Sigmoidal Salp Swarm Algorithm
Sandeep Kumar, Rajani Kumari, Anand Nayyar · 2020
The salp swarm algorithm (SSA) imitates the swarming behavior of salps while foraging in the deep ocean. This algorithm is equally good for single as well as multiobjective optimization problems. The position update process of leader salp depends on perturbation. Currently SSA used fixed value to decide perturbation that lacks diversity in solutions. In this research paper, we propose a new variant of SSA by introducing a new parameter for perturbation. The new parameter is inspired by sigmoidal decreasing function. Since it is nonlinear function; it gives improved results for complex optimization problems. The anticipated approach is named as sigmoidal salp swarm algorithm (S3A). The performance of S3A is tested over a set of fifteen benchmark problems and results proves that it improves results up to 60% in terms of mean value.