An Intelligent Power Flow Adjustment Method Integrating Deep Neural Network-based Convergence Judgment and Proximal Policy Optimization

As the structure of power systems becomes increasingly complex and renewable energy sources are widely integrated, traditional power flow adjustment methods are unable to effectively adapt to the actual operating modes of smart grids. Furthermore, these adjustment processes are inefficient, lack flexibility, and heavily rely on expert experience. To address these issues, this paper proposes an intelligent power flow adjustment method based on deep neural network power flow convergence judgment and proximal policy optimization algorithm. This method utilizes a deep neural network to determine power flow convergence and models the power flow adjustment problem as a multi-stage dynamic decision process. Then, the proximal policy optimization algorithm is employed to achieve intelligent adjustment of power flow convergence. Simulation results based on the IEEE 30-bus system demonstrate that the proposed method effectively learns adjustment policies, significantly improves power flow adjustment efficiency, and enhances the operational security of power systems.