OPTIMUM INDUSTRIAL PID CONTROLLER DESIGN FOR PROCESS WITH TIME DELAY VIA ADAPTIVE TABU SEARCH

摘要

The industrial process control systems usually encounter the time delay. Such the systems are commonly controlled by the industrial PID controllers consisting of proportional-integral (P-I) elements existing in forward path for input tracking and load regulating proposes, and derivative (D) element existing in feedback path to reduce noises. In order to find an alternative potential design procedure, this paper proposes design method to obtain an optimum industrial PID controller by the adaptive tabu search (ATS) for a process with time delay. The proposed ATS-based design method is compared with three conventional design methods, i.e. IAE, ISE, and ITAE. Results show that the system responses obtained by the ATS-based method are superior to those obtained by the conventional methods. Once parameter variation occurs in the control loop due to aging and working environment, robust stability and robust performance would be investigated. Based on the Monte Carlo method, it was found that the stability robustness of the controlled system is assured within ±30% of model parameter variations.