Fault Detection Of Active Front Wheel Steering System Based On Event-triggered Interval Observer

The vehicle active front wheel steering system realizes the normal steering operation through the active control of the vehicle,which influence directly the safety and stability of the driver in the process of driving the vehicle.It has important study meaning and is the main research topic of vehicle security technologies.Because of the complex structure of the system,long time operation will cause component loss,and greatly increasing the probability of failure in the process of vehicle operation,so in order to enhance the safety and stability of the system,it is necessary to detect the fault in time.The core of model-based fault detection strategy is residual generation and residual evaluation.Observer is an effective residual generator,and interval observer provides a new direction for fault detection.In this thesis,a fault detection strategy based on event-triggered interval observer is proposed for active front wheel steering system.Specific research work includes:Firstly,the dynamic analysis of active front wheel steering system is carried out,and the dynamic model of double-freedom active front wheel steering system is established.The actuator fault model is built when the actuator has offset fault.Secondly,a fault detection method based on event triggered interval observer is proposed.An event triggered mechanism is designed when aiming at the problems of communication burden and sampling mode.The event triggered interval observer is built when considering the factors such as event error and disturbance.Then,the ₁L and H_?performance indicators are introduced to improve the robustness and fault sensitivity of the residual interval.The relaxation variable is employed to decouple the system matrix and Lyapunov matrix,and the design conditions of the interval observer are obtained without restricting the structure of the Lyapunov matrix.On the basis of fault detection and analysis,the policy-making logic of fault detection is given to decrease the communication burden while ensuring the performance of fault detection.Furthermore,the simulation proves the effectiveness of the suggested method.Thirdly,the fault detection in the finite frequency domain is further studied for the actual operation of the vehicle,where faults in the active front wheel steering system usually occur in the low frequency range.By means of the equivalence between frequency domain and time domain in generalized KYP lemma,₁L and H_-performance indicators are introduced to improve the robustness and fault sensitivity of residual interval.The relaxation variable is used to decouple the system matrix and Lyapunov matrix,and the linear matrix inequality design condition of interval observer is obtained.The effectiveness of the proposed method is proved by simulation,and compared with the full frequency domain results,it is proved that the designed finite frequency interval observer improves the fault sensitivity and ensures better detection performance.