An Enhanced Swarm Method Based on the Locust Search Algorithm

Abstract

In the evolutionary methods, the optimal balance of exploration and exploitation performance in search strategies improve efficiency to found the best solution. In this chapter, an improved swarm optimization technique called Locust search II (LS-II) based on the desert locust swarm behavior, adapted to an emulation of a group of locusts which interacts to each other based on the biological laws of the cooperative swarm is proposed for solving global optimization problems. Such methodology combines a technique of exploration which avoids premature convergence effectively and a technique of exploitation able to intensify the global solutions. The proposed LS-II method was tested over several well-known benchmark test functions and engineering optimization problems and its performance was further compared against those of other state-of the-art methods such as Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC), Bat Algorithm (BA), Differential Evolution (DE), Harmony Search (HS) and the original Locust Search (LS). Our experimental results show LS-II to be superior to all other compared methods in terms of solution quality and as such proves to be an excellent alternative to handle complex optimization problems.