METHOD FOR DETECTING AND / OR ANALYSING A LEAK IN A LIQUID MEDIA PIPELINE, IN PARTICULAR A WATER PIPELINE

Description

[0001] The invention relates to a method for detecting and / or analyzing a leak in a line for fluid media, in particular a water line, especially preferably a drinking water line, for example an underground drinking water line.

[0002] Typically, noise and / or frequency loggers are used to locate leaks in a pipe network. These noise and / or frequency loggers can be installed at numerous measuring points and detect leak noises by recording the sounds generated by leaks at predetermined measurement intervals and processing this signal. Typically, such noise and / or frequency loggers are installed directly on the pipes carrying liquid media, particularly water pipes, especially drinking water pipes, or on pipe fittings such as valves, service valves, or underground or aboveground hydrants.

[0003] A leak in a pipe causes a pressure fluctuation at the leak point. This pressure fluctuation generates a vibration at the leak, which propagates through the pipe and can be recorded with sensitive sensors such as microphones or piezoelectric sensors. By analyzing the noise data, a leak in the pipe can be detected. A pipe without a leak does not produce any audible pressure fluctuations. The noise sensors are usually mounted on a slide bar that is mechanically connected to the pipe. Often, not only the noise generated by the leak is recorded, but also various background noises. These background noises are caused by pumping or humming from electrical equipment, such as transformer stations, substations, air conditioners, etc. The audio signal recorded by the sensor therefore usually consists of a mixture of the leak noise and background noise.Document EP2097728B1 (CYBERNETIX [FR]) dated November 9, 2016 (2016-11-09) describes a system for detecting and localizing events (mechanical impact, leak or pressure transient) in a fluid transport channel.

[0004] The object of the invention is therefore to provide a new method for detecting and / or analyzing a leak in a pipe for fluid media, in particular a water pipe, and in particular a method with which disturbance noises can be distinguished at least to a large extent from leak noises.

[0005] This problem is solved by a method for detecting and / or analyzing a leak in a (pipe) line for fluid media, in particular a water line, according to the features of claim 1. Advantageous embodiments and further developments are specified in the dependent claims.

[0006] The method according to the invention comprises the following steps: a) Providing at least one noise and / or frequency logger comprising a sensor and a processing unit, b) Attaching the noise and / or frequency logger to the pipeline or to fittings of the pipeline, c) Measuring a noise at the pipeline by the sensor and determining a noise level by means of the processing unit, c1) wherein a frequency analysis, in particular a Fourier analysis, preferably a Fast Fourier Transform, of the noise level is performed by means of the processing unit to determine a frequency spectrum, c2) wherein the determined frequency spectrum is divided into two or more frequency bands by means of the processing unit, c3) wherein a threshold value is set for each frequency band by means of the processing unit, in particular individually, and / or stored in the processing unit, which is greater than the maximum noise level in this frequency band when measured on the leak-free pipeline,c4) wherein a leak in the pipeline can be detected and / or analyzed by the computing unit based on the exceeding of one or more threshold values.

[0007] The sensor is or preferably comprises a noise sensor and / or a piezo sensor and / or a piezo microphone, in particular for measuring noise by means of vibration and / or structure-borne sound at the pipe wall, and / or a hydrophone for measuring noise directly at the water column.

[0008] The threshold values ​​stored in the processing unit can be set and / or adjusted by a user, or they can be set and / or adjusted manually. Furthermore, the threshold values ​​can also be set and / or adjusted automatically by algorithms and / or artificial intelligence.

[0009] The advantages of the invention lie particularly in the fact that the method according to the invention allows for better differentiation between leakage noise and background noise. In addition to the pure signal level, the frequency components of the signal are evaluated using the method according to the invention. The high frequency components are attenuated by the sound waves transmitted through the pipe. The closer the sensor is positioned to the leak, the more high frequency components are present. Low frequency components remain perceptible over a long distance. Typical leakage noises are often broadband in a frequency range of 150-600 Hz. Background noise from the power grid is usually found as narrowband frequencies at 50 / 60 Hz and 100 / 120 Hz.In known leak detection methods, the presence or absence of a leak is primarily determined by the noise level and by evaluating a single significant frequency value of the recorded noise. However, this becomes problematic as soon as an interfering noise generates a higher noise level than the leak noise itself. The invention allows conclusions to be drawn, based on the frequency value, as to whether the noise was primarily caused by a leak or by a network interference source. By setting the threshold values ​​for each frequency band, frequencies in which interference frequently occurs can be ignored.

