounted for in each scheme.
A third alternative would be to adjust the operational protocol for dispatching of engines
based upon the wind energy available; essentially using the wind energy to avoid starting
additional diesel engines. This would save more fuel but would require operators to track
the wind farm performance and have some predictive information on the wind and load to
make dispatch decisions.
The fuel savings increases predicted from this more risky scenario at the 3 MW size make
only marginal gains of 25,000 gallons/yr, which does not seem to justify the Operational
Protocol changes required.
Given the definition of the Low Impact scenario is appears that a wind farm up to 3.6 MW
could be viable with only active controlled power output control.
Distribution Limitations:
A preliminary analysis of the distribution system shows that currently the transformers,
not the power lines, are the limiting factor. Currently it appears that the transformer
would limit the max wind farm sizes at 3 MW for the city site and 2 MW for the KYUK site.
The limitation is imposed by the thermal capacity of the transformers. In actual operation
of the wind farm the electricity would be flowing back from the farm, feeding the loads on
the distribution before entering the powerhouse. Clearly the loads on the distribution will
need to be factored into the analysis to understand what additional wind curtailment might
-
100,000
200,000
300,000
400,000
500,000
600,000
700,000
800,000
0 2 4 6 8 10 12Thousands of Gallons/YrMW of Wind Farm
Utility Fuel Savings
AEA – 7040015 TDX Power, Inc.; Coleman
Page 12 of 23
occur, if during high wind power periods when the loads are low, the capacity of the
transformers might be exceeded.
Conclusion:
The Low Level wind farm should probably be limited to 3 MW is size and be located on one
distribution feeder. This is a relatively low risk, low cost option, for AVEC to pursue.
AEA – 7040015 TDX Power, Inc.; Coleman
Page 13 of 23
Medium Level Analysis
Requirements for Wind Farm Sizing
o Max amount of wind with minor changes to BEU system
Controllable wind turbine power adjustment optional
o Minor Allowed Changes
Allowed System Changes
Secondary Thermal Loads
Transformer upgrades
o Improved thermal handling performance
Fins, fans, pumps etc
NO
Power plant impacts
No distribution impact
Approach:
Install controllable Secondary Thermal Loads to limit wind power disrupting current
Operational Protocols or impacting diesel dispatch strategy. The excess wind energy
would not be curtailed but used with the current thermal heating loop to provide useful
energy.
Installation of the dispatchable thermal heaters into the thermal heating loop, after the
loop has been modified to include a heat exchanger to isolate the thermal loop into town
from the engine cooling water. Currently, the engines can deliver 6 to 7 MW of heat to the
thermal loop with average contribution of roughly 4 to 5 MW. The current thermal loops
efficiency and effectiveness was not addressed by this study, but it appears that not all the
available diesel engine waste heat is currently being utilized. The actual capacity limits,
customer demands and performance for the thermal loop need to be better defined, but for
the purposes of this study it appears viable that wind energy could augment the waste
diesel heat to provide usual thermal energy to customers in Bethel.
Secondarily, if the thermal loop were isolated with a heat exchanger than installing exhaust
gas heat recovery units on the existing diesel units might be another attractive approach to
increasing the efficiency of the power plant.
Currently, there appears to be a significant amount of excess heat being produced by the
current diesel genset configuration. The addition of a significant wind farm would lower
the amount of thermal engine available from the diesel engines due to their lower average
output. However, the combination of thermal output from the engines and the wind farm
would more than handle the current load and possibly even a doubling of the current
thermal load.
As the following graph illustrates, we believe the current diesel generated thermal energy
is more than double for the actual thermal energy being delivered to customers.
AEA – 7040015 TDX Power, Inc.; Coleman
Page 14 of 23
Current use is estimated at 12,000 MWH/YR, but clearly the current configuration, with or
without the wind turbines could provide 24,000 MWH/yr of heat.
If the thermal load were increased by double, the combined output of the recovered diesel
engine heat and wind energy would be satisfied and provide an economic benefit. There
will still be periods when there is too much thermal energy for the loads (no thermal
storage was envisioned in this exercise) in this graph, this is called EXCESS. There will also
be times when the system cannot meet 100% of the thermal demands, on cold nights when
utility loads are low and there is no wind. In this model, the Boiler is deployed to provide
this heat. However, the Bethel thermal contracts are not for firm delivery of heat, but only
for intermittent delivery, so there is no requirement for a boiler. If the Bethel utility were
to offer a firm supply of thermal energy, which may have a higher value to the customers,
than the addition of a small boiler would be advisable when the thermal load increase to
roughly double the current load.
