Abstract: In order to assess the annual peak regulation demand of power system, an annual peak regulation auxiliary service demand assessment model based on multi-time scale stochastic optimization is constructed, which covers three stages: the mid-and-long term stage, short term stage and peak regulation service stage. For the mid-and-long term stage, the randomness and the balance relationship of daily electricity, maximum and minimum load are considered to optimize the mid-and-long term plan for unit maintenance, cross-day start-stop and reservoir water level. For the short-term stage, the hourly power balance of source-load and its fluctuation is considered to optimize the units output on a ten-day basis according to the mid-and-long term plan. For the peak regulation stage, the start-up conditions of the auxiliary services are simulated through analyzing the daily peak regulation capacity and water curtailment, and the deep peak regulation and start-stop peak regulation capacity of thermal power units is considered to correct and optimize the short-term plan. Based on the actual data of a provincial power grid in China, a simulation analysis is conducted, which shows that the model can effectively simulate the randomness of source-load under multi-time scale, and realize the annual peak regulation demand assessment.

Key words: peak regulation auxiliary service market, annual power generation plan, multi-time scale optimization, stochastic optimization, demand assessment