Identification of fractional order non– homogeneous Hammerstein-Wiener MISO continuous systems

Abstract

Aiming at the problem of difficult identification of parameters in fractional non-homogeneous Hammerstein-Wiener nonlinear MISO time-delay continuous systems, an algorithm that can simultaneously identify coupled coefficients, orders, and time-delays in nonlinear fractional-order systems is designed. In this algorithm, a fractional order fitting model is constructed by using the Legendre wavelet operation matrix, and a full-parameter partial derivative model of the time-delay nonlinear system is derived. Second, the sequential quadratic programming (SQP) method is used to simultaneously estimate the non-homogeneous order, delay parameters, and coefficient parameters of the fractional order system. Finally, the effectiveness of the proposed algorithm is demonstrated through an academic example and a two-wall heat conduction system with a nonlinear fractional order system of MISO.