Method for detecting the fill level of a tank

Abstract

A method for detecting the fill level of a tank comprising an electrical conductor arranged in or on the tank wall in a coiled arrangement, the method comprising the steps of, at a given instant: Injecting (701) a signal at a first end of the electrical conductor, Measuring (702) the signal after its propagation in the electrical conductor, Analyzing (703) the signal measurement to identify a characteristic amplitude peak of reflection or transmission of the signal at a second end of the electrical conductor opposite the first end, measuring the time of flight associated with said peak, Comparing (704) the time of flight measured at the current instant with a time of flight measured on the same amplitude peak for a measurement taken at an earlier instant, Characterizing (705) the fill level of the tank from the comparison of the times of flight. Figure 7

Description

Title of the invention: Method for detecting the fill level of a tank

[0001] The invention relates to the field of pressure vessels and more specifically to methods of detecting the filling state of such a vessel.

[0002] The invention also relates to the field of deformation analysis techniques by reflectometry or transferometry which involve signal processing methods to analyze signals reflected or transmitted in an electrical conductor.

[0003] A general problem to be solved concerns the detection of the filling state of a tank by automated means without human intervention.

[0004] Monitoring the filling cycle of a tank is of particular, but not exclusive, interest in the fields of transport or energy production and storage. The tank may be, for example, a high-pressure gas tank or a tank containing a liquid.

[0005] Reference [1] describes a device for instrumenting a tank using optical fiber to characterize defects on a tank. A drawback of this solution is the high cost of the interrogators used and the limited bending of the optical fiber for cylindrical tanks.

[0006] Reference [2] proposes another solution based on guided wave sensors. The drawback of this solution is the complex distribution of the sensors required to ensure complete coverage of the tank. Furthermore, the two aforementioned solutions do not allow for determining the tank's fill level but only for characterizing deformation defects in the tank's structure.

[0007] The Applicant's patent application FR3028034 describes a method for measuring the volume of an inclined liquid in a tank using a reflectometry device. This method is based on the detection of interface points between the liquid and the air and is not applicable to gas tanks.

[0008] The invention aims in particular to propose a method for detecting the filling state of a tank which is of low complexity and which allows for automated detection in a precise manner.

[0009] The invention relates to a method for detecting the fill level of a tank comprising an electrical conductor arranged in or on the tank wall in a coiled arrangement, the method comprising the steps of, at a current instant: - Inject a signal at one end of the electrical conductor, - Measure the signal after it propagates through the electrical conductor. - Analyze the signal measurement to identify a characteristic amplitude peak indicating reflection or transmission of the signal at a second end of the electrical conductor opposite the first end, and measure the time of flight associated with said peak. - Compare the flight time measured at the current time to a flight time measured at the same amplitude peak for a measurement taken at a previous time. - Characterize the state of filling of the tank from the comparison of flight times.

[0010] According to a particular aspect of the invention, the step of characterizing the state of filling the tank includes comparing the difference between the flight time measured at the current time and the flight time measured at the time prior to at least a predetermined threshold.

[0011] According to a particular aspect of the invention, the measurement carried out at an earlier time is a reference measurement carried out respectively for an empty tank or a full tank.

[0012] In one embodiment, the method according to the invention includes the detection of a filled state of the tank when the reference measurement is carried out for an empty tank and the flight time measured at the current time is greater than the flight time measured at the previous time by at least the value of said threshold.

[0013] In one embodiment, the method according to the invention includes the detection of an empty state of the tank when the reference measurement is carried out for a full tank and the flight time measured at the current time is less than the flight time measured at the previous time by at least the value of said threshold.

[0014] In one embodiment, the method according to the invention includes the detection of a tank filling phase when the flight time measured at the current time is greater than the flight time measured at the previous time by at least a predefined value.

[0015] In one embodiment, the method according to the invention includes the detection of a tank emptying phase when the flight time measured at the current time is less than the flight time measured at the previous time by at least a predefined value.

[0016] In one embodiment, the method according to the invention further includes the comparison of the respective amplitudes of the peak amplitude measured at the current time and at an earlier time, the state of filling of the tank being further characterized from the comparison of said amplitudes.

