A Novel Self-adaptive Salp Swarm Algorithm for Dynamic Optimization Problems

摘要

Many real-world applications can be cast as dynamic optimization problems where it is required to locate and track the trajectory of the changing global optima while finding the global best solution in a dynamic and uncertain environment. In this article, we present a novel nature-inspired meta-heuristic optimizer to solve dynamic optimization problems, namely self-adaptive salp swarm algorithm. The self-adaptive parameter control technique is used with a multi-population and ageing mechanism, in which individuals have to maintain diversity during the optimization process in SA-SSA. The evaluation is conducted to examine the overall performance of SA-SSA on widely known generalized dynamic benchmark problems provided in the CEC'09 competition. Preliminary results showed that the proposed SA-SSA is promising.