PSO based optimized fuzzy controllers for decoupled highly interacted distillation process

Abstract

Normally, the highly interacted MIMO process – such as two coupled distillation columns – is decoupled into a group of independent loops and a conventional PID controller is assigned to control each loop. Tuning of conventional PID controllers is very difficult. Scientists consider tuning of PID controllers is an art more than science. In this paper, fuzzy PID controllers are proposed to replace the conventional ones. Moreover, the values of the parameters of the proposed fuzzy PID controllers are optimized using particle swarm optimization (PSO) technique. Sum square errors (SSEs) – for different loops – are used as fitness functions for PSO. SSEs minimization assures optimal values of different fuzzy PID controllers’ parameters. For the purpose of validation, PSO is also used to optimize the design of conventional PID controllers. The simulation of the proposed optimized fuzzy PID controllers proves their excellence in improving the transient and steady state characteristics.