[0017] According to a particular aspect of the invention, the step of analyzing the signal measurement is carried out by means of a reflectogram characterizing the back-propagated signal and measured at the first end of the electrical conductor after its reflection on the second end of the electrical conductor.

[0018] According to a particular aspect of the invention, the step of analyzing the signal measurement is carried out by means of a transferogram characterizing the signal transmitted through the electrical conductor and measured at its second end.

[0019] The invention also relates to a tank comprising an electrical conductor arranged in or on the wall of the tank in a spiral winding and a measuring device connected to at least one end of the conductor, the measuring device being configured to perform the steps of the method for detecting the state of filling of a tank according to the invention.

[0020] Other features and advantages of the present invention will become more apparent from the following description in relation to the following accompanying drawings.

[0021] [Fig. 1] represents a schematic diagram illustrating the principle of reflectometry,

[0022] [Fig.2] represents a tank equipped with a system for detecting the state of filling the tank according to a first embodiment of the invention,

[0023] [Fig.3] represents an example of a reflectogram for an empty tank and a full tank

[0024] [Fig.4] represents an enlargement of the area of ​​the cable end peak of the reflectogram of [Fig.3],

[0025] [Fig. 5] represents a tank equipped with a system for detecting the state of filling the tank according to a second embodiment of the invention,

[0026] [Fig.6] represents an example of a transferogram for an empty tank and a full tank

[0027] [Fig.7] represents a flowchart illustrating the steps for implementing a method for detecting the fill level of a tank according to the invention

[0028] Figure 1 illustrates the principle of reflectometry applied to the analysis of the condition of a cable, according to the prior art. The principle consists of injecting a signal with a controlled waveform at an injection point INJ of the cable. The incident wave OI propagates along the cable until it encounters an impedance discontinuity caused by the cable end FC. A portion of the wave is reflected back to the injection point INJ. Using measuring equipment, the reflected signal is measured. Analysis of the measured signal makes it possible to detect and locate any fault.

[0029] A correlator performs the cross-correlation between the generated signal and the received signal in order to produce a time-domain reflectogram R(t). If the reflectometry signal used is a simple time pulse, the correlator can be made optional.

[0030] The time reflectogram R(t) has a characteristic amplitude peak at the end of the cable at a time abscissa which is related to the distance between the signal measurement point and the end of the cable and to the speed of propagation of the signal in the cable.

[0031] As is known in the field of reflectometry diagnostic methods, the position dF c of the end of the cable, in other words its distance to the point of injection of the signal, can be directly obtained from the measurement, on the time reflectogram, of the duration tFC between the first peak of amplitude recorded on the reflectogram and the peak of amplitude corresponding to the signature of the end of the cable.

[0032] Various known methods are conceivable for determining the position dF c. One possible method consists of applying the relationship linking distance and time: dF c =Vt fc / 2 where V is the speed of propagation of the signal in the cable.

[0033] Reference [3] describes another method for analyzing impedance discontinuities in a cable, using transferometry. Unlike reflectometry, transferometry aims to measure the transmitted portion of the wave after the impedance discontinuity, this portion corresponding to the direct path of the signal. In this case, the signal is injected at one end of the cable and measured at the other end. Similarly, the measured transferogram has a characteristic amplitude peak at the end of the cable at a time abscissa that is related to the distance between the signal measurement point and the second measurement point located at the end of the cable, and to the signal propagation velocity in the cable.

[0034] Figure [Fig. 2] represents a reservoir 200, for example of cylindrical shape, comprising pressurized gas or a liquid. The tank 200 is equipped with an electrical conductor 202 arranged on the outer wall of the tank in a spiral winding. The electrical conductor 202 is connected at one end to a reflectometer 201 configured to measure a reflectogram by injecting and measuring a signal in the electrical conductor. The reflectometer 201 is further configured to implement a method for detecting the tank's fill level according to a first embodiment of the invention.

[0035] The electrical conductor 202 can be disposed on the outer wall, inside the wall or on the inner wall of the tank.

[0036] When the tank is filled under maximum pressure, an elongation of the electrical conductor is observed which is, for example, on the order of 1% of its length.

[0037] Reflectometry measurement makes it possible to detect this elongation by measuring the evolution of the characteristic peak at the end of the cable.

[0038] Fig. 3 shows an example of a reflectogram respectively for an empty tank (curve 300) and a full tank (curve 301) for an example of an electrical conductor having a length of 30 m.

