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

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