Font Size: a A A

Adaptive Fault Tolerant Control For Nonlinear Systems With Stochastic Actuator Failures

Posted on:2015-03-08Degree:DoctorType:Dissertation
Country:ChinaCandidate:B LiuFull Text:PDF
GTID:1228330428984306Subject:Control Science and Engineering
Abstract/Summary:PDF Full Text Request
The growing demand for reliability and safety in systems has motivated significant research in fault tolerant control. Fault tolerant control can accommodate faults among actuators, sensors or system components automatically to guarantee system stability and a desired level of performance. Over the past years, tremendous effort has been devoted to fault tolerant control research and many interesting results have been obtained. How-ever, most existing literatures focus on the case of finite number of actuator failures, that is, once an actuator fails, it will stay at the faulty mode during its rest operation. For the case of intermittent actuator failures which may cause infinite number of actuator failures, e.g. the failures caused by the electromagnetic wave interference from the outer space in spacecraft control system, or the failures caused by the communication network embed-ded into the control loops, the existing actuator failure compensation scheme will not work effectively. Meanwhile, the times, patterns and values of actuator failures are practically undetermined beforehand and are essentially stochastic. However, there are relatively few studies on fault tolerant control with stochastic actuator failures. As a result, the fault toler-ant control problem for several classes of nonlinear systems with stochastic actuator failures will be considered in our paper.First, for four classes of systems, that is, deterministic system with Markovian vec-tor, nonlinear system with Markovian vector and stochastic noise, nonlinear system with Markovian vector and constant time delay, nonlinear system with Markovian vector and time-varying time delay, some preliminaries including the joint transition probability, the infinitesimal generator, the existence and uniqueness of solution and the stability criterion are proposed respectively.Then, stochastic functions related to Markovian variables are introduced to denote the failure scaling factors for each actuators and take values between0and1, which can be classified into three cases. Different values of stochastic functions represent different modes of actuators. Therefore, the established actuator failure model is much more general and practical. Not only the partial loss of effectiveness and total loss of effectiveness, but also the stuck case is taking into account. Moreover, based on the obtained preliminaries, stabi-lization and tracking problems for deterministic systems, stochastic systems, constant time delay systems and time-varying time delay systems are concerned. Some state feedback con-trollers are designed by employing the adaptive backstepping method and the stability of all closed-loop signals are analyzed according to the Lyapunov stability theory. Some examples are presented to show the effectiveness of the proposed scheme. Besides, the extra transition rate related terms appearing in the infinitesimal generator of the Lyapunov function and the higher order Hessian term in the Ito stochastic differentiation are dealt with appropriately. It should be pointed out that no fault detection and diagnostic block is needed and the control law can be adapted automatically by taking account of the novel state information, which is easier to be implemented in practice.Finally, some open problems are pointed out in our future studies.
Keywords/Search Tags:Stochastic actuator failure, Fault tolerant control, Markovian vector, State feed-back, Adaptive
PDF Full Text Request
Related items