| 不确定性选址问题探讨 |
| |
| 引用本文： | 王亚.不确定性选址问题探讨[J].测绘科学,2003,28(3):46-48,51. |
| |
| 作者姓名： | 王亚 |
| |
| 作者单位： | 武汉大学遥感信息工程学院,,武汉,430079 |
| |
| 摘 要： | 阐述了GIS网络分析中不确定性选址问题的基本模型及特性。从问题的定义可知其为NP完备类问题。推导了最优解在紧条件的下界算法,并结合广义Powell算法及遗传算法,提出了不确定性选址问题的混合遗传算法,实验证明,在最优解的品质和收敛速度上都达到了比较好的效果。同时,实验的结果从另一个角度证明,如果兼顾收敛速度和解的品质这两个指标,单纯的遗传算法未必比其他搜索算法更优越,采用一些局部搜索性能较好的算法结合遗传算法,可以从两方面改善求解效果。
|
| 关 键 词： | GIS 不确定性选址 广义Powell算法 遗传算法 |
| 文章编号： | 1009-2307(2003)03-0046-04 |
Discussion for uncertain location problems |
| |
| WANG Ya.Discussion for uncertain location problems[J].Science of Surveying and Mapping,2003,28(3):46-48,51. |
| |
| Authors: | WANG Ya |
| |
| Abstract: | This paper presents an optimal algorithm, PGA algorithm based on the genetic algorithm, to deal with uncertain location problems. Firstly, a basic mathematical model of location problem is designed, and its simplest case is discussed. Some key lemmas about the optimal solution are deduced according to the model. Then this paper gives an algorithm to obtain the minimum of the solution. Secondly, it describes the processes of generalized Powell algorithm, which is introduced to the PGA algorithm to improve its efficiency in partial searching. The genetic algorithm is used to be the basic model of the PGA algorithm, generalized Powell algorithm is applied to control the number of generations for the PGA algorithm, and the algorithm to obtain the minimum of the solution is adopted to test the quality of the feasible solutions. The result of an experiment shows that the PGA algorithm is better than a pure genetic algorithm in regard to the quality of the solution and the speed of searching. |
| |
| Keywords: | GIS uncertain location problems generalized Powell algorithm genetic algorithm |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
