Research on the hybrid chaos-coud salp swarm algorithm

Abstract

Salp swarm algorithm (SSA) is a new swarm intelligence optimization algorithm, which has the advantages of simple structure, almost no parameter setting. However, SSA also has the shortcomings of slow convergence speed in the early stage and low optimization accuracy in the later stage when searching for the optimal solution. To address the problems, this study proposes a hybrid chaos-cloud salp swarm algorithm (CC-SSA). First, to accelerate the convergence speed, a positive normal cloud generator is used to perform local search on the superior salp individuals. Second, to enhance the diversity of CC-SSA and avoid it from falling into local optimum, the chaotic map is used to perform global search on the inferior salp individuals by adding global perturbation. Third, to control the execution ratio of global and local search, the mixed control parameter (MLMS) and population allocation coefficient (r) are introduced to organically combine three algorithms, SSA, chaotic map, and cloud model. Finally, to evaluate the performance of the proposed algorithm, it is compared with other 9 conventional algorithms on 12 classic functions. The experimental results show that CC-SSA has an average accuracy rate of 92.92 %, which ranks first. In addition, the average iteration of CC-SSA scores the third. Therefore, compared to conventional optimization algorithms, CC-SSA has better performance in terms of execution time and optimize accuracy.