Integrating Southwest Power Pool Wind to Southeast Electricity ...

More documents

Recommendations

Info

Wind Delivery (MW) 12000 10000 8000 6000 4000 2000 0 0:00 1:00 2:00 3:00 Time in hours 4:00 5:00 6:00 Figure 3-7 Worst short term event In addition to the wind production data, the dataset for this study also contained a day-ahead forecast of the wind production that exhibited the approximate error characteristics of current state of the art forecast models. Individual wind plant forecasts have an MAE (mean absolute error) of 17% on average for all of the sites used in the study. However, the aggregation effects discussed earlier apply to the forecasts as well. The statistics for the forecast for each of the regions is shown in Table 3-3. These DA forecasts of wind output are used in the day-ahead commitment process of the SCUC/SCED simulations ensuring that the actual DA forecast errors have to be accommodated during the real-time dispatch using resources online or expensive quick start generation. Table 3-3 Wind production forecast error statistics Worst Short Term Event ENTERGY SOCO SPP TVA Footprint MAE %Nameplate 8.3% 7.6% 8.6% 8.0% 7.7% MAE MW 653 1141 1268 829 3710 Max Error 37% 35% 44% 35% 37% Reserve Requirements The increased variability and uncertainty from wind power causes an increase in operating reserve requirements. Those requirements have to be provided by some combination of flexible generation and responsive load. Together, these contribute to the operating reserve that is available to help manage the wind and load variability. This reserve is calculated dynamically, and is a function of the observed variability of the wind power and the load. A methodology was developed to estimate the increased requirements for reserves with wind variability in the Eastern Wind Integration and Transmission Study (EWITS). Short-term variability is challenging because it is difficult to fully anticipate the scheduling changes and fluctuations that must be covered with reserves. In a system with 10-minute or faster dispatch update cycles, a typical approach is to forecast a flat value for wind output for the 3-8 SPP Entergy SoCo TVA
next interval based on the past 10 to 20 minutes (persistence forecast). The wind varies on that time scale, and an understanding is needed of how it will vary during the forecast interval. Figure 3-8 and Figure 3-9 illustrate how the forecast error is calculated for both 10-minute and 1hour dispatch schedules for this study. The forecast error is the difference between the actual data and the forecast value. Figure 3-8 Forecast for 10-minute dispatch Figure 3-9 Forecast for 1-hour dispatch made at 40 minutes prior to the beginning of the operational period An estimate of the reserve requirements can be made using a statistical approach. Based on detailed load, wind and forecast data, the standard deviation or other variability metric can be used to calculate this estimate. For this study, the reserve requirements are broken down into three classes by the types of resources required to fulfill them (Types 2 and 3 are both types of contingency reserves): 3-9
Page 1: DOE: Integrating Southwest Power Po
Page 4 and 5: DISCLAIMER OF WARRANTIES AND LIMITA
Page 6 and 7: EXECUTIVE SUMMARY This DOE-funded p
Page 8 and 9: Summary of average regulation requi
Page 10 and 11: generation in the 2022 Non-RES case
Page 12 and 13: System and Cost Impacts of High Win
Page 14 and 15: Figure 6-6: Individual Coal Capacit
Page 17 and 18: 1 INTRODUCTION AND BACKGROUND Proje
Page 19 and 20: 2 REPRESENTATION OF TRANSMISSION NE
Page 21 and 22: participants the possibility of a r
Page 23: assumption is affecting specific di
Page 26 and 27: Table 3-1 Wind energy targets for e
Page 28 and 29: site selection could have congregat
Page 30 and 31: Wind Production (MW) Wind Productio
Page 34 and 35: 1. Regulation is required to cover
Page 36 and 37: The polynomial shown in Equation 3
Page 38 and 39: Regulation (MW) Figure 3-12 Summary
Page 41 and 42: 4 HIGH WIND TRANSFER SCENARIO DECRI
Page 43 and 44: 5 HIGH WIND TRANSFER CASE RESULTS T
Page 45 and 46: decreasing smaller amounts. It is i
Page 47 and 48: Coal 3.6 3.6 3.6 3.6 3.6 GasOil 0.7
Page 49 and 50: Figure 5-3 Average hourly generatio
Page 51 and 52: Figure 5-6: Average hourly generati
Page 53 and 54: SBA. More details on these types of
Page 55 and 56: Figure 5-8: Total production costs
Page 57 and 58: The flow chart shown in Figure 5-9
Page 59 and 60: Table 5-10 shows the difference in
Page 61 and 62: more likely to be turned off during
Page 63 and 64: Figure 5-13: Average Regulation up
Page 65 and 66: Figure 5-16: Average Spinning reser
Page 67 and 68: Figure 5-18: SBA Capacity Factors b
Page 69 and 70: Figure 5-20: Capacity Factors for S
Page 71 and 72: Unit Startups Changing the balancin
Page 73 and 74: Gas Price Sensitivity Case Results
Page 75 and 76: 6 CONCLUSIONS AND RECOMMENDED NEXT
Page 77 and 78: generation characteristics across r
Page 79: overstates the ability of the syste
Page 82 and 83:
Figure 6-2: Individual coal capacit
Page 84 and 85:
in the 22% to 50% in Scenario 2. Th
Page 86 and 87:
Figure 6-8: individual coal capacit
Page 88 and 89:
able to carry some of the reserve t
Page 90:
Export Control Restrictions Access
show all

Integrating Southwest Power Pool Wind to Southeast Electricity ...

Create successful ePaper yourself

Delete template?

Save as template?