Fitness-Distance-Ratio Based Particle Swarm Optiniization

This paper presents a modification of tlte panicle swarm optimization algorithm (PSO) intended to combat the problem ofpremature convergence observed in many applications of PSO. The proposed new algorithm moves particles towards neorby particles of higher fitness, instead of amacting each panicle towards just the best position discovered so far by any particle. This is accomplished by using the ratio of the relative fitness and the distance of other particles to determine the direction in which each component of thepa&leposition needs to be changed. The resulting algorithm (FDR-PSO) is shown to perform significantly better than the original PSO algorithm and some of its variants, on many different benchmark optimization problems. Empiricol examination of the evolution of Ihepam’cles demonstrates tbot the convergence of the algorithm does not occur a1 an early phase of panicle evolution, unlike PSO. Avoiding premature convergence allows FDR-PSO to continue search for global optima in dificult multimodal optimization problems.