AFSAOCP: A novel artificial fish swarm optimization algorithm aided by ocean current power

Abstract

Artificial Fish Swarm is a kind of swarm intelligence algorithm, which focuses on the behavior of individual fishes and information interactions among them during foraging and preying on something in real environment. A novel Artificial Fish Swarm Optimization Algorithm Aided by Ocean Current Power (abbreviation for AFSAOCP) is proposed, which assumes the ocean current always causes certain influence on the fishes’ activity speed. Firstly, the computing model of ocean current is developed and constructed. Then the influence level of the ocean current on fishes is analyzed. If fishes are swimming along ocean current direction, ocean current will drive fishes’ speed increment, which is called positive influence; if fishes are swimming against ocean current, the current will hinder the fishes’ speed, which is called negative influence. In addition, fishes’ speed is not influenced by the ocean current, which is called merits offset faults. To sum this up, fishes in each group have different speed range, respectively. Grouping strategies can not only increase species diversity, but it can also make the algorithm escape from local optimal value in the iteration process. The proposed variant, AFSAOCP, is examined on several widely used benchmarked functions, and the experimental results show that the proposed AFSAOCP algorithm improves the existing performance of other algorithms when dealing with the different dimension and multimodal problems.