Using method of artificial intelligence to optimise and control chemical reactor

Abstract

The application domain of chemical reactions and reactors constitutes one of the backbones for interdisciplinary collaboration. In fact, the optimization of industrial chemical processes has drawn attention in recent years. In this paper, we have used a new method of artificial intelligence, namely the evolutionary algorithm SOW (self-organizing migrating algorithm), for static optimization of a continuous stirred tank reactor (CSTR), for the purpose of improve its behavior in the uncontrolled state and predictive control. The optimizations and control of the reactor have been performed in several ways, each one for a different set of reactor parameters or different cost function. The optimized reactor was used in a simulation with prediction control by evolutionary algorithms and the results are presented in graphs.