Proposals to eliminate the impact of curtailments on wind generators’ capacity calculations create many losers as well as winners, according to data presented to the Planning Committee Thursday.

Two proposals are being considered under a problem statement approved in April to protect intermittent generators from being assigned artificially depressed capacity values as a result of curtailments directed by PJM.

Under current policy, when wind generators are curtailed by PJM for any portion of a peak summer hour, the hour is excluded from the generator’s capacity credit calculation.

PJM staff conducted an analysis of the two alternative calculations using data for summer 2012, when 33 of 50 wind generators had at least one curtailment.

Alternative 1

Alternative 1 removed from the calculations only the five-minute periods in which a curtailment occurred rather than the full hour, as in current practice.

It increased capacity factors for 21 of the 33 units that experienced curtailments (64%).  Changes ranged from an increase of 2.6 percentage points to reductions of almost 2 points with a median increase of 0.2 points.

The biggest increases in capacity factors went to units curtailed most frequently — a 2 percentage point boost for most units with more than 150 curtailed five-minute periods.

Alternative 2

Alternative 2 reduced capacity factors for 21 of 33 curtailed generators. It estimates what the generator’s output would have been during curtailment by interpolating data between the five-minute periods before and after the interruption.

Alternative 2 showed a more normal distribution of impacts, with little correlation to the number of curtailment periods. It reduced capacity factors by a median of 0.2 percentage points, with some units losing almost 2 points while others increased by 2 points.

No Robust Solution

Steve Herling, PJM vice president of planning, said PJM limited the analysis to 2012 because data from prior years was not as reliable or complete.

“With only one year of data it’s going to be very difficult to come up with a solution that’s really robust,” Herling said. “That doesn’t mean we shouldn’t try to do something.”

The committee will be asked at its next meeting to decide whether to choose one of the alternatives or to leave the methodology unchanged.

PJM’s current procedure uses hourly integrated metered data. The two alternatives would use five-minute data from PJM’s state estimator. Because the hourly integrated data is more accurate, PJM plans to continue to use that data for the units with no curtailments, said PJM’s Tom Falin.

See “MRC Action: Calculating Capacity Values for Intermittent Resources.”