Sunday, June 24, 2018

MISO Ponders Reserve Scheduling Fix

By Amanda Durish Cook

MISO is considering the possibility of factoring transmission constraints into its reserve requirement modeling to help prevent occurrences of scarcity pricing.

Staff are evaluating changing the algorithm underpinning the minimum zonal reserve requirement to reflect energy flow constraints. Under current practice, MISO calculates reserve requirements for a single operating day three days in advance using an offline study to produce results to be posted in time for the day-ahead market.

But actual operating conditions can change from original study assumptions, said MISO Senior Manager of Transmission Security Planning Mike Mattox, adding that the three-day modeling timeline has been in place for about 10 years.

Chatterjee | © RTO Insider

Dhiman Chatterjee, director of market evaluation and design, said a reserve model that does not capture constraints in the interim can unnecessarily create scarcity pricing for reserves.

MISO staff began taking notice of the issue after events occurring April 1, when an offline study predicted an 84-MW minimum contingency reserve requirement for Zone 6, which encompasses Indiana and a slice of Kentucky. In reality, more reserves were needed to compensate for generation and transmission outages in southeast Louisiana, including the shutdown of the Waterford 3 nuclear unit. As a result, Zone 6 experienced an outflow of energy to the south, triggering operating reserve scarcity conditions for 152 five-minute intervals during the day, with spinning and supplemental reserves clearing at about $200/MWh during 131 intervals and $1,100/MWh during 21 intervals.

“We’re really focusing on the event in April. We saw some relatively high prices for the reserves. … In this case, we saw it was bit counterintuitive so we dug into that,” said Chatterjee during a July 13 Market Subcommittee meeting.

MISO Principal Advisor of Market Development and Analysis Yonghong Chen said the constraints can be accounted for in the day-ahead modeling process to address reserve deliverability. Chen said that had the modeling more explicitly accounted for constraints, there would have been more inexpensive prices and a reduction in post-reserve deployment congestion.

“There is really no need to go into scarcity pricing,” Chen said.

Chatterjee said MISO is in the early stages of analyzing an additional modeling step that it could implement by the end of this year or early next year. He said staff would return to the subcommittee next month to update stakeholders on the feasibility of the change.