HR: 0800h
AN: U21B-0812 [Abstracts]
TI: Water Leak Detection by Using Ground Penetrating Radar, Synthetic Simulation and Four-Dimensional
Visualization
AU: Al-Shukri, H
EM: alshukri@seismo.ualr.edu
AF: Department of Applied Science at University of Arkansas at Little Rock, 2801 South University Avenue,
Little Rock, AR 72204
United States
AU: * Eyuboglu, S
EM: eyuboglu.1@osu.edu
AF: Department of Applied Science at University of Arkansas at Little Rock, 2801 South University Avenue,
Little Rock, AR 72204
United States
AU: Mahdi, H
EM: mahdi@seismo.ualr.edu
AF: Department of Applied Science at University of Arkansas at Little Rock, 2801 South University Avenue,
Little Rock, AR 72204
United States
AB:
Many geophysical techniques have been suggested as candidates for detecting water leakage in water distribution system,
including ground penetrating radar (GPR), acoustic devices, and gas sampling devices. A series of laboratory experiments
were conducted to determine the validity and effectiveness of GPR in detecting water leakage in metal and plastic PVC pipes.
The goal was to derive a practical and robust procedure for detecting such leakage. Initially, prototype laboratory
experiments were designed to simulate leaks in both PVC and metal pipe. The experiments were very well controlled and
results obtained indicate that GPR is effective in detecting subsurface water leaks. This was followed by an outdoor life
size experiments. 50 feet by 30 feet by 5 feet test bed was constructed using local soil and commercial water distribution
pipes. A 400 MHz antenna was used to collect three-dimensional GPR data as a function of time for a number of experiments
using different type of pipes. Advanced imaging and visualization technology was used to further analyze the data. The UALR
Virtual Reality Center CAVE facilities were utilized to accomplish this test. Results obtained indicate that GPR is effective
in detecting subsurface water leaks in both pipes. Synthetic models of the GPR signals based on Finite Difference Time
Domain Method (FDTD) were built to help select an appropriate equipment configuration (frequency band, type of antenna, and
real-time imaging software) prior to data acquisition. The simulation software was used to determine the near-field
radiation characteristics of the GPR antenna. Different experimental models were adapted for which observational GPR data
was previously collected. Matlab regression analysis was used to generate the incident waves for each model to ensure highly
accurate and controlled experiments.
DE: 0530 Data presentation and visualization
DE: 0545 Modeling (4255)
DE: 0634 Measurement and standards
DE: 0644 Numerical methods
DE: 0910 Data processing
SC: Union [U]
MN: Fall Meeting 2005