[0039] It is observed that when the tank is full, the end-of-cable peak shifts to the right by a value approximately equal to 1% of the cable length compared to the end-of-cable peak measured for an empty tank.

[0040] Thus, the analysis of the cable end peak on the reflectogram makes it possible to characterize the state of filling of the tank, in particular whether it is empty or full.

[0041] Figure 4 shows an enlargement of Figure 3 in the area of ​​the cable end peak. It can be observed both that the position of the cable end peak shifts to the right (due to the elongation of the cable) and that the amplitude of the peak decreases (due to greater attenuation related to the increasing cable length).

[0042] Figure 5 shows a tank 500, for example cylindrical in shape, containing pressurized gas or a liquid. The tank 500 is equipped with an electrical conductor 502 arranged on the outer wall of the tank in a coiled configuration. The electrical conductor 502 is connected at one end to a transferometer signal transmitter 501 configured to inject a signal into the electrical conductor 501. The electrical conductor 502 is connected at the other end to a transferometer device 503 configured to measure the transmitted signal and to implement a method for detecting the tank's fill level according to a second embodiment of the invention.

[0043] Figure 6 shows, under the same conditions as the measurements in Figure 3, a transferogram obtained respectively for an empty tank (curve 600) and for a full tank (curve 601). A characteristic peak is also observed at the end of the cable, which shifts to the right when the tank is filled due to cable elongation.

[0044] Thus, the invention consists in particular of detecting the filling of the tank via the elongation of the conductor by means of a reflectometry or transferometry measurement.

[0045] Figure 7 illustrates, in a flowchart, the steps for implementing a method for determining the state of filling of a tank according to an embodiment of the invention.

[0046] The method is implemented using any one of the detection systems described in Figures 2 or 5.

[0047] In step 701, a reflectometry or transferometry signal is injected at one end of the electrical conductor. The signal is, for example, a time pulse or a more complex signal, for example, a multi-carrier signal.

[0048] In step 702, the signal reflected or transmitted in the electrical conductor is measured so as to generate a reflectogram of the type described in [Fig.3] or a transferogram of the type described in [Fig.6].

[0049] In step 703, the characteristic peak of the reflection of the signal on the end of the cable or of the direct path taken by the signal between the transmitter and the receiver is identified on the reflectogram or the transferogram.

[0050] In particular, its time abscissa, which corresponds to the time of flight of the signal, is identified, and optionally its amplitude. In the case of a reflectometry method, the time of flight corresponds to a round trip between the signal injection point and the end of the cable. In the case of a transferometry method, the time of flight corresponds to a direct path between these same points.

[0051] In step 704, the measured flight time is compared to a flight time achieved on a previous measurement for the same characteristic peak at the end of the conductor.

[0052] According to a first embodiment, the previous measurement is a reference measurement. For example, it is taken on an empty tank. Conversely, it can be taken on a full tank.

[0053] The comparison step 704 consists of comparing the difference in flight times to a predefined threshold.

[0054] In step 705, a state of filling of the tank is determined from the comparison.

[0055] If the reference measurement corresponds to an empty state of the tank and the flight time measured in step 703 is greater than the reference flight time by at least a predefined value, we conclude that the conductor has lengthened and therefore that the tank is in a full state.

[0056] If the reference measurement corresponds to a full state of the tank and the flight time measured in step 703 is less than the reference flight time by at least a predefined value, we conclude that the conductor is shortened and therefore that the tank is in an empty state.

[0057] The predefined value is for example on the order of 1% of the total length of the conductor, converted into time of flight as a function of the speed of propagation of the signal.

[0058] In one embodiment, the earlier measurement is not a reference measurement taken for a tank with a controlled fill level, but a measurement taken at an earlier time. In this case, a tank filling phase is detected in step 705 when the flight time measured at the current time is greater than the flight time measured at the earlier time by at least a predefined value. Conversely, a tank emptying phase is detected in step 705 when the flight time measured at the current time is less than the measured flight time. at the previous instant of at least one predefined value. One advantage of this variant is that it allows continuous monitoring of the tank's fill level over time.