0
5000
10000
15000
20000
25000
30000
35000
40000
45000
MWH/YR FOR HEATWIND FARM SIZE ( 0 TO 5 EWT TURBINES)
THERMAL ENERGY SOURCES
EXCESS
Boiler
WIND
Diesel Heat
AEA – 7040015 TDX Power, Inc.; Coleman
Page 15 of 23
Distribution Limits
Transformer upgrades would increase the transformer rating so that more power could be
brought back into the power plant from wind farms located on the distribution lines.
Currently it appears that the transformer would limit the max wind farm sizes at 3 MW for
the city site and 2 MW for the KYUK site. The distribution line limitations are
approximately twice this much. Increasing the capacity of the transformers could also be
considered. If both sites were developed then roughly 5 MW of wind turbine could be
installed.
If the transformers were upgraded or replaced, then the maximum capacity would be
approximately 3 to 3.5 MW on the KYUK feeder and 4 MW on the City feeder for a total of
7.5 MW.
Optional Configuration:
Incorporating dispatchable electrical heating equipment on the electrical distribution lines
may be a viable approach and may be economically competitive to the central thermal loop.
This approach would also eliminate the need to upgrade the distribution transformers as
the wind energy would be absorbed along the lines prior to entering the power plant. The
advantages of dispatchable electrical heating would be
a) No expansion of the current thermal loop customer base or thermal loop
infrastructure needed.
b) New customers not exclusively in the downtown area could take advantage of
wind energy for heating
c) The installation cost is very reasonable
d) Thermal storage can be part of the electrical heating package, Steffens heaters as
an example.
e) Increase utilization of the distribution grid with little or no expense.
For the Medium sized wind farm case, the size of the future wind plant is dependant on the
size of the thermal load and the thermal distribution system to be deployed .
The upper limit is defined by the current transformer ratings on both of the likely
distribution circuits. Deploying controllable power set point on the wind turbines to curtail
energy production could also help manage distribution system overloads during situations
of high wind and low load without needing to fully shut down the wind turbines.
Conclusion:
The Medium size wind farm is not a technical challenge and represents a project that would
have significant impact to the Bethel. It will require design changes to the thermal loop and
integration of a dispatchable secondary loads. This presumes that adequate additional
AEA – 7040015 TDX Power, Inc.; Coleman
Page 16 of 23
thermal customers are available and willing to pay a reasonable rate for thermal energy.
Depending on site(s) availability this option could be deployed with wind farm sizes from 3
to 4 MW. With transformer upgrades and utilizing both sites a total of about 6 to 7 MW
could be deployed. Clearly identification of thermal load customers and siting of the
turbines become the gating issues.
AEA – 7040015 TDX Power, Inc.; Coleman
Page 17 of 23
High
Define Limitations in the current system
Propose Work-Around for each Limitation
Define Size Limits of Wind Farm post change
Limitation Identified
Distribution System
Power Plant
o Diesel Gensets
o Controls
Thermal Sales
Not Considered Viable at this time
Diesel Off Mode Operations
Large Scale Electrical Storage
Limitation and Potential Solutions.
1) Distribution, Transformers, Protection, Bus Connection
If the two main feeders in the distribution system were brought up to their current line
capacity by changing or upgrading the thermal capacity of the transformers, and
modification of the protection equipment and bus connection, then it appears that the
maximum wind farm size as limited by only the conductors, at the current voltage, would
be
- 3.5 MW at KYUK site
- 4.5 MW at City site
If utilizing the maximum capacity of the feeder, active VAR support to maintain voltage
within an appropriate window will most likely be required. This capability might be an
internal function of the wind turbine, or a dedicated external device on the feeder itself.
For larger capacity wind farms, a trade-off analysis between rewiring, changing the
distribution voltage with the same wire, or building a new distribution line will need to be
considered. For a very large wind farm (10 to 12 MW) a new dedicated feeder run from the
wind farm to the powerhouse is probably advisable.
2) Automatically Dispatch Diesel Genset
As the wind farm increases in size, and with the caveat that the system will not be run in
Diesel-Off mode, there is a fundamental limitation in the speed of diesel engine dispatch,
that won’t be overcome until an automated genset is deployed.
The option of deploying a diesel genset with automatic dispatch, synchronization and
control capability would provide for faster dispatch control that would shrink the dispatch
control time frame to approximately 20 to 30 minutes. This has potential to save significant
AEA – 7040015 TDX Power, Inc.; Coleman
Page 18 of 23
fuel. This system would provide greater flexibility for the operators and eliminate the
need for fast dispatch scenarios for the EMD engines current in the power plant.
The EMD’s do not appear to be suitable for automatic dispatch due to their inherent start-
up requirements, which demand operator intervention. Once started the current
synchronization system is fully manual. An automatic synchronization system was tried in
the past to no avail. It was abandoned and the tried and true manual system was
reinstated. It is not clear that these units could not ever have automatic synchronization,
but from the operator’s point of view, it didn’t work successfully.
