Research On Fault-tolerant Control Of Actuator Failures In Distributed Electric Vehicle Drive Systems

Under the development trend of vehicle electrification,distributed electric vehicles have become a hot spot in the research of electric vehicles due to their unique driving arrangements and transmission methods.Compared with traditional electric vehicles,distributed electric vehicles have greater advantages in terms of steering,driving torque response,and vehicle handling.However,as the number of distributed electric vehicle actuators,sensors,etc.increases,the reliability of the driving systems will be reduced.Therefore,according to different situations of actuator failure of distributed electric vehicle driving systems,this dissertation conducts fault-tolerant control algorithm research to improve the stability and safety of faulty vehicles.The specific contents include:Firstly,a distributed electric vehicle dynamics model is built that meets the needs of the research,which includes seven degrees of freedom in the longitudinal,lateral,yaw and fourwheel rotation motions of the body.On the basis of this model,the "magic formula" tire model,the wheel rotation dynamics model,the wheel motor model and the longitudinal speed tracking control sub-model are built respectively.The accuracy of the model is verified by Car Sim under different operating conditions.Secondly,in view of the problem of actuator failure in distributed electric vehicle drive systems,a longitudinal and transverse force cooperative fault-tolerant control system is built using the idea of hierarchical modelling.The upper layer is the motion tracking control layer,which uses model predictive control theory to track the desired driving state of the vehicle and obtain the required longitudinal combined force,lateral combined force and yaw moment of the vehicle.The lower layer is the longitudinal and lateral force reconstruction optimization control distribution layer,where the total forces and moments in the upper layer are optimized with the minimization of the tire loading rate as the optimization objective function,and the required longitudinal and lateral forces for each wheel are optimized by considering the tire forces and adhesion circle simplification constraints.The required longitudinal and lateral forces are converted into drive torque and front wheel angle as actuator preparation inputs by means of the moment equation and the inverse model of tire lateral deflection characteristics.According to the fault detection information of the actuator,a failure condition is determined,and then different failure control strategies are used to reconfigure the control distribution to obtain the new drive torque and front wheel angle output,which guarantees the stability and dynamics of the vehicle after the failure.Finally,four different simulation conditions are set up for the three controllable failure situations,and the steering wheel angle is used as the system input,which verifies that the fault-tolerant control algorithm can ensure that the failed vehicle follows the driver’s operational intention to maintain stable driving.At the same time,a hardware-in-the-loop simulation test platform based on the NI-PXI real-time system is built to further verify the effectiveness of the fault-tolerant control algorithm.