Applied Mathematics & Information Sciences

Author Country (or Countries)



Standard particle swarm optimization algorithm has two drawbacks in engineering application when particles dimension was high; first is premature convergent and second is low convergent speed. Counting these drawbacks we proposed a novel algorithm with high convergent speed in high dimensional search place based on particle health degree, and we provided particle health degree concept and computation method. The algorithm through dynamic monitoring particle health when the particle health value was lower than given threshold value, we separately use mutation operation on these particles. This method can not only protect the health particles keep searching the optimum value but also therapy the ill-health particles and enhance the ability of searching optimum value and jumping out the local optimum. We used many benchmark functions to test our algorithm, and compete with Standard PSO algorithm and nonlinear inertia weight variation (WPSO). Test results show that the algorithm we proposed has higher convergent speed and searching efficiency.

Suggested Reviewers


Digital Object Identifier (DOI)