SYSTEM AND METHOD FOR MEASURING THE FILLING LEVEL OF A FLUID CONTAINER BY MEANS OF ACOUSTIC WAVES

Abstract

Disclosed is a system for measuring the filling level of a container, by means of acoustic waves, the system comprising at least three transducers configured to be positioned at different vertical heights on the outer face of the cylindrical portion of the casing, each transducer being configured to emit, upon receiving a first electric signal, an incident acoustic wave at the outer face of a wall of the casing, and to emit a second electric signal upon receiving a reflected acoustic wave generated by the reflection of the incident acoustic wave on the inner face of the wall, and at least one calculation unit configured to determine, based on the electric signals and physical properties of the monophase fluids, the presence of the first monophase fluid at each vertical height of the transducers.

Claims

1-11. (canceled)

12. An acoustic wave measurement system for measuring the fill level of a tank, said tank storing a first single-phase fluid having first physical properties and a second single-phase fluid having second physical properties, said first physical properties comprising a first density ρ.sub.1 and said second physical properties comprising a second density ρ.sub.2 strictly lower than the first density ρ.sub.1 so that the single-phase fluids are vertically superimposed in the tank, the first single-phase fluid being located in the lower part of the tank, the second single-phase fluid being located in the upper part of the tank, said first single-phase fluid and said second single-phase fluid being separated by a substantially horizontal interface, said tank comprising an envelope extending longitudinally along an axis X10, the envelope comprising an inner face in contact with the single-phase fluids and an outer face, the envelope comprising a cylindrical median portion, wherein the measurement system comprises: at least three transducers configured to be positioned at different vertical heights on the outer face of the cylindrical portion of the envelope, each transducer being configured, on the one hand, to emit, upon receiving a first electric signal, an incident acoustic wave to the outer face of a wall of the envelope and, on the other hand, to emit a second electric signal, upon receiving a reflected acoustic wave, corresponding to the reflection of the incident acoustic wave on the inner face of said wall, the reflected acoustic wave having passed only through the wall of the envelope, the second electric signal being a function of the difference in acoustic energy between the incident acoustic wave and the reflected acoustic wave, at least one calculation member configured to determine, from the electric signals and physical properties of the single-phase fluids, the presence of the first single-phase fluid at each vertical height of the transducers and to deduce the fill level therefrom.

13. The measurement system according to claim 12, wherein the transducers are aligned along a line in a vertical plane.

14. The measurement system according to claim 13, wherein the transducers are aligned along a rectilinear line.

15. The measurement system according to claim 12, wherein the transducers are configured to emit horizontal incident acoustic waves.

16. The measurement system according to claim 12, wherein the difference in acoustic energy is determined from acoustic impedances of the single-phase fluids.

17. The measurement system according to claim 12, wherein the calculation member is configured to compare the electric signals to a database comprising reference acoustic attenuations of the single-phase fluids for said tank to determine the presence of the first single-phase fluid at each vertical height of the transducers.

18. The measurement system according to claim 12, wherein, with the calculation member being configured to determine, for each transducer, a lower state in the presence of the first single-phase fluid or an upper state in the absence of the first single-phase fluid, the calculation member is configured to determine the height of the interface from the height of the two successive transducers, one of which is in a lower state and the other in an upper state.

19. The measurement system according to claim 12, wherein the transducers are configured to emit, further to the incident wave, a complementary incident wave in the wall of the envelope of the tank, the trajectory of this complementary incident wave being oriented by a measurement angle with respect to that of the incident acoustic wave so as to generate a complementary reflected acoustic wave which is received by a transducer adjacent to the transducer having emitted the incident waves.

20. The measurement system according to claim 12, wherein the calculation member is configured to measure the acoustic attenuation at each transducer to a database comprising reference acoustic attenuations of the single-phase fluids for said tank and for said measurement angle.

