Sensor of acoustic waves in a liquid and method for monitoring a network for conveying liquid

Abstract

According to the invention, a sensor (1) of acoustic waves propagating through a liquid is optically sensitive to variations in the density of the liquid that are produced by pressure variations related to the passage of an acoustic wave. The sensor is designed to be threaded through a wall (102) of a pipe, for example of a water distribution network. The sensor may be used to detect leaks of the liquid at distance from a location of installation of the sensor, or to adjust an intensity of an acoustic wave used to detach a deposit present on a surface making contact with the liquid.

Description

Description Title: Acoustic Wave Sensor in a Liquid and Method for Monitoring a Liquid Conveyance Network technical field

[0001] This description relates to a sensor for acoustic waves propagating in a liquid, and a method for monitoring a liquid conveying network. Previous technique

[0002] Liquid transport networks are used in many sectors of activity, for example the distribution of water in a geographical area for the daily consumption of inhabitants of that area, the transport of oil, regardless of its level of refining to produce petroleum fuel, and the transport of agri-food liquids, particularly in agri-food processing and / or manufacturing plants.

[0003] However, these liquid distribution networks are subject to numerous types of degradation, including leaks and the formation of deposits on internal network surfaces in contact with the liquid. A leak is an unintentional, localized flow of liquid outside the network and can be characterized by its location within the network and by an impedance value. The impedance is calculated by dividing the pressure drop along the leak by the leakage flow rate. The nature of the deposits that can form inside a network depends on the liquid. These deposits can be biofilms, such as bacteria or microalgae, or mineral deposits, such as limescale. These deposits locally reduce the available cross-sectional areas for the liquid, thus decreasing the network's efficiency in conveying it.

[0004] However, some liquid transport networks can be very extensive geographically and / or difficult to access, particularly with sections of the network that are buried. It is then laborious and very expensive to identify the location of a leak and / or a sealed deposit in the network by accessing it directly, in order to remedy it. Technical problem

[0005] One object of the present invention is to identify more easily, and in a more precise and reliable manner, a leak in a liquid conveying network.

[0006] Another goal is to enable the removal of a deposit that is present on an internal surface of the network in contact with the liquid, in a controlled manner, particularly with regard to the risks of damage to the network. Summary of the invention

[0007] To achieve at least one of these goals, or another, a first aspect of the invention proposes a sensor for acoustic waves propagating in a liquid, this sensor comprising: - an elongated rigid support, which extends between two ends, called respectively the proximal end and the distal end, and which is designed to form a first optical path between the two ends; - an optical source, preferably of the laser type, which is fixed to the support at the proximal end and oriented to emit radiation towards the distal end in the first optical path during use of the sensor; - a first retroreflector or retrodiffuser, which is fixed to the support at the distal end and oriented to reflect at least part of the radiation towards the proximal end, still in the first optical path; and - a first optical detection and analysis system, which is located at the proximal end and adapted to detect at least part of the reflected radiation, and which is configured to deduce values ​​of variations of an optical length of the first optical path.

[0008] According to a feature of the sensor of the invention, the support has a transverse passage located between its two ends, adapted to allow liquid to fill this passage, so that the liquid is traversed by the first optical path, thus the first optical detection and analysis system is sensitive to variations in liquid density produced by an acoustic wave propagating through the liquid via the passage.

[0009] According to another feature of the sensor of the invention, it further comprises: - sealing means, suitable for establishing a seal between the inside and outside of a pipe containing the liquid when at least part of the support, including the passage, is placed inside the pipe and is filled with the liquid.

[0010] The sensor of the invention is thus adapted to be threaded through a pipe wall without causing any leakage of the liquid at the point of insertion. In particular, the portion of the sensor that enters the pipe can be small. Furthermore, the sensor can be made up of components used in many other applications, such as laser rangefinding for short distances measured in the first optical detection and analysis system.

[0011] Preferably, the first optical detection and analysis system can be adapted to provide in real time or near real time the values ​​of the variations in the optical length of the first optical path.

[0012] The acoustic wave propagating through the liquid produces variations in the liquid's density, known as "piezo-optical" variations, which in turn cause variations in the liquid's optical refractive index. These refractive index variations modify the optical length of the first optical path, which is measured by the first optical detection and analysis system. From this, amplitude and phase values ​​of the sinusoidal components of the acoustic wave can then be deduced.

[0013] In specific embodiments of the sensor, the sealing means may include: - peripheral sealing means, which are located on the support between the proximal end and the passage, and which are adapted to form, after the support has been threaded through a pipe wall so that the passage is inside the pipe and the proximal end of the support is outside the pipe, a watertight seal between an external peripheral surface of the support and the pipe wall; and - optionally, a window which is transparent to radiation and located across the first optical path between the passage and the proximal end, and which is hermetically connected to the peripheral sealing means.

