利用数据流进行供水二次管网的漏损控制技术分析  

Data Flow is Used to Analyze the Leakage Control Technology of Secondary Water Supply Network

作　　者：岳雯 王建 王凯轩 刘自圣 闫硕杰 彭雅萱 YUE Wen;WANG Jian;WANG Kaixuan;LIU Zisheng;YAN Shuojie;PENG Yaxuan(Civil Engineering Department,North China Institute of Aerospace Engineering,Langfang Hebei065000,China;Langfang Qingquan Water Supply Co.,Ltd.,Langfang Hebei065000,China;Hebei University of Technology,Tianjin300400,China)

机构地区：[1]北华航天工业学院建筑工程学院,河北廊坊065000 [2]廊坊市清泉供水有限责任公司,河北廊坊065000 [3]河北工业大学,天津300400

出　　处：《北方建筑》2024年第4期77-82,共6页Northern Architecture

摘　　要：城市供水系统是城市基础设施的重要组成部分,保障供水系统的正常运行对于提高城市生活质量和促进经济发展至关重要。本文总结了实时数据分析技术在供水系统漏损管理中的应用,详细探讨了流量、压力、水质及地理信息数据的实时监测与采集方法。采集的数据通过回归分析、灰色预测模型及多种机器学习算法进行漏损预测,并通过物理硬件检测法和基于模拟计算的方法进行漏损定位,展示了贝叶斯理论、神经网络算法及支持向量机在漏损识别中取得的研究成果。最后,探讨了在数据和模型两方面所面临的挑战,并提出了相应的改进建议,为未来城市供水管网漏损治理提供了科学依据与实践指导。Urban water supply system is an important part of urban infrastructure,ensuring the normal operation of water supply system is very important to improve the quality of urban life and promate the economic development.This paper summarized the application of real-time data analysis technology in water supply system leakage management,and it discussed in detail the real-time monitoring and data collection methods for flow,pressure,water quality,and geographic information.The collected data is used for leakage prediction through regression analysis,grey prediction models,and various machine learning algorithms,and leakage location is identified using physical hardware detection methods and simulation-based approaches,showcasing research achievements in Bayesian theory,neural network algorithms,and support vector machines for leakage identification.Finally,the paper addressed challenges related to data and models and proposed corresponding improvements,providing scientific basis and practical guidance for future urban water supply network leakage management.

关 键 词：供水系统 实时数据分析 漏损预测 漏损定位 机器学习方法 

分 类 号：TK284.7[动力工程及工程热物理—动力机械及工程]