Research On Energy Management For Plug-in Hybrid Electric Vehicle Based On Prediction Of Driving Conditions

Equipped with the internal combustion engine and the electric motor, the mobility of the plug-in hybrid electric vehicle(PHEV) can be achieved by fuel and electricity. The PHEV combines the merits of conventional HEV and electric vehicle(EV), with higher battery capacity and the ability to be charged from an external outlet. Unlike the conventional HEV, the PHEV can cover a longer all electric range. So the PHEVs are expected to be the best solution before the technical breakthrough in EV. Energy management strategy of PHEV or HEV is an important factor for improving fuel economy. The goal is to split the power demand between the two power sources to achieve better vehicle performance.This thesis relied on National Natural Science Fund Project "Research on Components Sizing and System Control of The Dual Clutch Full Hybrid Electric Vehicle with Single integrated starter generator motor(ISG) Motor"(51305468). Energy management for plug-in hybrid electric vehicle was formulated based on prediction of driving conditions. The main works are as follows:① The configuration and the working principle of PHEV were analyzed. Components parameters of the power train were designed in this section. The parameters of the ISG motor were designed based on the required power analysis for different driving cycles to meet the requirements of most conditions. This method downsized the electric dive system and reduced the cost.② Rule based energy management was used to control the PHEV. Based on the systematic efficiency analysis of the power train, the driving mode shift schedule and the gear shift schedule were designed for Charge Depleting mode(CD) and Charge Sustaining mode(CS) respectively to ensure higher efficiency. The mode shift schedule was taken as the basis of global optimization of the next step.③ Driving cycles and velocity profiles have great impact on PHEV performance in terms of overall energy consumption and fuel economy. Seven representative driving patterns were selected, e.g., urban, suburban and expressway to calculate optimized control parameters for each representative driving cycle by off-line calculation. The genetic algorithms(GAs) were used to determine the optimal modes switch parameters for selected representative driving patterns with the goal of achieving the lowest cost of energy consumption. This offline optimization allows us to derive an adaptive control algorithm for real-time application. The multi-mode driving control was realized by switching the control parameters in each representative driving cycle. During vehicle driving, certain duration of the past cycle was chosen to calculate the feature vector to recognize which driving pattern it longs to. Then in the next step, the vehicle operated with the corresponding optimal control parameters adaptively.④ Road slope affects the required power of the vehicle. Predictive energy management strategy for PHEV was proposed based on the road slope prediction provided by the vehicle navigation system. In case of battery electricity shortage and over discharge during uphill, the knowledge of route data(trip distance, road slope,altitude) were used to predict the electricity consumption. Then State of Charge(SOC) trajectory was planned for CD mode and CS mode. With exact calculation of the charge timing, SOC was charged to the target value before uphill and reached to the threshold after the uphill. The proposed adaptive SOC control strategy improved vehicle performance and fully absorbed the regenerative braking energy.⑤ Backwards simulation model for plug in hybrid electric system was built based on the MATLAB/Simulink platform. The proposed energy management strategy was verified under selected driving cycle including driving pattern recognition and the prediction of energy consumption during uphill.