Skip to Main Content
In this paper, a novel constraint-handling mechanism based on multi-swarm is proposed. Different from the existing constraints handling methods, the sub-swarms are adaptively assigned to explore different constraints according to their difficulties. The new mechanism is combined in dynamic multi-swarm optimizer (DMS-PSO) for handling constrained real-parameter optimization problems and sequential quadratic programming (SQP) method is combined to improve its local search ability. The performance of the modified DMS-PSO on the set of benchmark functions provided by CEC2006  is reported.