GB2584633A

GB2584633A - Flow monitoring apparatus

Info

Publication number: GB2584633A
Application number: GB1907784.1A
Authority: GB
Inventors: Trow Stuart
Original assignee: Invenio Systems Ltd
Current assignee: Invenio Systems Ltd
Priority date: 2019-05-31
Filing date: 2019-05-31
Publication date: 2020-12-16
Anticipated expiration: 2039-05-31
Also published as: GB201907784D0; GB2584633B

Abstract

A flow monitoring apparatus comprising first 7 and second 8 audio transducers (e.g. microphones), one of the first and second transducers for location proximate a pipeline 2 (e.g. proximate (or on) a shut-off valve 1) and the other for location at a distance from the pipeline and the other transducer. A processing unit is connected to the transducers to receive signals therefrom, the processing unit connected to a power source and including a micro controller and data transfer means. It can record and process data so that it may be used to assess activity in the valve 1 and/or the pipeline 2. The signal from microphone 8 is representative of a background sound and provides a reference signal. The signal recorded by microphone 7 induces both background sound and sound associated with flow through the pipeline / valve. Data processing software may be used for processing data recorded by the apparatus and receive the recorded data from the memory of the apparatus, compare the received data with a database of data representing known fluid flows through known pipelines and/or valves, providing a classification of the received data.

Description

Flow Monitoring Apparatus

Field of the Invention

The present invention relates to an apparatus for monitoring the flow of fluid through a pipe. The apparatus is particularly useful in detecting leaks in buildings down stream of a stopcock.

Background of the Invention

Water is supplied to end user properties via a network of pipes and control valves. In many circumstances the property is provided with a meter which has a stopcock associated therewith, the stop cock being a valve which is operable to turn off the flow of water to the property. Not all properties are metered. Nevertheless, unmetered properties are provided with a stopcock that may be used to shut off supply to the property.

It is inevitable that water supply networks suffer from leaks. Leakage in the water supply network in the United Kingdom is estimated at around three million litres of water each day In order to reduce the amount of water leaked from the network, water companies are given targets for leakage reduction and are penalised if such targets are not met.

It has been suggested that a proportion of leakage from the network is not from the network itself, but occurs within the properties of end users. For example, water consumption may be due to leaking pipes, dripping taps, malfunctioning water tanks, cisterns, and domestic appliances and the like, or simply because a tap has been left on unintentionally Different types of apparatus for detecting leaks are known. For example, in GB2537013 the Applicant describes a method of monitoring water consumption that is based on measuring and monitoring over time the temperature of water flowing from the water network to end users.

Other leak detection systems use acoustic transducers to measure leakage. EP3112820 describes a consumption meter arranged to measure a flow rate of a fluid supplied in a flow tube, first and second ultrasonic transducers are arranged at the flow tube for transmitting and receiving ultrasonic signals transmitted through the fluid. A first control circuit is arranged for operating the first and second ultrasonic transducers, and generates a signal indicative of the flow rate of the fluid accordingly. A second control circuit arranged for operating a sensor arranged at the flow tube for detection of vibro-acoustic signals of the flow tube generates a signal indicative of noise level of the flow tube. This noise signal can be used to detect and locate leaks.

It would be desirable to provide an apparatus and method of operation of such apparatus capable of detecting consumption of water in an end user's property, that is down stream of a valve when the valve is open. Summary of the Invention According to a first aspect of the invention there is provided a flow monitoring apparatus comprising first and second audio transducers, one of the first and second transducers for location proximate a pipeline and the other of the first and second transducers for location at a distance from the pipeline and the transducer located proximate the pipeline, and a processing unit, wherein the first and second audio transducers are connected to the processing unit and receive signals therefrom, the processing unit including a micro-controller and data transfer means.

The processing unit may include a clock.

The first and second audio transducers may each comprise a microphone.

Preferably, the audio transducers are connectable to the processing unit by wires. Alternatively, the audio transducers may be connected to the processing unit wirelessly.

