A real-time energy management method for APTMS based on FCM-BP neural network

Abstract

Abstract In order to improve the real-time performance of APTMS energy management and ensure fuel economy, a real-time energy management method based on a fuzzy C-means clustering algorithm combined with a BP neural network is proposed. Firstly, the coupling simulation model of APTMS and engine energy efficiency is established. The multi-source energy configuration parameters are obtained by using a particle swarm optimization algorithm combined with an off-line simulation of various flight conditions, and the energy management rule sample library is formed. The fuzzy C-means clustering algorithm is used to classify the energy management rule sample library, and some samples are extracted based on the ranking of membership degrees for the training of the neural network. Finally, the trained neural network predictor is used to predict multi-source energy parameters according to real-time flight conditions. The simulation results show that the FCM-BP neural network method can approach the effect of off-line optimization of PSO and achieve good multi-source energy parameter prediction. The maximum prediction error is 1.91 %, and the FCM-BP neural network method shortens the simulation time by 97.31 %, which has the ability of an online application.