Research On State Estimation And Stability Control Of Electric Vehicle Driven By In-Wheel Motor

Electric vehicle driven by in-wheel motor can embed in-wheel motor into each wheel,so it has the advantages of independent torque control,flexible control strategy,energy saving and high efficiency and large space inside the vehicle.According to the characteristics of driving system,this paper systematically studies on the state estimation algorithm and driving stability control of this type of vehicle.The content of the research mainly includes the following points:(1)Building a simulation model of the vehicle.Co-simulation is performed using Carsim and Matlab/Simulink to verify the algorithm.Due to the lack of pure electric vehicle model in Carsim,it is necessary to improve the original fuel vehicle model.The parameters of the vehicle,suspension system and tires are set in turn.The power system of the original fuel vehicle is removed,and the four-wheel power input is changed to external input.At the same time the in-wheel motor model is built in Simulink.Four-wheel differential coordination algorithm and vehicle speed tracking controller based on PID algorithm are designed.Finally,the input and output variables of the vehicle model are set in Carsim,so Carsim and Matlab/Simulink can connect to each other to complete the construction of the model.The improved electric vehicle model is compared with Carsim’s own model to verify the rationality of the model.(2)Study the state estimation problem of vehicles.Accurate acquisition of vehicle state variables is the basis of vehicle stability control.The lateral speed and centroid side angle of the vehicle are difficult to obtain directly,especially the centroid side angle.If it is obtained directly by using sonsers,the cost is too high.Based on the three-degree-of-freedom vehicle model,an Extended Kalman Filter(EKF)is designed to estimate these state variables and satisfactory estimation results are obtained.However,the extended Kalman filter assumes that process noise and observed noise are known,so the Q and R matrices are set to fixed values when used.Under different working conditions,if the same set of parameters is used,the estimation effect may be poor.For this problem,a Fuzzy Adaptive Extended Kalman Filter(FAEKF)is designed.Fuzzy logic is used to adjust the Q and R matrices online and the algorithm is verified by multi-case simulation.(3)Study vehicle stability control issues.According to the driving characteristics of electric vehicle driven by in-wheel motors,combined with the state estimation results,the vehicle yaw moment controller is designed which has a layered structure.In the upper controller,the joint control method is adopted to control yaw angular velocity and centroid side angle.Two controllers based on fuzzy PID algorithm are designed to control them respectively.A fuzzy adjuster for adjusting the output torque scale factors of the above two controllers is designed.The two output torque values are summed using the adjusted scale factor to achieve decoupling control of the two.The lower controller uses a torque distribution algorithm based on minimum objective function optimization to reasonably distribute the additional yaw moment and drive torque.At last,genetic algorithm is used to optimize the yaw moment controller.The content of the optimization mainly includes the initial PID parameters and the parameters of the fuzzy controller’s membership function.The yaw moment controllers before and after optimization are simulated under various working conditions.The simulation results demonstrate the effectiveness and superiority of the optimized algorithm.