[0010] The spectrum of the leakage noise is calculated using an FFT (Fast Fourier Transform / frequency analysis) and consists of a relatively large number of values ​​(e.g., 4096). Since this is a large amount of data for a sensor, it makes sense to reduce the data, especially to 4 to 128 frequency bands, for example, to 16 frequency bands. This reduces transmission time and thus power and data volume. However, the range can also be chosen to be larger. It could also be, for example, 2 to 4096 frequency bands. The reduction is primarily done to decrease the amount of data.

[0011] According to a preferred embodiment of the invention, a predetermined number of frequency bands is defined by the processing unit and / or stored in the processing unit, wherein between 4 and 128 frequency bands are provided, particularly preferably 16 frequency bands. It can be provided that the defined and / or stored threshold value of each frequency band is between 1% and 100% greater than the maximum noise level in that frequency band when measured on the leak-free line.

[0012] Furthermore, it may be provided that the distance of the threshold value of each frequency band to the maximum noise level in this frequency band is individually determined for each frequency band by means of the processing unit during a measurement on the leak-free line and / or is stored in the processing unit.

[0013] According to a further development of the invention, the spacing in a frequency band is set larger in frequency ranges of known and / or stored interference noise than the spacing in a frequency band in frequency ranges without known and / or stored interference noise.

[0014] According to the invention, the threshold value of the frequency bands in frequency ranges where noise level increases due to leakage noise typically occur, i.e., the threshold value of the frequency bands in the frequency range 150Hz to 600Hz, is set and / or stored in the computing unit at a lower level above the maximum noise level in these frequency bands when measured on the leak-free line, than the threshold value of the frequency bands in the frequency range below 100Hz and / or above 800Hz.

[0015] The width of the frequency bands can be individually set using the processing unit and / or stored in the processing unit.

[0016] According to the invention, the width of the frequency bands specified or provided in the frequency range of typical leakage noises, i.e. in the frequency range of 150Hz to 600Hz, is less than the width of the frequency bands specified or provided in the usable range, in particular in the frequency range below 100Hz and / or above 800Hz.

[0017] Furthermore, according to the invention, the width of the frequency bands in frequency ranges of known and / or stored interference noises, in particular in the frequency range below 100Hz and / or above 800Hz, is selected or set to be smaller than the average width of the frequency bands in the frequency range below 100Hz and / or above 800Hz.

[0018] According to one embodiment of the invention, a warning signal is generated by the processing unit each time a threshold is exceeded. This warning signal is processed, in particular, by the processing unit and / or a central processing unit to detect or analyze a leak. Optionally, each warning signal is weighted by the processing unit, with the weight depending on the frequency band in which the threshold is exceeded. Furthermore, the percentage by which the threshold is exceeded and / or multiple exceedances of the threshold can also be included in or factored into the weighting.

[0019] Preferably, the weighting of exceeding the threshold for the frequency bands in the frequency range 150Hz to 600Hz is higher than the weighting of exceeding the threshold for the frequency bands in the frequency range below 100Hz and / or above 800Hz.

[0020] The computing unit may be configured to issue an alarm message or transmit it to a central computing unit if one or more warning signals are generated.

[0021] The system may also be configured to issue an alarm message or transmit it to a central processing unit if a predetermined overall weighting value is reached or exceeded after weighting the generated warning signals. This overall weighting value can be a measure of the probability of a leak. The more thresholds are exceeded and / or the higher the weighting of the corresponding frequency bands, the greater the probability of a leak.