The data transfer means may transfer data directly from the processing unit to a remote memory. Alternatively, the processing unit may comprise a memory and data may be transferred from the memory to another memory from time to time. The memory may be a removable memory device, such as a non-volatile memory card. The memory may be a flash memory.

The processing unit may be provided with at least one drive for connection to a peripheral device.

The processing unit may comprise a signal conditioner for conditioning electronic signals received from the audio transducers. The signal conditioner may be a codec.

The processing unit may include a positioning system, such as Global Positioning System (GPS). The function of the positioning system is to provide a location for recorded signals emanating from the audio transducers and/or the signal conditioning codec.

According to a second aspect of the invention there is provided a flow monitoring apparatus according to the first aspect of the invention wherein one of the first and second audio transducers for generating a flow signal is located proximate a pipeline, the pipeline configured for the flow of fluid therethrough, and the other of the first and second audio transducers for generating a reference signal is located at a distance from the pipeline and the audio transducer proximate the pipeline.

The pipeline may include a shut-off valve and the audio transducer located proximate the pipeline may be located proximate the shut-off valve and preferably on the shut-off valve. Where the audio transducer is located proximate the shut-off valve, rather than on the shut-off valve, it is preferred that the material on which the audio transducer is mounted is the same material from which the valve is formed and that there is an audio transmission pathway from the valve to the audio transducer.

The pipeline may pass through an access housing and the audio transducer located at a distance from the pipeline and the audio transducer proximate the pipeline may be located on or adjacent to a wall of the housing.

The pipeline may be connected to a property, the access housing being located upstream of the property and preferably, the wall of the housing on or next to which the audio transducer is mounted is a wall of the housing that is distal from the property.

The reference microphone may be positioned to the side of the pipeline.

According to a third aspect of the invention there is provided data processing software configured for processing data recorded by the apparatus of the second aspect of the invention, the data processing software configured to: receive the recorded data from the apparatus, from the memory of the apparatus, from external memory or by direct data transfer; compare the received data with a database of data representing known fluid flows through known pipelines and/or valves; and provide a classification of the received data according to the degree of similarity of the received data with data in the database.

The classification of the received data may include one or more of: an indication that fluid is flowing through the pipeline and/or valve for a period of time longer than associated with an intended use of the fluid; an indication that fluid is likely to be flowing through the pipeline and/or valve for a period of time longer than associated with an intended use of the fluid; an indication that fluid is unlikely to be flowing through the pipeline and/or valve; an indication that fluid is not flowing through the pipeline and/or valve; an indication that fluid is being used intentionally.

The data processing software may include a noise cancellation step. In the noise cancellation step the reference signal may be inverted.

The reference signal may be phase shifted. The extent of phase shift may be related to the distance between the first and second transducers and the materials proximate the transducers.

The inverted or inverted and phase shifted reference signal may be added to the signal from the audio transducer proximate the pipeline and/or valve to provide a noise cancelled wave form.

The noise cancelled wave form may be processed through one or more steps to provide a normalised spectral profile.

The reference signal and the signal from the audio transducer proximate the pipeline and/ or valve may be processed through one or more steps to provide normalised spectral profiles.

The processing steps may include one or more of: applying a low pass and/or high pass filter; splitting the signals into segments, such as 100 ms segments; applying a Fast Fourier transform to each segment; calculating a value in dB for each segment in a frequency range, for example between 100 and 3000Hz.

The processing steps may include calculating one or more of the following statistical data: mean dB level; the standard deviation and/or variance of the dB values; the average spectral level (dB); the Xth percentile; the spectral shape associated with the Xth percentile.

A best fit data analysis may be performed on the noise cancelled wave form or on data derived from the noise cancelled waveform and preferably on the normalised spectral profile. The best fit analysis may be performed using the calculated statistical data.

The result of the best fit data analysis may be compared with the database of data representing known fluid flows through known pipelines and/or valves.

