Spot market electricity seller quotation method based on reinforcement learning

• Main Authors: HE XIAOMIN, LIU QIUHUA, FANG ZHENGCONG
• Format: Patent
• Language: Chinese; English
• Published: 12.08.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention discloses a spot market electricity seller quotation method based on reinforcement learning, and the method comprises the steps: setting a spot market transaction rule, carrying out the market transaction through employing a bilateral centralized bidding mode of electricity purchasing and selling parties, and carrying out the clearing settlement through a time-phased stepped quotation mode and employing a unified marginal clearing mode; establishing an electricity seller income model, giving a target model of different types of electricity seller quotation strategies, and performing assumption condition and initial value setting on the model; a reinforcement learning algorithm is introduced, state spaces, action spaces and reward functions of all the electricity sellers in the model are defined according to the reinforcement learning algorithm, and a spot market electricity seller quotation model based on reinforcement learning is established; and solving by adopting an epsilon-greedy algorithm