Method for detecting position of water leakage in water pipe using smart water leakage detector and method for determining and processing level of water leakage in water pipe using same
The smart water leak detector system addresses the inefficiencies of conventional methods by using waveform processing and cross-correlation to accurately locate and assess the intensity of water leaks in pipes, enhancing detection precision and response times.
Patent Information
- Application Number
- PCT/KR2024/009086
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-07-13
- Filing Date
- 2024-06-28
- Publication Date
- 2025-01-16
AI Technical Summary
Conventional methods for detecting water leaks in pipes are costly, time-consuming, and lack accuracy in determining the exact location of leaks, leading to wastage of water and potential environmental hazards.
A smart water leak detector system that measures and processes waveforms from multiple detectors to calculate the time difference in sound arrival, allowing for precise leak location detection and intensity assessment using cross-correlation and noise filtering techniques.
The system provides accurate and rapid detection of water leaks, reducing noise interference and enabling quicker decision-making for repair, while also determining the leak level for appropriate measures.
Smart Images

Figure KR2024009086_16012025_PF_FP_ABST
Abstract
Description
A method for detecting the location of a water leak in a water pipe using a smart leak detector and a method for determining and handling the level of a water leak in a water pipe using the same.
[0001] The present invention relates to a method for detecting a location of a water leak in a water pipe and a method for determining and processing a level of a leak, and more particularly, to a method for detecting a location of a water leak in a water pipe using a smart leak detector capable of accurately identifying the location of a leak when a leak occurs in a water pipe, determining the intensity of the leak, and notifying a manager, etc., and a method for determining and processing a level of a water leak in a water pipe using the same.
[0002] Typically, water pipes are laid underground to supply water to houses, buildings, factories, and apartments, and clean water is supplied to various locations through branch pipes that branch off from the main pipe.
[0003] Although old water pipes are being replaced periodically, if leaks occur due to poor construction or failure to replace water pipes, not only will a lot of tap water be wasted, but there is also a risk of water pollution due to the inflow of foreign substances, ground subsidence, and changes in tributaries.
[0004] In the past, portable devices based on the listening principle were used to walk around the site to detect water leaks, but the listening detection method requires highly skilled techniques to determine water leaks with great care, and it has the disadvantage of being costly and time-consuming because it requires carrying listening devices and moving along underground pipes to detect them one by one.
[0005] Additionally, there is a problem that it is very difficult to accurately identify the actual location of the leak, that is, the location of the leak.
[0006] To solve these problems, various conventional methods for estimating the location of water pipe leaks have been developed, but there are still limitations in the accuracy of leak location.
[0007] The present invention is intended to solve the above-mentioned problems, and the purpose of the present invention is to provide a method for detecting a leak location in a water pipe using a smart leak detector, which can accurately detect the location of a leak when a leak occurs in a water pipe, and can easily determine the level of the leak, that is, the intensity of the leak, and quickly take appropriate measures accordingly, and a method for determining and processing a leak level in a water pipe using the same.
[0008] In order to achieve the above object, the present invention provides a method for detecting a location of a water pipe leak using a smart water leak detector, comprising: a measuring step of measuring a first original waveform of a water pipe leak sound using a first water leak detector during the same time period, and measuring a second original waveform of the water pipe leak sound using a second water leak detector arranged spaced apart from the first water leak detector; an absolute value generation step of converting the amplitude of the first original waveform into an absolute value to generate a first absolute waveform, and converting the amplitude of the second original waveform into an absolute value to generate a second absolute waveform; a section division step of dividing the first absolute waveform and the second absolute waveform into a plurality of sections in equal time units; an average calculation step of calculating a first average value which is an average of each section of the first absolute waveform, and a second average value which is an average of each section of the second absolute waveform; A first standard deviation calculation step of calculating a first standard deviation, which is the standard deviation of a plurality of first average values, and a second standard deviation, which is the standard deviation of a plurality of second average values; a range calculation step of calculating a first specific range by adding or subtracting the first standard deviation from the first average value of each section of the first absolute waveform, and calculating a second specific range by adding or subtracting the second standard deviation from the second average value of each section of the second absolute waveform; a filtering step of removing a first abnormal section of the first original waveform corresponding to the first average value section if there is a value among the plurality of first average values that deviates from the upper and lower amplitudes of the first specific range, and removing a second abnormal section of the second original waveform corresponding to the second average value section if there is a value among the plurality of second average values that deviates from the upper and lower amplitudes of the second specific range; A first correlation step of removing a section of the first original waveform corresponding to the second abnormal section from the first original waveform that has completed the filtering step, and removing a section of the second original waveform corresponding to the first abnormal section from the second original waveform that has completed the filtering step; A processed waveform generation step of generating a first processed waveform by connecting the first original waveforms that have completed the first correlation step, and generating a second processed waveform by connecting the second original waveforms that have completed the first correlation step;It is characterized by comprising a time difference calculation step of calculating a time difference (△T), which is the difference in time between the time the leak sound reaches the location where the first leak detector is installed and the time the leak sound reaches the location where the second leak detector is installed, by applying a cross-correlation operation to the first and second processed waveforms; and a leak location detection step of detecting a leak location in a water pipe using the time difference.
