Simple Decentralized PID Controller Design Method Based on Dynamic Relative Interaction Analysis

Abstract

This paper presents a simple yet effective method to design a decentralized proportional−integral−derivative controller for multivariable processes. On the basis of structure decomposition, the dynamic relative interaction is defined and used to derive the multiplicate model factor (MMF) for an individual control loop. The MMF is approximated by a time delay function at the neighborhood of individual control loop critical frequency. An equivalent transfer function for each control loop is then obtained by combining the original loop transfer function with the approximated MMF. Consequently, controller parameters for each loop are determined by employing the single-input−single-output controller tuning rules for the equivalent transfer function. A 2 × 2 process is used to demonstrate a step-by-step design procedure, and the simulation results for a 4 × 4 system show that the design technique results in a better overall control system performance than those of existing design methods.