Sensor Fault Tolerant Control For Strict Feedback Stochastic Nonlinear Systems

Complexity and large scale are the new development direction of modern industrial systems,the safe operation and reliability of the system are becoming more and more important.As the system becomes more and more complex,the possibility of system failure increases,so the importance of fault-tolerant control in realistic manufacturing environments is self-evident.Nonlinear components are generally included in industrial systems,so nonlinear systems are of great research interest.Actual engineering systems usually contain random noise,research on stochastic systems has also made great progress.Based on the current research of strict feedback nonlinear systems,and using fault-tolerant control technology,decoupling backstepping technique and parameter separating technology,this thesis studies the state feedback sensor fault-tolerant control problem for a class of stochastic nonlinear systems with unknown dynamics and decentralized sensor fault-tolerant control problem for a class of interconnected stochastic nonlinear large-scale systems.First of all,the problem of adaptive fault-tolerant control is considered for a class of singleinput and single-output nonlinear Ito stochastic systems with unknown dynamics and multiple sensor faults.Due to the partial loss of effectiveness of sensors,each measured state contains an unknown time-varying fault parameter.Then,to circumvent the main obstacle caused by the coupling of unknown fault parameters and real states,a new method of fault parameters separating of stochastic nonlinear systems is proposed.Combining with the modified backstepping design techniques,an adaptive state feedback controller is constructed recursively to estimate unknown fault parameters and guarantee the stabilization of the stochastic system.By using quartic Lyapunov functions,it is proved that all signals of the closed-loop system are bounded in probability.Secondly,the problem of sensor fault-tolerant control is considered for a class of strict feedback interconnected stochastic nonlinear large-scale systems with multiple sensor faults.The interconnection term depends on the output of each subsystem,and sensor failure may occur in all subsystems.Due to the failure of the sensor,there are also fault parameters in the interconnection term.In order to overcome the difficulties caused by the coupling of fault parameters and real states,a new method for fault separation of interconnected stochastic nonlinear large systems is proposed.Combined with decentralized control technology and decoupling backstepping design technology,an adaptive state feedback controller is designed to ensure the stability of the entire interconnected stochastic nonlinear large-scale system.By using quartic Lyapunov functions,it is proved that all signals of the closed-loop system are bounded in probability.Finally,for the considered two class of strict-feedback stochastic nonlinear systems,the simulation results are given to illustrate the effectiveness of the proposed scheme.