Fuzzy Control Systems with Reduced Parametric Sensitivity Design Based on Hybrid Grey Wolf Optimizer–Particle Swarm Optimization
Radu‐Codruţ David, Radu‐Emil Precup, Ștefan Preitl, Alexandra-Iulia Szedlak-Stinean, Raul‐Cristian Roman, Emil M. Petriu · 2020
This paper proposes an optimal tuning method for Takagi-Sugeno-Kang Proportional-Integral fuzzy controllers (TSK PI-FCs) based on a novel hybridization of Grey Wolf Optimizer (GWO) and Particle Swarm Optimization (PSO) algorithms referred to as hybrid GWO-PSO algorithm. The optimization problem defined for servo system processes controlled by TSK PI-FCs is solved by applying the hybrid form of GWO-PSO in the minimization of an objective function that depends on the output sensitivity function of the sensitivity model. The sensitivity analysis applied to these fuzzy control systems produces sensitivity models regarding the parametric variations of the dynamic processes subjected to control (i.e., the servo system). Solving the optimization problem implies the minimization of the objective function by means of the hybrid GWO-PSO algorithm. This hybrid variation of the two nature-inspired algorithms allows an increased control over the exploitation phase by inserting PSO search process features along with the exploration capabilities of GWO. The design method is validated using an experimental setup based on a nonlinear servo system.