A simple multi-chromosome genetic algorithm optimization of a Proportional-plus-Derivative Fuzzy Logic Controller

Abstract

In this paper, a genetic algorithm is used to optimize the input and output fuzzy sets of a proportional-plus-derivative fuzzy logic controller (PDFLC). The center points of these sets are organized into "chromosomes," and then bred and mutated in a genetic algorithm to produce a population of offspring. The offspring are then put through a fitness algorithm to determine which of them survive to breed the next generation. This iterative process results in a solution optimized toward the definition of a "fit" design. This design method is illustrated with a numerical example.