摘要
针对长江干线LNG动力船加注站的选址问题,将改进(引入遗传变异)的蚁群算法和聚类分析进行结合,聚类分析可以解决选址问题中的不确定性和模糊性,而引入遗传变异的蚁群算法又可以有效地解决传统聚类分析因初始点选取不当易陷入局部最优解的问题。实例验证表明:基于遗传变异的蚁群聚类与普通蚁群聚类分析相比缩短了迭代次数,减少了计算量,与传统聚类分析相比更接近于全局最优。
Aiming at the location of LNG powered ship bunkering station in the main line of the Yangtze River,the improved ant colony algorithm（ considering genetic variation） and the clustering analysis were combined. The clustering analysis can solve the problem of uncertainty and ambiguity in location selection,and the ant colony algorithm with genetic variation can effectively solve the problem of local optimal solution,due to the improper selection of the initial point in the traditional clustering analysis. The case studies show that comparing with the general ant colony clustering analysis,the ant colony clustering based on genetic variation shortens the number of iterations and reduces the amount of calculation; it is much closer to the global optimum,comparing with the traditional clustering analysis.
作者
杨勇生
周亚民
许波桅
YANG Yongsheng ZHOU Yamin XU Bowei(Logistics Research Centre,Institute of Logistics Science ＆ Engineering, Shanghai Maritime University, Shanghai 201306, P. R. China)
出处
《重庆交通大学学报（自然科学版）》
CAS
北大核心
2016年第6期141-147,共7页
Journal of Chongqing Jiaotong University(Natural Science)
基金
高等学校博士学科点专项科研基金项目(20133121110005)
上海市科委科技创新行动计划项目(14170501500)
上海市科委自然科学基金项目(15ZR1420200)
教育部人文社会科学研究青年基金项目(15YJC630145
15YJC630059)
上海海事大学研究生创新基金项目(YXR2015038)
关键词
交通运输工程
LNG加注站
长江干线
遗传变异
蚁群聚类
选址问题
traffic and transportation engineering
LNG bunkering
main line of the Yangtze River
genetic variation
ant colony clustering
location problem
