Accelerating Gaussian bare-bones differential evolution using neighbourhood mutation

Gaussian bare-bones differential evolution (GBDE) is a new DE algorithm which employs Gaussian random sampling to generate mutant vectors. Though this method can maintain population diversity and enhance the global search ability, it may result in slow convergence rate. In this paper, we present an improved GBDE (IGBDE) algorithm by using neighbourhood mutation to accelerate the evolution. Moreover, a modified parameter control method is utilised to adjust the crossover rate (CR). To verify the performance of our approach, 13 well-known benchmark functions are tested in the experiments. Simulationresults show that IGBDE outperforms the original GBDE in terms of solution accuracy andconvergence speed. Copyright © 2013 Inderscience Enterprises Ltd.