A new method to estimate a first-order plus time delay model from step response

In this paper an identification method to estimate the parameters of a first-order plus time delay model is proposed. Such a method directly obtains these parameters using a new linear regression equation. No iterations in calculation are needed. Moreover, a simple true/false criterion to establish if the hypothesis on the process type is correct can be easily derived. The proposed method shows an acceptable robustness to disturbances and measurement noise as it is confirmed by several simulated experiments.