[0014] In general, for the invention, the passage can be designed to allow the liquid to flow through the support, obliquely or perpendicularly to a longitudinal direction of the support, without altering the average direction of liquid flow in the pipe compared to the pipe without the sensor. Thus, installing the sensor in the pipe minimally disturbs the liquid flow.

[0015] In preferred embodiments, known as "dual-beam" embodiments, the sensor may further include: - a second optical path which also extends between the two ends of the support, parallel to the first optical path; - a radiation entry into the second optical path, which is located at the proximal end and configured to introduce additional radiation into this second optical path towards the distal end; - a second retro-reflector or retro-diffuser, which is fixed to the support at the distal end, and oriented to reflect at least part of the additional radiation towards the proximal end, in the second optical path; - a second optical detection and analysis system, which is located at the proximal end and adapted to detect at least part of the additional reflected radiation, and configured to deduce variations in an optical length of the second optical path; - a separation tube, which contains the second optical path and is either empty or filled with a transparent gas, and which is designed to remain liquid-tight between the radiation entry into the second optical path and the second retroreflector or retrodiffuser while the liquid is in the passage, so that the second optical detection and analysis system detects variations in the length of the medium; and - a data processing module, configured to remove a contribution from the variations in length of the medium, to the variations in optical length of the first optical path, so as to isolate a contribution from the acoustic wave which propagates in the liquid through the passage. The characterization of the acoustic wave provided by the sensor can be more precise when two beams are thus implemented, and a difference is calculated.

[0016] Preferably, the first and second optical detection and analysis systems, as well as the data processing module, can be adapted to provide in real time or near real time the values ​​of the contribution of the acoustic wave in the variations of the optical length of the first optical path.

[0017] Advantageously, and more generally for the invention, the sensor may optionally include, in addition, at least one of the following additional modules: • a self-contained power supply module, comprising: - a turbine, which is positioned outside the passage, and so as to be driven into rotation by a flow of liquid; - a dynamo, which is coupled to the turbine to generate electrical energy; and - an electrical energy management system, which is connected to the dynamo's input to collect electrical energy, and configured to power the sensor in such a way as to allow for energy-autonomous operation of that sensor; and • a radio transmit-receive module, configured to transmit to an external receiver, radio signals which contain first measurement results produced by the first optical detection and analysis system, and, where applicable, second measurement results produced by the second optical detection and analysis system, or a result of removing the contribution of the length variations of the medium to the optical length variations of the first optical path. Each of these modules helps to facilitate the use of the sensor in areas of a network that are difficult or impossible to access, so that acoustic wave measurements can be carried out economically and quickly.

[0018] A second aspect of the invention proposes a method for monitoring a liquid conveyance network, in particular a water distribution network, a network or pipeline for transporting oil or petroleum products, or a network or pipeline for transporting an agri-food liquid, according to which a sensor conforming to the first aspect of the invention is threaded in a sealed manner through a wall of a network piping, so that the passage is inside the piping. This process includes the following steps: / i / generate at least one acoustic wave in the liquid, preferably in the form of a liquid-hammer wave; then / ii / after the acoustic wave has propagated through at least part of the network, detect the acoustic wave using the sensor.

[0019] Such a monitoring method can have many applications, including detecting at least one leak of liquid to the outside of the network, or unclogging a deposit that is present on a surface internal to the network.

[0020] When used to detect at least one leak of the liquid outside the network, the method may have the following additional features: at step / i / : the acoustic wave may be generated in the liquid so as to generate acoustic components which propagate in the network according to natural modes; and step / ii / is then carried out after the acoustic components have undergone internal reflections in the network, including reflections at the leak.

[0021] At step / i / , the acoustic wave can be generated in the liquid in the form of a water hammer. In this case, it is the superposition of acoustic components that propagate simultaneously in the network.

[0022] Alternatively, the acoustic wave can be generated in the liquid at step / i / in such a way as to successively form acoustic components with varying acoustic frequency values. In this case, the acoustic wave can be generated using a sinusoidal acoustic wave generator and by varying the frequency of the generated sinusoidal acoustic wave.

[0023] For the leak detection application, the process further includes the following step: / iii / deduce a value of a leakage impedance and a location of it in the network from changes in respective natural frequency values ​​of the natural modes. The use of modifications to the respective natural frequency values ​​of the modes own is particularly advantageous, because such a measurement result is independent of how the acoustic wave was generated at step / i / , in particular independent of an amplitude profile of the acoustic wave with respect to time at the time of its generation.

[0024] This method is capable of detecting one or more leaks located some distance from the sensor's installation point in the network. This greatly simplifies network monitoring for leaks.

