Low Carbon Economic Dispatch of Integrated Electricity-Gas Energy System Based on Proximal Policy Optimization

Jiaming Lei,Aihua Jiang,Xinfei Wu

DOI: https://doi.org/10.2139/ssrn.4051657

2022-01-01

SSRN Electronic Journal

Abstract:Integrated electricity-gas energy system (IEGES) optimization plays a key role. The current reliance on model-based intelligent algorithms for low-carbon economic dispatch problem within IEGES leads to weak ability in solving uncertain problem because of their poor adaptability to the changes of structure and devices. In response to this phenomenon, this paper uses data-driven deep reinforcement learning (DRL) to transfer the low-carbon economic dispatch problem into the discrete Markovian decision process (MDP) in the IEGES environment with the carbon capture system (CCS), then the experience is gained by simulating stochastic data to train the decision-making capacity under uncertain conditions. In order to prevent the difficulty in choosing the appropriate step length and the difference between the old and new policies is too large for learning during training, so the proximal policy optimization (PPO) algorithm is adopted to solve the problem, which has a better sample complexity and can make the policy update more accurate. Simulation results show that with the introduction of CCS and carbon emission target, the total cost of IEGES can be reduced, and the carbon emission decreased by 23.7%, compared with Deep Deterministic Policy Gradient (DDPG) algorithm, the average daily cost is also dropped by 5.9%.

English Else