Improved Salp Swarm Algorithm with mutation schemes for solving global optimization and engineering problems

Abstract

Salp Swarm Algorithm (SSA) is a recent metaheuristic algorithm developed from the inspiration of salps’ swarming behavior and characterized by a simple search mechanism with few handling parameters. However, in solving complex optimization problems, the SSA may suffer from the slow convergence rate and a trend of falling into sub-optimal solutions. To overcome these shortcomings, in this study, versions of the SSA by employing Gaussian, Cauchy, and levy-flight mutation schemes are proposed. The Gaussian mutation is used to enhance neighborhood-informed ability. The Cauchy mutation is used to generate large steps of mutation to increase the global search ability. The levy-flight mutation is used to increase the randomness of salps during the search. These versions are tested on 23 standard benchmark problems using statistical and convergence curves investigations, and the best-performed optimizer is compared with some other state-of-the-art algorithms. The experiments demonstrate the impact of mutation schemes, especially Gaussian mutation, in boosting the exploitation and exploration abilities.