[0009] In the above leak location detection step, the leak location (L) is detected by the following equation:
[0010] Leakage location (L) = center point (d / 2) + (△T * vibration sound transmission speed) / 2
[0011] The center point (d / 2) is a point half the distance (d) between the first and second leak detectors, and the vibration sound transmission speed is determined by variables including the material and diameter of the water pipe. If the calculated leak location is positive, it is a location closer to the second leak detector from the center point, and if the calculated leak location is negative, it is a location closer to the first leak detector from the center point.
[0012]
[0013] In addition, in order to achieve the above object, the method for determining and processing a water leak level of a water pipe using a smart water leak detector of the present invention comprises: a second correlation step of detecting a location of a water leak in a water pipe by the detection method of claim 1, and then generating a correlation graph representing a correlation between the first processing waveform and the second processing waveform; a second standard deviation calculation step of calculating a third standard deviation which is a standard deviation of the correlation graph; a detection standard step of calculating a plurality of detection standard values by multiplying the third standard deviation by a plurality of reference level values; a leakage level designation step of determining a plurality of leak levels corresponding to the reference level values in advance; a leakage level determination step of comparing the maximum amplitude value of the correlation graph with the detection standard value to obtain a detection standard value corresponding to the maximum value among the detection standard values exceeded by the maximum amplitude value of the correlation graph, and determining a leakage level corresponding to the reference level value multiplied to obtain the detection standard value in the detection standard step as a final leakage level; It is characterized in that it comprises a notification step for notifying preset measures according to the final leakage level, and in the leakage level designation step, even if the leakage level is the same, the reference level value is changed according to the material of the water pipe so that the same leakage level is designated.
[0014] At the same leakage level, the reference level value is greater when the material of the water pipe is made of metal than when it is made of plastic.
[0015] According to the method for detecting a water leak location in a water pipe using the smart water leak detector of the present invention as described above and the method for determining and processing a water leak level in a water pipe using the same, the following effects are obtained.
[0016] By using the smart water leak detector of the present invention, the water leak location of the water pipe is detected using only waveforms with almost no noise after removing unnecessary noise, so the water leak location can be detected more accurately than with existing methods, and additional measures can be taken quickly accordingly.
[0017] In addition, by using the smart water leak detector of the present invention to determine and handle the water leak level of a water pipe, a manager can easily determine the level of the leak, i.e. the intensity of the leak, when a water leak occurs in a water pipe, and quickly take appropriate measures.
[0018] Figure 1 is a flowchart of a method for detecting a leak location in a water pipe using a smart leak detector according to an embodiment of the present invention.
[0019] Figure 2 illustrates the original waveform by the measurement step according to an embodiment of the present invention.
[0020] Figure 3 illustrates an absolute waveform by an absolute value generation step according to an embodiment of the present invention.
[0021] Figure 4 is a diagram for explaining the interval division step, the average calculation step, the standard deviation first calculation step, and the range calculation step according to an embodiment of the present invention.
[0022] FIG. 5 is an original waveform diagram for explaining a filtering step according to an embodiment of the present invention;
[0023] Figure 6 is a diagram of a processing waveform generated by a processing waveform generation step according to an embodiment of the present invention.
[0024] Figure 7 is a drawing for explaining a leak location detection step according to an embodiment of the present invention.
