Research On Integral Sliding Mode Control For Singular Systems

In the practical engineering applications, there exist unknowns and uncertainties between mathematical model and the real plant because of the modeling error, system perturbation, and external disturbance. When using the controller designed based on the precise model, the uncertainties may degrade the performance. Sliding mode control which has evolved into an important branch of nonlinear robust control theory, has attracted more and more attention as the result of its perfect features of arbitrary robustness and complete adaption to matching uncertainties and external disturbances. As a recently proposed design concept, the integral sliding mode control which eliminates the reaching phase guarantees that system variables will start in the manifold from the first time instant.There widely exist singular systems in the area of constrained robot manipulator, electric power system and aerospace system, but the research results on uncertain singular systems are not very rich. Therefore some problems of integral sliding mode control for uncertain singular system are studied in this thesis, and the primary results can be described as follows,(1) The synthetic design method of integral sliding mode control for the singular system has been presented, which includes how to design the sliding manifold and the controller.(2) To solve the problem of the reaching phase and of the robustness against unmatched perturbations simultaneously , the combination of integral sliding mode control and H_∞control has been proposed. After the decomposition, the matched uncertainty is canceled by the discontinuous sliding mode controller and the unmatched part is treated by the H_∞control. Besides, a method of selecting the projection matrix is proposed, which minimizes the effect on the sliding mode. So the controller is optimized due to the introduction of the integral sliding mode control. (3) A performance oriented robust controller is proposed for a class of singular systems with matched and unmatched nonlinear uncertainties. The control approach combines integral sliding mode control (ISM), additional nonlinear control with composite nonlinear feedback control (CNF). The ISM control term rejects matched uncertainties completely and drives the system trajectory to the ideal sliding mode. The additional nonlinear control term attenuates the unmatched nonlinear uncertainties which exist on ideal sliding mode and may affect the performance of the sliding mode. The CNF control term guarantees the output tracking of reference input signal according to the prescribed specifications.(4) An uncertain singular system model for the 3 DOF helicopter is established by the approximation. The numerical examples are given to illustrate the effectivity of the proposed approach.The conclusion and perspective are given at the end of the dissertation.