Abstract
Most of the water/sewer pipes in Japan were constructed after the high economic growth that occurred in the 1970s. Those pipes are now aging, and the water business is shifting from construction to rehabilitation. NJS, as a consulting company in Japan, has had to shift focus to reflect the increasing need for rehabilitation in the water sector and the limited budget available. Doing business efficiently and better managing water assets has become a critical part of the business. NJS has been searching for a way to inspect sewer pipes efficiently, not just a few times in a service life but in increased frequency – in the range of every year or two. During this process, a focus on UAVs (Unmanned Aerial Vehicles) was considered beneficial. However, one that could stably fly through a confined space, such as a sewer, did not yet exist. Hence, NJS decided to develop a UAV for this purpose in cooperation with a partner company. To evaluate the performance of the UAV, the NJS team has inspected an actual sewer section, in cooperation with Yokosuka city, Kanagawa prefecture, Japan. The result shows that the UAV is capable of inspecting more than 1,000 m a day without personnel needing to enter a manhole. The team has also inspected the same section by CCTV and manhole camera to compare the inspection speed, operator's safety and the reliability of data. From the obtained result, it can be said that UAV is an effective screening method to efficiently conduct CCTV inspection; in other words, to prioritize the sections that need a detailed inspection.
INTRODUCTION
Currently, the main method of inspecting sewer pipes is CCTV. To raise the efficiency of CCTV inspection, NJS has begun to consider utilizing unmanned aerial vehicles (UAVs) as a screening method, to be used prior to CCTV. This will help to detect pipe abnormalities before their condition deteriorates significantly. However, the existing models of UAV on the market were not capable of flying stably in such a confined space. Therefore, NJS decided to develop a unique UAV that could fly stably in such a space and conducted an evaluation test in the city of Yokosuka. The result of the evaluation test showed that, although there were some items that showed insufficient data compared to CCTV, the overall data were sufficient to detect the existence of abnormalities. Especially, to be able to confirm the inspection speed of 1,000 m/day without entering a manhole was a breakthrough. From these results, NJS considers UAV inspection as a new pipe inspection method for assessing sewer condition, which is the critical element of asset management.
BACKGROUND
The majority of sewers in Japan were constructed after the 1970s. In 2017, sanitation coverage reached approximately 80%. Along with the growth of coverage rate, the sewer stock has grown steadily, reaching approximately 470,000 km in the recent statistics (Figure 1).
The procedure of maintaining assets such as sewers typically has the following steps: (1) analyze the present state; (2) evaluate the risk; (3) simulate a fiscal plan in the long-term; and (4) set the inspection/rehabilitation plan. Among those steps, securing budgets for inspection is the most difficult part, requiring a way to develop a new method that can inspect pipes efficiently without losing quality.
Currently, small diameter pipes are commonly inspected by a vehicle-mounted CCTV. This method requires an operator to assemble the CCTV on site and usually to enter a manhole, which takes a certain amount of time before completing the section. To take measures against the growing amount of aging pipes, reducing inspection time was crucial. To solve this problem, NJS started to look for a new inspection method and began to examine the possibility of utilizing a UAV.
EVALUATION BY COMPARISON TO OTHER INSPECTION METHODS
Method of evaluation
To examine the reliability of data obtained by UAV, the team inspected the same section with CCTV and manhole camera.
Inspected section
The sections used for evaluation were as follows:
Pipe usage: combined sewer
Pipe diameter: 400 mm to 600 mm
Number of lines: 30
Total length: 1,324.8 m
Device specification
The devices used for inspection were CCTV, manhole camera and UAV. CCTV had a resolution of 410,000 pixels and the manhole camera, 2,000,000 pixels, both used commonly in Japan. The specification of the UAV can be found in Table 1 and its appearance in Figure 2. This model was jointly developed by NJS and Autonomous Control Systems Laboratory Ltd. (ACSL), a Japanese UAV company.