The data processing software may be configured to generate graphic outputs for display on a visual display unit for example. The graphic outputs may be used by an operator for the purpose of review.

The software may provide for a user to ascribe an assessment of flow to recorded data. According to a fourth aspect of the invention there is provided the combination of the apparatus of the second aspect of the invention, a computer device, the data analysis software of the third aspect of the invention operating on the computer device, the computer device adapted to receive data recorded by the apparatus of the second aspect of the invention and to process the received data according to the data processing steps of the data processing software.

According to a fifth aspect of the invention there is provided a method of assessing flow in a pipeline or valve in a pipeline comprising the steps of: 1. providing the apparatus of the second aspect of the invention; 2. operating the apparatus provided in step 1 for a period of time; 3. inputting the data recorded in step 1 to a computer device programmed with the data processing software of the third aspect of the invention; 4. processing the inputted date according to the steps of the data processing software; 5. providing an indication related to flow of fluid through the pipeline and/or valve.

Brief Description of the Drawings

In the Drawings, which illustrate preferred embodiments of the invention, and which are by way of example: Figure 1 is a schematic representation illustrating the transducers of the invention; Figure 2 is a block diagram illustrating the components of the apparatus of the invention; Figure 3 is a series of graphs representing data recorded by or produced by the apparatus of the invention illustrated in Figures 1 and 2.

Detailed Description of the Preferred Embodiments

Referring now to Figure 1, there is shown a valve 1, also known as a stop cock, which forms part of a pipeline 2. The pipeline 2 connects one side of the valve 1 to a property 3. The pipeline 2 to the other side of the valve 1 is connected to the water main 4. The valve 1 is situated in a housing 5. The purpose of the housing is to provide access to the valve 1, which typically is located below ground, and to protect the valve 1 and the pipeline 2 against damage. The housing 5 may take numerous forms. For example the housing 5 may be a plastic pipe or chamber, a chamber formed of bricks, a concrete ring or chamber.

The apparatus of the invention includes two audio transducers 7, 8. The audio transducer 7 is mounted on the valve 1. The other transducer 8 is mounted on a wall 6 of the chambers. In the illustrated example, the transducer 8 is spaced apart from both the valve 1 and the pipeline 2. The transducer 8 lies on an axis that passes through the valve at an angle of approximately 45 degrees to a vertical axis y-y passing through the valve 1. Whilst it is essential that the transducers 7, 8 are spaced apart from one another, it is also preferred that the distance between the sensors is 200mm or less. In the illustrated embodiment the transducer 8 is spaced apart from the valve 7 by approximately 200mm. The distance between the two transducers must be sufficient to ensure that there is a clear measurement of an activity at the valve versus background activity. Hence, the reference transducer needs to be sufficiently distant from the valve transducer for any activity in the valve not to be detected by the reference transducer The function of the audio transducer 7 is to sense audio signals emanating from the pipeline 2.

Referring now to Figure 2, the block diagram illustrates the components of the apparatus of the invention.

The audio transducers 7, 8 are piezo electric microphones. Each microphone 7, 8 inputs to an audio input signal conditioning codec 9 which converts the output signals of the microphones 7, 8 into a digital signals. These digital signals are stored in a lossless wave file format as channel 0 and channel 1.

The output of the signal conditioning codec 9 forms an input to the processing unit 10. The processing unit 10 comprises a micro controller, a memory, a real time clock (RTC) and Global Positioning System (GPS) for the purposes of time stamping and providing a location for recorded signals emanating from the signal conditioning codec 9. The memory may be a removable nonvolatile memory card (known as an SD card) or alternatively the processing unit 10 may be provided with a flash memory. It is desirable to log activity with the microphones for a period of 30 seconds. At the 44.1 kHz sampling rate of the microphones 7, 8 for the period of 30 seconds, a memory of at least 4GB is needed. It may be desirable to provide a larger memory in order to facilitate either longer sampling times or faster sampling rates. For example, a memory of 16GB may be provided.

