Overcoming Transmission Constraints: Energy …

[Pages:5]Overcoming Transmission Constraints: Energy Storage and Wyoming Wind Power1

Mindi Farber-DeAnda (SAIC, McLean, Virginia, USA); farbermj@ Victor Gorokhov (SAIC); Mark Kuntz (VRB); Bradley Williams (PacifiCorp)

Energy storage can help wind-farm developers facing transmission and other constraints. This is a key finding of an investigation by SAIC, VRB, PacifiCorp, and the Wyoming Business Council, under a Special Energy Project/State Energy Program Grant, funded through the Energy Storage Systems Program of the U.S. Department of Energy. This investigation is unique in that it examined actual operating data from functioning facilities, including:

? A year of hourly operating data from the Foote Creek Rim I windfarm, near Laramie, Wyoming - A substation owned and operated by PacifiCorp - 69 wind turbines and six meteorological stations owned by PacifiCorp and Eugene Water and Electric Board, and operated by SeaWest

? Selected months of operating data (ten-minute interval) of the VRB vanadium redox flow battery installed at PacifiCorp's Castle Valley substation near Moab, Utah

? A year of hourly transmission flow data from PacifiCorp for the TOT 4A and 4B paths that cut diagonally across the state of Wyoming

? A negotiated tariff for a windfarm developer

Turbine Output (kW) Wind Speed (m/s)

Figure 1. Wind Speed and Turbine Output in Winter (December 14-20, 2003)

650

30

550

25

450 20

350 15

250

10 150

50

5

-50 12/14

12/15

12/16

12/17

12/18

12/19

Turb 1 Turb 32

Turb 2 Turb 44

TurbDa6te Turb 55

Turb 16 Turb 66

Source: Calculated from SeaWest data on wind speeds at turbines and meteorological stations.

12/20

0 12/21

Turb 22 Wind (Right Axis)

The sheer volume of data examined in this project required the development of analytical models. Foote Creek Rim runs almost three miles atop a ridge with elevation varying from 7,950 feet down to 7,750 feet. SAIC built and tested Excel models to validate wind speed measurements across the 69 wind turbines and six meteorological stations and wind generation output from actual wind speeds and turbine operating algorithms. Significant seasonal variations were found, but wind speed measurements across the 69 turbines and six meteorological stations were found to be sufficiently similar to permit use of wind generation at the substation as the input into another model (see Figure 1).

1Funded by the Energy Storage Systems Program of the U.S. Department Of Energy (DOE/ESS) as a State Energy Grant.

Winter is the season with the highest wind speeds in Wyoming, averaging 13 meters/second. The highest wind speeds also occur late at night or early in the morning. The Mitsubishi turbines used in Foote Creek Rim do not generate power at wind speeds below 5 meter/second or above 27 meters/second (if temperatures are very cold, the turbines will cut off at lower speeds). These cut off speeds are shown in red on Figure 1. There is another blue line at the bottom of the graph, which indicates that turbine #44 was out of service (probably for scheduled maintenance) during the week of December 14-20, 2003. Summer wind speeds rarely reached the 27 meters/second cut-off, hovering instead around the 5 meters/second mark (see the red line in Figure 2). On average, summer wind speeds were 6 meters/second. During this week, turbine #66 was out of service, as seen by the aqua line at the bottom of the graph.

Turbine Output (kW) Wind Speed (m/s)

Figure 2. Wind Speed and Turbine Output in Summer (July 25-31, 2003)

450

400

350

300

250

200

150 100

50

0

-50

7/25

7/26

7/27

7/28

7/29

7/30

Turb 1 Turb 32

Turb 2 Turb 44

TurbD6ate Turb 55

Turb 16 Turb 66

Source: Calculated from SeaWest data on wind speeds at turbines and meteorological stations.

14

12

10

8

6

4

2

0

7/31

8/1

Turb 22 Wind (Right Axis)

Analysis of the wind generation data revealed one-to-three day periods in which wind speed and temperature forced the turbines to be shut down, especially in the summer. Such lengthy periods of forced turbine outage exceed the discharge duration of the flow batteries, which operate best in the 6 ? 10 hour discharge range. This eliminated the potential use of energy storage to firm wind generation profiles.