21. An assembly of a tank and the measurement system according to claim 12.

22. An acoustic wave measurement method for measuring the fill level of a tank, implemented by means of the measurement system according to claim 12, the method comprising: a step of emitting by each transducer an incident acoustic wave to the outer face of a wall of the envelope following the reception of a first electric signal, a step of receiving a reflected acoustic wave by each transducer, generated by the reflection of the incident acoustic wave on the inner face of said wall, the reflected acoustic wave having passed only through the wall of the envelope, the second electric signal being a function of the difference in acoustic energy between the incident acoustic wave and the reflected acoustic wave, a step of determining the presence of the first single-phase fluid at each vertical height of the transducers from the electric signals and the physical properties of the single-phase fluids, and a step of determining the fill level as a function of the presence of the first single-phase fluid at each vertical height of the transducers.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The invention will be better understood upon reading the following description, which is given solely by way of example, and referring to the appended drawings given as non-limiting examples, in which identical references are given to similar objects and in which:

[0036] FIG. 1 is a schematic representation of a first acoustic wave measurement method for measuring the fill level of a fluid tank according to prior art,

[0037] FIG. 2 is a schematic representation of a second acoustic wave measurement method for measuring the fill level of a fluid tank according to prior art,

[0038] FIG. 3 is a schematic representation of a third acoustic wave measurement method for measuring the fill level of a fluid tank according to prior art,

[0039] FIG. 4 is a schematic representation of the acoustic wave measurement method upon detecting the presence of fluid at several vertical heights for a horizontal tank,

[0040] FIG. 5 is a cross-section view of the tank of FIG. 4,

[0041] FIGS. 6A and 6B are schematic representations of the incident wave emitted to the outer face of the wall and the wave reflected from the inner face of said wall at several vertical heights,

[0042] FIGS. 7 and 8 are close-up schematic representations of an emission of a horizontal incident wave and a complementary incident wave,

[0043] FIG. 9 is a schematic representation of the acoustic wave measurement method upon emitting a complementary incident wave,

[0044] FIG. 10 is a schematic representation of the acoustic wave measurement method upon detecting the presence of fluid at several vertical heights for a vertical tank, and

[0045] FIG. 11 is a schematic representation of the steps of the acoustic wave measurement method for measuring the fill level of a fluid tank.

[0046] It should be noted that the figures disclose the invention in detail to implement the invention, said figures can of course be used to better define the invention if necessary.

DETAILED DESCRIPTION

[0047] With reference to FIG. 4, an acoustic wave measurement system for measuring the fill level N.sub.R of a tank 10 is schematically represented.

[0048] In the following, the vertical direction is defined as the direction of the axis of gravity and the horizontal direction as the direction perpendicular to the vertical. The terms “low,” “high,” “upper,” and “lower” are determined with respect to the vertical direction.

[0049] As illustrated in FIG. 4, there is represented a tank 10 storing a first single-phase fluid F.sub.1, a liquid phase refrigerant for example, and a second single-phase fluid F.sub.2, in this example gas phase refrigerant, in particular, the same.

[0050] The first single-phase fluid F.sub.1 has first physical properties P.sub.1, especially a first density ρ.sub.1, and the second single-phase fluid F.sub.2 has second physical properties P.sub.2, especially a second density ρ.sub.2 strictly lower than the first density ρ.sub.1, so that the first single-phase fluid F.sub.1 is located in the lower part of the tank 10 and the second single-phase fluid F.sub.2 is located in the upper part of the tank 10. The theoretical speeds of propagation V.sub.1, V.sub.2 of an acoustic wave in the single-phase fluids F.sub.1, F.sub.2 are also known. As will be set forth later, the tank 10 further comprises a temperature probe (not represented) so as to determine temperature T.sub.1, T.sub.2 of the fluids F.sub.1, F.sub.2.

