摘要
采用遗传算法建立了公路路政管理站点规划选址优化模型,研究了路网中规划选址优化的多变量算法不易收敛问题,应用受限P 中心问题以减少管理站的备择点数目,从而减少遗传算法中染色体的基因长度,并按不同策略进行初始群体和种群的选择,加快收敛速度,从而提高算法的运算效率。针对高速公路巡视路线优化中最大巡路长度问题,采用贪婪算法和邻域搜索算法的结合求解最佳巡视路线,给出了问题的多目标函数优化模型和算法。
In this paper, authors have presented the genetic algorithm (GA) in order to get the optimal model of planning location to highway management stations and inspection route and have studied the hard-limitation of multi-variable algorithm in the planning location to management stations in the highway net. Theory of limited P-center is applied to reduce the number of stations, which reduces the genetic length of the chromosome of GA. During the calculation, different strategic choices of initial groups and seed groups improve the limiting speed in order to improve the calculation efficiency of the algorithm. As to the longest way of freeway inspection route, which is furthest and its station number is small as soon as possible, the connection of greedy algorithm and local search algorithm is used to assign the optimal inspection route. And authors present the optimal model and algorithm of multi-objection function.
出处
《中国公路学报》
EI
CAS
CSCD
北大核心
2005年第2期69-73,共5页
China Journal of Highway and Transport
基金
教育部博士学科点基金项目(20030056007)
关键词
交通工程
优化模型
遗传算法
规划选址
巡视路线
traffic engineering
optimization model
genetic algorithm
programming location
inspection route