[0025] Figure 8 is a flowchart of a method for determining and processing a water leak level in a water pipe using a smart leak detector according to an embodiment of the present invention.
[0026] FIG. 9 is a drawing for explaining a method for determining and processing a water leak level in a water pipe using a smart leak detector according to an embodiment of the present invention.
[0027] The method for detecting a leak location in a water pipe using a smart leak detector of the present invention comprises, as illustrated in FIG. 1, a measurement step (S1), an absolute value generation step (S2), an interval division step (S3), an average calculation step (S4), a first standard deviation calculation step (S5), a range calculation step (S6), a filtering step (S7), a first correlation step (S8), a processed waveform generation step (S9), a time difference calculation step (S10), and a leak location detection step (S11).
[0028] All of the above steps are done through computer programs, etc.
[0029] The above measurement step (S1) is a step of measuring the original waveform of the water leak sound of the water pipe using the first leak detector and the second leak detector.
[0030] Figure 2 shows the original waveform measured by the above measurement step (S1).
[0031] The first and second leak detectors are installed in the water pipe at a constant distance from each other along the length of the water pipe, as shown in Fig. 8.
[0032] Specifically, the first original waveform of the water pipe leak sound is measured using the first leak detector, and the second original waveform of the water pipe leak sound is measured using the second leak detector.
[0033] The above first and second leak detectors measure leak noise through separate inputs and automatically generate the first and second original waveforms.
[0034] Measurement of the first source waveform using the first leak detector and measurement of the second source waveform using the second leak detector are performed simultaneously at the same time.
[0035] In FIGS. 2 to 6, only one waveform is shown for convenience of explanation, but in reality, there is a first waveform expressed through the first leak detector and a second waveform expressed through the second leak detector.
[0036] The above absolute value generation step (S2) is a step of generating a first absolute waveform by converting the amplitude of the first original waveform into an absolute value using a computer program, and generating a second absolute waveform by converting the amplitude of the second original waveform into an absolute value.
[0037] Therefore, both the first absolute waveform and the second absolute waveform have waveforms converted to values greater than or equal to 0.
[0038] Figure 3 shows an absolute waveform converted by the absolute value generation step (S2).
[0039] The above section division step (S3) is a step of dividing the first absolute waveform and the second absolute waveform into multiple sections with the same time unit.
[0040] This section division step (S3) is illustrated as being divided into multiple sections in Fig. 4.
[0041] For example, if the first original waveform and the second original waveform are measured for 16 seconds in the above measurement step (S1) and converted into the first absolute waveform and the second absolute waveform in the above absolute value generation step (S2), the first absolute waveform and the second absolute waveform are each divided into 32 sections in 0.5 second units in the above section division step (S3).
[0042] The above average calculation step (S4) is a step of calculating a first average value, which is an average of each section of the first absolute waveform, and calculating a second average value, which is an average of each section of the second absolute waveform.
[0043] That is, when the first absolute waveform is divided into 32 sections, the average value of each of the 32 sections is calculated and set as the first average value for each of the 32 sections.
[0044] Therefore, the first average values for each of the 32 sections may be the same or different.
[0045] In addition, when the second absolute waveform is divided into 32 sections, the average value of each of the 32 sections is calculated and set as the second average value for each of the 32 sections.
[0046] Therefore, the second average values for each of the 32 sections may be the same or different.
[0047] The average value calculated in this average calculation step (S4) is displayed as an average value by section in Fig. 4.
[0048] The above standard deviation first calculation step (S5) is a step of calculating the first standard deviation, which is the standard deviation of the plurality of first average values calculated in the above average calculation step (S4), and calculating the second standard deviation, which is the standard deviation of the plurality of second average values.
[0049] The above range calculation step (S6) is a step of calculating a first specific range by adding or subtracting the first standard deviation from the first average value of each section of the first absolute waveform, and calculating a second specific range by adding or subtracting the second standard deviation from the second average value of each section of the second absolute waveform.
[0050] The graph that has completed the standard deviation first calculation step (S5) and range calculation step (S6) is shown in Figure 4.
[0051] The above filtering step (S7) is a step for filtering out unnecessary noise, etc. generated during waveform measurement.
[0052] This filtering step (S7) removes the first abnormal section of the first original waveform corresponding to the first average value section if there is a value among the plurality of first average values that deviates from the upper and lower amplitudes of the first specific range.
