Design of photovoltaic-energy storage-load forecasting and optimal control method for distributed distribution network based on reinforcement learning

Le résumé fourni par la source

The safe and stable operation of distribution network can improve the economic benefit and operational performance of power system. In order to improve the photovoltaic-energy storage-load effect of distribution network, a method of photovoltaic-energy storage-load prediction and optimal control of distributed distribution network based on reinforcement learning is proposed. Calculate the dynamic photovoltaic-energy storage-load and static photovoltaic-energy storage-load of distribution network, determine the change of the ratio of static and dynamic photovoltaic-energy storage-load of distribution network, and comprehensively predict the photovoltaic-energy storage-load of distribution network; Then, reinforcement learning is used to optimize the photovoltaic-energy storage-load control of distributed distribution network. The experimental results show that the proposed method shows better prediction accuracy and effect after optimization, and has higher control accuracy for photovoltaic output power, energy storage power and load power of distributed distribution network.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.