Generation capacity far exceeds instate customer demand in Wyoming, one of the lowest population density states in the nation. A number of very large coal generating plants were built in the state in the 1970s. In recent years, assisted by a sales tax exemption on the purchase of renewable energy, the state has been home to most major windfarm development in the U.S. The state legislature granted a. Foote Creek Rim I, at 41.4 MW, was the first large windfarm developed in Wyoming in 1999. Along Foote Creek Rim are four other windfarms owned and operated by a number of utilities and developers. Other windfarms in southwestern and southeastern Wyoming contribute to the 285 MW of installed wind generation that ranked Wyoming fifth among the states.

Wyoming is a significant electricity exporter to Western states. Electricity is transmitted primarily from the Northeast to the Southwest along TOT 4A, as the transmission line is known. A smaller capacity transmission path in the opposite direction is known at TOT 4B. Constraints along the PacifiCorp TOT 4A and 4B transmission lines occur in the early morning hours, not because of excess demand in the Western states. Rather, PacifiCorp schedules power flows into the Northeastern portion of the state in anticipation of major power plant outages for maintenance. Transmission constraints can last for a few hours when they are scheduled (see Figure 3).

Figure 3. Wyoming TOT4 System Load Duration

100

80

Load (%)

60

40

20

0 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 Annual H ours

Source: Calculated from PacifiCorp's end of hour TOT 4A and 4B load data in Wyoming, 2003.

8 ,0 0 0

9 ,0 0 0

In 2003, there were 187 hours spread over 47 days in which load duration exceeded 90% along TOT4. While this is little over 2% of the hours in a year, the constraints did occur on 13% of the days in a year. This indicates a strained system, unable to accommodate the additional wind and coal generation projects proposed. As a result, Wyoming and Utah launched the Rocky Mountain Area Transmission Study (RMATS) in August 2003 to identify the incidence and duration of congestion, estimate the resulting congestion costs, and perform sensitivity analysis of different renewable energy cases. RMATS participants concluded the need for the Frontier Line, a multi-billion-dollar transmission line from the inter-mountain western states to the denselypopulated coast.

The megawatt size of the windfarm and duration of the transmission congestion periods necessitate a large multi-MWh energy storage device that can store and discharge energy over hours. A flow battery offered a very attractive technological solution. PacifiCorp was already testing a 250-kWh vanadium redox flow battery at a substation in Castle Valley in Moab, Utah. The VRB battery was being used in a different application, at the end of a long feeder line strained by significant seasonal demand fluctuations. Based on its experience in Australia, VRB felt that its battery would interface well with multi-MW windfarms. They scaled up the capacity of the battery, considering different conceptual plans, including changes in the orientation of the electrolyte tanks. The basic module was a 1-MW, capable to discharge up to eight hours.

There is no Independent System Operator or Regional Transmission Operator covering Wyoming and the tariffs in the state do not encourage renewable energy development. PacifiCorp is very supportive of renewable energy, however, the utility offers a conservative energy charge with no time-of-day or seasonal variation and no capacity charge for renewable generation. This investigation required a more aggressive tariff, with seasonal and time-of-day variation to encourage discharge of the wind energy stored in order to maximize revenues.

As a result, the analysis examined tariffs used in California with capacity and energy charges and time-of-day factors as a proxy to determine the optimal size and discharge duration the energy storage facility that would provide to the 41.4 MW wind farm. As can be seen in Tables 1 and 2, the tariff was rather restrictive, especially in the winter when the wind turbines generated the most energy.

The tariffs reveal a six-hour period on weekdays in the summer (June ? September) when selling wind generation to the California utility can be quite lucrative. Each MWh could be paid $255.10 plus 1.4251 times the base cost of energy. Another nine hours each weekday in the summer are considered mid peak and subject to "Sched" energy factor, which is a weighted average time-of-use factor calculated at the end of each month. But the summer is only four months long, leaving the majority of the year in the off peak period. In fact, there is only one hour a weekday in which generation qualifies as mid peak.

Table 1. Seasonal Time-of-Use Periods

Time-of- Summer (June-Sept) Winter (Oct-May)

Use Period start

finish

start finish

On-Peak

12:00 18:00

Mid Peak

8:00 12:00 20:00 21:00

18:00 23:00

Off-Peak

23:00

8:00

6:00 20:00

21:00 0:00

0:00

0:00

6:00 0:00

Super Off

0:00 6:00

Days/ week M-F M-F M-F M-F M-F Sa-Su M-Su

Table 2. Seasonal Time-of-Use Energy Factors and Capacity Charges

Time-of-

Energy Factors

Capacity Charges/MWh

Use Period Summer

Winter

Summer

Winter

On Peak

1.4251

$255.10

Mid Peak

Sched

1.2185

$0.26

$11.63

Off Peak

0.8526

Sched

$0.26

$0.73

Super Off

0.7766

$0.69

SAIC built a model to calculate battery cycling based on actual transmission congestion, wind turbine output, and tariff. The model examined wind-energy storage combinations against potential wind generation without congestion or storage in a series of sensitivity runs: number of charge/discharge cycles per day, storage capacity (MW) and discharge duration (hours), different salvage and rebate values, life expectancy and interest rates, etc. Table 3 indicates the Inputs for this model.