[0022] One embodiment of the invention provides that several noise and / or frequency loggers are arranged, in particular at predetermined intervals, on the line for fluid media, in particular the water line, or on fittings of the line, wherein each noise and / or frequency logger performs the following steps: a) Measuring a noise at the line by the sensor and determining a noise level using the processing unit, c1) wherein a frequency analysis, in particular a Fourier analysis, of the noise level is carried out using the processing unit to determine a frequency spectrum, c2) wherein the determined frequency spectrum is divided into two or more frequency bands using the processing unit, c3) wherein a threshold value is set for each frequency band using the processing unit, in particular individually, and / or stored in the processing unit, which is greater than the maximum noise level in this frequency band when measured on the leak-free line, c4) wherein a leak at the line can be detected and / or analyzed using the processing unit based on the exceedance of one or more threshold values.

[0023] According to further training, the computing units of the noise and / or frequency loggers transmit alarm messages to a central computing unit, for example via a cloud, particularly in the event of a leak, with each alarm message including the noise level and the frequency of the noise level when the threshold is exceeded.

[0024] It may be provided that the alarm messages, in particular alarm messages from several computing units, are processed by the central computing unit to detect a leak and / or to analyze it, in particular with regard to the size and / or position of the leak.

[0025] Furthermore, it may be provided that the position of each of the noise and / or frequency loggers and the route of the pipe, in particular the water pipe, is stored in the central computing unit, and that the central computing unit determines the size and / or location of the leak based on the position of the noise and / or frequency loggers and the route of the pipe as well as the alarm messages.

[0026] The noise and / or frequency loggers can be positioned, for example, at intervals of 50–200 m along the pipeline. The closer the sensor is to the leak, the better the leak noise can be detected, resulting in a higher signal level (loudness of the noise). The signal level is also higher the larger the leak. In particular, the signal also contains more high-frequency components.

[0027] The invention is further explained below with regard to other features and advantages by means of a description of an exemplary embodiment and with reference to the accompanying schematic drawing. FIG. 1 Figure 1 shows an example of a frequency spectrum of a noise level which is divided into several frequency bands according to the inventive method and wherein a threshold value is defined for each frequency band.

[0028] The method according to the invention is intended for detecting and / or analyzing a leak in a pipe carrying fluid media. Water pipes, in particular drinking water pipes, are used as an example below. According to the method according to the invention, in step a) at least one noise and / or frequency logger is provided, comprising a sensor for measuring noise and a processing unit.

[0029] According to step b), the noise and / or frequency logger is attached to a water pipe, in particular a drinking water pipe, or to a fitting of the water pipe, for example a valve rod, which is mechanically connected to the water pipe.

[0030] In step c), the noise at the water pipe is measured by the sensor and a noise level of 10 is determined using the computing unit.

[0031] Using the processing unit, a frequency analysis, in particular a Fourier analysis, for example a Fast Fourier Transform, of the noise level 10 is then performed in step c1) to determine a frequency spectrum 20. This frequency spectrum 20 is in FIG. 1 depicted.

[0032] In step c2), the determined frequency spectrum 20 is subdivided into 16 frequency bands 21 by the processing unit. This subdivision into 16 frequency bands is preset. The width 22 of these frequency bands 21 is individually defined for each frequency band 21 by the processing unit. The intended width 22 of the frequency bands 21 can be stored in the processing unit. The width 22 of the frequency bands 21 defined or intended in the frequency range 150 Hz to 600 Hz is set smaller than the width 22 of the frequency bands 21 defined or intended in the frequency range below 100 Hz and / or above 800 Hz. This is based on the fact that leakage noise typically occurs in the frequency range 150 Hz to 600 Hz and should be detected particularly efficiently there.In addition, the width 22 of the frequency bands 23 in frequency ranges of known and stored interference noises, in the frequency range below 100Hz and / or above 800Hz, is set smaller than the average width 22 of the frequency bands in the frequency range below 100Hz and / or above 800Hz.