[0051] As illustrated in FIG. 4, the first single-phase fluid F.sub.1 and the second single-phase fluid F.sub.2 are thus separated by a substantially horizontal interface I. The height of this interface I corresponds to the fill level N.sub.R of the tank 10 with the first single-phase fluid F.sub.1. In this example, the first single-phase fluid F.sub.1 and the second single-phase fluid F.sub.2 are respectively a liquid and a gas, but of course they could be in the form of two different gases or even two different liquids.

[0052] The tank 10 extends longitudinally along an axis X10 which is, in this first embodiment, horizontal. For the sake of clarity and brevity, a tank having a longitudinal axis extending vertically will be referred hereinafter to as a “vertical tank” and a tank having a longitudinal axis extending horizontally will be referred to as a “horizontal tank”.

[0053] With reference to FIG. 5, the tank 10 comprises an envelope 11 having a wall comprising an inner face 14 in contact with the single-phase fluids F.sub.1, F.sub.2 and an outer face 12 opposite to the inner face 14. The tank 10 comprises two ends and a cylindrical median portion 13 that extends along the axis X10. In the following, a cylindrical median portion 13 having an annular cross sectional area will be set forth, being particularly adapted to distribute pressure forces, but of course it could be different. Such a tank 10 is known to the person skilled in the art and will not be set forth in more detail.

[0054] A system for measuring the fill level N.sub.R of the tank 10 according to the invention will now be set forth with reference to FIGS. 4 and 5.

[0055] In this example, the measurement system 20 comprises a plurality of transducers 22a-22g positioned on the tank 10 at different vertical heights as well as a control member 21 and a calculation member 23 which are connected to the transducers 22a-22g.

[0056] Each transducer 22a-22g is configured, on the one hand, to emit, upon receiving a first electric signal U.sub.1, an incident acoustic wave O.sub.1 to the outer face 12 of a wall of the envelope 11 and, on the other hand, to emit a second electric signal U.sub.2, upon receiving a reflected acoustic wave O.sub.2, corresponding to the reflection of the incident acoustic wave O.sub.1 on the inner face 14 of said wall, the reflected acoustic wave O.sub.2 having passed only through the wall of the envelope 11, the second electric signal U.sub.2 being a function of the difference in acoustic energy between the incident acoustic wave O.sub.1 and the reflected acoustic wave O.sub.2. In other words, unlike prior art which teaches to measure the acoustic attenuation in a fluid between the walls of the envelope 11, the present invention suggests to focus only on the acoustic attenuation of the wall of the envelope 11, that is, within the wall thickness. The reflected acoustic wave O.sub.2 is received faster than in prior art and has a greater acoustic power, which facilitates its processing since it traveled a shorter distance. This significantly increases accuracy.

[0057] Unlike prior art which teaches to measure acoustic attenuation directly in a fluid, the present invention is directed to an indirect measurement by analyzing the acoustic attenuation of the wall of the envelope 11. Advantageously, the applicant has realized that the acoustic attenuation of the wall of the envelope 11 depends on the presence or absence of fluid on the inner face 14 of the wall. Advantageously, each transducer 22 is of the piezoelectric type and allows an electric signal to be converted into a mechanical stress (vibration) and vice versa. However, other types of transducers 22a-22g could of course be used, for example, PZT ceramics, PVDF polymers, etc. Preferably, the incident acoustic wave O.sub.1 is a sinusoidal pulse.

[0058] Each transducer 22a-22g thus enables the acoustic attenuation between the incident wave O.sub.1 and the reflected wave O.sub.2 to be measured through the inner face 14 of the wall of the envelope 11.

[0059] As will be set forth later, the calculation member 23 is configured to determine, from the electric signals U.sub.1-U.sub.2 and the physical properties P.sub.1-P.sub.2 of the single-phase fluids F.sub.1, F.sub.2, the presence of the first single-phase fluid F.sub.1 at each vertical height of the transducers 22a-22g and to deduce the fill level N.sub.R therefrom.

