Penalty function method is one of the most popular used Constraint Handling Techniques for Evolutionary Algorithms (EAs) solution selecting, whose performance is mainly determined by penalty parameters. This paper tries to study the penalty parameter from the aspect of problem characteristics, i.e., to construct a corresponding relationship between the problems and the penalty parameters. The experimental results confirm the relationship, which provides valuable reference for future algorithm design.
Abstract. Penalty function methods have been widely used for handling constraints, but it's still a challenge about how to set the penalty parameter effectively though many related methods have been proposed. In this paper, the penalty parameter is firstly analyzed systematically by introducing four rules. Based on this analysis, a new Dynamic Penalty Function (DyPF) is proposed by adjusting penalty parameter in three different situations during the evolution (i.e., the infeasible situation, the semi-feasible situation, and the feasible situation). The experiments are designed to verify the effectiveness of our newly proposed DyPF. The results show that DyPF presents a better overall performance than other five dynamic or adaptive state-of-the-art methods in the community of constrained evolutionary optimization.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.