[0033] According to step c3), a threshold value 30 is individually defined for each of the 16 frequency bands by the processing unit. This threshold value can, for example, be stored in the processing unit. The threshold value 30 of each frequency band 21 is higher than the maximum noise level 11 in that frequency band 21 when measured on the leak-free water pipe. Furthermore, the threshold value 30 of the frequency bands 21 in the frequency range 150 Hz to 600 Hz is set by the processing unit at a lower level above the maximum noise level 11 in these frequency bands 21 when measured on the leak-free water pipe than the threshold value 30 of the frequency bands 21 in the frequency range below 100 Hz and / or above 800 Hz. This is based on the fact that leak noises typically occur in the frequency range 150 Hz to 600 Hz and should be detected particularly efficiently there.

[0034] By individually setting the threshold 30, the distance 31 between the threshold 30 of each frequency band 21 and the maximum noise level 11 in that frequency band 21 is also individually determined for each frequency band 21 by means of the processing unit. In a frequency band 23, the distance 31 is set larger in frequency ranges containing known and stored interference noise than in a frequency band 21 in frequency ranges without known interference noise. In this way, the "fineness" of the threshold 30 can be adjusted. The threshold is set less finely in frequency bands 21 where known interference noise is possible or present, in order to at least largely avoid a warning signal caused by interference noise.

[0035] In step c4), a leak in the water pipe is detected and / or analyzed by the processing unit based on the exceedance of one or more threshold values ​​30. Each time a threshold value 30 is exceeded, a warning signal is generated by the processing unit, which is then processed by the processing unit or a central processing unit, in particular to detect or analyze a leak.

[0036] Each warning signal can be weighted by the processing unit, with the weight depending on the frequency band 21 in which the threshold 30 is exceeded. Furthermore, the percentage by which the threshold 30 is exceeded and / or multiple exceedances of the threshold 30 can also be included in the weighting. The weighting for exceeding the threshold 30 in frequency bands 21 in the frequency range of 150 Hz to 600 Hz is higher than the weighting for exceeding the threshold 30 in frequency bands 21 in the frequency range below 100 Hz and / or above 800 Hz. This is based on the fact that leakage noises typically occur in the frequency range of 150 Hz to 600 Hz and should be detected particularly efficiently there.

[0037] The processing unit issues an alarm message or transmits an alarm message to a central processing unit if either one or more warning signals are generated or if, after weighting the generated warning signals, a predetermined total weighting value is reached or exceeded.

[0038] A further development of the procedure provides that several noise and / or frequency loggers are arranged at predetermined intervals, for example 50m - 200m, on the water pipe or on fittings of the water pipe, with each noise and / or frequency logger performing steps c) to c4).

[0039] The processing units of the noise and / or frequency loggers transmit an alarm message to a central processing unit based on the criteria described above and / or in the event of a leak. Each alarm message includes the noise level (10) and the frequency of the noise level (10) when the threshold (30) is exceeded. The central processing unit processes the alarm messages to detect a leak and analyze its size and location. The position of each noise and / or frequency logger and the pipe layout are stored in the central processing unit's database. Using this information, along with the alarm messages, the central processing unit determines the size and location of the leak. Reference symbol list

[0040] 10 Noise level 11 Maximum noise level 20 Frequency spectrum 21 Frequency band 22 Width 23 Frequency band in the frequency range of a known disturbance 30 Threshold 31 Distance

