Deep Reinforcement Learning Based on Binary Search for Distribution Network and Microgrid Cooperative Optimal Scheduling

The integration of massive small and micro resources brings high transaction costs and significant regulatory challenges to the distribution network. To address this issue, a distribution network and microgrid cooperative optimal scheduling model is established. By combining binary search with the proximal policy optimization algorithm, a deep reinforcement learning algorithm that can autonomously adjust resource behavior to meet grid demands is formed. The effectiveness of this algorithm has been validated through case studies.