Specification of the UAV
| Weight | 1.5 kg (includes a Li-Po battery) |
| Size | W:250 mm, L:570 mm, H:120 mm |
| Flight duration | Approx. 5 min |
| Flight speed | 0.5–3 m/sec |
| Camera | SONY DSC-RX0 |
| Resolution (video shooting) | Approx. 2,000,000 pixels |
| Viewing angle | 84° |
| Weight | 1.5 kg (includes a Li-Po battery) |
| Size | W:250 mm, L:570 mm, H:120 mm |
| Flight duration | Approx. 5 min |
| Flight speed | 0.5–3 m/sec |
| Camera | SONY DSC-RX0 |
| Resolution (video shooting) | Approx. 2,000,000 pixels |
| Viewing angle | 84° |
New sewer construction and accumulated length by year (Japan Sewage Works Association 2015).
New sewer construction and accumulated length by year (Japan Sewage Works Association 2015).
Appearance of the UAV.
COMPARISON OF THE INSPECTED RESULTS
The comparison of the three inspection methods was evaluated by (1) inspection speed, (2) operator's safety and (3) reliability of data. Details of each evaluation are described below.
Inspection speed
The indexes of the ‘inspection speed’ were the number of days required for inspection, the number of manhole openings, the average inspection path per manhole (Figure 3) and the average of inspected length. Table 2 shows the result of inspection by each method. The number of days required to inspect the total length of 1,324.8 m were CCTV = 4 days, manhole camera = 2 days and UAV = 1 day. The UAV's average inspection time per manhole was 12 min 23 seconds. The total duration time that UAV took to inspect the whole length was 4 hours and 18 min, including preparation, tuning of devices and clean up.
Comparison of the three inspection methods
| CCTV | Manhole camera | UAV | |
|---|---|---|---|
| Number of days needed | 4 | 2 | 1 |
| Number of manhole openings | 30 | 36 | 15 |
| Average inspection path per manhole | 1.0 | 0.8 | 2.0 |
| Average of inspected length per manhole | 44.16 | 36.8 | 88.32 |
| CCTV | Manhole camera | UAV | |
|---|---|---|---|
| Number of days needed | 4 | 2 | 1 |
| Number of manhole openings | 30 | 36 | 15 |
| Average inspection path per manhole | 1.0 | 0.8 | 2.0 |
| Average of inspected length per manhole | 44.16 | 36.8 | 88.32 |
Manhole openings and the inspected paths.
The number of manhole openings required to inspect the whole length was, CCTV = 30 openings, manhole camera = 36 openings and UAV = 15 openings, showing UAV to have the least number of openings of all. This is owing to the fact that UAV was able to inspect one path in the upstream direction and two paths in the downstream direction, resulting in a maximum of three paths from a single opening. The average number of inspection paths per manhole was CCTV = 1.0 path, manhole camera = 0.8 path and UAV = 2.0 paths. The average length of inspection per manhole was CCTV = 44.16 m, manhole camera = 36.8 m and UAV = 88.32 m. The results indicate that UAV inspection is the most efficient of the three methods tested, in terms of the number of days required, average number of inspection paths and inspection length per manhole.
Operator's safety
The index for the operator's safety was set to be the number of manhole openings that operators had to enter. The numbers were CCTV = 30 openings, manhole camera = 0 opening and UAV = 1 opening. The one opening recorded for UAV had happened when the operator misguided the UAV to a deep gap and had to enter the manhole to retrieve it. This could have been prevented by examining the construction drawings prior to the inspection. Basically, an inspection by UAV was able to carry out the whole process from above ground, including preparation, inspection and clean up (Figure 4).
Operators working from above ground in UAV inspection.
(Left) Crack obtained by UAV, (Right) crack obtained by manhole camera.
Reliability of data
The reliability of the data obtained by UAV was examined by the number of abnormalities matched to the CCTV results; that is, setting CCTV as a benchmark. The results of CCTV are shown in Table 3 and the levels of severities (A-C) are explained in Table 4, with A being the most serious level. The criteria shown in Table 4 were taken from the Guideline for sewer management (Japan Sewage Works Association 2014), issued by Japan Sewerage Works Association (JSWA) in 2014. A manhole camera was also used to inspect the same section and the result was compared to that of CCTV. The visibility (screen clarity) was evaluated by how many abnormalities the method detected for each item (Table 5). Figure 5 shows a crack image obtained by UAV and a manhole camera.