[0025] Advantageously, step / iii / can be executed using a linear matrix relationship between the changes, or increments, of the respective natural frequency values ​​of the natural modes and leakage impedances that are assigned respectively to locations in the network.

[0026] According to an improvement for leak detection, a steady-state liquid pressure in the network can be successively established at several different values, and steps / i / , / ii / , and / iii / can be performed for each steady-state pressure value, with steps / i / and / ii / occurring while that steady-state pressure value is in effect. The impedance value of the leak and its location in the network can then be deduced from the changes in the respective frequency values ​​of the natural modes, obtained for the plurality of steady-state pressure values. The steady-state liquid pressure that is varied between successive iterations of steps / i / , / ii / , and / iii / can be a static or quasi-static liquid pressure, i.e., when the liquid flow in the network corresponds to zero or low flow, or a pressure that is associated with a steady or quasi-steady liquid flow in the network.Using several steady-state pressure values ​​helps to make the results obtained for the impedance and location of each detected leak more reliable.

[0027] It is possible to simultaneously detect several leaks affecting the network through the same sequence of execution or repetition of steps / i / , / ii / , and / iii / . The method of the invention is therefore particularly effective in its ability to simultaneously detect potentially multiple leaks.

[0028] When the acoustic wave is generated in step / i / in the form of a water hammer, the natural frequency values ​​of the natural modes can be identified in step / iii / such as frequency values ​​for which a spectrum of the acoustic wave as detected in step / ii / exhibits peaks.

[0029] When the acoustic wave is generated in step / i / using a sinusoidal acoustic wave generator and varying a frequency of the sinusoidal acoustic wave that is generated, the natural frequency values ​​of the natural modes can be identified in step / iii / as values ​​of the frequency of the generated sinusoidal acoustic wave that correspond to amplitude resonances of the acoustic wave as detected in step / ii / .

[0030] When the method of the invention is used to unclog at least one deposit present on at least a portion of an internal surface of the network, by means of shearing produced by the acoustic wave inside the deposit, step / ii / can be used to adjust the intensity of the acoustic wave generated during a repetition of step / i / . The method of the invention avoids using an excessively intense acoustic wave to unclog the deposit, thanks to the monitoring provided by the sensor, which ensures that the network is not unnecessarily damaged, particularly at points of weakness it may present.

[0031] More effective unclogging can be achieved if the acoustic wave is generated in the liquid at step / i / in the form of a water hammer.

[0032] In particular, the method of the invention makes it possible to adjust the intensity of the acoustic wave which is used to produce the unclogging, by installing the sensor at a distance from the place from which the acoustic wave is generated in the network, and also at a distance from the place where the deposit to be unclogged is located. Brief description of the figures

[0033] The features and advantages of the present invention will become clearer in the following detailed description of non-limiting embodiments, with reference to the accompanying figures, among which:

[0034] [Fig. 1] represents a liquid conveying network in which a sensor according to the invention can be used;

[0035] [Fig. 2] is a longitudinal cross-sectional view of a sensor according to the invention, according to a first embodiment called a single-beam sensor;

[0036] [Fig. 3] corresponds to [Fig. 2] for a second embodiment of a sensor according to the invention, called a two-beam sensor;

[0037] [Fig. 4] is a partial perspective view illustrating an optional improvement of a sensor according to the invention, which can be combined with embodiments of [Fig. 2] and [Fig. 3]; and

[0038] [Fig. 5] is a step diagram of a process for detecting leaks present in the network of [Fig. 1], Detailed description of the invention

[0039] For clarity, the dimensions of the elements shown in these figures do not correspond to actual dimensions or ratios of actual dimensions. Furthermore, some of these elements are represented only symbolically, and identical references shown in different figures designate identical elements or elements with identical functions.

[0040] [Fig. 1] schematically represents a portion of a liquid conveyance network 100, such as, for example, a domestic drinking water distribution network. Such a network 100 consists of pipes 101 that are filled with water. These pipes 101 connect successive branch nodes 110 in the network 100 or, when a pipe 101 is a terminal pipe in the network 100, it connects a node 110 to a water consumption point 111. The state of such a water consumption point 111 can vary over time, corresponding to varying values ​​of water flow rate and impedance for that consumption point. The impedance at a point in the network 100 is defined as the ratio between the steady-state water pressure value and the water flow value, as these values ​​exist at the point considered in the network 100.The 100 network can also exhibit leaks, which are each characterized by an impedance value and its location in the network.

