Distributed Dynamic Economic Optimal Scheduling Method for Microgrid Based on Deep Learning

Deep learning is a research hotspot in the field of artificial intelligence. It has developed rapidly in recent years and can be seen participating in many fields. With its powerful automatic feature extraction capabilities, deep learning has powerful processing capabilities for data fitting and rapid data processing and analysis. In recent years, there have been countless application studies of its application in microgrids. The main purpose of this paper is to conduct optimal scheduling research on the distributed dynamic economy of microgrids based on deep learning. Aiming at the problem of dynamic economical scheduling of microgrids, this paper establishes a dynamic economical scheduling model for microgrids that considers various microgrid costs. The decoupling method of the boundary of each region is proposed, and the synchronous ADMM algorithm is used to calculate the distributed dynamic economic dispatch model of the distribution network of the microgrid, and compares the results with the centralized optimal dispatch method. The research shows that the error of the combined forecasting method proposed in this paper is obviously reduced and has advantages, which can effectively reduce the forecasting error and improve the forecasting accuracy.