Table 3. Inputs to the Model Battery Capacity Charge Duration Discharge Duration Round-trip Efficiency Losses Minimum Discharge Level Tariff - Energy Charge Tariff - Capacity Charge - Up to 5 different rate classes - Up to 2 seasons (Summer/Winter) Base Cost of Energy

Battery Capital and O&M Costs Interest Rate Battery Expected Life Span Battery Salvage Value Rebate/Utility Assistance Charging/Discharging Schedule

- Up to 2 seasons (Summer/Winter) - Weekday/Weekend - Up to 2 cycles/day - Start/Finish at end of hour

The analysis compares the current wind farm operation without storage against alternate applications of a large flow battery. Sensitivity runs were performed to find the optimal battery capacity and charge/discharge duration at Foote Creek Rim I. Aligning the battery charge/discharge profile with tariff time-of-use factors increased the MWh discharged and the capacity charges. A flow battery with a 10-hour charge and 8-hour discharge aligned with on-peak and mid-peak tariff periods, maximized annual additional revenue beyond a No Battery case. Altering battery capacity from 8 MWh to 48 MWh while holding 10-hour charge and 8-hour discharge periods revealed 24 MWh as the optimal size. Energy discharged and revenue per MWh discharged are both optimized (see Figure 4).

Since presenting the project before the ESA 2005 meeting in May 2005, changes were made to the model to augment its flexibility (e.g., accommodate more than one charge/discharge cycle per day, introduce alternate tariffs with up to five different time-of-day periods and two seasons, and incorporate salvage value and rebates). The team expected multiple charge/discharge cycles to improve the economic analysis, however the restrictive tariff made more than one discharge/day uneconomic. It was not the intention of the team to create tariffs, rather to work off existing data, and as a result, further analysis on multiple discharge cycles was not performed.

MWh $/MWh

Figure 4. Finding the Optimal Battery Capacity

7,500 6,500 5,500

$68

Discharged

MW h

$67

$66

4,500 3,500

$65

Revenue/

$64

MWh Discharged

2,500

$63

1,500

$62

500

$61

8 MWh 16 MWh 24 MWh 32 MWh 40 MWh 48 MWh

Two different contributions to cost reduction were examined: salvage of spent vanadium and rebates on new installations. The vanadium used in the electrolyte can be recovered at the end of battery life. VRB has estimated this value at $76/kWh of energy storage capacity. Salvage value reduces cost by 19%, a very significant contribution. If a smaller salvage value were considered, e.g., $50/kWh, there still is a 13% reduction in cost.

No state or utility currently offers rebates for energy storage systems, however, the California Energy Commission has instituted a Pilot Performance-Based Incentive Program for renewable energy systems. The 50?/kWh incentives are for the first three years only and require CEC monitoring of system performance. While energy storage is specifically called out as ineligible, the team feels that CEC should consider sharing a small portion of the benefit afforded the renewable energy system with energy storage. If 5% of the rebate or 1?/kWh was made available to storage, costs would reduce by over 2%. If 10% of the rebate were shared, it would reduce costs by almost 12%.

Other contributions to cost reduction and benefit enhancement will be considered during the final stages of this project. The analysis is not yet complete and additional sensitivity runs will be documented in the final report.

References

California Energy Commission, Emerging Renewables Program, Decisions on Pilot Performance-Based Incentive Program, CEC-300-2005-002-CMF, January 2005.

Mitsubushi Heavy Industries, Specification of MWT ? 450/600 kW of Wind Turbine Generator for Foote Creek Project, WM-97-007 Rev.1, March 1997.

Rocky Mountain Area Transmission Study, Draft Report and January 2004 Base Case 2008 presentation.

Proprietary data from SeaWest on Foote Creek Rim 1 operations, 2003.

Proprietary data from PacifiCorp on TOT 4A and 4B transmission, 2003.

Proprietary data from VRB on Castle Valley battery operations, selected months 2004-2005. Craig Quist, "Transmission Service Provide Needs and Perspectives," NREL Wind Turbine Generator Modeling Workshop, January 15-16, 2003.

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download