考虑等效竞争对手的发电商日前市场竞价策略优化OA

In the context of the continuously expanding electricity market trading scale,the scientific and reasonable formulation of market bidding strategies for power generators has become an urgent issue to solve.This paper proposes a day-ahead market bidding decision method for power generators participating as price makers,which considers equivalent competitors.The method first standardizes historical bidding data of power generators using Chameleon clustering to extract typical bidding patterns for different types of generating units.To manage the uncertainty of competitors'bidding behaviors,the concept of equivalent competitors is introduced,and a two-stage long short-term memory(LSTM)model is established to estimate and predict the cumulative bidding curve of these equivalent competitors.Based on this,a bi-level optimization model is constructed to determine the optimal bidding strategy for power generators participating in the day-ahead market.Case studies validate the effectiveness of this method,demonstrating good simulation results for real electricity markets and improved market returns for power generators,thereby providing effective guidance for quantity and price bidding decisions of power generators.