[0041] According to [Fig. 2], a sensor, which is designated globally by reference numeral 1, comprises an elongated rigid support 10, for example a hollow tube extending between a first end 11, called the proximal end, and a second end 12, called distal end. L denotes the longitudinal axis of the support 10. The distance between the two ends 11 and 12 can be approximately 90 mm (millimeter), by way of example, and the internal diameter of the tube 10 can be 10 mm, also by way of example. A coherent optical source 13, for example of the laser diode type or alternatively a quantum cascade laser or external cavity laser, is fixed to the tube 10, inside it, at the proximal end 11, and a retroreflector or retroscatterer 14 is fixed to the tube 10, also inside it but at the distal end 12. An emission wavelength of the optical source 13 can be substantially equal to 850 nm (nanometer), 1300 nm or 1550 nm, by way of non-limiting examples. During operation of the sensor 1, the optical source 13 emits radiation R towards the retro-reflector or retro-diffuser 14, so that this radiation is retro-reflected towards the optical source 13.The round trip between the optical source 13 and the retroreflector or retroscatter 14 has been called the first optical path in the general part of this description and is designated by C1 in the figures. An optical detection and analysis system 15 is integrated into the sensor 1 to measure the optical path length C1. Such systems are known, with different operating principles. According to one of these principles, the optical path C1 constitutes an external optical cavity to a laser that constitutes the source 13, and a change in the optical path length C1, while the wavelength of the radiation R remains constant, produces a change in light intensity that is measured by a detector of the optical detection and analysis system 15. An analysis module of this system 15 can directly correlate the measured change in light intensity to the change in optical path length C1.However, other models of optical detection and analysis systems 15 can be used alternatively.

[0042] A passage 16 is provided through the tube 10, between the optical source 13 and the retroreflector or retrodiffuser 14. This passage 16 may have oblong opening sections through two diametrically opposed portions of the tube wall 10. These opening sections may be identical on both opposite sides of the tube, and the tube 10 retains sufficient rigidity through its residual wall sections between the two opening sections. The passage 16 is preferably perpendicular to the longitudinal axis L of the tube 10, and its opening sections may have a length approximately 30 mm parallel to this longitudinal axis L, and approximately 6 mm wide. When using the sensor 1, the passage 16 is intended to be filled with liquid, which may be stationary or flow through the passage 16 between the two opposite lateral sides of the tube 10. The optical path C1 is then located in the water, and the optical detection and analysis system 15 measures the instantaneous values ​​of the optical length of this optical path C1, between the system 15 and the retroreflector or retroscatter 14. Preferably, the optical detection and analysis system 15 has a maximum sampling frequency greater than 10 MHz (megahertz), for example, 100 MHz. Such values ​​for the maximum sampling frequency make it possible to characterize, in amplitude and phase, acoustic components with frequency values ​​between 100 Hz (hertz) and 10 4 Hz.

[0043] The sensor 1 is threaded or pushed through the wall 102 of a pipe filled with liquid, for example a water pipe 101 forming part of the network 100 of [Fig. 1]. For this, a hole is first drilled through the wall 102 of this pipe 101, with a diameter that is equal to or just greater than that of the tube 10, then the sensor 1 is threaded into the hole by the distal end 12 until the passage 16 is entirely inside the pipe 101 while the proximal end 11 of the tube 10 remains outside the pipe. A collar of sealing material 17, for example of thermosetting resin or other, constituting peripheral sealing means, may have been initially provided around the tube 10, between the passage 16 and the proximal end 11. The tube 10 is pushed into the pipe 101 until the collar 17 comes to rest against the wall 102.The collar 17 is then heated to establish a seal between the inside and outside of the pipe 101 around the tube 10. A window 18 can also be provided inside the tube 10, between the passage 16 and the optical source 13. This window 18 is perpendicular to the longitudinal axis L of the tube 10 and is transparent to the radiation R from the optical source 13. The window 18 is effective in ensuring a seal inside the tube 10. In particular, it can prevent the optical source 13 and the optical detection and analysis system 15 from being immersed while the optical path C1 between the window 18 and the retroreflector or retrodiffuser 14 is in water. A peripheral edge of the window 18 can be bonded watertight to the inner surface of the tube wall 10. However, other mountings of the window 18 are possible. The sealing material collar 17 and the sight glass 18 are an example of an embodiment of the sealing means mentioned in the general part of this description, but any other means may also be used. [Fig. 2] shows the sensor 1 which has been installed in the water pipe 101. It is not necessary for the tube 10 to pass completely through the pipe 101, so the distal end 12 can be a free end located in the water.

