System identification based on closed-loop step response data

Abstract

This paper presents a new identification algorithm for processes operating in closed-loop. The algorithm involves fitting two Laguerre models directly to the control signal and the process output signal generated by a step change in the setpoint. The algorithm does not require any prior assumptions about the structure of the process model. The method generates an estimate of the process model in the frequency domain, and then transforms this frequency domain estimate into a continuous-time Laguerre model. Simulation results are used to illustrate the accuracy of the algorithm.