Binary distillation column top and bottom compositions control using two PI controllers tuned with GA

Abstract

This paper outlines the results that can be obtained when the top and bottom compositions of a propylene/propane distillation column are controlled using two Proportional-Integral (PI) controllers tuned with Genetic Algorithms (GA). The process is nonlinear and the PI controllers must consider this aspect, to get the best performance. In this paper the case of multi-model control is considered as the process nonlinear model is a reunion of linear models, for the different process operating ranges. For each process operating range different sets of controllers' tuning parameters are computed using GA tuning method. The experimental results are compared in terms of control system's static and dynamic performance with the ones obtained when the proposed top and bottom compositions PI controllers are tuned using Ziegler-Nichols (ZN) method, which is the widest used tuning method.