MULTI-LAGUERRE APPROACH TO IDENTIFY PLANT WITH WIDELY SPREAD POLES

Abstract An identification method for a plant which has fast and slow poles is presented. The extended Laguerre model is used to estimate the parameters of the continuous-time ARX model with discrete-time input and output signals. To detect both the fast and slow poles of the plant, several cost functions are prepared for each corresponding frequency region and summed to estimate the parameters with least squares. To achieve bias-free identification, the IV approach is used. With the proposed method, the sampling period can be arbitrary short without making the identification accuracy worse.