南方电网技术2024，Vol.18Issue(1)：94-101,8.DOI:10.13648/j.cnki.issn1674-0629.2024.01.010
基于强化学习的电-气-热多微网系统定价策略
Pricing Strategy for Electric-Gas-Heat Multi-Microgrid System Based on Re-Inforcement Learning
摘要
Abstract
With the gradual marketization of energy trading,the retail price pricing strategy of microgrid service provider in a multi-microgrid system including electric-gas-heat will affect the operation of the system and the interests of all participants.In order to study the pricing strategy of microgrid service providers,this paper firstly describes the internal transaction process of the electric-gas-heat multi-microgrid system and establishes the system model.This pricing problem is then described as a Stackelberg game,and it shows that there is a unique equilibrium point for this game.In order to protect the privacy of each subject,this paper proposes a solu-tion method based on reinforcement learning to solve the Stackelberg game with time coupling.The case study shows that this method can accurately and effectively solve the proposed pricing strategy problem,and the microgrid service providers and all the microgrids have adopted effective strategies to ensure their own interests.At the same time,the method effectively protects the privacy of market participants and exhibits good computing performance.关键词
多微网系统/定价策略/斯塔克尔伯格博弈/强化学习Key words
multi-microgrid system/pricing strategy/Stackelberg game/reinforcement learning分类
信息技术与安全科学引用本文复制引用
李媛,迟昆,王洲,彭婧,贾春蓉,刘炳文..基于强化学习的电-气-热多微网系统定价策略[J].南方电网技术,2024,18(1):94-101,8.基金项目
国家自然科学基金资助项目(52177112). Supported by the National Natural Science Foundation of China(52177112). （52177112）
