Torque Distribution Strategy Of ISG Hybrid Electric Vehicles Based On Model Predictive Control

With the booming automobile industry,the contradiction between serious environmental pollution,energy shortages and huge market demand problems is extremely sharp,as the most promising development of hybrid electric vehicles have become the focus of new energy vehicles to research.The power source of hybrid electric vehicle includes motor and engine,Reasonably use and distribution the torque between two power sources is an important way to improve the fuel economy,and has become a research hotspot in the field of energy management of hybrid electric vehicle.Firstly,for a single-shaft parallel integrated starter generator hybrid electric vehicle(ISG-HEV),with a single-motor dual-clutch structure of the drive system,the control models of the main components such as motor,engine and battery are established by the method of combining numerical modeling and theoretical modeling,the energy flow and torque distribution of the transmission system under different working modes are analyzed,which is the basis of designing the torque distribution strategy of ISG hybrid system based on model prediction control.Secondly,the model prediction control strategy based on exponential function is designed.the vehicle velocity is predicted by exponential function,the vehicle demand torque prediction model is established,and the objective function is established based on the output state of the forecasting model,then using dynamic programming algorithm to solve the objective function of minimum fuel consumption,the torque between the engine and the motor is reasonably distributed to improve the fuel economy of the vehicle.Thirdly,stochastic model predictive control torque distribution strategy is designed.Markov prediction principle is applied to the torque distribution strategy of single-shaft parallel ISG-HEV,combined with the driver's behavior for stochastic learning control.The stochastic model predictive control combined with the dynamic optimization algorithm is more reasonable to distribution the torque between the power source engine and the motor,the vehicle fuel economy is further improved.Finally,a simulation model of single-shaft parallel ISG-HEV is established in MATLAB and ADVISOR software,then the simulation and comparison analysis of control strategy embedded vehicle model is made in this paper.The simulation results show that comparing the static logic threshold control strategy,engine efficiency of the model predictive control strategy based on exponential function is increased 6.9%,motor efficiency is optimized to improve 7.3%,and the fuel consumption is reduced by 5.7%.On the basis of exponential function model predictive control strategy,stochastic model predictive control torque distribution strategy improves the vehicle's power and fuel economy greatly,engine and motor efficiency is respectively improved by 2.4% and 4.3%,the fuel consumption is reduced by 4.1% in 100 km,and the final value of the battery SOC is almost the same as the initial value.