A new approach for parameter identification of time-varying systems via generalized orthogonal polynomials

Abstract

Abstract A very effective method of using the generalized orthogonal polynomials (GOP) for identifying the parameters of a process whose behaviour can be modelled by a linear differential equation with time-varying coefficients in the form of finite-order polynomials is presented. It is based on the differentiation operational matrix of the GOP, which can represent all kinds of individual orthogonal polynomials. The main advantage of using the differentiation operational matrix is that parameter estimation can be made starting at any time of interest, i.e. without the restriction of starting at zero time. In addition, the present computation algorithm is simpler than that of the integration operational matrix. Using the concept of GOP expansion for state and control functions, the differential input-output equation is converted into a set of linear algebraic equations. The unknown parameters are evaluated by a weighted least-squares estimation method. Very satisfactory results for illustrative example are...