Multi-objective optimization design and control of plug-in hybrid electric vehicle powertrain for minimization of energy consumption, exhaust emissions and battery degradation

Abstract

Plug-in hybrid electric vehicles (PHEV) offer an attractive alternative to achieve the ambitious goals set by strong policies focused on near-term air quality and fuel-efficient road transportation. This paper provides a comprehensive study regarding the PHEV’s optimum powertrain design, by means of a multi-criteria analysis carried out by the interactive adaptive-weight genetic algorithm approach. The optimization aims to simultaneously minimize the PHEV’s fuel consumption, exhaust emissions, electric powertrain size, battery state of health, charging time and costs. To achieve these objectives, several PHEV’s design parameters are optimized such as in-wheel electric motors’ torque curves, battery voltage and capacity. The drivetrain is also optimized according to the determination of the best configuration of gearbox and differential gear ratios, taking into account constructive constraints. Furthermore, the fuzzy logic controllers responsible for the engine/electric motors power-split management and gear shifting control are included in the multi-objective optimization in order to define the best membership functions, rules and respective weights. To guarantee robust solutions, the PHEV is optimized under different driving conditions, which is given by the combination of the FTP-75, HWFET and US06 driving cycles. To evaluate the optimum PHEV performance, it is also simulated under the WLTC driving cycle and a real-world driving cycle based on the Campinas city. The best trade-off configuration results in 39.57% decrease in vehicle travel cost along with 43.39% carbon monoxide (CO), 45.13% unburned hydrocarbons (HC) and 72.64% nitrogen oxides (NOx) emissions reduction under the combined driving cycle.