现代电力2025，Vol.42Issue(2)：314-321,8.DOI:10.19725/j.cnki.1007-2322.2023.0058
基于孪生延迟DDPG强化学习的电-热耦合系统低碳经济调度
Low-carbon Economic Dispatch of Electric-thermal Coupling System Based on Twin Delayed DDPG Reinforcement Learning
陈思畏 1李建军 2邹信迅 2罗旭 1崔希1
作者信息
- 1. 江西江投集团能源技术研究院,江西省 南昌市 330096
- 2. 江西赣能股份有限公司,江西省 南昌市 330096
- 折叠
摘要
Abstract
For the electric-thermal coupling system with re-newable energy access,a reinforcement learning method is pro-posed for low-carbon economic dispatch of electric-thermal coupling systems.Firstly,a low-carbon economic dispatch model of the electric-thermal coupling system is established with both the economy and carbon emissions taken into ac-count.The low-carbon economic dispatch process of the elec-tric-thermal coupling system containing renewable energy is subsequently transformed into a Markov decision process(MDP).With the aim of minimizing both the economy and car-bon emissions,a multi-objective reward function is designed by combining the penalty constraint mechanism.Additionally,based on the improved algorithm of deep deterministic policy gradient(DDPG),a twin delayed DDPG algorithm is utilized to train reinforcement learning agents interactively.Finally,the numerical result demonstrates that the agent trained by the pro-posed method can respond to the uncertainty of renewable en-ergy and electric/thermal load in real time,enabling the optim-ization of the low-carbon economic scheduling for the electric-thermal coupling system containing renewable energy online.关键词
电-热耦合系统/低碳经济调度/强化学习/孪生延迟DDPGKey words
electric-thermal coupling system/low-carbon economic dispatch/reinforcement learning/twin delayed DDPG分类
动力与电气工程引用本文复制引用
陈思畏,李建军,邹信迅,罗旭,崔希..基于孪生延迟DDPG强化学习的电-热耦合系统低碳经济调度[J].现代电力,2025,42(2):314-321,8.
