Integrated Electricity-Gas System Optimal Dispatch Based on Deep Reinforcement Learning
Authors:
- 23 December 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1082-1086
Abstract
Optimal dispatch plays a significant role in the efficient work of Integrated Electricity-Gas System (IEGS). This paper employs the Deep Reinforcement Learning (DRL) method to solve the IEGS optimal dispatch problem to improve the efficiency of decision-making. Mathematical modeling for the power network, gas network, and coupling equipment in the IEGS is first conducted. The objective function and constraints of the IEGS optimal dispatch model considering the economy and low carbon emission are proposed. Then, the Soft Actor-Critic algorithm (SAC) is selected to solve the optimal dispatch problem. A case on a modified IEGS combining IEEE 9-node with Gas 9-node is introduced to verify the effectiveness of DRL on the optimal dispatch problem.Keywords
Funding Information
- National Natural Science Foundation of China
- Natural Science Foundation of Jiangsu Province
This publication has 3 references indexed in Scilit:
- On-Line Building Energy Optimization Using Deep Reinforcement LearningIEEE Transactions on Smart Grid, 2018
- Computational Methods for Residential Energy Cost Optimization in Smart GridsACM Computing Surveys, 2016
- Dynamic Pricing and Energy Consumption Scheduling With Reinforcement LearningIEEE Transactions on Smart Grid, 2015