Identification of a time-delayed process model using an overparameterisation method

An extensive literature exists on the estimation of the model parameters of time-delayed processes. This paper briefly reviews the main overparameterisation methods for the Single-Input, Single Output (SISO) and Multi-Input, Multi-Output (MIMO) model parameter (and time delay) estimation problem. Subsequently, two algorithms, and their modifications, are discussed, and their performances are assessed in simulation. Results indicate that both the algorithms can suitably identify the model parameters, including the time delay. The third author has taught the method successfully to students taking advanced level control courses at DIT for a number of years; these experiences will be briefly reported. The paper also discusses the advantages and disadvantages of the overparameterisation approach.