The processor 10 is connected to a power source 11. In the illustrated example this is a lithium battery with a self-contained charging circuit. Also connected to the processor 10 is a graphic display. In the illustrated example the graphic display is a dot matrix liquid crystal display.

The main function of the apparatus is to record data and to process that data so that it may be used to assess activity in the valve 1 and/or the pipeline 2. It is preferred that the data is off-loaded from the processor 10 for subsequent analysis. To this end the apparatus includes a means of connecting to the processor 10. This may be a USB interface 13 and/or a wireless interface 14. Where a wireless interface is provided this may operate on a Bluetoothe protocol or a WiFi protocol.

The USB interface 13 provides not only for the offloading of data but also software and firmware updates, system configuration and battery charging.

A near field communication reader 15 may be provided. This would allow an RFID tag to be read and recorded by the processor 10. Such an RFID tag could be used to identify the property 3. Alternatively, the property may be identified by other means, such as a barcode which is scanned using a bar code reader and linked to the data recorded by the processor. The bar code reader may attach to the processor 10 by means of a USB port or other suitable connection. When a recording is taken the bar code on the tag in the chamber is scanned and the bar code is automatically associated with the audio recordings through the firmware in the device.

The data that is recorded by the processor 10 is processed on another computer device that is programmed with data analysis software. The function of the data analysis software is to provide visualisation of the recorded data. The software may also provide predictive scores indicating degrees of confidence of flow through a valve 1 and may also permit the data to be reviewed by a data analyst.

The data analysis software processes the data downloaded from the processor 10 in two stages.

Stage 1 -Data processing The two-channel 0 and 1 stored in the wave file are split into separate files. One file represents the signal from the microphone 7. This signal is representative of flow through the valve 1 and pipeline 2. The other file represents the signal from the microphone 8. This signal is representative of background sound and provides a reference signal. It is assumed that the signal recorded by the microphone 7 induces both background sound and sound associated with flow through the pipeline/valve.

In stage 1, other information may be extracted from the data including information relating to the property the flow to which is being monitored and which typically comes from a barcode, an RFID tag or other manually entered data. This location information is typically saved within the filename of the measurement file. A GPS location may also be extracted from the data, the GPS location usually being stored in the header information of the data file.

From the above and in stage 1, the software creates an index file which contains the data path, that is information relating to the storage of the data in the device memory and information representing the measured audio signals at each of the microphones.

The index file is used in subsequent stages of processing.

Stage 2 -Noise Cancellation The software performs a noise cancellation process on the data recorded in the two separate files. The steps of the noise cancellation process are as follows: 1. The data from the file that is associated with the microphone 8, that being the reference data is inverted so that positive values become negative and negative values positive; 2. A time-shift is applied to the reference data. In the illustrated example the phase shift applied was 20ms. A time shift is required because there is a separation distance between the reference microphone 8 and the microphone 7 associated with the valve. Hence, the background sound detected by both microphones will reach the reference microphone 8 first. The software assumes that street level sound and vibration detected by the microphones 7, 8 will have travelled predominantly through materials such as soil, plastic, metal, clay, concrete, asphalt, stone, brick, etc, and that the speed of a sound wave through these materials is in the order of 80 to 300 ms-1.

3. The inverted and phase shifted reference wave form is mathematically summed with the valve wave form. The output of this step is a noise cancelled wave form.

4. High and low pass filters are applied to the reference wave form, the valve wave form and the noise cancelled waveform. Frequencies below 3kH and above 100kHz are removed from the three signals. Frequencies below 3kHz and above 100 kHz are not usually associated with known flows within a valve.

The noise cancellation stage removes background sounds which means that only the sound and activity occurring within the valve 1 are analysed. It is common practice to conduct leak detection work at night because the background noise at night is much lower than during daylight hours. The apparatus and method of the invention removes the need to conduct leak detection work at night.

