Microcomputer Systems Allow a Real-Time Parameter Estimation for the Control of a Solar Plant

Abstract This paper describes a study of the applicability of identification LS and GLS recursive algorithms to solar energy systems; a microcomputer structure is implemented. A priori physical knowledge of the considered process is utilized to determine the model structure. The recursive character of the methods is able to continuously evaluate the characteristic parameters of a linearized, time-varying, model derived from the non-linear original one. The comparison between LS and GLS algorithms is reported; the simulation of the identified models results in good agreement with the real data in spite of the introduced simplifications. The microcomputer system designed allows advantages in the cost, flexibility and adaptability in the development of effective devices for non-conventional control.