New Predictor and 2DOF Control Scheme for Industrial Processes With Long Time Delay

Abstract

To address the difficulty of controlling industrial processes with long time delay, a novel design of dead-time compensator (DTC) is introduced, which can be used to predict the undelayed output response of any process (no matter stable or unstable) such that the control design may be focused on the delay-free part of the process for performance optimization. Based on the undelayed output estimation, a two-degree-of-freedom (2DOF) control scheme is analytically developed for optimizing the set-point tracking and disturbance rejection, respectively. By proposing the desired transfer functions, the corresponding controllers are analytically derived based on commonly used low-order process models. A notable advantage is that there is a single adjustable parameter in the proposed DTC, as well as in each controller, which can be monotonically tuned to meet a good tradeoff between the prediction (or control) performance and its robustness. Illustrative examples from the literature and a practical application to a temperature control system of a jacketed reactor are used to demonstrate the effectiveness of the proposed predictor-based control scheme.