[0060] The transducers 22a-22g are positioned at different vertical heights so as to detect the presence of the first single-phase fluid F.sub.1 at different heights and thereby deduce the fill level N.sub.R therefrom. Advantageously, the number of transducers 22a-22g is chosen as a function of the accuracy desired. The transducers 22a-22g are positioned in contact with the outer face 12 of the cylindrical portion 13 of the envelope 11 so as to optimally emit/receive into the wall of the envelope 11.

[0061] The control member 21 is in the form of a calculation unit configured to emit the first electric signal U.sub.1 at predetermined times. For this purpose, the control member 21 comprises a clock.

[0062] In a similar manner, the calculation member 23 is in the form of a calculation unit configured to receive the second electric signal U.sub.2 and to determine the time of reception. For this purpose, the calculation member 23 comprises a clock.

[0063] Preferably, the control member 21 and the calculation member 23 are integrated in a same calculation module, for example, an electronic board. Preferably, the calculation module comprises a battery for powering the control member 21, the calculation member 23 and the transducers 22.

[0064] Preferably, the measurement system 20 further comprises a communication unit for communicating, in a wired or wireless manner, the fill level N.sub.R which has been determined. This is particularly advantageous when the measurement system comprises a signaling device as disclosed in patent application FR1871656.

[0065] Preferably, the measurement system 20 comprises a flexible support to which the transducers 22a-22g are mounted. This allows the transducers 22a-22g to be accurately positioned with respect to each other, thereby improving the accuracy of the measurement of the fill level N.sub.R. Preferably, the flexible support is connected to the tank by bonding, magnetization, or the like.

[0066] The calculation member 23 is configured to determine the presence of the first single-phase fluid F.sub.1 at each vertical height of the transducers 22a-22g from electric signals U.sub.1, U.sub.2 from each transducer 22a-22g. Thereafter, when a transducer 22a-22g detects the presence of the first single-phase fluid F.sub.1, it is considered to be in a lower state ET.sub.1, (presence of fluid) while it is considered to be in an upper state ET.sub.2 otherwise (absence of fluid).

[0067] Thus, the calculation member 23 makes it possible to define a first group of transducers in the lower state ET.sub.1, and a second group of transducers in the upper state ET.sub.2. The calculation member 23 can thus conveniently and quickly deduce the interface height I at the interface between both groups of transducers. The height of interface I is higher than any transducer 22 in the lower state ET.sub.1 (presence of fluid) and lower than any transducer 22 in the upper state ET.sub.2 (absence of fluid). In this example, with reference to FIG. 4, transducers 22a-22d are in the lower state ET.sub.1 while transducers 22e-22g are in the upper state ET.sub.2. Thus, it can be deduced therefrom that the interface I is located between transducer 22d and transducer 22e and thus the fill level N.sub.R can be determined.

[0068] In the first embodiment of FIG. 4, the tank 10 is oriented horizontally and the transducers 22a-22g are distributed on the half-circumference of the outer face 12 at the cylindrical median portion 13, in other words along a line curved in a plane transverse to the axis X10 of the tank 10. In this example, the calculation member 23 is configured to determine the presence of the first single-phase fluid F.sub.1 as a function of acoustic attenuation.

[0069] Preferably, the transducers 22a-22g are advantageously equidistant from each other so that the fill level N.sub.R of the tank 10 can be determined with an accuracy calibrated to the pitch of the transducers 22a-22g. The density of transducers 22 could of course be higher near certain critical fill levels.

[0070] In this example, the transducers 22a-22g are aligned along a single plane transverse to the axis X10 of the tank 10 but, of course, in a second configuration, they could extend along a plurality of transverse planes. In particular, the transducers 22a-22g may extend along two transverse planes spaced apart by the order of the size of a transducer 22a-22g and distributed in a staggered manner so that each transducer 22a-22g has a different vertical position. Such an arrangement is advantageous for possessing significant accuracy without being limited by the dimensions of a transducer 22a-22g. Thus, the accuracy can be higher than the size of the transducer 22a-22g.