[0044] As illustrated in [Fig. 4], the sensor 1 can further include a micro-turbine 19a, which is designed to be driven in rotation by the water in the pipe 101 when that water is flowing. This micro-turbine 19a is coupled to the rotor of a dynamo 19b, which is fixed by its stator to the tube 10. In an optimized configuration, the micro-turbine 19a and the dynamo 19b can be arranged around the tube 10, for example coaxially with it, at the passage 16. The micro-turbine 19a is open between its blades, and the rotor and stator of the dynamo 19b are open between their magnetic elements, so that the passage 16 remains essentially permanently open for water to flow through it.Such an arrangement of the micro-turbine 19a and the dynamo 19b does not disturb acoustic waves propagating in water, because the wavelength of such acoustic waves, typically greater than 1 m, is much greater than the external dimensions of the sensor 1. Other arrangements of the micro-turbine 19a with respect to the tube 10 are also possible, for example the micro-turbine 19a arranged beyond the distal end 12, in the extension of the tube 10.

[0045] The flow of water in pipe 101 thus produces electricity that can be used to power all the components of sensor 1, including the optical source 13 and the optical detection and analysis system 15. For this purpose, an energy management system, not shown, can also be incorporated into sensor 1. Such a system may include a battery, a capacitor, or any other equivalent means for storing electrical energy between the time it is produced by the dynamo 19b and a later time it is consumed by the operation of sensor 1. Typically, such an energy management system can be configured to adapt the voltage levels between the dynamo 19b and the electrically consuming components of sensor 1. Finally, sensor 1 may also include a radio transceiver module 19c (see [Fig. 2]), for example located at The proximal end 11 is used to communicate with an external radio receiver, such as a handheld receiver used by a network monitoring and maintenance technician 100. Such a module 19c allows operating instructions to be transmitted to sensor 1 and measurement results to be collected, without requiring direct physical access to sensor 1. The radio transceiver module 19c can also be powered by the generator 19b. Sensor 1 thus operates in an energy-autonomous manner, meaning it does not need to be connected to a power source, and can be operated remotely.

[0046] When installed in the pipe 101 as described above, and when the distance is constant between the optical source 13 and the retroreflector or retro-diffuser 14, and also between this retroreflector or retro-diffuser 14 and the optical detection and analysis system 15, the sensor 1 is sensitive to variations in the optical refractive index of the water in the passage 16. Now, the optical refractive index of water depends on its instantaneous density, which is modified by an acoustic wave that can propagate through the water via the passage 16. Thus, the sensor 1 makes it possible to measure in real time such an acoustic wave propagating inside the pipe 101.In particular, when the optical detection and analysis system 15 is configured to analyze such an acoustic wave by Fourier decomposition, it provides as measurement results the frequency, phase, and amplitude values ​​of sinusoidal components that, by superposition, constitute the acoustic wave. The embodiment of [Fig. 2] is called a single-beam sensor.

[0047] However, the support 10 can undergo transient length variations parallel to the axis L; that is, variations in geometric length as opposed to variations in optical refractive index, which contribute to variations in the optical path length C1. These length variations of the support 10 between the optical source 13 and the retroreflector or retrodiffuser 14 can have various causes. Measuring these length variations of the support 10 between the optical source 13 and the retroreflector or retrodiffuser 14 allows us to remove, from the measured variations in the optical path length C1, a contribution from the length variations of the support 10, thus isolating the contribution of the acoustic wave. A more accurate and detailed characterization of the latter can therefore be obtained. Possibly, the two contributions can be separated from each other by a differential approach.

[0048] The improved sensor 1 shown in [Fig. 3] allows for the measurement of variations in the length of the support 10. For this purpose, a separating tube 10' can be placed inside the tube 10 to create a sealed enclosure around a second optical path, designated C2, which extends between a second optical source 13' and a second retroreflector or retrodiffuser 14'. The separating tube 10' can be empty or permanently filled with air or a transparent gas to provide additional radiation R' produced by the optical source 13'. It is arranged in the tube 10 so as to extend next to the passage 16, preferably without causing an obstruction to the flow of water in this passage 16. Preferably also, the optical source 13' can be located at the same level as the optical source 13 and the retroreflector or retrodiffuser 14' at the same level as the retroreflector or retrodiffuser 14, in projection on the longitudinal axis L of the tube 10.The separation tube 10' has a material and / or a mounting, for example a compression-seal mounting, adapted so that the geometric length variations of the second optical path C2 are identical at every instant to those of the first optical path C1. A second optical detection and analysis system 15' is associated with the second optical path C2, in a way that can be identical to that of the optical detection and analysis system 15 for the first optical path C1. Possibly, the optical source 13' can be replaced by a system for sampling a portion of the radiation R produced by the optical source 13, and a system for injecting this sampled portion of the radiation R into the second optical path C2. In this case, the optical source 13' is replaced by an input into the second optical path C2 of the sampled portion of the radiation R in order to constitute the radiation R'.

