OPTIMAL DESIGN OF C ONTINUOUS S TIRRED T ANK R EACTOR C ONTROL S YSTEM , U SING E VOLUTIONARY O PTIMIZATION M ETHODS B ASED O N T EACHING - L EARNING A ND G ENE TICS

Continuous s tirred tank reactor (CSTR) is an important issue in chemical processes and offers a diverse range of research on chemical control engineering and a variety of control methods in CSTR is used to control its parameters. CSTR various scientific control methods are used to control the p arameters. In this paper, two different control method is performed, an optimization algorithm based on training - learning and other genetic optimization algorithm. The goal is to control CSTR temperature in the presence of a set of point changes that factories are design its control and simulation it in MATLAB and also plans to design proposed control function based on the least square error (MSE) and standard deviation (SDC). The results clearly show that both control strategies provide acceptable pe rformance due to functional changes. In other words it is tracking the robustness of the proposed methods in the face of uncertainty across achieving the reference signal in calculation. In addition, the algorithm based on training - learning has better di spersion than genetic algorithm. Genetics also has better rate than education ? learning