Open AccessDesign optimisation for component sizing using multi‐objective particle swarm optimisation and control of PEM fuel cell‐battery hybrid energy system for locomotive application
Author(s) -
Sarma Upasana,
Ganguly Sanjib
Publication year - 2020
Publication title -
iet electrical systems in transportation
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.588
H-Index - 26
eISSN - 2042-9746
pISSN - 2042-9738
DOI - 10.1049/iet-est.2018.5053
Subject(s) - automotive engineering , engineering , particle swarm optimization , powertrain , proton exchange membrane fuel cell , sizing , battery (electricity) , state of charge , energy management , power balance , fuel efficiency , power (physics) , control engineering , computer science , energy (signal processing) , fuel cells , statistics , physics , mathematics , quantum mechanics , machine learning , chemical engineering , thermodynamics , art , torque , visual arts
This study presents a design optimisation (DO) approach for optimal component sizing of the proton exchange membrane fuel cell (PEMFC)‐battery hybrid energy system (HES) and a controller design approach for DC‐bus voltage regulation and load sharing control for the HES so that the proposed HES is capable of replacing the diesel locomotive WDM‐3D used to haul intercity passenger trains in India. The DO approach is formulated as a multi‐objective optimisation problem with objective functions of simultaneous minimisation of the total cost of HES and total fuel consumption under the operational constraints of battery state‐of‐charge limit and instantaneous power balance. Two energy management strategies (EMSs) are proposed for instantaneous power balance between the HES and locomotive power demand. The DO problem is solved for a chosen drive cycle using strength Pareto evolutionary algorithm‐2‐based particle swarm optimisation, in which an EMS is used as a subroutine. The Pareto front obtained with the multi‐objective DO problem consists of a set of solutions with different sizes of PEMFC and battery. A MATLAB–Simulink model is developed with the proposed control strategy. The simulation results show that the designed controllers satisfactorily work under the dynamically varying load similar to a locomotive.