An Improved Firefly Algorithm Based on An Attraction Switch

Firefly algorithm(FA) is a new swarm intelligence optimization algorithm. For any two firefly individuals that are far apart.the attraction module of the FA at this moment loses its attraction during each iteration. This is a poor convergence of the F A,and it is easy to fall into a local optimum. In order to overcome the shortcoming, based on the standard firefly algorithm and LF-FA, the paper proposes a improved firefly algorithm(AS-IFA) that switches attractive modules. The AS-IF A has a strong attraction for any firefly that is far away. At the same time.it also has obvious attraction to the closer fireflies. The experimental results show that compared with the standard firefly algorithm and its variant algorithm, the AS-IFA has the best convergence behavior, and its global exploration efficiency is the best.