含泄漏供水管道瞬变流动特征及泄漏定位  

Transient Flow Characteristics and Leakage Location of Water Supply Pipeline with Leakage Points

作　　者：杨振东 曹亚龙 张巧玲 赵思茂 曹佳豪 吴峰 李国栋[1] YANG Zhen-dong;CAO Ya-long;ZHANG Qiao-ling;ZHAO Si-mao;CAO Jia-hao;WU Feng;LI Guo-dong(Institute of Water Resources and Hydro-Electric Engineering,Xi'an University of Technology,Xi'an 710048,China)

机构地区：[1]西安理工大学水利水电学院,陕西西安710048

出　　处：《中国给水排水》2021年第15期35-40,共6页China Water & Wastewater

基　　金：陕西省教育厅基金资助项目(16JK1542);国家自然科学基金资助项目(51706180、51906201)。

摘　　要：针对长距离供水管道泄漏问题,采用CFD软件研究了供水管道内瞬变流动特性,分析了管道泄漏点位置的检测机制。基于流体的瞬变模型法及径向基函数(RBF)神经网络方法,开展了供水管道泄漏定位研究。利用Flowmaster仿真软件中的水力模型建立长度为1000 m的一维管路系统,并针对此系统进行了不同泄漏位置下的数值仿真计算以产生训练样本,借助RBF神经网络开展了泄漏工况下的网络训练和预测。结果表明,瞬变流过程中泄漏孔的存在对直管道水锤波的传播周期影响较小;管道泄漏孔的存在引起压力波传播畸变,在周期首相的压力幅值变化更为明显;在出口位置设置激励和监测点的条件下,泄漏孔距离出口越近,检测点信号衰减越快;RBF神经网络具有较强的监测能力及抗噪性能。In order to solve the leakage problem of long-distance water supply pipeline,transient flow characteristics in a water supply pipeline were explored by using CFD software,and the detection mechanism of the pipeline leakage point was analyzed.Based on fluid transient model method and RBF neural network method,leakage location of the water supply pipeline was explored.A one-dimensional pipeline system with a length of 1000 m was established by hydraulic transport model in Flowmaster simulation software.For this system,numerical simulation calculations at different leakage locations were carried out to generate training samples,and network training and prediction under leakage condition was carried out by RBF neural network.The presence of leakage holes in transient flow process had little effect on propagation cycle of water hammer waves in straight pipes.However,it would cause distortion of pressure wave propagation,and the change of pressure amplitude was more obvious at the beginning of a cycle.When excitation and monitors were set at the exit position,the closer the leak hole was to the outlet,the faster the signal attenuation at the detection point would be.The RBF neural network had a strong monitoring capability and anti-noise performance.

关 键 词：供水管道 泄漏定位 瞬变流 RBF神经网络 

分 类 号：TU991[建筑科学—市政工程]