The result of the CCTV inspection
| Corrosion | Sagging in vertical direction | Breakage | Cracks | Displaced joints | Infiltration | Extrusion of lateral pipes | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| A | B | C | A | B | C | a | b | c | a | b | c | a | b | c | a | b | c | a | b | c |
| 0 | 1 | 2 | 0 | 0 | 0 | 3 | 8 | 6 | 1 | 5 | 8 | 0 | 0 | 3 | 0 | 0 | 1 | 0 | 0 | 1 |
| Root intrusion | Attached deposit (mortar) | Others | Total | |||||||||||||||||
| a | b | c | a | b | c | a | b | c | A | B | C | Total | a | b | c | Total | ||||
| 0 | 3 | 0 | 0 | 0 | 4 | 0 | 0 | 1 | 0 | 1 | 2 | 3 | 4 | 16 | 24 | 44 | ||||
| Corrosion | Sagging in vertical direction | Breakage | Cracks | Displaced joints | Infiltration | Extrusion of lateral pipes | ||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| A | B | C | A | B | C | a | b | c | a | b | c | a | b | c | a | b | c | a | b | c |
| 0 | 1 | 2 | 0 | 0 | 0 | 3 | 8 | 6 | 1 | 5 | 8 | 0 | 0 | 3 | 0 | 0 | 1 | 0 | 0 | 1 |
| Root intrusion | Attached deposit (mortar) | Others | Total | |||||||||||||||||
| a | b | c | a | b | c | a | b | c | A | B | C | Total | a | b | c | Total | ||||
| 0 | 3 | 0 | 0 | 0 | 4 | 0 | 0 | 1 | 0 | 1 | 2 | 3 | 4 | 16 | 24 | 44 | ||||
Explanation for each level A(a) - C(c) is in Table 4.
Whole length assessment: A = most severe → C = less serious.
Pipe-wise assessment: a = most severe → c = less serious.
Levels of severities (Japan Sewage Works Association 2014)
| Levels Sewer conditions | A | B | C | ||
|---|---|---|---|---|---|
| Whole length assessment | Corrosion of pipes | Steel reinforcement visible | Aggregate visible | Surface roughness | |
| Deformed in vertical direction | (Inner diameter) less than 700 mm | 100 % or greater of inner diameter | 50 % or greater of inner diameter | less than 50 % of inner diameter | |
| (Inner diameter) between 700 mm and 1,650 mm | 50 % or greater of inner diameter | 25 % or greater of inner diameter | less than 25 % of inner diameter | ||
| (Inner diameter) between 1,650 mm and 3,000 mm | 25 % or greater of inner diameter | 12.5 % or greater of inner diameter | less than 12.5 % of inner diameter | ||
| Levels Sewer conditions | a | b | c | ||
| Local (pipe-wise) assessment | Breakage/longitudinal crack | Reinforced concrete pipe, etc. | Collapsed | Longitudinal crack 2 mm or greater in width | Longitudinal crack less than 2 mm in width |
| Longitudinal crack 5 mm or greater in width | |||||
| Clay pipe | Collapsed | Longitudinal crack less than 50 % of pipe length | – | ||
| Longitudinal crack 50 % or greater of pipe length | |||||
| Circumferential crack | Reinforced concrete pipe (RC) etc. | Circumferential crack 5 mm or greater in width | Circumferential crack 2 mm or greater in width | Circumferential crack less than 2 mm in width | |
| Vitrified clay pipe (VC) | Circumferential crack two-thirds or greater of the circumferential length | Circumferential crack less than two-thirds of circumferential length | – | ||
| Displaced joints | Extruded joints | Reinforced concrete pipe etc.: 70 mm or greater | Reinforced concrete pipe etc.: less than 70 mm | ||
| Vitrified clay pipe: 50 mm and over | Vitrified clay pipe: less than 50 mm | ||||
| Infiltration | Gushing | Running | Seeping | ||
| Extrusion of lateral | 50 % or greater of inner diameter | 10 % or greater of inner diameter | Less than 10 % of inner diameter | ||
| Attached deposit, grease | Blockage of 50 % or greater | Blockage of less than 50 % of inner diameter | – | ||