Claims

1. Method for determining and / or analyzing a leak in a line for fluid media, in particular a water line, the method comprising the following steps: a) providing at least one noise and / or frequency logger comprising a sensor and a computing unit, b) attaching the noise and / or frequency logger to the line or to fittings of the line, c) measuring noise on the line via the sensor and determining a noise level (10) by means of the computing unit, c1) wherein a frequency analysis of the noise level (10) is performed by the computing unit to determine a frequency spectrum (20), c2) wherein the determined frequency spectrum (20) is divided into two or more frequency bands (21) by means of the computing unit, c3) wherein a threshold value (30) is determined for each frequency band by means of the computing unit and / or is stored in the computing unit, which threshold value is greater than the maximum noise level (11) in this frequency band (21) during a measurement of the leak-free line, c4) wherein, based on one or more threshold values (30) being exceeded, a leak in the line can be determined and / or analyzed or is detected and / or analyzed by means of the computing unit, characterized in that that the threshold value (30) of the frequency bands (21) in the frequency range of 150Hz to 600Hz is set by means of the computing unit and / or is stored in the computing unit to be not as high above the maximum noise level (11) in these frequency bands (21) during a measurement of the leak-free line than the threshold value (30) of the frequency bands (21) in the frequency range below 100Hz and / or above 800Hz, and that the width (22) of the frequency bands (21) set or provided in the frequency range of 150 Hz to 600 Hz is less than the width (22) of the frequency bands (21) set or provided in the frequency range below 100 Hz and / or above 800 Hz.

2. Method according to claim 1, characterized in that a provided number of frequency bands (21) is set by means of the computing unit and / or is stored in the computing unit, wherein between 4 and 128 frequency bands (21), preferably 16 frequency bands (21), are provided.

3. Method according to claim 1 or 2, characterized in that the set and / or stored threshold value (30) of each frequency band (21) is between 1% and 100% greater than the maximum noise level (11) in this frequency band (21) during a measurement of the leak-free line.

4. Method according to one of the preceding claims, characterized in that the distance (31) of the threshold value (30) of each frequency band (21) at the maximum noise level (11) in this frequency band (21) during a measurement of the leak-free line is set individually for each frequency band (21) by means of the computing unit and / or is stored in the computing unit.

5. Method according to one of the preceding claims, characterized in that the distance (31) in a frequency band (23) in frequency ranges of interference noises that are known and / or stored in the computing unit is set greater than the distance in a frequency band (21) in frequency ranges without interference noises that are known and / or stored in the computing unit.

6. Method according to one of the preceding claims, characterized in that the width (22) of the frequency bands (21) is set individually by means of the computing unit and / or is stored in the computing unit.

7. Method according to one of the preceding claims, characterized in that the width (22) of the frequency bands (23) in frequency ranges of interference noises that are known and / or stored in the computing unit is selected or set to be less on the average than the width (22) of the frequency bands in the frequency range below 100 Hz and / or above 800 Hz.

8. Method according to one of the preceding claims, characterized in that a warning signal is generated by the computing unit each time a threshold value (30) is exceeded.

9. Method according to claim 8, characterized in that each warning signal is given a weighting by the computing unit, wherein the weighting depends on the frequency band (21) in which the threshold value (30) is exceeded.

10. Method according to claim 9, characterized in that the weighting of exceeding the threshold value (30) for the frequency bands (21) in the frequency range 150 Hz to 600 Hz is higher than the weighting of exceeding the threshold value (30) for the frequency bands (21) in the frequency range below 100 Hz and / or above 800 Hz.

11. Method according to one of claims 8 to 10, characterized in that the computing unit outputs an alarm message or transmits it to a central computing unit if one or more warning signals are generated.

12. Method according to one of claims 8 to 11, characterized in that the computing unit outputs an alarm message or transmits it to a central computing unit if, after weighting the generated warning signals, a provided total weighting value is reached or exceeded.

13. Method according to one of the preceding claims, characterized in that a plurality of noise and / or frequency loggers are arranged on the line or on fittings of the line, each noise and / or frequency logger performing steps c) to c4).

14. Method according to claim 13, characterized in that the computing units of the noise and / or frequency loggers transmit alarm messages to a central computing unit, each alarm message comprising the noise level (10) and the frequency of the noise level (10) when the threshold value (30) is exceeded.

15. Method according to claim 14, characterized in that the alarm messages are calculated together by the central computing unit in order to determine and / or analyze a leak.

16. Method according to claims 13 to 15, characterized in that the position of each of the noise and / or frequency loggers and the line characteristic is stored in the central computing unit, and the central computing unit determines the size and / or position of the leak based on the position of the noise and / or frequency loggers and the line characteristic as well as the alarm messages.