Font Size: a A A

Study Of Fault Diagnosis And Fault Tolerant Control Based On Adaptive Control Technique

Posted on:2016-08-04Degree:DoctorType:Dissertation
Country:ChinaCandidate:Q K ShenFull Text:PDF
GTID:1108330503976018Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
In order to improve system reliability and to guarantee system stability in all situations, many effective fault tolerant control(FTC) approaches have been proposed in literature. Among the faults occurred in the controlled systems, the actuator faults and sensor faults are common. Up to now, for the actuator faults and sensor faults, many relevant results have been obtained in literature. However, these theoretical studies are not perfect, and there still are problems of actuator faults and sensor faults, which are worth to be further investigated. This paper, combining adaptive control technique with the other control technique or approaches, investigates the problem of fault diagnosis(FD) and fault tolerant control for uncertain dynamic systems including linear and nonlinear systems with or without time delay. First of all, a novel general composite fault model is proposed, which can deal with both time-varying gain and bias faults, and FD and FTC for nonlinear systems with such faults occurred in one or multiple actuators or sensors are investigated, respectively. Then, by using the implicit function theorem and exploring the useful property of the basis function of the radial basis function neural network, FD and fault compensation for un-modeled faults are discussed. Next, the fault detection problem of uncertain time-delay systems is considered, and a novel adaptive fault detection observer is proposed, which can estimate the unknown time delay. Further, the time delay due to fault diagnosis and isolation(FDI) and its influence on the controlled systems’ performance are quantitatively analyzed. Finally, based on the above results, the cooperative adaptive fault tolerant fuzzy tracking control problem of networked nonlinear multi-agent with time-varying actuator faults is studied. This study is divided into the following sections:Firstly, the problem of fault tolerant control for T-S fuzzy systems with single actuator faults is addressed. First, a general actuator fault model is proposed, which integrates time-varying bias faults and time-varying gain faults. Then, sliding mode observers(SMOs) are designed to provide a bank of residuals for fault detection and isolation(FDI), and a novel fault diagnostic algorithm is proposed, which removes the classical assumption that the time derivative of the output errors should be known as in some existing work. Further, a novel fault estimation observer is designed. Utilizing the estimated actuator fault, an accommodation scheme is proposed to compensate for the effect of the fault. In addition, a sufficient condition for the existence of SMOs is derived according to Lyapunov stability theory. Finally, simulation results of near space hypersonic vehicle(NSHV) are presented to demonstrate the efficiency of the proposed approach..Secondly, a practical issue in fault diagnosis and estimation for a near space hypersonic vehicle(NSHV) with single sensor faults is investigated. First, the Takagi–Sugeno(T–S) fuzzy model is established to represent the NSHV nonlinear dynamics. Then, a novel sensor fault model is proposed, which contains time-varying bias faults and time-varying gain faults. Further, based on the Lyapunov stability theory, novel SMO-based fault diagnostic algorithms are given to detect, isolate and estimate the sensor fault, which removes the two classical assumptions in existing works that the time derivative of the output errors should be known and that the system states should be bounded before and after fault occurrence. Finally, simulation results are presented to demonstrate the efficiency of the proposed approach.Thirdly, the problem of fault tolerant control(FTC) for near space vehicle(NSV) attitude dynamics with multiple actuator faults is considered, which is described by a T-S fuzzy model. Firstly, a general actuator fault model is proposed, which integrates varying bias and gain faults, which are assumed to be dependent on the system state. Then, sliding mode observers(SMOs) are designed to provide a bank of residuals for fault detection and isolation(FDI). Based on Lyapunov stability theory, a novel fault diagnostic algorithm is proposed, which removes the classical assumption that the time derivative of the output error should be known. Further, for the two cases where the state is available or not, two accommodation schemes are proposed to compensate for the effect of the faults, where fuzzy logic systems are used to approximate the unknown state-dependant faults. These schemes do not need the condition that the bounds of the time derivative of the faults should be known. In addition, a sufficient condition for the existence of SMOs is derived according to Lyapunov stability theory. Finally, simulation results of NSV are presented to demonstrate the efficiency of the proposed FTC approach.Fourthly, the problem of fault-tolerant control(FTC) for a class of uncertain nonlinear systems with actuator faults is discussed, and an observer-based FTC