Adaptive Control For A Class Of Systems With Time-varying Parametric Uncertainties

Time-varying parametric uncertain system is an extension oftime-invariant parametric uncertain system, which has wide practical background.However, due to there are essential differences between time-varying andtime-invariant parametric uncertain systems, the existing adaptive control theory isnot perfect, and many important problems have not been settled, especially theconvergence problem of tracking error hasn’t been solved very well. In this paper,adaptive control for several classes of systems with time-varying parametricuncertainties is studied and some meaningful results have been obtained as follows.1. Adaptive control of high order time-varying uncertain systems with unknowncontrol gain. The uncertainties of system are time-invariant and time-varyingparameters. In previous researches, we needed to introduce saturation controltechnology in order to guarantee that the tracking error converges to zero, however, itneeded to know the upper bound of uncertainty while employing saturation controlmethod. Without the use of saturated control, a new adaptive controller was designedin this paper by using backstepping method, the designed controller guarantees thatthe tracking error can converge to zero.2. Semi-global adaptive control for nonlinear systems withnon-parametric uncertainties and periodic disturbance. Periodic disturbanceis a common type of disturbance, and a typical kind of time-varyinguncertainties. In order to design adaptive controller, first, the neural networksare used to approximate non-parametric uncertainties in the system. Then, byconstructing a barrier Lyapunov function, a semi-global adaptive controller isdesigned based on neural network. The previous method of neural networkneeded to assume that the track of the closed loop system remains in thebounded domain and it can only achieve approximate control. In our method,we can make the state of system adjust to zero asymptotically without makingthe priori assumption and can reach precise control.3. As an application, we consider the tracking control problem of electro-hydraulic servo system model without considering external leak case. Byusing the adaptive control method that puts forward in second chapter, the parameteruncertainties and disturbance have been satisfactorily solved, and the target trackingis realized as well, as a result, the effectiveness of that algorithm can be proved.All the adaptive control algorithms proposed in this paper are simulated bymatlab/simulink, simulation results demonstrate the effectiveness and feasibility ofcontrol methods.