摘要
为实时诊断给水管网故障,提出了一种基于支持向量机(SVM)的宏观关系模型,用以建立实时监测故障前后给水管网3个节点的水压变化与管网中其他所有未监测节点水压变化的非线性关系,通过对水压变化和等值线分析,可以快速和准确地诊断出故障位置和故障程度。最后,运用给水管网局部破坏状态下水力分析的算法求解一个小型给水管网。
A Support vector machine (SVM) based macro relationship model was established for realtime diagnostic of municipal water distribution networks. The water pressure deficits at three nodes in the network were monitored realtimely, and the nonlinear relationships of water pressure examinations of these three points and all the nomonitored points before and after troubles had been recognized. By the way of water pressure deficit monitoring and the isogram analysis, the location and the severity of the happening troubles could be found out quickly and exactly. Also a small network was calculated by hydraulic analysis of local break downed network.
引文
1赵胜跃,赵新华,张丽,等.供水SCADA系统的研究与建立.中国给水排水,2002,18(1):50~53
2Liggett J A,Chen L C.Inverse transient analysis in pipe networks.Journal of Hydraulic Engineering,ASCE,1994,120:934~955
3梁建文,肖笛,赵新华,等.给水管网故障实时诊断方法.水利学报,2001,12:40~47
4邓乃扬,田英杰.数据挖掘中的新方法——支持向量机.北京:科学出版社,2004
5俞亭超,柳景青,张土乔.改进支持向量机的管网状态模型.浙江大学学报(工学版),2005,39(6):858~862
6严煦世,刘遂庆.给水排水管网系统.北京:中国建筑工业出版社,2005
7周建华,赵洪宾,胡文茜.城市供水管网事故时的水力分析.给水排水,2004,24(8):104~107
8韩阳.城市地下管网系统的地震可靠性研究:[学位论文].大连:大连理工大学,2002
9梁建文,肖笛,张宏伟,等.给水管网故障实时诊断方法(Ⅱ).自然科学进展,2002,12(9):970~974