Gradient-based step response identification of overdamped processes with time delay

Abstract

In this paper, a step response identification method is proposed for overdamped industrial processes with time delay from sampled data by developing a gradient searching approach to minimize the output prediction error. Based on establishing a least-squares fitting of the time domain expression of a low-order process model, i.e. first-order-plus-dead-time (FOPDT) or second-order-plus-dead-time (SOPDT), with respect to a step change, the rational model parameters together with the delay parameter can be simultaneously estimated, while the computation effort can be significantly reduced compared to the existing step identification methods based on using the time integral approach for model fitting. Both cases of repetitive poles and distinct poles are considered for the identification of an SOPDT model, along with a guideline for a suitable choice of the model structure for practical applications. The convergence and accuracy of the proposed algorithms are analysed with a strict proof. Four illustrative examples from recent references are used to show the effectiveness of the proposed method.