[0071] As illustrated in FIGS. 4 and 5, for a horizontally oriented tank 10, each transducer 22a-22g is positioned along the tangent to the cylindrical median portion 13 so as to emit the incident wave O.sub.1 and receive the reflected wave O.sub.2 along a direction normal to the tangent to the positioning point of said transducer 22a-22g. As illustrated in FIG. 5, the reflected wave O.sub.2 reflects primarily along the same direction as the incident wave O.sub.1. Because the transducers 22 are at different vertical heights, the incident acoustic waves O.sub.1 propagate in different parts of the tank 10.

[0072] Reflected acoustic waves O.sub.2 are generated by reflection of the incident acoustic waves O.sub.1 on the inner face 14 of the emission wall and propagate in an identical direction but reverse sense to that of the incident acoustic waves O.sub.1 as illustrated in FIG. 4. The reflected acoustic wave O.sub.2 can thus be conveniently received by the transducer 22 having emitted the incident acoustic wave O.sub.1. Because of the proximity of the outer face 12 and the inner face 14, the propagation time is short and the measurement is accurate, as the misalignment between the incident acoustic wave O.sub.1 and the reflected acoustic wave O.sub.2 is small.

[0073] An example of an implementation of a method for measuring the fill level is shown in FIG. 11. The method comprises: [0074] a step of emitting E.sub.1 by each transducer 22 an incident acoustic wave O.sub.1 to the outer face 12 of a wall of the envelope 11 following the reception of a first electric signal U.sub.1, [0075] a step of receiving E.sub.2 a reflected acoustic wave O.sub.2 by each transducer 22 generated by the reflection of the incident acoustic wave O.sub.1 on the inner face 14 of said wall, the reflected acoustic wave O.sub.2 having passed only through the wall of the envelope 11, the second electric signal U.sub.2 being a function of the difference in acoustic energy between the incident acoustic wave O.sub.1 and the reflected acoustic wave O.sub.2 [0076] a step E.sub.3 of determining the presence of the first single-phase fluid F.sub.1 at each vertical height of the transducers 22 from the electric signals U.sub.1, U.sub.2 and the physical properties P.sub.1, P.sub.2 of the single-phase fluids F.sub.1 F.sub.2, in particular, by comparing the acoustic powers, and [0077] a step of determining E.sub.4 the fill level N.sub.R as a function of the presence of the first single-phase fluid F.sub.1 at each vertical height of the transducers 22.

[0078] According to the invention, the incident acoustic waves O.sub.1 propagate only in the wall of the envelope 11 in the direction normal to the plane tangent to the outer face 12. In other words, the incident acoustic waves O.sub.1 propagate in the thickness of the wall of the envelope 11. The reflected acoustic waves O.sub.2 are generated by the reflection of the incident acoustic waves O.sub.1 on the inner face 14 of the wall of the envelope 11 and propagate in an identical direction but reverse sense to that of the incident acoustic waves O.sub.1 as illustrated in FIG. 7. In this way, each reflected acoustic wave O.sub.2 is generated by reflection on the inner face 14 in contact on the other side with only one of the single-phase fluids F.sub.1, F.sub.2. As a function of the single-phase fluid F.sub.1, F.sub.2 in contact with the inner face 14, acoustic energy of the reflected wave O.sub.2 is different as illustrated in FIGS. 6A and 6B. To detect the relevant reflected acoustic wave O.sub.2, it is sufficient to monitor the reflected wave O.sub.2 that is received within a predetermined time interval Δt as illustrated in FIGS. 6A and 6B and measure its amplitude.

[0079] As an example, each transducer 22a-22g is a piezoelectric having a diameter of 10 mm, preferably between 5 mm and 20 mm. Each transducer 22a-22g has a frequency between 0.5 MHz and 2 MHz, preferably in the order of 1 MHz. The amplitude is between 1 and 50V, preferably in the order of 10V. Preferably, the emission steps are spaced from 10 to 100 ms, preferably, in the order of 20 ms.

