Intelligent energy management for a parallel hybrid electric vehicle in customer use

This thesis addresses an investigation on energy management measures to reduce the fuel consumption for a hybrid electric vehicle. The first step is to develop simulation models to evaluate main performance characteristics. The experimental tests are accelerated using a validated co-simulation tool based on high resolution, complex thermal and kinematics models. Instead of considering only one standard cycle (e. g. New European Driving Cycle (NEDC)) in calculating optimal energy management, several customer representative cycles with the help of the 3D (Driver, Driving environs, Driven vehicle) method are considered in order to cover all driving conditions. One of the main tasks is to determine how such conditions affect fuel consumption and how this information can be used to optimize the control strategy. Furthermore, some vehicle energy management measures, such as engine compartment insulation and grille shutter are investigated. In addition, due to increased attention in renewable energy sources, solar cells have been integrated into the vehicle. An intelligent approach is applied to a control strategy using a fuzzy logic controller and a genetic algorithm to optimize the energy distribution in a HEV. A multi-objective genetic algorithm is introduced to optimize the control parameters. An additional focus is on the combination of the energy management strategies. After having developed the optimal control strategy and incorporated it with other energy management measures in the vehicle co-simulation, the simulation results reveal a reasonable potential of fuel reduction that vary between 1.5\% and 4\% for a conventional vehicle. A HEV shows a greater fuel reduction potential that lies between 2 \% and 8.6 \%.