[0053] And, if there is a value among the plurality of second average values that is outside the upper and lower amplitude of the second specific range, the second abnormal section of the second original waveform corresponding to the second average value section is removed.
[0054] The above first and second abnormal sections refer to sections of the original waveform that contain unnecessary noise.
[0055] Figure 4 shows a section that deviates from a specific range of upper and lower amplitudes among the average values, and Figure 5 shows an abnormal section of the original waveform corresponding to the waveform shown in Figure 4, and such an abnormal section is deleted.
[0056] Therefore, through the filtering step (S7), sections with large noise are removed from the first and second original waveforms.
[0057] The above first correlation step (S8) is a step of removing a section of the first original waveform corresponding to the second abnormal section from the first original waveform that has completed the filtering step (S7), and removing a section of the second original waveform corresponding to the first abnormal section from the second original waveform that has completed the filtering step (S7).
[0058] In other words, in the first correlation step (S8), if a second abnormal section exists in the second original waveform, the same section of the first original waveform corresponding to the time zone where the second abnormal section exists, i.e., the section in the same time zone, is removed from the first original waveform, and if a first abnormal section exists in the first original waveform, the same section of the second original waveform corresponding to the time zone where the first abnormal section exists, i.e., the section in the same time zone, is removed from the second original waveform.
[0059] The above-mentioned processing waveform generation step (S9) is a step of generating a first processing waveform by connecting the first original waveforms that have completed the first correlation step (S8) and generating a second processing waveform by connecting the second original waveforms that have completed the first correlation step (S8).
[0060] Through this, both the first original waveform and the second original waveform are correlated so that all sections corresponding to the first and second abnormal sections are removed, and then converted into waveforms of the same length, i.e., for the same time.
[0061] Figure 6 shows a processed waveform that has completed the first correlation and processed waveform generation step (S9).
[0062] The above time difference calculation step (S10) calculates a time difference (△T), which is the difference between the time (T1) at which the leak sound reaches the point where the first leak detector is installed and the time (T2) at which the leak sound reaches the point where the second leak detector is installed, by applying a cross-correlation operation to the first and second processed waveforms.
[0063] The specific method of cross-correlation operation is disclosed in Patent Publication No. 10-2022-0165690, etc., and since it is sufficient to use conventional technology, a detailed description thereof is omitted.
[0064] The above-mentioned water leak location detection step (S11) is a step for detecting the water leak location of a water pipe using the above-mentioned time difference.
[0065] In the above leak location detection step (S11), the leak location (L) is detected by the equation below.
[0066] Leakage location (L) = center point (d / 2) + (△T * vibration sound transmission speed) / 2
[0067] Here, the center point (d / 2) is half the distance (d) between the first and second leak detectors, and the vibration sound transmission speed is determined by variables including the material and diameter of the water pipe.
[0068] The above variables include water temperature, inner surface resistance of the water pipe, water pressure, water velocity, and corrosion status of the water pipe.
[0069] To explain more easily, the contents related to leak location detection are shown in Fig. 7.
[0070] For example, if the water pipe is SP (steel pipe) and has a diameter of 2200 mm, the vibration sound transmission speed is 910 m / s.
[0071] If the leak location calculated in the above leak location detection step (S11) is 0, the leak location becomes the center point (d / 2).
[0072] And, if the leak location calculated in the above leak location detection step (S11) is positive, it means a location closer to the second leak detector from the central point, and if the calculated leak location is negative, it means a location closer to the first leak detector from the central point.
[0073] All of the above steps of the present invention are performed through a computer program.
[0074] By using the method for detecting a leak location in a water pipe using the smart leak detector of the present invention as described above, unnecessary noise (e.g., first and second abnormal sections) is clearly removed and only waveforms with almost no noise are used to detect a leak location in a water pipe, so that the leak location can be detected more accurately than with existing methods, and additional measures can be taken quickly accordingly.
[0075]
[0076] Meanwhile, the method for determining and processing a water leak level in a water pipe using a smart water leak detector of the present invention is performed after detecting a water leak location in a water pipe using the method for detecting a water leak location in a water pipe using the smart water leak detector described above.
