Adaptive Control of a Plant with Time-varying Structured Uncertainties - A New Approach

摘要

The physical parameters determining the coefficients of the mathematical model used to represent a plant, are affected by some uncertainties, which leads to a difference between actual and theoretical values of the coefficients and, consequently, to a difference between the actual and ideal outputs of the plant In this paper, a model reference adaptive controller is designed for a linear single variable plant where the parameters present in the differential equation representing the plant contain time-varying structured uncertainties. In this approach, the magnitudes of the uncertainties are estimated on-line first to obtain the controller parameters and then Lyapunov's theory is used to achieve bounded response of the plant, when the model reference adaptive controller is used. Finally the adaptive controller is simulated for a continuously stirred tank reactor (CSTR) in presence of sinusoidal and random uncertainties and the results are compared with those obtained by applying the existing technique of sigma-modification.