摘要
在深入剖析已有大型供水管网漏损成因的基础上,提出了采用BP神经网络深度学习的方法预测漏损点位置,构建了供水管网漏损模拟模型,通过管网发生漏损时5个位置的流量变化和17个位置的压力监测点变化的相关性分析,利用人工神经网络深度学习来诊断漏损点所在管网中位置,并以实验室自搭建的小型供水管网为例对漏损定位的研究方法进行了验证。结果表明,所提方法是一个实时诊断的快速又有效的方法,可实现较为准确的漏损点定位。
Based on the in-depth analysis of the causes of leakage of existing large-scale water supply network,a BP neural network deep learning method is proposed to predict the location of leakage points.A simulation model of water supply pipe network leakage is constructed.Correlation analysis of flow changes at five locations and changes in pressure monitoring points at 17 locations is implemented when leakage occurs in the pipe network.The artificial neural network deep learning is used to diagnose the location of the leakage in the pipe network.The research method of leakage location is verified by taking the small water supply pipe network built in the laboratory as an example.The results show that the proposed method is a fast and effective method for real-time diagnosis,which can achieve more accurate leakage point location.
引文
[1] Vitkovsky J P,Liggett J A,Simpson A R,et al.Optimal Measurement Site Locations for Inverse Transient Analysis in Pipe Networks[J].Journal of Water Resources Planning and Management,2003,129(6):480-492.
[2]曹晓莉,江朝元,甘思源.基于GSM/GPRS的供水管网泄漏监测与定位系统[J].仪器仪表学报,2004,25(4):173-175.
[3]马涛.供水管网在线测压点布置的探讨[J].黑龙江水利科技,2006(1):77-79.
