Parameter identification of systems with preload nonlinearities based on the finite impulse response model and negative gradient search

摘要

This paper considers identification problems of a class of nonlinear systems. By introducing a switching function and the finite impulse response model, we obtain the identification model of a preload nonlinear system and develop a stochastic gradient algorithm to the parameters of the system. Furthermore, a stochastic gradient algorithm with a forgetting factor is given to improve the identification accuracy. A simulation results indicate that the forgetting factor can improve the parameter estimation accuracy.