Application of self-tuning control based on generalized minimum variance method in evaporator For ORCS

In order to solve uncertainty and time-varying problems of evaporator which is non-minimum phase object, direct self-tuning control based on generalized minimum variance method combined with the forgetting factor recursive least squares(FFRLS)parameter estimation is investigated in this paper. Firstly, self-tuning control has the ability to tune its own parameters. It is well suited for handling time-varying and uncertain system. Secondly, Generalized minimum variance control(GMV) is one of the most flexible and successful approaches to self-tuning control. GMV employs a cost function that incorporates system inputs, outputs and reference signals, and thus enables controller to provide an elegant way of tracking reference signals and handling non-minimum phase systems. Thirdly, forgetting factor recursive least squares(FFRLS) parameter identification method is very simple, fast and suitable for time-varying system. This algorithm is applied to control the temperature of evaporator's outlet under various disturbances. Simulations indicate good stationary and dynamic performance along with good tracking property.