Distribution based artificial fish swarm in continuous global optimization

Abstract

Distribution based artificial fish swarm (DbAFS) is a new heuristic for continuous global optimization. Based on the artificial fish swarm paradigm, the new algorithm generates trial points from the Gaussian distribution, where the mean is the midpoint between the current and the target point and the standard deviation is the difference between those two points. A local search procedure is incorporated into the algorithm aiming to improve the quality of the solutions. The performance of the proposed DbAFS is investigated using a set of small bound constrained optimization problems.