The Community for Technology Leaders
2015 11th International Conference on Computational Intelligence and Security (CIS) (2015)
Shenzhen, China
Dec. 19, 2015 to Dec. 20, 2015
ISBN: 978-1-4673-8659-3
pp: 274-277
ABSTRACT
Adaptive differential evolution algorithm based on gradient and polar coordinates search strategies (ADE) is proposed in this paper. In order to improve the precision of solutions, gradient and polar coordinates search strategies are introduced. Since the gradient search strategy generates offsprings using the derivative definition, it will accelerate the convergence speed. Polar coordinates search strategy can help the algorithm jump out of the local optimization and avoid continuously searching in wrong direction. The simulation results show that the proposed algorithm has better results compare to SaDE, NSDE and CMAES for benchmark functions 1-14 in CEC2005.
INDEX TERMS
Search problems, Optimization, Linear programming, Sociology, Statistics, Convergence, Algorithm design and analysis
CITATION

J. Yang, J. Wei and J. Liu, "Adaptive Differential Evolution Algorithm Based on Gradient and Polar Coordinates Search Strategies," 2015 11th International Conference on Computational Intelligence and Security (CIS), Shenzhen, China, 2015, pp. 274-277.
doi:10.1109/CIS.2015.74
93 ms
(Ver 3.3 (11022016))