CONCENTRATION CONTROL OF AN ISOTHERMAL CSTR BY SOFT COMPUTING TECHNIQUES

Abstract

Continuous Stirred Tank Reactor plays a very important role in all chemical reactions and it is difficult to control because of its highly non linear system. Since it is difficult to find the optimal or PID parameters near optimal values by conventional tuning methods like Ziegler Nichols method, stochastic optimization methods like Particle Swarm Optimization method and Bacterial Forging Optimization algorithms methods are used for PID tuning. The primary objective of this work is to analyze the implementation of PSO method and BFO method for PID controller optimal tuning for concentration control of CSTR. The comparative study is done on the basis of step response analysis. Next the stochastic methods are applied for finding optimal tuning PID parameters using error indices as objective functions like integral time absolute error, integral square error, etc for concentration control of CSTR. First, PSO is utilized to optimize PID parameters for concentration control of CSTR. Secondly, the Bacterial Forging algorithm is implemented to obtain the optimized PID parameters for concentration control of CSTR . The results obtained by implementation of PSO and BFO algorithm on CSTR process are compared on the basis of step responses. MATLAB software is used for all calculations and analysis. The conventional tuning methods are not very efficient in this case because there are non linear ties in the system. The output of CSTR concentration control system tuned by conventional method i.e. Ziegler Nichols has a high overshoot and settling time. The main objective of this work is to apply PSO and BFO to design PID controller for an isothermal CSTR to get an output with better step response.