[0049] In the embodiment of [Fig. 3], called a two-beam sensor, the optical detection and analysis system 15' measures the instantaneous optical length of the second optical path C2, which is equal to the geometric length of the corresponding portion of the support 10. A data processing module 20 subtracts the instantaneous variations in optical length of the second optical path C2 from those of the first optical path C1, for the same measurement instants. The result of the subtraction characterizes the instantaneous density of water in the passage 16, such that This density is affected by any acoustic wave that passes through this passage. A series of successive measurements taken during the propagation of the acoustic wave provides a record of the variations in water density caused by this acoustic wave.

[0050] We now describe a method for monitoring network 100 of [Fig. 1] using sensor 1 of [Fig. 2] or [Fig. 3]. For this, this sensor 1 is installed in a pipe 101 in the manner described above.

[0051] An acoustic wave is then generated in the network 100, at a point in the network not too far from the sensor 1 so that the acoustic wave passes through the sensor via its passage 16 before being absorbed. Preferably, this acoustic wave is generated in the form of a water hammer, corresponding to a sudden change over time in the momentum associated with a mass of water inside the network 100. For example, such a water hammer can be generated by rapidly interrupting an exhaust flow of water to the outside of the network 100, notably using a quick-closing valve.

[0052] As is known, such a water hammer acoustic wave has a very broad Fourier spectrum. In other words, the sinusoidal components that make up this acoustic wave can have frequencies ranging from 100 Hz (hertz) to 10 kHz (kilohertz).

[0053] Furthermore, it is also known that a sinusoidal acoustic wave can only propagate through water within network 100 according to natural modes determined by the network configuration, including the points in this network that produce acoustic reflections. Water consumption points 111 constitute acoustic reflection points, as do points where leaks occur. Analyzing the acoustic natural modes of network 100, and especially analyzing the frequency changes of these natural modes, provides information on the evolution of acoustic reflection points. Each leak can thus be characterized as an acoustic reflection point by its location in network 100 and a value of its leakage impedance.

[0054] From the moment the water hammer is generated, sensor 1 is activated to capture instantaneous variations in water density in passage 16, for a sufficient recording duration. A Fourier transform with respect to time is applied to this recording, which reveals the acoustic eigenmodes of the network 100 as spectral amplitude peaks at respective frequency values. In other words, the water hammer excited the acoustic eigenmodes of the network 100, and these appear in the captured recording as spectral acoustic components.

[0055] Furthermore, the propagation of sinusoidal acoustic waves in the network 100 can be modeled numerically using a map of the network 100 that includes its division nodes 110 and its water consumption points 111, the latter including the locations of suspected leaks. Such a model provides the respective natural frequency values ​​of the acoustic natural modes as a function of the respective impedance values ​​of the water consumption points 111. More precisely, starting from reference values ​​for the respective impedances of the water consumption points 111, deviations from these reference impedance values ​​are linearly related by a matrix to respective variations in the natural frequency values ​​of the natural modes.By introducing into this linear matrix relationship the respective variations of the natural frequency values ​​of the natural modes, we obtain the differences in impedance values ​​of the water consumption points 111, including those corresponding to suspected leaks. A leak that has increased in size is thus revealed by a decrease in its impedance value.

[0056] [Fig. 5] summarizes the steps of the leak detection process, as just described, with the following references: 51: installation of sensor 1 in one of the pipes 101 of network 100; 52: generation of water hammer in network 100; 53: recording of temporal variations in water density using sensor 1; 54: calculation of the Fourier transform of the recording with respect to time; S5: Identification of the peaks of the acoustic eigenmodes of the 100 network in the spectrum obtained in step S4; M1: modeling of the 100 network and the natural modes of acoustic propagation in this 100 network; M2: matrix calculation; M3: calculation of the impedance values ​​of the water consumption points 111 by combining the matrix obtained in step M2 with the differences in natural frequency values ​​of the natural modes identified in step S5; and R: result step in the form of an identification and a balance of leaks according to the impedance values ​​calculated in step M3.

[0057] Such a method can initially be applied to a portion of the network 100 that is limited in size while still including the water hammer generation point and the sensor 1 installation point. Preferably, the number of water consumption points 111, including those corresponding to suspected leaks, that are considered is equal to the number of eigenmodes used, the latter being selected so that the matrix coefficients are sufficiently large. Possibly, the portion of the network 100 modeled in this way can be increased in size until the matrix coefficients combining the added water consumption points 111 with eigenmodes perceptible from the locations used to generate the water hammer and to install sensor 1 are small enough to no longer have a significant influence on the results of the matrix relationship.

