Recursive subspace identification of Hammerstein-type nonlinear systems under slow time-varying load disturbance

In this paper, a recursive subspace identification method is proposed for Hammerstein-type nonlinear systems that is affected by slow time-varying load disturbance. Based on superposition principle, the system output response is divided into three parts, i.e. disturbed, stochastic and deterministic components. Correspondingly, the disturbance response is regarded as a time-varying variable to be identified. In order to achieve the goal of identifying the system matrices accurately, a recursive leastsquares (RLS) identification method is proposed. The peculiarity of this method lies in the introduction of forgetting factors. One of the forgetting factors quickly tracks the load disturbance response, the other are for recursive estimation. An illustrative example is presented to demonstrate the effectiveness and merit of the proposed identification method.