Signal Synthesis Model Reference Adaptive Controller with Genetic Algorithm for a Control of Chemical Tank Reactor

Modelling and controlling of Continuous Stirred Tank Reactor (CSTR) is one of the significant problems in the process industry. Chemical reactions inside the CSTR depend on the reference value of the temperature. Design and implementation of suitable control device for such system is a challenge for researchers. This paper proposes the Model Reference Adaptive Control (MRAC) based control strategy as a solution to control problem of CSTR. An enhancement of Signal Synthesis MRAC scheme has been proposed in this study to improve the steady-state and transient-state performance of CSTR. Genetic Algorithm (GA) based controller parameter tuning method is employed to obtain the optimal performance of the controller. This paper presents the design and implementation of conventional Proportional-Integral-Derivative (PID) tuned with Ziegler-Nichols (ZN) tuning method, PID tuned with GA, MRAC, and GA-MRAC for CSTR. Detailed comparison based on simulation studies is also presented to show the improved transient and steady state response with GA-based improved MRAC scheme.