Optimal Control Algorithm for Subway Train Operation by Proximal Policy Optimization

With the increasing scale of urban subway, total energy consumption subway has increased dramatically and poses a great challenge to comfort passengers punctuality train operation. In order ensure on-time operation passenger comfort, at same time reduce operation, this paper proposes Proximal Policy Optimization (PPO)-based optimization algorithm for optimal control Firstly, reinforcement learning architecture is constructed with position speed as state, objectives, constraint. The proposed model trained by PPO algorithm, reward scaling added training process accelerate improve efficiency algorithm. experimental results show that can effectively while ensuring