An Improved Shuffled Frog Leaping Algorithm to Optimize the Parameters of PID

Abstract

As to the nonlinear, time-varying, and large delay control system, the conventional proportional-integral-derivative (PID) control effect is not ideal; thus, PID parameters based on shuffled frog leaping algorithm (SFLA) are presented. Concerning the problems of the SFLA such as slow convergence rate and local optimality, we propose the improved shuffled frog leaping algorithm (ISFLA).This algorithm uses reverse selection mechanism in the evolutionary process so as to keep the diversity of population and improve the ability of evolution algorithm. Introducing the linear decreasing adaptive inertia weight to correct the poor frog update strategy can balance the global search and local search. The simulation results of experiments on the two classical control system show that the ISFLA, when compared with SFLA and particle swarm optimization (PSO), can balance the global search and local search with a smaller number of iterations and stronger optimization ability and is more suitable for the tuning of the PID parameters optimization.