Particle Swarm Optimization 论文

摘要

Abstract—Nature inspired algorithms implement successful optimization and adaptation strategies observed in the nature. Various bio-inspired algorithms mimic the behavioural patterns of plants, animals, their communities and their evolution. Surpris-ingly, the behavioural patterns and survival strategies of protozoa, one of the most prevalent and successful species on Earth, did not receive significant attention from the bio-inspired computing community until present time. This study proposes a new variant of Particle Swarm Optimization incorporating behaviour inspired by protozoa and evaluates the performance of such an algorithm on a set of well known test functions. Index Terms—Bio-inspired algorithms; particle swarm opti-mization; protozoic behaviour I.