عنوان

A Simple Hybrid Particle Swarm Optimization.

9537619117

9789537619114

b600293

A Simple Hybrid Particle Swarm Optimization.

[Book]

Wei-Chang Yeh

INTECH Open Access Publisher

2008

A successful evolutionary algorithm is one with a proper balance between exploration (searching for good solutions), and exploitation (refining the solutions by combining information gathered during the exploration phase). In this study, a new hybrid version of PSO called HPSO is proposed. The HPSO constitutes a vector based PSO method with the linearly varying inertia weight, along with a local search. A novel, simpler, and efficient mechanism is employed to move the gBest to its next position in the proposed HPSO. The HPSO combines the population-based evolutionary searching ability of PSO and local searching behavior to effciently balance the exploration and exploitation abilities. The result obtained by HPSO has been compared with those obtained from traditional simple PSO (SPSO) and improved PSO (IPSO) proposed recently. Computational results show that the proposed HPSO shows an enhancement in searching efficiency and improve the searching quality. In summary, the results presented in this work are encouraging and promising for the application of the proposed HPSO to other complex problems. Further analysis is necessary to see how other soft computing method (e.g., the genetic algorithm, the taboo search, etc.) react to local searches for future researchers who may want to develop their own heuristics and to make further improvements. Our research is still very active and under progress, and it opens the avenues for future efforts in this directions such as: how to adjust parameters, increase success rates, reduce running times, using other local search, and the aggregation of different and new concepts to PSO.

Open Access Collection.

Wei-Chang Yeh

Wei-Chang Yeh

[Book]

Y