Clustering Multi-objective Wolf Pack Algorithm for Constrained Optimization Problems

In order to solve the problems of lack of diversity, difficulty in getting rid of local optimization and less consideration of constraint conditions in the optimization process of multi-objective wolf pack algorithm, a clustering multi-objective wolf pack algorithm for constrained optimization problems (CMOWPA-C) was proposed. Combining the self-adaptive penalty model and the adaptive tradeoff model, a new method is proposed to transform the constrained problem into the unconstrained problem. Random disturbance factor was introduced to increase the moving step of the population and prevent the population from falling into local optimization. The K-means clustering algorithm is used to group the population, and the population is divided into different clusters according to the distance between the population and the cluster center, so as to ensure that there are individuals around each cluster center to be associated with it, so as to increase the diversity of the population. In order to verify the performance of the algorithm, 9 new algorithms are compared on the benchmark problem, and 9 constrained multi-objective evolutionary algorithms are compared on the practical constraint problem. The results show that the diversity of CMOWPA-C is significantly increased and can effectively avoid falling into local optimality.