Multi-objective transmission network expansion planning based on Reinforcement Learning

Abstract

With the rapid expansion of power system scale and the increase of complexity, the application effect of conventional power grid planning method is limited due to the use of artificial judgment and engineering experience. In this paper, the power grid planning algorithm based on reinforcement learning is used to plan the power grid, which has the characteristics of continuous interactive and autonomous learning with the environment, so as to realize large-scale and complex transmission network expansion planning. Firstly, the characteristics of reinforcement learning and Markov decision-making process are summarized; secondly, the digital model of power grid planning is established, and the multi-layer and multi-objective grid evaluation system is constructed through the lack of power, the number of electrical mediums and the cost of grid construction; finally, the accuracy and effectiveness of this method are verified by the planning calculation of IEEE garver-6 section.