Analysis and simulation of the continuous stirred tank reactor system using genetic algorithm

Abstract

Continuous Stirred Tank Reactor (CSTR) is broadly used in process industry as well as chemical industry to higher biomass productivity and production efficiency. Reactor temperature is deviated from the reference value because dynamic characteristic of CSTR is highly nonlinear. Chemical reaction depends on the reactor temperature and due to temperature deviation may degrade the quality of the product. Designing a control system for the ability to reduce the effects of disturbances is a challenging control problem. In this paper, tuning of three Proportional–Integral–Derivative (PID) controller gains is carried out by Genetic Algorithm (GA). In our work, controller gains can be derived suitably by means of minimizing the objective function. The proposed schemes are exposed with the hybrid combinations PID-GA and schemes are compared with the conventional PID tuning method. Studies of proposed techniques give fast computation and reduction in the error. The CSTR system is simulated by the proposed hybrid control schemes. The performance analysis comprises study of without controller, conventional PID and PID-GA in Matlab environment. The simulation results represent the substantial improvement in terms of response of CSTR, Mean Square Error (MSE) and Integral Time Absolute Error (ITAE).