A single automatically dispatchable genset (or two units so there would be one backup
unit) in the 1 to 2 MW size range is probably appropriate.
A new automatic dispatchable genset coupled with dispatchable load control would be very
appropriate system architecture for the HIGH case. The difference between an
automatically dispatchable diesel vs. the current manual dispatch approach would be the
additional fuel saving and lowering of diesel run times. The reduction in run time would be
most prevalent when previously the diesels would be running at or near there minimum
load level, where their fuel efficiency is lowest.
3) Thermal Sales
Thermal sales by the utility represent a significant opportunity in Bethel. Historically the
utility has provided building heat at a competitive price from diesel engine waste heat.
Using wind energy, when the wind contribution exceeds the allowable instantaneous
penetration levels, for heat is a stable configuration. The value of the wind energy as heat
isn’t as high as it would be if utilized for electrical energy but it still has a significant value.
Wind power used to offset diesel fuel in electrical generation is equal to roughly the price
of fuel and the diesel engine conversion rate. At 14 kwh/gallon and $5/gallon the wind
energy on just a fuel-offset value is worth 36 cents. (utility bulk delivered price)
With resistive heating the value of wind generated electricity is about 16 cents/kwh if oil
were selling for $6/gallon. (Customer delivered retail price)
If heat pumps were deployed with an average Coefficient of Performance in the 2.5 range,
than the wind energy would be worth close to 40 cents a kWh using diesel fuel $6/gallon.
Clearly, the value of wind energy as a heating option appears to be a viable, economically
attractive reason to expand the wind farm size beyond that which might be justified by the
reduction in electrical generation cost alone.
From the thermal consumer point of view, if one can heat their home or business for less by
using wind energy than why wouldn’t they.
AEA – 7040015 TDX Power, Inc.; Coleman
Page 19 of 23
Value of Heat
To End Customer
Fuel Oil Electricity
Resistive Heat Pump
BTU/gallon 138,700 COP=3
BTU/kWh 3416 10248
Heating Efficiency 85% 100% 100%
Cost $6 $0.10 $0.10
Gallon Kwh Kwh
Value
$/Therm $5.09 $2.93 $0.98
Therm=100,000 BTUs
The graphic, “Value of Heat To End Customer” normalized the value of thermal energy
measured in $/therm for three difference sources of heat. For market based economics, it’s
clear that electricity to heat would be competitive to $6 fuel oil if the electricity were
valued at 16 cents using resistive heating. Any value less would be economically attractive
to the customer.
The use of a heat pump, which might be practical in a number of discreet loads in Bethel,
has enhanced economics due to the significant efficiency gain from that technology.
Development of viable heat pump options for rural Alaskan applications needs to be
undertaken to consider this option viable. It clearly has promise, but at this time, the
technology is not ready for commercial deployment. TDX would recommend a
development and demonstration program be undertaken by AEA to validate the use of heat
pumps, especially in medium to high penetration wind farm applications.
The capacity of the current thermal loop is not clearly defined, nor is the current thermal
load and possible expansion of thermal load with new customers. AVEC is currently
evaluating the current thermal load profile, customer base and thermal system
configuration. The unknowns of the Bethel thermal energy market will need additional
study to evaluate the business options, there does appear to be an attractive opportunity to
exploit wind energy for serving thermal loads.
4) Alternative Uses
As the wind farm size increases there will significantly more wind energy available for
alternative loads such as electrical transportation. When dispatchable wind energy is
available a majority of the time, than incorporating electric vehicles transportation options
could be viable in Bethel.
AEA – 7040015 TDX Power, Inc.; Coleman
Page 20 of 23
Conceptualization of a 10 MW Wind Farm
Currently there is one identified site west of the airport with sufficient land for a 10 MW
wind farm. Given the location and distance from the powerhouse, a separate MV
distribution feeder is a reasonable assumption. To take advantage of the wind power, a
major expansion of the thermal energy delivery system will be required. It may involve
both thermal loop expansion and distributed dispatchable electric heating appliances. For
the purposes of this study, we will assume sufficient viable loads exist.
The powerhouse can remain in it’s current configuration, using only dispatchable load
controls backed up with wind turbine set point control to provide sufficiently robust and
fast load matching for the wind energy. In this configuration, one 2.2 MW EMD will always
be running to provide voltage and frequency reference, therefore not requiring a newer
automatically dispatchable diesel engine nor electrical storage, however there is legitimate
concern that the governor response on the EMD’s will remain stable. This issue clearly
needs more research and investigation before contemplating this large a wind farm.
A 10 MW wind farm will supply roughly 25% of the electricity needs of Bethel and can
provide significant thermal energy as well.
