A comparative study of population-based optimizations for tuning PID parameters

Abstract

The tuning and optimization of Proportional-Integral-Derivative (PID) parameters have always been a complicated but important issue in the field of automatic control. The recent optimization design methods are frequently difficult to consider the system requirements for speed, consistency and robustness. In this paper, methods of PID parameters using population-based heuristic optimization are presented. Some quantitative and qualitative comparisons are given together with their computational time. Simulations with Matlab have showed that Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) are better based on the performance index than that of the traditional Ziegler-Nichols (Z-N) method, and are methods which have superior practical value of the PID parameter tuning and optimization.