计算机应用

• 数据挖掘 • 上一篇    下一篇

点和边有容量约束的网络最大流新算法

厍向阳 罗晓霞   

  1. 西安科技大学
  • 收稿日期:2007-07-19 修回日期:2007-10-08 发布日期:2008-01-01 出版日期:2008-01-01
  • 通讯作者: 厍向阳

New algorithm for maximum flow algorithm in network with both node and edge capacity confined

<a href="http://www.joca.cn/EN/article/advancedSearchResult.do?searchSQL=(((xiangyang SHE[Author]) AND 1[Journal]) AND year[Order])" target="_blank">xiangyang SHE</a>   

  • Received:2007-07-19 Revised:2007-10-08 Online:2008-01-01 Published:2008-01-01
  • Contact: xiangyang SHE

摘要: 针对目前网络最大流算法存在的问题,研究一种适应性更广的新算法。定义了有向路径和残量网络的概念,依据可行流分解定理,引入人工智能中搜索的方法,以邻接矩阵为网络数据存储结构,提出条件约束下的网络最大流新算法。最后,通过实例进行了算法测试和比较。算法测试表明:点和边有容量约束的网络最大流新算法是完全可行和有效的。

关键词: 邻接矩阵, 网络最大流, 可行流, 容量约束, 残量网络

Abstract: Concerning the problems in flow algorithm on network, a new algorithm with strong applicability was studied. The orientation path and residual network were defined, the new maximum flow algorithm in network with both node and edge capacity confined was put forward based on feasible flow decompose theorem and by way of the search in artificial intelligence. In the end, the algorithm was validated and compared by examples. Algorithm testing shows: The new maximum flow algorithm in network with both node and edge capacity confined is completely feasible and availability effective.

Key words: adjacency matrix, maximum flow in network, feasible flow, confined capacity, residual network