Research On Anti-slip Regulation System Of In-wheel Motor 4 Wheel Drive Electric Vehicle

4 in-wheel motor drive electric vehicles is an important development direction of new energy vehicles.When the vehicles start on the road with low adhesion coefficient and speed up,the wheels tend to slip.excessive trackslip will reduce the longitudinal adhesion force,affecting vehicle drive performance and stability.Anti-slip Regulation(ASR)which has been widely studied in the traditional vehicles,is the main safety system to solve the problems of wheel slip,but can't be applied directly to 4 in-wheel motor drive electric vehicles.It's an important research field how to design and implement ASR based on 4 in-wheel motor drive electric vehicles.In this paper,we mainly study the acquisition of wheel slip rate and the optimal slip rate recognition of the anti-skip system and design the corresponding anti-skip control strategy.Specific research work is as follows:Firstly,the GPS / INS integrated navigation device is used in the acquisition of the wheel slip rate to design the slip rate acquisition system cause there are some complicated shortcomings in the vehicle parameters and model establishment while using commonly vehicle model estimation method.The acquisition system uses the Kalman filter to fuse the velocity signal collected by the GPS and the longitudinal acceleration signal collected by the accelerometer to obtain the longitudinal vehicle speed.Furthermore,The BP neural network is used to assist the acquisition system to avoid the problem that the GPS module in GPS/INS integrated navigation device is easy to lock out in bad working condition for Kalman filter to output the longitudinal speed and make sure that the system still estimates the vertical speed in case of GPS unlocked.An experimental platform is built through the actual collected data to verify the effectiveness of the method.The results show that the slip rate based on the GPS / INS combination device has good accuracy in the case of GPS lock.When the GPS is not locked,the slip rate calculated by the BP neural network assisted method also has good accuracy.Secondly,we use the Unknown Inputs Observer(UIO)method to observe adhension coefficient and design UIO base on the establishment of the wheel motor model and wheel dynamics model.To obtain optimal slip rate of road,RLS is used to fit the magic function using the adhension coefficient collected by UIO and slip rate collected by the acquisition system.In order to verify the correctness and validity of the method,the vehicle model was established in CarSim according to the actual parameters,and the wheel motor model,UIO and RLS observer were established in Simulink.The feasibility and accuracy of the UIO observation wheel using the adhesion coefficient are verified by the simulations of Simulink and CarSim,and the recognition accuracy of the optimal sliding rate of the RLS observer is verified either.The simulation results show that the output of UIO can effectively and accurately follow adhesion coefficient and the RLS observer can well identify the optimal slip rate of the road.This method is more accurate than the current method of using the wheel speed differential,so that the accuracy of the optimal slip rate recognition is higherThirdly,complete the slip rate of the anti-skip system by PI controller.The slip rate calculated is taken as the feedback and the optimal slip rate of the RLS observation is taken as the target.The deviation of them is input to the PI controller,the output torque is adjusted by the output of PI controller in real time to control the slip rate.Finally,the article summarizes the work done in the whole paper.There are still many problems though some achievement has made and shortcomings will be improved in the following study.