计算机应用研究2016，Vol.33Issue(6)：1656-1661,6.DOI:10.3969/j.issn.1001-3695.2016.06.013
求解大规模优化问题的正交反向混合差分进化算法
Hybridization differential evolution algorithm of orthogonal crossover and opposition-based learning for large-scale optimization problem
摘要
Abstract
Differential evolution is simple and efficient.However,when solving the large-scale optimization problems,the performance decreases rapidly.To overcome this problem,this paper proposed a hybridization differential evolution algorithm of orthogonal crossover and opposition-based learning.In the hybrid algorithm,it used orthogonal crossover to enhance the ex-ploitation ability and adopted opposition-based learning to adjust the diversity of population.Thus it could balance the local and global search ability efficiently.It tested the new algorithm on 11 standard benchmark problems and compared with other four famous variants of differential evolution.The results show that performance of the algorithm is better than those of the com-pared algorithms in terms of accuracy and speed.Thus,it can be an efficient algorithm for large scale optimization problems.关键词
大规模优化问题/差分进化/正交交叉/反向学习Key words
large-scale optimization problems/differential evolution/orthogonal crossover/opposition-based learning分类
信息技术与安全科学引用本文复制引用
董小刚,邓长寿,谭毓澄,彭虎..求解大规模优化问题的正交反向混合差分进化算法[J].计算机应用研究,2016,33(6):1656-1661,6.基金项目
国家自然科学基金资助项目（61364025）；武汉大学软件工程国家重点实验室开放基金资助项目（SKLSE 2012-09-39）；江西省教育厅科学技术资助项目（GJJ13729，GJJ14742）；九江学院科研资助项目 （）
