Computer Engineering and Applications ›› 2017, Vol. 53 ›› Issue (3): 87-94.DOI: 10.3778/j.issn.1002-8331.1506-0284

Previous Articles     Next Articles

Clustering genetic algorithm based on complex method

QIAN Wuwen1, CHAI Junrui1, 2   

  1. 1.State Key Laboratory Base of Eco-hydraulic Engineering in Arid Area(Xi’an University of Technology), Xi’an 710048, China
    2.College of Hydraulic and Environmental Engineering, Three Gorges University, Yichang, Hubei 443002, China
  • Online:2017-02-01 Published:2017-05-11

基于复合形法的聚类遗传算法

钱武文1,柴军瑞1,2   

  1. 1.西北旱区生态水利工程国家重点实验室培育基地(西安理工大学),西安 710048
    2.三峡大学 水利与环境学院,湖北 宜昌 443002

Abstract: To solve the premature convergence of the standard genetic algorithm and improve its local convergence ability, a new clustering genetic algorithm based on the complex method is proposed in this paper. The traditional genetic algorithm is improved by using the complex method and clustering niche technique. A clustering genetic algorithm(NCGA)based on complex method is obtained. The algorithm is programmed using FORTRAN language, and its performance is tested by using three kinds of complex test functions, and the performance comparison of the Adaptive Genetic Algorithm(AGA)is carried out. It also analyzes the influence of the initial population’s quality on the performance of the algorithm. Test results show that the improvement effect of genetic algorithm is obvious. The local search ability of the genetic algorithm can be significantly enhanced by the integration of complex operation in genetic algorithm. And the global search ability of genetic algorithm is significantly enhanced after using clustering technique. The stability of the algorithm is enhanced by the addition of backward learning operation. The improved genetic algorithm is better than the traditional genetic algorithm.

Key words: genetic algorithm, niche technique, clustering analysis, early convergence

摘要: 针对标准遗传算法的未成熟收敛问题和局部收敛能力不佳等情况,提出一种基于复合形法的聚类遗传算法。通过使用复合形法结合聚类小生境技术对传统的遗传算法进行改进,得到基于复合形法的自适应聚类遗传算法(NCGA)。该算法使用FORTRAN语言进行编程,通过使用三种复杂的测试函数对其性能进行测试,并与自适应遗传算法(AGA)进行了性能比较,还分析了初始种群的优劣对算法性能的影响。测试结果表明:对于遗传算法的改进效果明显,在遗传算法中融入复合形操作能明显增强遗传算法的局部搜索能力,且聚类技术使得遗传算法的全局搜索能力得到显著增强,反向学习操作的添加能增强算法的稳定性。改进后的遗传算法的性能明显好于传统的遗传算法。

关键词: 遗传算法, 小生境技术, 聚类分析, 早熟收敛