Study On Driving Force Allocation Strategy Of Four-wheel Independent Drive Electric Vehicle

In order to reduce the dependence on petroleum resource and improve the environmental quality of human existence,countries all over the world are advocating the development of new energy vehicles strongly.As a new layout of new energy vehicle,the four-wheel independent drive electric vehicle simplifies the vehicle transmission system,improves the transmission efficiency and the controllability of the vehicle.It is a key technical problem that four-wheel independent drive electric vehicles need to solve urgently by controlling the output torque of each in-wheel motor accurately to improve the handling stability of the vehicle during the driving process.Aiming at this key technical issue,this paper completes the driving force distribution strategy design and experimental verification through the following specific work.Firstly,the multi-degree-of-freedom nonlinear system of four-wheel independent drive electric vehicle is simplified and a seven-degree-of-freedom vehicle dynamics model is built in Simulink.The structure and performance parameters are matched with the mature vehicle model in Carsim.The accuracy and effectiveness of the vehicle simulation platform are verified by simulation under the same working conditions.Secondly,the driving force distribution strategy is designed with the hierarchical structure.By leveraging the learning ability of the neural network in upper layer,the parameters of the fuzzy rules and membership function in the fuzzy controller are optimized,and the additional yaw moment and longitudinal force required for the stable driving of the vehicle are calculated.In the lower layer,the distribution method and the optimal distribution method targeted for the minimum tire load rate are adopted in specific driving conditions to distribute the driving torque.Then,according to the GB/T6362-2014 automobile handling stability test method,the designed driving force distribution strategy is simulated and vertified through Carsim/Simulink co-simulation in the steering wheel step condition,double-line shifting condition on the high-adhesion road and snake-shape conditions on the low-adhesion road.Finally,a hardware-in-the-loop test platform is built through the real-time simulation platform dSPACE and the controller DSP28335.The controller collects the driver’s manipulation information and the vehicle status information that outputs from the dSPACE vehicle real-time model,and performs hardware-in-the-loop tests under steering wheel step conditions.