关键词: 樽海鞘优化算法, 统计引导, 多项式差分学习, 开启时机

Abstract: The Salp Swarm Algorithm(SSA)has better robustness and optimization capacity than the traditional swarm intelligence optimization algorithm. However, there are still some defects such as the limited global optimization ability, low execution efficiency and easily running into the local extreme value. In allusion to these problems, this paper proposes a new method of polynomial differential learning strategy to distinguish and improve the traditional linear difference method, and also designs a random population partition method to make the information transfer uniformly in the neighborhood topology. In addition, we define the global exploration operator and local development operator of polynomial difference learning, and introduce the statistical guiding coefficient A to develop different polynomial learning methods, thereby further improving the global search ability and optimization precision of the algorithm. Finally, the numerical comparison experiments and practical problem tests are made to confirm the superiority and robustness of the improved algorithm, and expands the application range of the original algorithm.

Key words: salp swarm algorithm, statistical guidance, polynomial differential learning, opening time