| Roots intrusion | Blockage of 50 % or greater | Blockage of less than 50 % of inner diameter | – | ||
| Attached deposit, mortar | Blockage of 30 % or greater | Blockage of 10 % or greater | Blockage of less than 10 % | ||
| Levels Sewer conditions | A | B | C | ||
|---|---|---|---|---|---|
| Whole length assessment | Corrosion of pipes | Steel reinforcement visible | Aggregate visible | Surface roughness | |
| Deformed in vertical direction | (Inner diameter) less than 700 mm | 100 % or greater of inner diameter | 50 % or greater of inner diameter | less than 50 % of inner diameter | |
| (Inner diameter) between 700 mm and 1,650 mm | 50 % or greater of inner diameter | 25 % or greater of inner diameter | less than 25 % of inner diameter | ||
| (Inner diameter) between 1,650 mm and 3,000 mm | 25 % or greater of inner diameter | 12.5 % or greater of inner diameter | less than 12.5 % of inner diameter | ||
| Levels Sewer conditions | a | b | c | ||
| Local (pipe-wise) assessment | Breakage/longitudinal crack | Reinforced concrete pipe, etc. | Collapsed | Longitudinal crack 2 mm or greater in width | Longitudinal crack less than 2 mm in width |
| Longitudinal crack 5 mm or greater in width | |||||
| Clay pipe | Collapsed | Longitudinal crack less than 50 % of pipe length | – | ||
| Longitudinal crack 50 % or greater of pipe length | |||||
| Circumferential crack | Reinforced concrete pipe (RC) etc. | Circumferential crack 5 mm or greater in width | Circumferential crack 2 mm or greater in width | Circumferential crack less than 2 mm in width | |
| Vitrified clay pipe (VC) | Circumferential crack two-thirds or greater of the circumferential length | Circumferential crack less than two-thirds of circumferential length | – | ||
| Displaced joints | Extruded joints | Reinforced concrete pipe etc.: 70 mm or greater | Reinforced concrete pipe etc.: less than 70 mm | ||
| Vitrified clay pipe: 50 mm and over | Vitrified clay pipe: less than 50 mm | ||||
| Infiltration | Gushing | Running | Seeping | ||
| Extrusion of lateral | 50 % or greater of inner diameter | 10 % or greater of inner diameter | Less than 10 % of inner diameter | ||
| Attached deposit, grease | Blockage of 50 % or greater | Blockage of less than 50 % of inner diameter | – | ||
| Roots intrusion | Blockage of 50 % or greater | Blockage of less than 50 % of inner diameter | – | ||
| Attached deposit, mortar | Blockage of 30 % or greater | Blockage of 10 % or greater | Blockage of less than 10 % | ||
The number of abnormalities detected by each inspection method
| Abnormalities | CCTV [1] | Manhole camera [2] | UAV [3] | Manhole camera | UAV | |
|---|---|---|---|---|---|---|
| Matching rate to CCTV | Matching rate to CCTV | |||||
| [2]/[1] | [3]/[1] | |||||
| Defects (critical) | 1. Corrosion | 3 | 0 | 3 | 0% | 100% |
| 2. Sagging in vertical direction | – | – | – | – | – | |
| 3. Breakage | 17 | 3 | 14 | 18% | 82% | |
| 4. Cracks | 14 | 0 | 7 | 0% | 50% | |
| 5. Displaced joints | 3 | 0 | 3 | 0% | 100% | |
| 6. Infiltration | 1 | 0 | 1 | 0% | 100% | |
| Subtotal | 38 | 3 | 28 | 8% | 74% | |
| Defects (managerial) | 7. Extrusion of lateral pipes | 1 | 1 | 1 | 100% | 100% |
| 8. Attached deposit (grease) | – | – | – | – | – | |
| 9. Root intrusion | 3 | 0 | 1 | 0% | 33% | |
| 10. Attached deposit (mortar) | 4 | 1 | 2 | 25% | 50% | |
| 11. Blockage of lateral | 1 | 0 | 0 | 0% | 0% | |
| Subtotal | 9 | 2 | 4 | 22% | 44% | |
| Total | 47 | 5 | 32 | 11% | 68% | |
| Abnormalities | CCTV [1] | Manhole camera [2] | UAV [3] | Manhole camera | UAV | |
|---|---|---|---|---|---|---|