The apparatus is isolated from background noise to the greatest extent possible, because all the components required for measurement are placed in the chamber where the valve 1 is located. It is preferred that the chamber lid that is removed for fitting of the apparatus is replaced during monitoring. Other known devices that monitor valves protrude from the chamber 5, allowing background noise to have a greater effect on the noise signal detected at the valve 1. Nevertheless, replacement of the chamber lid during monitoring is not essential since the noise cancellation step removes the background noise.

Stage 3 -Statistical Review and Visualisation of Data The output signals from stage 2 are the valve signal, the reference signal and the noise cancelled signal. These signals undergo the following processing steps and statistical analysis. The processing steps in Stage 3 are: 1. the wave forms are split into 100ms segments; 2. each 100ms segment of the wave form is subject to a Fast Fourier Transform (FFT); 3. A noise (dB value is calculated for each of the 100ms segments for all energy shown the FFT over the frequency range 100Hz to 3000Hz; and 4. the following statistics are generated across all 100ms segments: a. mean dB level b. standard deviation and variance in the dB levels; c. average spectral level; d. Xth percentile values are generated; e. spectral shape associated with Xth percentile values generated.

The graphs shown in Figures 3a to 3e illustrate the differing processing steps and resulting outputs encompassed within stage 3.

Figure 3a illustrates the reference signal (top row of graphs) and valve signal (middle row of graphs) that are used in the noise cancellation stage, and the noise cancelled signal (bottom row of graphs).

The leftmost column of graphs are histograms showing signal magnitude in dB during each 100ms segment.

The second column in from the left illustrates graphs of signal magnitude in dB for each 100ms segment as a function of time. The noise cancellation signal shows the difference between the reference signal and the signal from the microphone at the valve. It is the noise cancelled signal that represents flow through the valve 1.

From the first and second columns the following statistics are calculated for each row: The mean dB level; The standard deviation and variance of the dB levels; The Xth percentile level.

The centre column illustrates the respective signals following a Fast Fourier Transform (FFT) for each 100ms segment.

In the fourth column the spectral profiles have been normalised by setting them to the same overall dB level.

The statistical analysis calculations performed on the data from the first and second columns is used to find the best fit to the normalised spectral profile shown in the fourth column. In the normalised spectral profiles shown in the fourth column of Figure 3, the best fit is shown by the black line.

The data analysis software is connected to a database which includes a set of normalised spectral profiles for known flows. This is known as the confirmed dataset. The confirmed dataset is created by monitoring a flow through a pipeline (or a valve that is part of a pipeline) with the apparatus of the invention, performing the data analysis as set out above, and measuring the flow using other apparatus. In this way a database of normalised spectral profiles can be built up for different flows through difference pipelines (or different valves).

The best fit for normalised spectral profile of the noise cancelled signal in column 4 of Figure 3 and illustrated by the black line is compared with the normalised spectral profiles of noise cancelled signal in the confirmed dataset. The normalised spectral profile in the confirmed data set to which the black line is the best fit is considered to represent the flow through the pipeline or valve thereof.

The apparatus, data processing software and method of the invention allow flow measurements to be taken and assessed in a particularly cost effective manner

