Process identification using polynomial models

Deals with the identification of linear systems using input-output response data. Specifically we focus on nonparametric (finite impulse or step response, FIR or FSR) models widely used in model predictive control. A polynomial model is proposed to reduce the number of parameters needed to represent the model. This leads to parsimonious, yet extremely robust models. The time delay and response time of the process can be explicitly included as parameters in the model. Various properties of this model including the variance of parameter estimates are given in the paper. Simulation and experimental results show the superiority of this approach over conventional methods especially at low signal/noise ratios, when other conventional techniques fail. Most remarkably, no prefiltering of the noise is necessary using this method. The polynomials act as an adaptive filter to remove the noise.