·Table of Contents
·Methods and Instrumentation
Case Studies of Buried Anomaly by Ground Penetrating RadarMs. Jakyung-Hahn, Korea Telecom
Tel :+82- 042- 866-3074, Fax :+82- 042- 866-3065
Ground penetrating radar(GPR) uses radio waves to detect buried objects in any non-metallic material. Initially it was used to detect structures in ice. GPR has evolved to include the penetration of soils, rocks and man-made structures.
Measurements must be made at many points to ensure accurate representation of the area being examined. A computer records and merges the GPR data to create image. Hyperbola(arches) displayed on this image indicate the placement and depth of objects in the survey area.
The depth of exploration and image definition depend on the radio frequency used. Low frequencies are used for deep geological mapping. High frequencies are necessary for high definition imaging, such as delineating reinforcing bars in concrete.
The followings are GPR applications in the field of buried objects
GPR for utility detection
In the recent, GPR(surveying) methods have been employed to minimize the risks during construction. GPR(Ground Penetrating Radar) is applied to outside plant telecommunication facilities such as cable tunnels, and underground conduits. The thickness and soundness of tunnel lining can be evaluated, and the location of rebars and steel ribs can be also found effectively. The location of conduits and other pipes also can be found.
The followings are case studies of GPR being applied to detect buried anomaly and model tests which is for making out each frequency character and effect of data processing.
1. civil structure safety diagnosis
Tests were run along a profile to examine position of buried anomaly in lining concrete.
The configuration of cable tunnel which was constructed by NATM is shown in Fig 1. The main purpose of this survey was to distinguish rebar part from no rebar part of sidewall and finding cavity of crown part, leaking water points. For crown part survey, we used antenna-aid-instrument, is developed by us, to reduce time and manpower(Fig 1). Targets could be found with 450MHz and 900MHz antennas effectively.
|Fig 1: The cross section of cable tunnel(NATM)||Fig 1: The scene of crown survey by using antenna-aid instrument|
Fig 2. shows the radar section obtained from side wall part of cable tunnel. In the area of 1m position, remarkable hyperbolic reflections which are from rebars appeared. This section coincided with design map well. Fig 3 shows the result of crown part test. considering the condition of inside the lining, and the appearance of water leakage, the hyperbolic reflection patterns in the position 2.2m~2.8m was estimated cavity.
|Fig 2: Radar section (sidewall of cable tunnel)||Fig 2: Water leakage from crown part of cable tunnel|
|Fig 3: Radar section (crown part of cable tunnel)|
The following result(Fig.4) was collected using 450MHz antennas to find two sewage pipes which are buried in 0.4m and 0.6m depth and in diameter 300mm. This section shows two pipes (which are buried different position) very well. This result was also examined with the naked eye and identified.. This section also shows interface of compaction soil(which is for pavement construction)and heterogenious ground. By the way, GPR section marked more narrow spacing of two objects than they are.
|Fig 4: Two sewage pipes buried in ground||
Fig 5: Configuration of the test site
2. Anomaly model test
GPR operating skill is very important. It is no exaggeration that survey results are dependent on operator's efficiency. Operator has to set up survey design cautiously. It is important to choose antenna frequency, time window, velocity, survey direction, sampling interval and spacing of transmitter-receiver in operating GPR.
|Fig 6: The configuration of model test|
This model(Fig. 6) test was for finding out the relation of antenna resolution to target depth and effect of Data processing(migration). In general, high resolution detection is for shallow targets and low resolution detection is for deep targets. The data set was collected using 900MHz and 1200MHz antennas. Table1 & Fig. 6 show test parameters and setup in detail.
|Test No.||Anomaly size(m)||Freq. (MHz)||Depth(m)
|Table 1: Model test parameters|
CaseA ~ CaseL are the results of model tests. These tests were executed by changing three different parameters such as antenna frequency, target depth and time window. Left sections(CaseA, CaseC,...,CaseK) are compared with right sections(CaseB, CaseD,...,CaseL) in a difference with time window. Using 1200MHz antenna frequency(CaseA ~ CaseF), echo was found in 20ns time window. Echo means abnormal waves difficult to analyse and it is a kind of noise. In case of 900MHz antenna frequency(CaseG ~ CaseL), echo was found in 25ns time window. Thus, proper choice of time window is very important to obtain a good result.
|Fig 7: Model test results (section A ~ L)|
The influence of a depth ratio(ratio of target depth to target width) on the results is shown in Case A, Case C and Case E when 1200MHz antenna frequency was used. In case depth ratio was 1.0 or over, target could be observed to parabolic shape. But in case it was below 1.0, parabolic shape was not clear. These results was obtained in other time window(Case B, Case D and Case F). Using 900MHz antenna frequency(Case G ~ Case L), similar tendency also could be seen according to depth ratio. But in comparison with 1200MHz antenna frequency, parabolic shape became so rough. Thus, in 900MHz antenna frequency, proper range of depth ratio is much higher than that in 1200MHz antenna frequency. But the upper limit of depth ratio was not clarified in both case.
The followings are the results of model tests.
The main advantage of this migration method is that the entire waveform is used rather than just select events. The main limitations are that the survey line topography must be reasonably flat, the geometrical spreading is 2- dimensional and attenuation is not compensated for.
Fig. 8 ~ Fig.10 shows a several examples of migration and Table 2 shows test source information.
|Figure No.||Anomal y size(m)||Freq. (MHz)||Depth(m) / (D/Width ratio)||Time Window (ns)|
|Table 2: Test Source Information|
In the before migration section, It shows hyperbolic reflector bigger than it's real size(30cm width) that is due to wave diffraction. It is also show linear structures (V-shape and horizontal thick line from bottom part) which are by noise effect that is reflected by both sidewalls and bottom of the box.
On the other hand ,in after migration section, hyperbola signal became small and linear noise are disappeared.
Fig. 8 shows radar section in grey mode and Fig. 9 is a converted section by color mode. Fig.10 is the best section acqired by data processing including migration. It shows exact positon and size of buried anomaly.
|Fig 8: Effect of migration (Grey mode)||Fig 9: Effect of migration(Color mode)||Fig 10: The two radar sections same buried anomaly.|
The ability of GPR to locate underground utilities quickly and accurately helps to maintain existing utilities and identify potentially hazardous areas before damage occurs. The location of underground pipes as well as concrete rebars can be found effectively and then the results of these tests gave good coincidence with utility maps.
but GPR can not give information of exact buried object's size(pipe diameter, etc) and it's type(shape, material, etc). Through the post-data processing, partial problem can be solved. It is also important to use proper antenna frequency and time window to acquire better images.
However, it is still difficult to analyse GPR section because there are various conditions in the field and GPR system's own characteristics.
In order to solve these problems, it is require to develop data processing and other helpful survey techniques.
With this, GPR could be useful solutions to locate buried objects and detect a defect in underground facilities.
|© AIPnD , created by NDT.net|||Home| |Top||