Claims

Claims 1. A flow monitoring apparatus comprising first and second audio transducers, one of the first and second transducers for location proximate a pipeline and the other of the first and second transducers for location at a distance from the pipeline and the transducer located proximate the pipeline, and a processing unit, wherein the first and second audio transducers are connected to the processing unit and receive signals therefrom, the processing unit connected to a power source and including a micro-controller, and data transfer means.
2. A flow monitoring apparatus according to Claim 1, wherein the first and second audio transducers each comprise a microphone.
3. A flow monitoring apparatus according to Claim 1 or 2, wherein the audio transducers are connectable to the processing unit by wires, or the audio transducers are connected to the processing unit wirelessly.
4. A flow monitoring apparatus according to any preceding claim, wherein the data transfer means is configured to transfer data from the processing unit to a remote memory.
5. A flow monitoring apparatus according to any preceding claim, wherein the processing unit comprises a memory.
6. A flow monitoring apparatus according to Claim 5, wherein the memory is a removable memory device, or an integrated memory.
7. A flow monitoring apparatus according to any preceding claim, wherein the processing unit is provided with: at least one drive for connection to a peripheral device; and/ora near field communication reader; and/ora barcode reader interface; and/or data input means; and/or a display.
8. A flow monitoring apparatus according to any preceding claim, wherein the processing unit comprises a signal conditioner for conditioning electronic signals received from the audio transducers.
9. A flow monitoring apparatus according to any preceding claim, where the processing unit includes a positioning system and/or a clock.
10. A flow monitoring apparatus according to Claim 9, wherein processing unit attaches a position location to the recorded signals emanating from the audio transducers and/or the signal conditioner.
11. A flow monitoring apparatus according to any of Claims 1 to 10, wherein one of the first and second audio transducers is located proximate a pipeline, the pipeline configured for the flow of fluid therethrough, and the other of the first and second audio transducers is located at a distance from the audio transducer proximate the pipeline.
12. A flow monitoring apparatus according to Claim 11, wherein the pipeline includes a shutoff valve and the audio transducer located proximate the pipeline is located proximate the shut-off valve.
13. A flow monitoring apparatus according to Claim 12, wherein the audio transducer located proximate the pipeline is located on the shut-off valve.
14. A flow monitoring apparatus according to any of Claims 11 to 13, wherein the distance between the first and second audio transducers is between 100mm and 200mm.
15. A flow monitoring apparatus according to any of Claims 11 to 14, wherein the pipeline passes through an access housing and the audio transducer located at a distance from the audio transducer proximate the pipeline is located on or adjacent to a wall of the housing, said wall being upstream of the audio transducer located proximate the pipeline.
16. A flow monitoring apparatus according to Claim 15, wherein the housing has a cover and the cover is in place when the first and second audio transducers record signals.
17. Data processing software configured for processing data recorded by the apparatus according to any of Claims 11 to 16, the data processing software configured to: receive the recorded data from the memory of the apparatus; compare the received data with a database of data representing known fluid flows through known pipelines and/or valves; providing a classification of the received data according to the degree of similarity of the received data with data in the database.
18. Data processing software according to Claim 17, wherein the classification of the received data includes one or more of: an indication that fluid is flowing through the pipeline and/or valve for a period of time longer than associated with intended use of the fluid; an indication that fluid is likely to be flowing through the pipeline and/or valve for a period of time longer than associated with intended use of the fluid; an indication that fluid is unlikely to be flowing through the pipeline and/or valve; an indication that fluid is not flowing through the pipeline and/or valve; an indication that fluid is being used intentionally.
19. Data processing software according to Claim 17 or 18, wherein the data processing software includes a noise cancellation step.
20. Data processing software according to Claim 19, wherein in the noise cancellation step the reference signal is inverted.
21. Data processing software according to any of Claims 18 to 20, wherein the reference signal is phase shifted, and the extent of phase shift is related to the distance between the first and second transducers.
22. Data processing software according to Claim 20 or 21, wherein the inverted or inverted and phase shifted reference signal is added to the signal from the audio transducer proximate the pipeline and/or valve to provide a noise cancelled wave form.
23. Data processing software according to Claim 22, wherein the noise cancelled wave form, and/or the reference signal and the signal from the audio transducer proximate the pipeline and/or valve are processed through one or more steps to provide a normalised spectral profile.
24. Data processing software according to any of Claims 17 to 23, wherein the software performs a best fit data analysis and compares the best fit with the database of data representing known fluid flows through known pipelines and/or valves.
25. The combination of the apparatus according to any of Claims 11 to 16, a computer device, the data analysis software of any of Claims 17 to 24 operating on the computer device, the computer device adapted to receive data recorded by the apparatus according to any of Claims 11 to 16 and to process the received data according to the data processing steps set out in any of Claims 17 to 24.