Diversity enhanced particle swarm optimization with neighborhood search

Particle Swarm Optimization (PSO) has shown an effective performance for solving variant benchmark and real-world optimization problems. However, it suffers from premature convergence because of quick losing of diversity. In order to enhance its performance, this paper proposes a hybrid PSO algorithm, called DNSPSO, which employs a diversity enhancing mechanism and neighborhood search strategies to achieve a trade-off between exploration and exploitation abilities. A comprehensive experimental study is conducted on a set of benchmark functions, including rotated multimodal and shifted high-dimensional problems. Comparison results show that DNSPSO obtains a promising performance on the majority of the test problems.

► PSO tends to suffer from premature convergence because of the loss of diversity.
► The diversity enhanced mechanism could slow down the diversity of swarm.
► The neighborhood search strategy could accelerate the convergence rate.
► The proposed approach achieves a balance between the exploration and exploitation.