Actuators Fault Diagnosis Of ESC System Considering Crosswind Disturbance

Vehicle safety refers to the performance of vehicles in avoiding accidents and ensures the safety of pedestrians and occupants when driving.Generally,it is divided into active safety,passive safety,accidental safety and ecological safety.In road traffic accidents,the safety of the car is a factor that cannot be ignored.A Car with perfect safety performance will avoid the occurrence of accidents or reduce the degree of casualties.Electronic Stability Control(ESC)is a kind of active safety system,which can effectively improve the stability of the vehicle operation and prevent the occurrence of skidding and tailgating.To a certain extent,ESC system can ensure human safety.When the ESC system fails,the stability of the vehicle is difficult to guarantee.If ESC failure is not promptly diagnosed,some serious things will happen.Therefore,fault diagnosis of ESC system has become the focus of research in recent years.First of all,in order to realize the vehicle fault diagnosis and consider the subsequent fault-tolerant control,this paper introduces the actuator fault and crosswind disturbance on the basis of seven-degree-of-freedom vehicle model.Considering the current general layered algorithm of vehicle stability control,this paper selects the longitudinal force,lateral force and lateral moment of convergence of the vehicle as the virtual input of the system.In order to omit the complicated lower torque distribution in fault-tolerant control,this paper will describe the actuator failure with the change of the tire's force.Through the aerodynamic analysis of the vehicle,the aerodynamic drag,lateral force and yaw moment generated by the crosswind are introduced into the seven-degree-of-freedom vehicle model.Secondly,this paper studies the fault diagnosis of ESC system actuators under special crosswind conditions.In order to implement fault diagnosis algorithm decoupling disturbance,we adopt the differential homeomorphism transformation method to obtain the maximum observable space of the system and realize the transformation of the system.The resulting subsystem can be seen as a system that is affected only by faults and known inputs.In order to obtain fault information,this paper will establish the augmentation system with the fault value as an augmented state based on the subsystem.Using Lyapunov to design a nonlinear fault diagnosis observer for augmented subsystems to realize fault diagnosis of a single actuator.Using Matlab / Simulink to build models and fault diagnosis observers,the simulation results show that the fault can be detected and isolated.Finally,taking into account that the fault diagnosis algorithm proposed under the special crosswind situation cannot be widely used,this paper designs the ESC system actuator fault diagnosis algorithm for the general side wind disturbance situation.Taking into account the nonlinear of the seven-degree-of-freedom vehicle model considering cross-wind disturbances and actuator faults,and the nonlinear observer design method for fault diagnosis with disturbance and fault is complex,we make the model linearized.In order to realize the linearization of all states,the feedback linearization method based on differential geometry is adopted in this paper,that is,the transformation is performed by using the differential homeomorphic mapping and the state feedback is introduced at the same time.After linearization,fault and state are coupling.Disturbance and state are also coupling.In order to get the fault information,this paper establishes the augmented system as the augmented state.Considering that the disturbance cannot be completely decoupled,this paper adopts the idea of robust observer design.By selecting the observer gain matrix,we can ensure the convergence of the state estimation error as soon as possible while achieving progressive convergence of the observation error.The fault diagnosis observer with suppressed disturbance is obtained and we can realize fault diagnosis when a single actuator failure occurs.Finally,using MATLAB/SIMULINK to build the model and fault diagnosis observer.The simulation results show that the fault diagnosis algorithm can detect and isolate faults in the case of no-fault and gradual change and abrupt change of the actuator.