摘要

The paper proposes a novel neighborhood search operation of particle swarm in the numerical objective space (eg. Rn), and employs the clonal selection strategy which leads the particle swarm to find the optima of the objective space using the neighborhood search operation. So, a novel particle swarm algorithm based on the clonel selection strategy (NPSA/CS) is proposed. In the test experiment, 6 unconstrained benchmark functions are used to demonstrate the performance of NPSA/CS, and compared with CPSO, PSO, GA, DE. The results show that NPSA/CS can find the optimal or close-to-optimal solutions of those benchmark functions efficiently.

全文