On the Cooperation Between Evolutionary Algorithms and Constraint Handling Techniques

During the past few decades, many Evolutionary Algorithms (EAs) together with the Constraint Handling Techniques (CHTs) have been developed to solve the constrained optimization problems (COPs). To obtain competitive performance, an effective CHT needs to be in conjunction with an efficient EA. In the previous paper, how the Differential Evolution influence the relationship between problems and penalty parameters was studied. In this paper, further study on how much can be improved through good evolutionary algorithms, or whether a good enough EA can make up the shortcoming of a simple CHT, and which factors are related will be the focus. Four different EAs are taken as an example, and Deb’s feasibility-based rule is taken as the CHT for its simplicity. Experimental results show that better performance in EAs is not necessarily the reason for the improved performance of constrained optimization evolutionary algorithms (COEAs), and the key point is to find the shortcoming of the CHT and improve the shortcoming in the corresponding revision of EA.