Coordination Optimization Operation Strategy of Distribution Network Based on Long Short-Term Memory - Proximal Policy Optimization Algorithm

The widespread adoption of electric vehicles poses certain challenges to the safe and stable operation of the distribution network. In response to the interaction between electric vehicle clusters and the distribution network, this paper proposes an interaction method between electric vehicles and the distribution network based on deep reinforcement learning. First, a mathematical model of electric vehicles, photovoltaics, and energy storage is constructed; then, considering the interaction between photovoltaics, energy storage, and the distribution network, a Long Short-Term Memory Recurrent Neural Network (LSTM)-Proximal Policy Optimization (PPO) algorithm is used to construct an orderly charging strategy for electric vehicles aimed at meeting user charging needs, reducing charging station operating costs, and promoting the consumption of renewable energy. Finally, taking the electric vehicles and distribution network of a certain city as an example, the effectiveness and superiority of the interaction strategy between electric vehicles and the distribution network based on this algorithm are verified.