Optimization of PI Controller for Coupled Tank Systems Using Genetic Algorithm

In present work, Genetic Algorithm (GA) is utilized to construct a Proportional-Integral (PI) controller for a Coupled Tank System (CTS). Its effectiveness is evaluated and compared with a conventional PI controller. The performance of the proposed GA-PI controller is also benchmarked against previous research. The CTS, a standard process control system, is employed to regulate liquid levels in tanks. The controller based on GA approach offers a viable alternative to traditional PI controller tuning techniques that may not be appropriate for this task. Here, PI controller parameters have been optimized using the GA to obtain the desired control performance. Comparative analysis is conducted between the traditional PI controller and the GA-PI controller. The steady-state response analysis reveals that the PI controller tuned via GA surpasses the performance of the traditional PI controller. Additionally, the research findings indicate that utilizing a GA for PI controller tuning results in superior control performance compared to conventional methods. Additionally, it is noted that the suggested GA-based PI controller demonstrates superior performance compared to previous studies in terms of rise time, settling time, and overshoot, particularly with respect to minimizing the Integral of Absolute Error.