A Random Spatial lbest PSO-Based Hybrid Strategy for Designing Adaptive Fuzzy Controllers for a Class of Nonlinear Systems

In this paper, a new variant of particle swarm optimization (PSO), called random spatial lbest PSO model, is proposed and implemented for designing a newly devised stable adaptive hybrid fuzzy controller. The newly developed concurrent hybrid strategy for designing fuzzy controllers utilizes the conventional Lyapunov theory and the proposed PSO-based stochastic approach. The objective is to design a self-adaptive fuzzy controller online, optimizing both its structures and free parameters such that the designed controller can guarantee the desired stability and simultaneously provide satisfactory transients performance. The global version and two different lbest variants of PSO schemes and the proposed random spatial lbest model of PSO are employed for three popular, challenging, and nonlinear processes, and the proposed controller emerges as the superior algorithm in terms of tracking performance overall. These results aptly demonstrate the usefulness of the proposed approach.