Integrated Fault Diagnosis Of Actuator And Sensor For Vehicle Stability Control System

In recent years,with the increasingly traffic congestion,complex road,and traffic accident,electronic stability system of vehicle as the standard has gradually entered the production of automobile.ESC detects the state of the vehicle to identify the dangerous conditions through sensor with high sensitivity,and determine a viable intervention to restore the vehicle to a stable state in advance.Therefore,it is the key problem of the vehicle stability control system that the fault of sensor and actuator in ESC system can be diagnosed quickly and the corresponding fault tolerance is realized.It is of great significance to study the joint fault diagnosis of actuator and sensor.According to the nonlinear vehicle model of eight degrees of freedom,a virtual input model is proposed.The nonlinear part satisfies the Lipschitz condition.A fault diagnosis and estimation method based on virtual model is proposed.The fault of actuator in virtual model is as a part of the augmented state vector to construct a nonlinear actuator augmented system.The fault of sensor will be transformed into the fault of actuator with simple filter to construct the nonlinear sensor augmented system.The nonlinear actuator augmented system and the nonlinear sensor augmented system are proved that they satisfies the necessary condition of observable.For the nonlinear actuator augmented system,observer is designed by using Lyapunov theory.Linear matrix inequalities making observation error convergent is obtained.The convergence speed of observer is guaranteed.The fault estimation of actuator and the fault diagnosis of sensor are realized.Sliding mode observer for sensor augmented system is designed.The fault estimation of sensor is realized.The actuator is virtual actuator in virtual model,and the virtual actuators have a linear relationship.The fault information of wheel is realized by the fault information of virtual actuator.The maximum Lipschitz constant is designed by binary search method.Finally,the gain matrix is gained through the YALMIP toolbox.The vehicle model of eight degree of freedom and observer model are built by MATLAB-SIMULINK.The fault of the electronic stability control system,such as longitudinal force,lateral force,yaw moment,longitudinal velocity sensor and yaw rate sensor are diagnosed.The simulation results show that two observers can detect the fault information of the actuators and sensors of vehicle stability control system,and the safe driving of the vehicle is well guaranteed.