An Optimization Energy Management Strategy for Hybrid Emergency Power Supply System of More Electric Aircraft Based on Machine Learning

This article proposes a machine learning (ML) based energy management strategy (EMS) optimization method for a hybrid emergency power supply system (HEPSS) of more electric aircraft (MEA). This HEPSS contains a fuel cell (FC) and a supercapacitor (SC). The proposed EMS can calculate the optimal current distribution of the FC and SC at the next processing time under the current operation of the power supply system, so as to prolong the cycle life of the system and increase the average efficiency of the system. The membership function in the fuzzy logic controller of EMS is optimized by ML to obtain the best control effect. Similarly, the optimization module with large amount of calculation is improved through ML to improve the EMS control speed. In this paper, the performance of the EMS before and after optimization is compared, the simulation results show that the proposed method has better dynamic performance.