A comparative study of PID based temperature control of CSTR using Genetic Algorithm and Particle Swarm Optimization

Abstract

Industrial processes are highly integrated processes and Continuous Stirred Tank Reactor (CSTR) is an integral part of it. To work efficiently in a system certain parameters of the CSTR needs to be controlled. This paper presents a model of the PID controller using Zeigler Nichols tuning for the temperature control of the CSTR. Optimization techniques Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) have been used to optimize the controller parameters and a comparative study is done taking the transient and steady-state characteristics into considerations. It is found that the controller parameters are best optimized by Particle Swarm Optimization irrespective of the disturbances than other optimization techniques.