Differential Evolution with Division of Labor

摘要：

The advantages and disadvantages of rand/1 and best/1 mutation strategies are pointed out respectively.A new differential evolution algorithm with division of labor is proposed.In the new algorithm,exploitation and exploration tasks are assigned to all individuals according to their fitness during evolution.Better individuals choose best/1 to be assigned exploitation tasks and the other individuals choose rand/1 to be charged with exploration tasks.The balance between local search and global search is achieved by division of labor,consequently the algorithm’s performance is improved.Experiments on benchmark functions show that the new algorithm is simple and easy to be implemented and performs better than DE/rand/1 /bin and DE/best/1/bin.

关键词： differential evolution division of labor exploration exploitation.

作者: Li-qiang Jiang Guang-bin Liu Zhi-yong Yu

作者单位: Department of 303,Xian Research Inst.of High-tech Hongqing Town,710025 ShanXi,China Research Inst.of Missile,Xian Research Inst.of High-tech Hongqing Town,710025 ShanXi,China

会议类型: 国际会议

会议名称: International Conference on Modelling,Identification and Control(模拟、鉴定、控制国际会议)

会议地点: 上海

会议语种:英文

在线出版日期: 2008-06-29（万方平台首次上网日期，不代表论文的发表时间）