[0058] Each matrix used as described above depends on a steady-state pressure value for network 100. If it is possible to increase this steady-state pressure value, for example by using pressurization equipment connected to network 100, the process can be repeated for several different steady-state pressure values. The matrix relationships involved for the different steady-state pressure values ​​are independent, thus providing multiple evaluations of impedance values, including those of suspected leaks. A reliable evaluation of these leaks can therefore be obtained.

[0059] According to an alternative implementation of the monitoring method of the invention, the eigenmodes of the network 100 can be detected using an excitation mode different. An acoustic wave generator can be installed, for example submerged, at a point in the network 100 not too far from sensor 1. This sensor is then controlled to emit a sinusoidal acoustic wave of variable frequency into the water, for example, with a frequency value that gradually increases over time, and with a controlled amplitude. The acoustic wave detected by sensor 1, in the form of sinusoidal temporal variations in water density in passage 16, exhibits resonance peaks whenever the acoustic emission frequency is equal to that of one of the eigenmodes of the network 100. The rest of the process remains identical to the previous implementation.

[0060] Another use of sensor 1 in network 100 is the unclogging of a deposit present on the inner surface of one or more of the pipes 101. The deposit may be organic in nature, for example, consisting of microalgae agglomerated on the inner surface of the pipe. Unclogging is triggered by shear stresses produced in the deposit when an acoustic wave propagates along it. It is more effective when the intensity of the acoustic wave is high, and / or when the acoustic wave has a steep wavefront, particularly when the acoustic wave has been generated as a water hammer. However, an excessively powerful acoustic wave could damage network 100, especially at its seals or fittings. It may therefore be important to control and adjust the intensity of the acoustic wave, particularly the water hammer.Sensor 1 can be used for this purpose, allowing the amplitude of the acoustic wave to be measured both at a distance from its point of generation and from the location of the deposit to be cleaned. A precise relationship between the amplitude of the acoustic wave measured by sensor 1 and its amplitude at the deposit location can be obtained through theoretical and / or numerical modeling. Such modeling can be performed by decomposing the pressure wave into its natural vibrational modes and convolving it with respect to time using a valve closing law that generates the pressure wave. Once the device is installed, the numerical model is calibrated by determining the valve closing law using a standard deconvolution technique that utilizes the known impulse response and is applied to the pressure signal collected by the acoustic wave sensor.Once the installation is calibrated, the prediction can then be adapted to the specific flow rate case. used. Depending on this flow rate in the pipeline, we can indeed adapt the calibrated prediction allowing us to know at every point not only the pressure as a function of time, but also the instantaneous shear rate which is precisely the mechanical quantity related to unclogging.

[0061] It is understood that the invention can be reproduced by modifying minor aspects of the embodiments and implementations described in detail above, while retaining at least some of the advantages mentioned. Furthermore, all numerical values ​​cited are for illustrative purposes only and may be changed depending on the application.

Claims

Demands

1. A sensor (1) of acoustic waves propagating in a liquid, comprising: - an elongated rigid support (10), which extends between two ends, respectively called proximal end (11) and distal end (12), and which is designed to form a first optical path (C1) between the two ends; - an optical source (13), which is fixed to the support (10) at the proximal end (11) and oriented to emit radiation (R) towards the distal end (12) in the first optical path (C1) during use of the sensor (1); - a first retroreflector or retrodiffuser (14), which is fixed to the support (10) at the distal end (12) and oriented to reflect at least part of the radiation (R) towards the proximal end (11), still in the first optical path (C1); and - a first optical detection and analysis system (15), which is located at the proximal end (11) and adapted to detect at least a part of the reflected radiation (R), and which is configured to deduce values ​​of variations of an optical length of the first optical path (C1), the support (10) having a transverse passage (16) which is located between the two ends (11, 12) of the support, and adapted to allow the liquid to fill the passage, so that the liquid is traversed by the first optical path (C1), such that the first optical detection and analysis system (15) is sensitive to variations in liquid density produced by an acoustic wave propagating in the liquid through the passage, characterized in that the sensor (1) further comprises: - sealing means, suitable for establishing a seal between an inside and an outside of a pipe (101) which contains the liquid when at least part of the support (10), including the passage (16), is placed inside the pipe and is filled with the liquid.

2. Sensor (1) according to claim 1, wherein the sealing means comprise: - peripheral sealing means (17), which are located on the support (10) between the proximal end (11) and the passage (16), and which are adapted to form, after the support has been threaded through a wall (102) of the pipe (101) so that the passage (16) is inside the pipe and the proximal end (11) of the support is outside the pipe, a watertight connection between an external peripheral surface of the support and the pipe wall; and - optionally, a window (18) which is transparent to radiation (R) and located across the first optical path (C1) between the passage (16) and the proximal end (11), and which is hermetically connected to the peripheral sealing means (17).