Impact of 10 MW wind farm on the Bethel utility compared to current configuration
- Diesel fuel reduction ------- 738,000 gallons
- Diesel run time reduction ------ 1,500 hours
- Wind Energy for thermal loads -----11,470 MWH/yr
From the thermal energy side, the wind energy that is not used directly for electricity is
shunted off for thermal needs. This analysis assumes only resistive heating technologies
are deployed, whether within the thermal loop or via distributed resistance heaters.
Clearly developing appropriate heat pumps for Alaskan conditions would enhance the wind
energy available for thermal heating.
The wind generated thermal energy, while only available when there is excess wind, is
available about 40% of the year at a pretty significant rate and available 60% of the time at
some level. Essentially if the wind is blowing there is sufficient capacity in the wind farm to
fulfill thermal needs. In the 10 MW wind farm case the utilization of the wind energy is
roughly 50/50 between electrical and thermal.
Preliminary Financial Evaluation
AEA – 7040015 TDX Power, Inc.; Coleman
Page 21 of 23
Preliminary financial review of the three level of wind farm integration into Bethel has
begun on a macro level to provide a gross over view of the economics governing the
potential projects. It will be important to revisit this analysis as more accurate and
comprehensive cost analyze are available. However, for the purposes of helping determine
the most appropriate first project for Bethel, this analysis provides high level guidance.
The assumption used in this analysis needs to be fully reviewed, but the tool seems
adequate for a first look for comparison sake.
Major Economic Variables:
Prices scale down as project size increases. Especially true for Alaskan sites, there is a
significant cost reduction on the price per MW for larger project because many of the fixed
project costs (soft project costs, project management, construction equipment deployment,
logistics and transportation) are spread over a larger project. Therefore larger projects
will most certainly have cost advantages over smaller projects.
The electrical infrastructure modification for each level of wind farm has been estimated
and needs a fully review.
The value of delivered diesel fuel is well understood and can be fined tuned. Longer-term
projections are valuable here since the true economics of a wind projects are only fully
understood when viewed from a 20-year lifetime point of view. The simple Pay Back
Period analysis provides a very simple comparative value, but under values the long term
benefits of the project.
Resistance heating vs. Heat pumps: the significant performance advantage that heat pumps
have over simple resistive heating makes this option important for these types of systems.
This is especially true when the energy for thermal loads becomes a common occurrence
and utilization factors are 50% or more are possible.
AEA – 7040015 TDX Power, Inc.; Coleman
Page 22 of 23
Bethel Wind Project Feasibility Project
Preliminary Financial Evaluation 16-Feb-15
Turbine installation Pricing
1 MW
COSTS $4,500,000
Bethel Energy
Values Diesel Fuel $5 per gallon
Electricity for
Heat $150 kWH
LOW MEDIUM HIGH
Wind Farm Size MW 3 6 10 10
Thermal Process None Resistive Resistive Heat Pumps
COP 2.5
Relative Project
Costs 100% 95% 90% 90%
Turbines Installed
Cost $13,500,000
$27,000,000
$45,000,000
$45,000,000
Utility Modifications $- $500,000 $2,500,000 $3,500,000
Total Project Costs $13,500,000 $27,500,000 $47,500,000 $48,500,000
Value Stream
Diesel Saving:
Gallons 383,200 581,200 738,000 738,000
Fuel $$ Saved $1,916,000 $2,906,000 $3,690,000 $3,690,000
Wind MWH to Heat - 4,500 11,400 28,500
Heat $$ Generated $- $675,000 $1,710,000 $4,275,000
Total Revenue $1,916,000 $3,581,000 $5,400,000 $7,965,000
Simple Pay Back
Period 7.0 7.7 8.8 6.1
This simplistic financial model is only valuable for rough guidance on a macro scale.
However, it illustrates one major point: Large wind farms that provide a significant
amount of thermal energy could be as financially viable as smaller wind farms. This fact is
driven by the lower installed cost per MW for larger project, which is especially true in
rural Alaska, the high current cost for thermal energy and the use of heat pump technology.
The most significant financial drivers for the Bethel project will be the project costs and the
value of the diesel fuel saved and the thermal energy from the wind turbines. This chart
uses our current best guess for these values. For larger sized wind farms the dominant
AEA – 7040015 TDX Power, Inc.; Coleman
Page 23 of 23
economic driver will be the value of thermal energy and the cost of delivering that product.
For this exercise we used $150/MWH for thermal energy, which equates to $4.50/Therm.
Recently the utility sold thermal energy for $3/therm or about 60% of the current price of
heating oil. A thermal energy marketing study for Bethel might provide more realistic
long-term thermal energy pricing estimates.
Conclusions:
The economic viability of the 10 MW wind farm appear attractive and certainly justifies
further exploration. However, the sheer size of the project with nearly a $50 million price
tag may be currently out of reach. Other hurdles include the necessity of a land deal
adequate to site a 10 MW wind farm, developing a thermal market for the wind power and
design and permitting for a new distribution feeder.