scheme is proposed. Adaptive fuzzy observers are designed to provide a bank of residuals for fault detection and isolation(FDI). Using a backstepping approach, a novel fault diagnosis algorithm is proposed, which removes the classical assumption that the time derivative of the output error should be known. Further, an accommodation scheme is proposed to compensate for the effect of the fault, where it is not needed to know the bounds of the time derivative of the fault. The proposed controller guarantees that all signals of the closed-loop system are semi-globally uniformly ultimately bounded(SGUUB) and converge to a small neighborhood of the origin by appropriately choosing designed parameters. Finally, a numerical example is used to demonstrate the effectiveness of the proposed FTC approach.Fifthly, the problem of adaptive active fault-tolerant control for a class of high-order nonlinear systems with unknown actuator fault is investigated. The actuator fault is assumed to have no traditional affine appearance of the system state variables and control input. The useful property of the basis function of the radial basis function neural network, which will be used in the design of the fault tolerant controller, is explored. Based on the analysis of the design of normal and passive fault tolerant controllers, by using the implicit function theorem, a novel neural networks-based active fault-tolerant control scheme with fault alarm is proposed. Comparing with results in literature, the fault-tolerant control scheme can minimize the time delay between fault occurrence and accommodation that is called the time delay due to fault diagnosis, and reduce the adverse effect on system performance. Furthermore, the fault-tolerant control scheme requires no additional fault detection and isolation model which is necessary in the traditional active fault-tolerant control scheme. Finally, simulation results are presented to demonstrate the efficiency of the developed techniques.Sixthly, the problem of fault detection for a class of nonlinear time-delay systems is investigated. The time-delay under consideration is assumed to be an unknown constant. In order to overcome the shortcoming that the fault diagnosis observers contain unknown time delay in the existing works, novel adaptive fuzzy fault diagnosis observer(AFFDO), which can estimate the time delay online, is introduced. The observer error is ensured to be bounded and then the convergence of the AFFDO is shown. In addition, sufficient conditions for the existence of AFFDO are derived. Finally, simulation results are presented to illustrate the effectiveness of the proposed algorithm.Seventhly, it is well known that there always exists some level of time delay between fault occurrence and fault accommodation, which is called as the time delay due to fault diagnosis(TDDTFD) in this paper. TDDTFD may cause severe loss of system performance and stability. This paper investigates TDDTFD’s adverse effect on system performance. First, a fault diagnosis(FD) model is constructed to diagnose sensor faults which integrate time-varying gain and bias faults, where a novel FD algorithm is proposed. Meanwhile, the time spent at each step in FD and its analytical expression is derived strictly. Further, the analysis of system performance degraded by TDDTFD is developed, and the conditions under which the magnitudes of sensor faults should be satisfied such that the state of the faulty system controlled by the normal controller remains bounded during TDDTFD are derived. In addition, the corresponding solutions are proposed to minimize the adverse effect of the time delay. Finally, simulation results of near space vehicle(NSV) attitude dynamics are presented to demonstrate the efficiency of the proposed approach.Finally, The cooperative adaptive fault tolerant fuzzy tracking control(CAFTFTC) problem of networked uncertain high-order nonlinear multi-agent with time-varying actuator faults is studied, and a novel CAFTFTC scheme is proposed to guarantee that all follower nodes asymptotically synchronize a leader node with tracking errors converging to a small adjustable neighborhood of the origin in spit of actuator faults. The communication network is also assumed to be a weighted directed graph with a fixed topology. Moreover, the adaptive compensation term of the optimal approximation errors and external disturbances is adopted to reduce the effects of the errors and disturbances, which removes the assumption that the upper bounds of unknown function approximation errors and disturbances should be known. In addition,analysis of stability and parameter convergence of the proposed algorithm are conducted based on algebraic graph theory and Lyapunov theory. Comparing with results in literature, the CAFTFTC scheme can minimize the time delay between fault occurrence and accommodation, and reduce its adverse effect on system performance. The FTC scheme requires no additional fault isolation model which is necessary in the traditional active FTC scheme. Finally, an example is provided to validate the theoretical results.
Keywords/Search Tags:Fault detection, fault estimation, fault accommodation, fault tolerant control, adaptive control
PDF Full Text Request
Related items