Page 38 - 2021年第52卷第7期

［ 13］ AYALA-CABRERA D，HERRERA M，IZQUIERDO J，et al . Location of buried plastic pipes using multi-agent
support based on GPR images［J］. Journal of Applied Geophysics，2011，75（4）：679-686 .
［ 14］ AYALA-CABRERA D，HERRERA M，IZQUIERDO J，et al . GPR-based water leak models in water distribu⁃
tion systems［J］. Sensors，2013，13：15912-15936 .
［ 15］ LAI W L，CHANG K W，SHAM F C，et al . Perturbation mapping of water leak in buried water pipes via laborato⁃
ry validation experiments with high-frequency ground penetrating radar（GPR）［J］. Tunnelling and Underground
Space Technology，2016，52：157-167 .
［ 16］ LAI W L，CHANG K W，SHAM F C . A blind test of nondestructive underground void detection by ground pene⁃
trating radar（GPR）［J］. Journal of Applied Geophysics，2018，149：10-17 .
［ 17］ SANTOS V R N，TEIXEIRA F L . Application of time-reversal-based processing techniques to enhance detection
of GPR targets［J］. Journal of Applied Geophysics，2017，146（11）：80-94 .
［ 18］ XUE W，ZHU J，RONG X，et al . The analysis of ground penetrating radar signal based on generalized S trans⁃
form with parameters optimization［J］. Journal of Applied Geophysics，2017，140（5）：75-83 .
［ 19］ SILVIA J O L，ELIZABETH P C A，DAVID A C，et al . 3D Model evolution of a Leak Based on GPR image inter⁃
pretation［J］. Water Science and Technology：Water Supply，2015，15（6）：1312-1319 .
［ 20］ ALHARTHI A，LANGE J . Soil water saturation：dielectric determination［J］. Water Resources Research，1987，
23（4）：591-595 .
［ 21］ 刘之平，付辉，郭新蕾，等 . 冰水情一体化双频雷达测量系统［J］. 水利学报，2017，48（11）：1341-1347 .

GPR-based leak detection technology in water pipelines：
principle and phenomenon mechanism

GUO Xinlei，ZHENG Feifei，MA Chaomeng，CHAI Duanwu，FU Hui，CUI Haitao
（1. State Key Laboratory of Simulation and Regulation of Water Cycle in River Basin，
China Institute of Water Resources and Hydropower Research，Beijing 100038，China；
2. Dalian Zhongrui Science and Technology Development Company，Dalian 116002，China）

Abstract： Ground penetrating radar （GPR）has recently evolved as a promising non-destructive leak detec⁃
tion method for water pipelines. It is at the initial stage in China， and there is still a lack of research
and development of instruments and comparative study on the leakage phenomenon， mechanism and feed⁃
back of the images. An experimental tank for pipeline leakage detection was setup by using a newly devel⁃
oped GPR instrument. The iron pipe leak scenario was simulated and the laboratory tests which including
intact， leaking and after leakage state were performed to extract features from the obtained GPR images.
The results show that GPR identifies leaks through detecting water circulation from leaks causing voids or
anomalies in pipe depth caused by changes of dielectric constant of surrounding soil. The reflection of the
pipe move down as V type from the horizontal profiles and the slope of the hyperbola is obviously ob⁃
served which is helpful for the approximate leak location. The arrow-type reflection and under-hole anoma⁃
lous reflection is the key typical features to distinguish between leaking state and after leakage， especially
the latter when the leakage is relatively larger. As long as these two anomalous reflection from the horizon⁃
tal and vertical profiles are identified， the leaks can be detected. Characteristics of all cases are obtained
and it can be the basis for further processing for the automatic classification of leaks and equipment devel⁃
opment.
Keywords： water pipeline； leakage； ground penetrating radar； image identification； under-hole anomalous
reflection

（责任编辑：李福田）

— 792 —