群体区域搜索算法.pdfVIP

第28 卷 第8 期 控 制 与 决 策 2013 年 8 月 Vol. 28 No. 8 Control and Decision Aug. 2013 文章编号: 1001-0920 (2013) 08-1235-07 群体区域搜索算法 刘昌军, 卫军胡, 王 虹, 高一星, 孙国基 (西安交通大学机械制造系统工程国家重点实验室，西安710049) 摘 要: 借鉴自然界群居生物的搜索行为模式, 提出一种群体区域搜索算法. 该算法在优化过程中逐步收缩个体搜 索半径并进行适度随机调整, 引入巡游追随机制, 以一种简单而自然的方式有效地实现了算法广域探索能力与局部 开发能力之间的平衡. 算法结构简单、易实现, 易与其他算法相结合. 通过6 个典型测试函数的实验结果表明, 该算法 全局优化能力强、收敛精度高、稳定性好、总体性能优, 适用于复杂函数优化问题的处理. 关键词: 群体智能；进化计算；巡游追随机制；群体区域搜索算法；全局优化 中图分类号: TP13 ；TP18 文献标志码: A Group area search for optima LIU Chang-jun, WEI Jun-hu, WANG Hong, GAO Yi-xing, SUN Guo-ji (State Key Laboratory for Manufacturing Systems Engineering，Xi’an Jiaotong University ，Xi’an 710049 ，China. Correspondent ：LIU Chang-jun ，E-mail ：chjliu@) Abstract: A novel swarm intelligence optimization algorithm, group area search(GAS), is proposed, which mimics the searching behavior patterns of gregarious creatures. In the algorithm, the search radius of each member is gradually shrunk and moderately adjusted in the optimization process. Coupled with a cruising-following mechanism, GAS can achieve a good balance between global exploration and local exploitation in a natural way. With the characteristics of robustness and parallelism in nature, GAS is simple to be implemented and can easily be combined with other optimization techniques. The test results on six benchmark functions show that the proposed algorithm has excellent global optimization capability, high convergence accuracy and stability, which outperforms the other eight nature-inspired algorithms in general and can cope with heterogeneous complicated function optimization problems. Key words: swarm intelligence；evolutionary comp