3. Sensor (1) according to claim 1 or 2, wherein the passage (16) is provided to allow the liquid to flow through the support (10), obliquely or perpendicularly to a longitudinal direction (L) of the support, without changing an average direction of flow of the liquid in the pipe (101) relative to said pipe without the sensor.

4. Sensor (1) according to any one of the preceding claims, further comprising: - a second optical path (C2) which also extends between the two ends (11, 12) of the support (10), parallel to the first optical path (C1); - a radiation entry in the second optical path (C2), which is located at the proximal end (11) and configured to introduce additional radiation (R') into said second optical path in the direction of the distal end (12); - a second retro-reflector or retro-diffuser (14'), which is fixed to the support (10) at the level of the distal end (12), and oriented to reflect at least part of the additional radiation (R') towards the proximal end (11), in the second optical path (C2); - a second optical detection and analysis system (15'), which is located at the proximal end (11) and adapted to detect at least part of the additional reflected radiation (R'), and configured to deduce variations in an optical length of the second optical path (C2); - a separation tube (10'), which contains the second optical path (C2) and is either empty or filled with a transparent gas, and which is designed to remain liquid-tight between the inlet of radiation in the second optical path and the second retroreflector or retrodiffuser (14') while said liquid is in the passage (16), so that the second optical detection and analysis system detects variations in the length of the support (10); and - a data processing module (20), configured to remove a contribution from the variations in length of the support (10), to the variations in optical length of the first optical path (C1), so as to isolate a contribution from the acoustic wave which propagates in the liquid through the passage (16).

5. Sensor (1) according to any one of the preceding claims, further comprising at least one of the following additional modules: • a self-contained power supply module, comprising: - a turbine (19a), which is arranged outside the passage (16), and so as to be driven in rotation by a flow of the liquid; - a dynamo (19b), which is coupled to the turbine (19a) to generate electrical energy; and - an electrical energy management system, which is connected at the input to the dynamo (19b) to collect electrical energy, and configured to power the sensor (1) so as to allow energy-autonomous operation of said sensor; and • a radio transmit-receive module (19c), configured to transmit to an external receiver, radio signals which contain first measurement results produced by the first optical detection and analysis system (15), and where applicable second measurement results produced by the second optical detection and analysis system (15'), or a result of removing the contribution of the length variations of the support (10) to the optical length variations of the first optical path (C1).

6. A method for monitoring a liquid conveyance network (100), in particular a water distribution network, a network or pipeline for the transport of oil or petroleum products, or a network or pipeline for the transport of an agri-food liquid, wherein a sensor (1) conforming to one of the preceding claims is threaded in a sealed manner through a wall (102) of a pipeline (101) of the network, such that the passage (16) is inside the pipeline, and including the following steps: / i / generate at least one acoustic wave in the liquid; then / ii / after the acoustic wave has propagated through at least part of the network (100), detect the acoustic wave using the sensor (1).

7. A method according to claim 6, used to detect at least one leak of the liquid to the outside of the network (100), wherein: in step / i / : the acoustic wave is generated in the liquid so as to generate acoustic components that propagate in the network (100) according to natural modes; and step / ii / is carried out after the acoustic components have undergone internal reflections within the network (100), including reflections at the leak, the method further comprising the following step: / iii / deduce a value of a leakage impedance and a location of said leakage in the network (100) from changes in respective natural frequency values ​​of the natural modes.

8. A method according to claim 7, wherein step / iii / is performed using a linear matrix relationship between the changes in respective natural frequency values ​​of the natural modes and leakage impedances which are assigned respectively to locations in the network (100).

9. A method according to claim 7 or 8, wherein a steady-state pressure of the liquid in the network (100) is established successively at several different values, and steps / i / , / ii / and / iii / are carried out for each value of the steady-state pressure, steps / i / and / ii / while said steady-state pressure value is effective, and according to which the leak impedance value and the location of said leak in the network (100) are deduced from the changes in the respective frequency values ​​of the natural modes, obtained respectively for the plurality of steady-state pressure values.

10. A method according to any one of claims 7 to 9, wherein the acoustic wave is generated in step / i / using a sinusoidal acoustic wave generator and by varying a frequency of the sinusoidal acoustic wave which is generated, and wherein the natural frequency values ​​of the natural modes are identified in step / iii / as values ​​of the frequency of the generated sinusoidal acoustic wave which correspond to amplitude resonances of the acoustic wave as detected in step / ii / .

11. A method according to claim 6, used to unclog at least one deposit present on at least a part of a surface internal to the network (100), by a shear which is produced by the acoustic wave inside the deposit, and according to which step / ii / is used to adjust an intensity of the acoustic wave generated during a repetition of step / i / .

Images