A novel hybrid genetic algorithm for global optimization
Shuihua Wang, Lenan Wu · 2010 Sixth International Conference on Natural Computation · 2010
In order to propose a more effective function optimization method, a novel algorithm named HGPSA was proposed which integrates the powerful global search ability of GA and the excellent local search ability of PS. The experiments of 10 runs on three test functions (Powell function, Rosenbrock function, and Schaffer function) demonstrate that the proposed algorithm is superior to both GA and PS with respect to the successful rate. Therefore, the proposed algorithm is valid.