Soft Computing

, Volume 21, Issue 3, pp 667–685

Interval type-2 fuzzy logic for dynamic parameter adaptation in the bat algorithm

  • Jonathan Perez
  • Fevrier Valdez
  • Oscar Castillo
  • Patricia Melin
  • Claudia Gonzalez
  • Gabriela Martinez
Focus

DOI: 10.1007/s00500-016-2469-3

Cite this article as:
Perez, J., Valdez, F., Castillo, O. et al. Soft Comput (2017) 21: 667. doi:10.1007/s00500-016-2469-3
  • 147 Downloads

Abstract

We describe in this paper a proposed enhancement of the bat algorithm (BA) using interval type-2 fuzzy logic for dynamically adapting the BA parameters. The BA is a metaheuristic algorithm inspired by the behavior of micro bats that use the echolocation feature for hunting their prey, and this algorithm has been recently applied to different optimization problems obtaining good results. We propose a new method for dynamic parameter adaptation in the BA using interval type-2 fuzzy logic, where an especially design fuzzy system is responsible for determining the optimal values for the parameters of the algorithm. Simulations results on a set of benchmark mathematical functions with the interval type-2 fuzzy bat algorithm outperform the traditional bat algorithm and a type-1 fuzzy variant of BA. The proposed integration of the type-2 fuzzy system into the BA has the goal of improving the performance of BA for the future applicability of the algorithm in more complex optimization problems where higher levels of uncertainty need to be handled, like in the optimization of fuzzy controllers.

Keywords

Optimization Dynamic parameter adaptation Bat algorithm Type-1 fuzzy logic Type-2 fuzzy logic Benchmark mathematical functions 

Copyright information

© Springer-Verlag Berlin Heidelberg 2016

Authors and Affiliations

  • Jonathan Perez
    • 1
  • Fevrier Valdez
    • 1
  • Oscar Castillo
    • 1
  • Patricia Melin
    • 1
  • Claudia Gonzalez
    • 1
  • Gabriela Martinez
    • 1
  1. 1.Tijuana Institute of TechnologyTijuanaMexico

Personalised recommendations