PSO and GSA algorithms for fuzzy controller tuning with reduced process small time constant sensitivity

This paper discusses implementation aspects related to a Particle Swarm Optimization (PSO) algorithm, a Gravitational Search Algorithm (GSA) and a hybrid PSO-GSA. These evolutionary optimization algorithms are applied to the optimal tuning of Takagi-Sugeno-Kang PI-fuzzy controllers (T-S-K PI-FCs) for a class of nonlinear second-order processes with an integral component. The parameters of T-S-K PI-FCs are variables in the optimization problems with objective functions which depend on the output sensitivity function with respect to the small time constant of the process, and T-S-K PI-FCs with a reduced sensitivity with respect to the process small time constant are proposed. A comparison of PSO, GSA and PSO-GSA from algorithms' accuracy point of view is presented in the framework of a case study accompanied by digital simulation results.