[0080] As an example, with reference to FIG. 5, the acoustic energy of the reflected wave O.sub.2 is smaller when the inner face 14 is in contact with the first single-phase fluid F.sub.1 (FIG. 6A) than with the second single-phase fluid F.sub.2 (FIG. 6B). Indeed, advantageously, the wall of the envelope 11 forms an energy filter that allows the presence of the first single-phase fluid F.sub.1 with the inner face 14 to be characterized. Such a detection of the nature of the single-phase fluid directly at the measurement location allows to dispense with drawbacks related to the reflection in a transit time measurement through the whole tank.

[0081] The calculation member 23 makes it possible to determine the state ET.sub.1/ET.sub.2 of a transducer 22 from the difference in acoustic energy between the emission of the incident acoustic wave O.sub.1 and reception of the reflected acoustic wave O.sub.2 that it has received. In other words, the calculation member 23 determines the acoustic attenuation Att corresponding to the reflected energy divided by the incident energy as illustrated in FIGS. 6A and 6B. Indeed, the difference in acoustic power varies as a function of the single-phase fluid F.sub.1, F.sub.2 in contact with the inner face 14.

[0082] In practice, the calculation member 23 is configured to compare the difference in acoustic energy at each transducer 22 to a database comprising reference acoustic attenuations of the single-phase fluids F.sub.1 F.sub.2 for said tank 10.

[0083] In this example, the reference acoustic attenuations of the single-phase fluids F.sub.1, F.sub.2 are determined empirically or theoretically from the acoustic impedances Z.sub.1, Z.sub.2 of the single-phase fluids F.sub.1, F.sub.2, size of the tank 10, thickness of its envelope 11, nature of its envelope 11, etc. Preferably, during the installation of the measurement system 20, reference acoustic attenuations of the single-phase fluids F.sub.1, F.sub.2 are determined by the installer, for example, during a calibration phase.

[0084] More precisely, the state ET.sub.1/ET.sub.2 of a transducer 22 is determined from the amplitude attenuation of the reflected acoustic wave O.sub.2 that it has received with respect to that of the incident acoustic wave O.sub.1 that it has emitted. Indeed, when an acoustic wave is reflected on any interface, the fluid located on the other side of the interface absorbs part of the energy of the acoustic wave, which decreases its amplitude. The part of absorbed energy depends on the resistance of the fluid, that is its acoustic impedance, and differs for two different single-phase fluids.

[0085] In practice, the acoustic impedances Z.sub.1, Z.sub.2 of the single-phase fluids F.sub.1, F.sub.2 are calculated from the physical properties P.sub.1, P.sub.2 of the single-phase fluids F.sub.1, F.sub.2, that is, their theoretical densities ρ.sub.1, ρ.sub.2, their theoretical speeds of propagation V.sub.1, V.sub.2 and their temperatures T.sub.1, T.sub.2, measured in the tank 10. However, of course the acoustic impedances Z.sub.1, Z.sub.2 of the single-phase fluids F.sub.1, F.sub.2 can be obtained in a different way. In this example, the temperatures T.sub.1, T.sub.2 are measured by a temperature sensor and transmitted to the communication unit connected to the calculation member 23. Preferably, the step of determining the acoustic impedances Z.sub.1, Z.sub.2 is repeated periodically as the temperatures T.sub.1, T.sub.2 of the single-phase fluids F.sub.1, F.sub.2 change over time.

[0086] The acoustic attenuation is determined according to the following formula:

Att=(Z.sub.2−Z.sub.1).sup.2/(Z.sub.2+Z.sub.1).sup.2  [Math. 1]

[0087] In practice, in the presence of a first single-phase fluid F.sub.1 which is liquid, the reflected acoustic energy is in the order of 97%. Conversely, in the presence of a second single-phase fluid F.sub.2 which is gaseous, the reflected acoustic energy is in the order of 99.8%.

