Research On Fuzzy Robust Tolerant Control For Nonlinear Systems

The fuzzy robust tolerant control of complex nonlinear systems with time delays, uncertainties, disturbances and faults is studied in this thesis. Several stabilization and tracking control schemes are presented by combining the fuzzy control technique with the robust control, adaptive control and the sliding mode control techniques, so that the closed-loop systems are ensured to be stable and satisfy the desired steady and dynamic control performances.At first, the robust output feedback control problem for a class of T-S fuzzy systems with time-varying state delay and disturbances both in the inputs and outputs is discussed. By defining a delay-dependent Lyapunov function, the L_∞robust control technique is extended to the time-delay systems, the synthetical design procedures of the static output feedback controllers are derived in terms of the LMIs, and the L_∞robust performance is optimized by the iteration algorithms. This method can efficiently attenuate the peak of the persistent bounded disturbances.Secondly, the L_∞robust tolerant control scheme is proposed, and the steady and dynamic performances of the T-S fuzzy systems are restricted in the D-region even when the structural failures of the sensors or the actuators occurred. Then, the additional matrix is introduced to increase the extra degree of the freedom according to the projection lemma in the design process, so that the closed-loop nonlinear systems are not only ensured to be robust and reliable but also have satisfied performances. Meanwhile, the problem of robust tolerant control for T-S fuzzy systems with time-varying state delay and actuator saturation is investigated. By defining a class of delay-dependent Lyapunov function and a given ellipsoid with state constraint, the set invariance condition and sufficient conditions of the existence of the robust dissipative tolerant controllers are given. The design results of the control scheme can not only provide a unified framework for H_∞control and passive control, but also ensure the closed-loop systems to be stable and reliable when the actuator faults occur.In order to further remove the hidden danger, the robust tolerant controllers with integrity are designed for a class of fuzzy singular time-delay systems, and the closed-loop systems are ensured to be stable even when the actuator and (or) sensor failures occurred simultaneously. Then, the equivalent conditions are introduced to relax the constraint in the process of solving LMIs. The conclusion can be extended to design robust tolerant controller with integrity for nonsingular systems. Furthermore, in fact, the T-S fuzzy model is difficult to describe the nonlinear systems with uncertainties, disturbances and faults in each equilibrium point, but the adaptive fuzzy logic systems can approximate the unknown system model dynamically, so, by combining the adaptive fuzzy logic systems with the sliding mode control technique, this paper presents an adaptive sliding mode output feedback tracking control scheme for a class of time-delay nonlinear systems with sector nonlinear inputs containing non-symmetric saturation and deadzones. When the states are unmeasurable, and under the assumption that the system mathematical models are known or unknown, the synthetical design schemes for sliding mode controllers are presented based on the adaptive observer and the fuzzy observer, respectively, and the parameter adjusting laws of the adaptive controller and the fuzzy logic system are derived by defining a class of Lyapunov-Krasovskii function. The proposed method not only has the advantage of little dependence on system models, but also enhances the robustness to disturbances, and ensures the stability of the closed-loop tracking error systems.In succession, an adaptive robust tolerant positioning control scheme is produced for a class of multivariable nonlinear systems with modeling errors and unknown uncertainties. The multivariable systems are composed of some subsystems, and the uncertainties and faults are assumed to be norm-bounded, but the upper bound is unknown. Based on the theory of Lyapunov and the Barbalat lemma, the synthetical design procedures of the reduced-order robust sliding mode controllers are given, and the fuzzy adaptive controllers in this scheme is used to cancel the effect caused by the modeling error, disturbances and unknown actuator faults. This positioning control method reduces the traditional and contrived estimation of the uncertainty, which does not satisfy the so-called matching condition.At last, the proposed adaptive fuzzy sliding mode tolerant control schemes are successfully applied to the simulation of the ship with dynamic positioning systems, and the simulation results indicate the validity of the proposed methods.