| Matching rate to CCTV | Matching rate to CCTV | |||||
| [2]/[1] | [3]/[1] | |||||
| Defects (critical) | 1. Corrosion | 3 | 0 | 3 | 0% | 100% |
| 2. Sagging in vertical direction | – | – | – | – | – | |
| 3. Breakage | 17 | 3 | 14 | 18% | 82% | |
| 4. Cracks | 14 | 0 | 7 | 0% | 50% | |
| 5. Displaced joints | 3 | 0 | 3 | 0% | 100% | |
| 6. Infiltration | 1 | 0 | 1 | 0% | 100% | |
| Subtotal | 38 | 3 | 28 | 8% | 74% | |
| Defects (managerial) | 7. Extrusion of lateral pipes | 1 | 1 | 1 | 100% | 100% |
| 8. Attached deposit (grease) | – | – | – | – | – | |
| 9. Root intrusion | 3 | 0 | 1 | 0% | 33% | |
| 10. Attached deposit (mortar) | 4 | 1 | 2 | 25% | 50% | |
| 11. Blockage of lateral | 1 | 0 | 0 | 0% | 0% | |
| Subtotal | 9 | 2 | 4 | 22% | 44% | |
| Total | 47 | 5 | 32 | 11% | 68% | |
The matching rate (of 11 items) to CCTV was 11% for manhole camera and 68% for UAV. Out of these results, the rate for critical defects (item 1–6) was 8% for manhole camera and 74% for UAV and managerial results (item 7–11), 22% and 44%.
THE WAY FORWARD
UAV matching results at 74% for critical defects that affect structural integrity can be considered as a high matching rate. Among the items, corrosion, displaced joints and extrusion of lateral pipes scored a 100% match to the result of CCTV inspection. This score owes to the UAV's capability of being able to fly to the center of the sewer section, collect images from a close distance and detect abnormalities, whereas a manhole camera could only collect images from a manhole. On the other hand, cracks, root intrusion and mortar deposit showed a matching rate of only 50% or less. The reasons for this low matching rate could be the low visibility that UAV occasionally encounters – the downwards airflow generated by the UAV splashes the water and the water sticks to the lens. Or, the narrow viewing angle may have blurred the image and made it difficult for the operator to detect the abnormalities. Therefore, camera selection will be an issue for future development. The result that lateral blockage showed as low as 0% matching came from the UAV not having a lateral vision. Without this capability, detecting lateral blockage was difficult.
To summarize, it can be said that even though some items showed less accuracy to CCTV, the overall score was better than that of a manhole camera. Therefore, UAV can be seen as a possible screening method for pipe inspection.
CONCLUSION
NJS has been searching for a way to raise the efficiency of CCTV inspection. During this process, NJS has focused on UAV as a possible screening method to be used prior to CCTV and has worked to develop a UAV for this purpose. To evaluate the performance of this UAV, NJS inspected a 1,324.8 m length of combined sewer with Yokosuka city. The same section was inspected by CCTV and manhole camera also and the result was compared to that of UAV.
The result by UAV inspection was as follows:
UAV detected as many abnormalities as CCTV in many inspection items
In the standard inspection procedure, there is no need for personnel to enter a manhole
Over 1,000 m of pipes can be inspected in a day
UAV inspection has a low detection rate for cracks, root intrusions, etc.
Detection of lateral blockage remains to be solved
From this result, it can be concluded that NJS has established a new method of inspection with UAV being a method to detect the presence of abnormalities but not the severity level in detail. NJS will continue to work on improving the image quality, which will lead to a higher detection rate.
%20cropped.png?versionId=7149)