Optimization of PID controller with first order noise filter

Abstract

Measurement noise is one of the most widely and frequently experienced problems of an industrial working environment. The effects of measurement noise can be reduced by filtering the measurement signal. The aim of the current work is to explore (a) the optimum location of filter in the control loop and (b) the controller tuning in presence of measurement noise. In the present work, two locations are considered for placing the first order noise filter. One, in the feedback path after the sensor and, other cascaded with the derivative term. The controller and filter parameters were tuned using Particle Swarm Optimization (PSO) technique for a combined performance criterion composed of Maximum Overshoot and Integral Absolute Error (IAE). For studying the effectiveness of the proposed scheme, two simulations and a hardware based experiment on a speed control of a DC motor are considered. Based on these investigations, it can be concluded that the filter located in feedback path gives better performance as compared to the filter cascaded with the derivative term in a noisy environment.