[0077] The method for determining and processing a water leak level in a water pipe using a smart water leak detector of the present invention is composed of the method for detecting a location of a water leak in a water pipe using the smart water leak detector described above, and further including a second correlation step (S21), a second standard deviation calculation step (S22), a detection reference step (S23), a leak level designation step (S24), a leak level determination step (S25), and a notification step (S26) as shown in FIG. 8.
[0078] All of the above steps are executed through a computer program.
[0079] FIG. 9 is a drawing illustrating a method for determining and handling a water leak level in a water pipe using a smart water leak detector of the present invention.
[0080] The above second correlation step (S21) is a step of generating a correlation graph representing the correlation between the first processed waveform and the second processed waveform.
[0081] Figure 9 shows such a correlation graph.
[0082] In Fig. 9, the above-described time difference (△T) is shown, and since this time difference is a negative number (-3.7498 sec), it can be seen that the leak location is located close to the first leak detector from the center point.
[0083] The method for generating a correlation graph showing the correlation between two waveforms can be sufficiently achieved by using a conventionally known method, so a detailed description thereof is omitted.
[0084] The above standard deviation second calculation step (S22) is a step for calculating the third standard deviation, which is the standard deviation of the correlation graph.
[0085] In Figure 9, the positive third standard deviation and the negative third standard deviation are indicated by horizontal lines.
[0086] The above detection standard step (S23) is a step of calculating a plurality of detection standard values by multiplying a plurality of standard level values by the third standard deviation.
[0087] The above leakage level designation step (S24) is a step of determining a plurality of leakage levels corresponding to the above reference level value in advance.
[0088] Figure 9 shows a table for each leakage level.
[0089] For example, if the water pipe is made of plastic, the leakage level is set to level 1 when the reference level value is 7, and the leakage level is set to level 2 when the reference level value is 8.
[0090] In the above leakage level designation step (S24), even if the leakage level is the same, the standard level value is changed depending on the material of the water pipe, and the same leakage level is designated.
[0091] At the same leakage level, the reference level value is greater when the water pipe material is made of metal than when it is made of plastic.
[0092] For example, if the water pipe is made of plastic, the leakage level is designated as level 1 when the reference level value is 7, and if the water pipe is made of metal, the leakage level is designated as level 1 when the reference level value is 8.
[0093] As above, even for the same leakage level, the standard level value is set differently depending on the material of the water pipe because the transmission speed of vibration sound is different depending on the material of the water pipe.
[0094] Therefore, a water pipe made of a material with a fast vibration sound transmission speed will have the same leakage level as a water pipe made of a material with a slow vibration sound transmission speed due to a larger reference level value.
[0095] The above leakage level determination step (S25) is a step of comparing the maximum amplitude value of the correlation graph with the detection reference value to obtain a detection reference value corresponding to the maximum value among the detection reference values exceeded by the maximum amplitude value of the correlation graph, and determining the leakage level corresponding to the reference level value multiplied to obtain the detection reference value in the detection reference step (S23) as the final leakage level.
[0096] In Fig. 9, it is shown that the maximum amplitude value of the correlation graph exceeds the stage 1 detection criterion value, and through this, it can be seen that the current leakage level is stage 1.
[0097] By the above leakage level determination step (S25), the preset leakage level can be automatically determined according to the material of the water pipe.
[0098] The above notification step (S26) is a step for notifying the outside of various preset measures according to the final leakage level.
[0099] By using the above method, when a leak occurs in a water pipe, the manager can easily determine the level of the leak, that is, the intensity of the leak, and quickly take appropriate measures.
[0100] In addition, by using data previously stored through experience, etc., it is possible to predict in advance the time it will take for a level 1 leak to change to level 2 or 3, and take appropriate precautions in advance.
[0101] The method for detecting a leak location in a water pipe using a smart leak detector of the present invention and the method for determining and processing a leak level in a water pipe using the same are not limited to the above-described embodiments, and can be implemented by various modifications within the scope permitted by the technical idea of the present invention.
[0102] The present invention is recognized as having industrial applicability because it can be applied to a method for detecting a location of a water leak in a water pipe and a method for determining and treating a level of a water leak in a water pipe.
