PID Flow-Level Control Tuned by Genetic Algorithm and Harmony Search Algorithm

Abstract

PID controller is widely used in industries because of its simplicity and robustness. A Simple approach to improve regulatory control performance is to design an advanced controller based on specific desired criteria. In this paper, a theoretical closed loop model of the Level-flow control trainer is developed using standard TF estimation method. The tuning of Single loop PID controller is done in MATLAB by Genetic Algorithm & Harmony Search Algorithm and the latter one gives the better performance. So, HSA is chosen for the advanced controller tuning. The values obtained from HSA are used for getting the real-time response of the system for cascade control and feed forward control strategy. A faster response in case of feed forward and a better disturbance rejection in case of cascade controller are found. The frequency response (Bode Plot) of single variable process and multivariable process are compared and comment on relative stability has been made based on the Gain margin obtained from the frequency response analysis.