[0059] In another embodiment, step 704 further consists of comparing the amplitude of the characteristic peak at the end of the conductor to a previous amplitude measurement of the same characteristic peak. The amplitude of the end-of-conductor peak is typically lower when the tank is full than when it is empty (as illustrated in [Fig. 4]).

[0060] Thus, when the amplitude of the peak measured at the current time is lower, by a predefined value, than the amplitude of the corresponding peak in a previous measurement, this is characteristic of a tank filling phase or even a filled state. According to one embodiment of the invention, the predefined value is determined based on the aging state of the material constituting the tank wall.

[0061] Conversely, if the amplitude of the peak measured at the current time is greater, by a predefined value, than the amplitude of the corresponding peak on a previous measurement, this is characteristic of a phase of emptying the tank or even of an empty state.

[0062] According to one embodiment of the invention, the predefined value is determined based on the aging state of the material constituting the tank wall. References

[0063] [1] OFDR Distributed Strain Measurements for SHM of Hydrostatic Stressed Structures: An Application to High Pressure Hydrogen Storage Type IV Composite Vessels - H2E Project (LSPM, CEA le Ripault) (2014)

[0064] [2] Damage localization in hydrogen storage vessel by guided waves based on a real- time monitoring System, Bin Yang, School of mechanical and power engineering

[0065] [3] Transferometry: a new tool for complex wired networks diagnosis, F. Auzanneau, Progress in Electromagnetics Research B, vol 70, 87-100,2016

Claims

Demands

1. A method for detecting the filling state of a tank comprising an electrical conductor arranged in or on the wall of the tank in a winding of turns, the method comprising the steps of, at a current time: - Injecting (701) a signal at a first end of the electrical conductor, - Measuring (702) the signal after its propagation in the electrical conductor, - Analyzing (703) the signal measurement to identify an amplitude peak characteristic of a reflection or transmission of the signal on a second end of the electrical conductor opposite to the first end, measuring the time of flight associated with said peak, - Comparing (704) the time of flight measured at the current time to a time of flight measured on the same amplitude peak for a measurement made at an earlier time, - Characterizing (705) the filling state of the tank from the comparison of the times of flight.

2. Method for detecting the state of filling of a tank according to claim 1 wherein the step of characterizing (705) the state of filling of the tank includes comparing the difference between the flight time measured at the current time and the flight time measured at the previous time to at least a predetermined threshold.

3. Method for detecting the state of filling of a tank according to claim 2 wherein the measurement taken at an earlier time is a reference measurement taken respectively for an empty tank or a filled tank.

4. Method of detecting the fill state of a tank according to claim 3 comprising detecting a filled state of the tank when the reference measurement is performed for an empty tank and the flight time measured at the current time is greater than the flight time measured at the previous time by at least the value of said threshold.

5. A method for detecting the fill level of a tank according to claim 3, comprising detecting an empty tank when the reference measurement is performed for a full tank and the flight time measured at the current time is less than the flight time measured at the previous time by at least the value of said threshold.

6. Method of detecting the filling state of a tank according to claim 2 comprising detecting a tank filling phase when the flight time measured at the current time is greater than the flight time measured at the previous time by at least a predefined value.

7. Method of detecting the filling state of a tank according to claim 2 comprising detecting a tank emptying phase when the flight time measured at the current time is less than the flight time measured at the previous time by at least a predefined value.

8. Method for detecting the state of filling of a tank according to any one of the preceding claims further comprising the comparison of the respective amplitudes of the peak amplitude measured at the current time and at an earlier time, the state of filling of the tank being further characterized from the comparison of said amplitudes.

9. Method of detecting the filling state of a tank according to any one of the preceding claims wherein the analysis step (703) the signal measurement is carried out by means of a reflectogram characterizing the back-propagated signal and measured at the first end of the electrical conductor after its reflection on the second end of the electrical conductor.

10. Method for detecting the filling state of a tank according to any one of claims 1 to 8 wherein the step of analyzing (703) the signal measurement is carried out by means of a transferogram characterizing the signal transmitted through the electrical conductor and measured at its second end.

11. Tank (200,500) comprising an electrical conductor (202,502) arranged in or on the wall of the tank (200,500) in a coiled arrangement and a measuring device (201,501,503) connected to at least one end of the conductor, the measuring device being configured to perform the steps of the method for detecting the fill state of a tank according to any one of the preceding claims.

Images