Claims
1. A measuring step of measuring a first original waveform of a water pipe leak sound using a first leak detector during the same period of time, and measuring a second original waveform of the water pipe leak sound using a second leak detector placed apart from the first leak detector; An absolute value generation step of generating a first absolute waveform by converting the amplitude of the first original waveform into an absolute value and generating a second absolute waveform by converting the amplitude of the second original waveform into an absolute value; An interval division step of dividing the first absolute waveform and the second absolute waveform into multiple intervals with the same time unit; An average calculation step of calculating a first average value, which is an average of each section of the first absolute waveform, and calculating a second average value, which is an average of each section of the second absolute waveform; A first standard deviation calculation step of calculating a first standard deviation, which is the standard deviation of a plurality of first average values, and calculating a second standard deviation, which is the standard deviation of a plurality of second average values; A range calculation step of calculating a first specific range by adding or subtracting the first standard deviation from the first average value of each section of the first absolute waveform, and calculating a second specific range by adding or subtracting the second standard deviation from the second average value of each section of the second absolute waveform; A filtering step of removing a first abnormal section of a first original waveform corresponding to the first average value section if there is a value among a plurality of first average values that is outside the upper and lower amplitude of the first specific range, and of removing a second abnormal section of a second original waveform corresponding to the second average value section if there is a value among a plurality of second average values that is outside the upper and lower amplitude of the second specific range; A first correlation step of removing a section of the first original waveform corresponding to the second abnormal section from the first original waveform that has completed the filtering step, and removing a section of the second original waveform corresponding to the first abnormal section from the second original waveform that has completed the filtering step; A processed waveform generation step of generating a first processed waveform by connecting first original waveforms that have completed the first correlation step, and generating a second processed waveform by connecting second original waveforms that have completed the first correlation step; A time difference calculation step for calculating a time difference (△T), which is the difference in time between the time the leak sound reaches the point where the first leak detector is installed and the time the leak sound reaches the point where the second leak detector is installed, by applying a cross-correlation operation to the first and second processed waveforms; A method for detecting a leak location in a water pipe using a smart leak detector, characterized in that it comprises a step of detecting a leak location in a water pipe using the above time difference.
2. In claim 1, In the above leak location detection step, the leak location (L) is detected by the following equation: Leakage location (L) = center point (d / 2) + (△T * vibration sound transmission speed) / 2 The center point (d / 2) is half the distance (d) between the first and second leak detectors, and the vibration sound transmission speed is determined by variables including the material and diameter of the water pipe. A method for detecting a leak location in a water pipe using a smart leak detector, characterized in that if the calculated leak location is positive, it is a location closer to the second leak detector from the central point, and if the calculated leak location is negative, it is a location closer to the first leak detector from the central point.
3. After detecting the location of a water leak in a water pipe by the detection method of claim 1, A second correlation step of generating a correlation graph representing the correlation between the first processing waveform and the second processing waveform; A second standard deviation calculation step for calculating the third standard deviation, which is the standard deviation of the above correlation graph; A detection standard step of calculating a plurality of detection standard values by multiplying a plurality of standard level values by the third standard deviation; A leakage level designation step for determining a plurality of leakage levels corresponding to the above reference level value in advance; A leakage level determination step of comparing the maximum amplitude value of the correlation graph with the detection criterion value to obtain a detection criterion value corresponding to the maximum value among the detection criterion values exceeded by the maximum amplitude value of the correlation graph, and determining the leakage level corresponding to the criterion level value multiplied to obtain the detection criterion value in the detection criterion step as the final leakage level; It is comprised of a notification step that notifies preset actions according to the final leakage level; A method for determining and processing a water pipe leak level using a smart leak detector, characterized in that in the above-mentioned leak level designation step, even if the leak level is the same, the reference level value is changed depending on the material of the water pipe so that the same leak level is designated.
4. In claim 3, A method for determining and processing a water leak level of a water pipe using a smart water leak detector, characterized in that, at the same water leak level, the reference level value is greater when the material of the water pipe is made of metal than when the material is made of plastic.
Citation Information
Patent Citations
Leakage signal analysis method
KR101447925B1
ICT Real-time leak detection system for ICT-based water distribution
KR101899112B1
Semiconductor device and method for fabricating the same
KR1020240081932A
Spicy noodles taste powdered natural salad seasoning
KR102344663B1
Method of detecting location of water leakage in water pipe, and Method of determining and processing water leakage level in water pipe using the method
KR102572640B1
