ABSTRACT
This paper presents an analysis of the sensitivity function and complementary sensitivity function for stock consistency process models using Particle Swarm Optimization-(PSO) based conventional Proportional-Integral (PI) and Proportional-Integral-Derivative (PID) controllers. The analysis of these functions is required to test the effect of disturbances, noise, and variation in process dynamics on the process output, i.e., to check the robustness of the controllers. The PI controller has been designed using a fuzzy login-based PSO algorithm. The objective is to determine the optimal controller tuning parameters using a proposed algorithm such that the effect of process variations and disturbances is minimum on the process output. The peak values of sensitivity and complementary sensitivity functions have been obtained through MATLAB along with the closed-loop bandwidth of the proposed control system. Also, the time taken by proposed controllers to suppress the effect of disturbance and noise has also been evaluated. The comparison has been done between PSO–PI and PID controllers on the basis of the above-mentioned criterion. From the comparative analysis, it has been observed that the PSO–PI controller exhibits better robustness as compared to the PSO–PID controller.
