Abstract:Pipe bursts are a universal problem in water supply systems, which can severely disrupt daily life and urban development. This study correlates flow measurements collected from inlets and outlets of a district metering area (DMA) to quickly identify bursts. The analysis uses a clustering algorithm to detect abnormal flow data i.e., outliers. Then, some outliers are identified as bursts according to the inlet and outlet flow fluctuation characteristics. The results indicate that the number and locations of the inlets and outlets affect the detection performance. The system can accurately detect bursts and insure a low false positive rate when there are a relatively small number of inlets and outlets and the locations are well placed.
刘书明, 吴以朋, 王晓婷, 刘友飞, 李佳杰. 应用聚类算法识别供水管网爆管事故[J]. 清华大学学报(自然科学版), 2017, 57(10): 1096-1101.
LIU Shuming, WU Yipeng, WANG Xiaoting, LIU Youfei, LI Jiajie. Clustering algorithm for burst detection in water distribution systems. Journal of Tsinghua University(Science and Technology), 2017, 57(10): 1096-1101.
Mounce S R, Khan A, Wood A S, et al. Sensor-fusion of hydraulic data for burst detection and location in a treated water distribution system[J]. Information Fusion, 2003, 4(3):217-229.
[2]
Mounce S R, Boxall J B, Machell J. Development and verification of an online artificial intelligence system for detection of bursts and other abnormal flows[J]. Journal of Water Resources Planning and Management, 2010, 136(3):309-318.
[3]
Romano M, Kapelan Z, Savic D A. Automated detection of pipe bursts andother events in water distribution systems[J]. Journal of Water Resources Planning and Management, 2014, 140(4):457-467.
[4]
Ye G L, Fenner R A. Kalman filtering of hydraulic measurements for burst detection in water distribution systems[J]. Journal of Pipeline Systems Engineering and Practice, 2011, 2(1):14-22.
[5]
Ye G L, Fenner R A. Weighted least squares with expectation-maximization algorithm for burst detection in U.K. water distribution systems[J]. Journal of Water Resources Planning and Management, 2014, 140(4):417-424.
[6]
凌文翠, 张涛, 强志民, 等. 城市供水管网独立计量区域的研究与应用进展[J]. 中国给水排水, 2011, 27(13):46-50.LING Wencui, ZHANG Tao, QIANG Zhimin, et al. Research and application of district metering area for urban water distribution network[J]. China Water & Wastewater, 2011, 27(13):46-50. (in Chinese)
[7]
Wu Y, Liu S, Wu X, et al. Burst detection in district metering areas using a data driven clustering algorithm[J]. Water Research, 2016, 100:28-37.
[8]
Tan P N, Steinbach M, Kumar V. Introduction to Data Mining[M]. New Jersey:Addison-Wesley, 2005.
[9]
Rodriguez A, Laio A. Clustering by fast search and find of density peaks[J]. Science, 2014, 344(6191):1492-1496.
[10]
Loureiro D, Amado C, Martins A, et al. Water distribution systems flow monitoring and anomalous event detection:A practical approach[J]. Urban Water Journal, 2016, 13(3):242-252.
[11]
Metz C E. Basic principles of ROC analysis[J]. Seminars in Nuclear Medicine, 1978, 8(4):283-298.