Fault Estimation And Fault-tolerant Control For Markov-type Stochastic Active Suspension System

Suspension system,as the transmission element between the frame and the body of a vehicle,plays a key role in the external vibration reduction of the vehicle.For this reason,suspension control has become a new hot topic of automobile chassis all over the world in recent years.In the process of vehicle usage,components will inevitably produce fatigue,aging,and other phenomena,which may lead to sensor and actuator faults,once fault occurs,the control effect of the controller will be partially or completely lost.Therefore,it is necessary to study the control strategy of faulty suspension system.This paper aims to solve the problem of nonfragile fault-tolerant control for a class of Markov-type suspension systems with faults and disturbances.The main work is as follows:(1)Considering the faults and the suspension system states delay caused by the fault,the dynamic modeling of the faulty active suspension system was carried out,and its state space equation was then derived.Afterwards,the faulty active suspension system was reconstructed as Markov jump system,which could accurately capture the dynamic behavior of the vehicle suspension system,and was convenient for the design of the faults observers and fault-tolerant controllers.(2)In order to repair the faulty active suspension system,Firstly,an adaptive observer is designed to accurately estimate the fault and suspension system status.The fault estimation error of the sensor is between ±0.004 and±0.005,and the fault estimation error of the actuator is between±0.2 and ±0.3.Second,the state feedback fault-tolerant controller is designed to compensate the suspension system faults,to ensure the stability of vertical and pitch motion of the suspension system.After that,the stability of the whole system is analyzed by using Lyapunov theory.Finally,compared with the faulty system,the state feedback fault-tolerant controller improves the system performance by more than 17%,indicating the effectiveness of the proposed state feedback fault-tolerant controller.(3)Considering the time delay of the faulty active suspension system,the dynamic model of the faulty active suspension system was established in the first place,and then a robust stochastic observer was designed to estimate the system state and unknown system faults,both the sensor fault estimation error and the actuator fault estimation error are between ±0.01.In addition,a non-fragile fault tolerant controller based on output feedback technology was developed to enhance the robustness of the vehicle suspension system performance.Finally,through the simulation of compound fault compensation control under different road excitation,the output feedback fault-tolerant controller improves the system performance by more than 48%compared with the faulty system,indicating the effectiveness of the proposed output feedback fault tolerant control method.(4)According to the fault-tolerant control requirements of active suspension,MATLAB APP Designer is used to design the intelligent control performance analysis software of active suspension,and the software functions are described.The developed software is helpful for automobile engineers and the related researchers.