[0088] Advantageously, the acoustic attenuation measured at the lowest positioned transducer 22a can be equated to the acoustic attenuation of the first single-phase fluid F.sub.1. Similarly, the acoustic attenuation measured at the highest positioned transducer 22g can be equated to the acoustic attenuation of the second single-phase fluid F.sub.2. These acoustic attenuations are obtained by a calibration step. Preferably, the calibration step is repeated periodically since the impedances are a function of the temperatures T.sub.1, T.sub.2 of the single-phase fluids F.sub.1, F.sub.2 which change over time.

[0089] Preferably, the acoustic attenuation is calculated from the reflected energy received by the transducer having emitted the incident wave O.sub.1. As illustrated in FIGS. 7 and 8, in order to increase accuracy, following the emission of an incident wave O.sub.1 by a determined transducer 22, the reflected energy received by the transducer 22 located directly below the determined transducer 22 is also measured.

[0090] Still referring to FIGS. 7 and 8, at least one transducer 22e is configured to emit, further to the incident wave O.sub.1 (FIG. 7), a complementary incident wave O.sub.3 (FIG. 8) into the envelope 11 of the tank 10 following the reception of a third electric signal U.sub.3 emitted by the control member 21. In this example, the curvature of the wall has been ignored for the sake of clarity. Of course the invention also applies to a curved wall. The trajectory of this complementary incident wave O.sub.3 is oriented at a measurement angle β with respect to that of the incident acoustic wave O.sub.1 (horizontal direction), so as to generate a complementary reflected acoustic wave O.sub.4 which is received by an adjacent transducer 22d (located directly below). As illustrated in FIG. 9, this transducer 22d is in turn configured to emit a fourth electric signal U.sub.4 to the calculation member 23, upon receiving the second reflected acoustic wave O.sub.4. Similarly to previously, the calculation member 23 is configured to measure the acoustic attenuation at each transducer 22 (between the third electric signal U.sub.3 and the fourth electric signal U.sub.4) to a database comprising the reference acoustic attenuations of the single-phase fluids F.sub.1, F.sub.2 for said tank 10 and for said measurement angle β. Similarly, said reference acoustic attenuations of the single-phase fluids F.sub.1, F.sub.2 are determined empirically or theoretically.

[0091] Preferably, the measurement angle β is between 1° and 15° so as to allow the interface I to be accurately detected between two transducers 22e-22d having different states ET.sub.1/ET.sub.2. This advantageously allows to determine whether the fill level N.sub.R is closer to the transducer 22d in the lower state ET.sub.1, or to the transducer 22e in the higher state ET.sub.2. In other words, the accuracy of measurement of the fill level N.sub.R is increased by this additional measurement.

[0092] Advantageously, the measurement system 20 thus allows a double measurement of the fill level N.sub.R of the tank 10, allowing a gain in both accuracy and reliability.

[0093] In other words, the reflected energy from an adjacent transducer is measured in order to more accurately determine the interface level I, in particular, when a transducer is located at the interface.

[0094] A horizontal tank with an acoustic energy attenuation measurement system has been set forth, but it is understood that such a measurement system 20 is adapted for a vertical tank 10 as illustrated in FIG. 10. In the case of a vertical tank 10, the transducers 22a-22g are distributed along the cylindrical median portion 13 along the length. Instead of being arranged along a curved line as in the case of a horizontal tank 10 described above, the transducers 22a-22g are positioned along a vertical rectilinear line, parallel to the axis X10 of the vertical tank 10.

[0095] As in the case of a horizontal tank 10, of course the transducers 22a-22g may be positioned along a plurality of vertical rectilinear lines, in particular two vertical rectilinear lines spaced apart in the order of one transducer 22a-22g and distributed in a staggered pattern. This has the advantage of being able to arrange a greater number of transducers 22a-22g closer together and thus of increasing accuracy of the measurement.