Using Dfferential Evolution Techniques for Management of a Hybrid-Electric Propulsion System

Manuel  A. Rendón; Carlos  D. Sanchez R.; Josselyn  Gallo M.; José  F. Garcia S.; Ning  Xiong; Patricia  Habib Hallak; Nícolas  Lima Oliveira; Yipsy  Roque Benito; Janderson  Mazzine Afonso

doi:10.48011/asba.v2i1.1727

Manuel A. Rendón Federal University of Juiz de Fora
Carlos D. Sanchez R. Escuela Superior Politécnica del Litoral
Josselyn Gallo M. Escuela Superior Politécnica del Litoral
José F. Garcia S. Escuela Superior Politécnica del Litoral
Ning Xiong Mälardalen University
Patricia Habib Hallak Universidade Federal de Juiz de Fora
Nícolas Lima Oliveira Universidade Federal de Juiz de Fora
Yipsy Roque Benito Universidade Federal de Juiz de Fora
Janderson Mazzine Afonso Universidade Federal de Juiz de Fora

DOI: https://doi.org/10.48011/asba.v2i1.1727

Keywords: Evolutionary algorithms for optimal control, Optimal control of hybrid systems, Real time optimization and control

Abstract

The air transportation industry contributes with 2% of the total greenhouse gas emissions, and there is a demand from global aviation regulators for reducing this percentage. Hybrid-electric propulsion systems (HEPS) for aircraft is an area of increasing interest for achieving these goals. It is a multidisciplinary research that involves internal combustion engines (ICE), electric motors (EM), power electronic converters, energy storage devices, propeller design, monitoring and control systems, management, etc. The Electromechanical Energy Conversion Group (GCEME) in Brazil developed a complete HEPS test bench and Laboratory. The facility will be able to test three different topologies: Series, full-electric and turbo-electric. The present work employs a detailed model for the test bench, and given a mission prole and the energy consumption as optimization function, it applies Dierential Evolution (DE) techniques in the energy management code. The results highlight the nonlinear nature of the HEPS model, and the worth of this methodology in looking for an optimal solution to reduce the computing processing time.