Method and system for determining the level of a liquid in a container and / or a pipe, in particular fuel in a tank of an aircraft

Abstract

One aspect of the invention relates to a method for determining the level of a liquid in a container and / or a pipe (1), for example a tank, comprising at least a first and a second sensor (11) positioned respectively above and below the liquid level (5), wherein the liquid level is determined from the propagation speeds of a first and a second propagation mode and a propagation time of a guided elastic wave that propagates through the wall of the container or of the pipe according to the first mode above the liquid level (5) and according to the second mode below the liquid level (5).

Description

[0001]DESCRIPTION TITLE: Method and system for determining a level of a liquid in a container and / or a conduit, in particular a fuel in a tank of an aircraft TECHNICAL FIELD OF THE INVENTION The technical field of the invention is that of measuring a level and / or a mass of a liquid in a container and / or a conduit, in particular a fuel on board a tank of an aircraft. The present invention relates to an ultrasonic gauging method and system for determining a level and / or a mass of liquid in a container and / or a conduit, in particular a fuel in a tank of an aircraft. TECHNOLOGICAL BACKGROUND OF THE INVENTION In aeronautics, the mass of fuel on board a tank is essential safety information for the pilot, in particular to know the flight autonomy of the aircraft,and for balancing the aircraft mass by compensating the center of gravity relative to the center of thrust. Existing fuel gauging systems rely on complex measurement technologies of sensors installed inside the tank. Such measurement systems, since they require an energy input into the tank, for example in electrical or optical form, are therefore subject to stringent requirements and standards, designated by the acronym ATEX, for ATmosphère Explosive, for installation and maintenance, in order to be compatible with use in such an atmosphere while being resistant to chemical attacks. Most fuel gauging systems include at least two separate measurement modules: - One to measure high and low fuel levels,indicating that the filling valves must be closed (high level) or warn the pilot that a fuel reserve has been reached (low level); and - The other to estimate an on-board fuel mass from two different devices, measuring respectively a density and a fuel level, the fuel mass being deduced from a height / volume law knowing the geometry of the tank and the attitude of the aircraft (yaw, pitch, roll, etc.). Regarding the measurement of point levels, the most common technologies are based on intrusive technologies such as a float, a self-heating thermistor or a capacitive sensor. Different devices for measuring the fuel density exist, such as a densimeter, an immersed float possibly associated with a counterweight,a mechanical oscillator or an immersed plate capacitor based on the Clausius-Mossotti law. The most common technology for measuring fuel level in the tank uses capacitive sensors immersed in the fuel. There are also approaches using waves guided along a waveguide immersed in the fuel, for example for electromagnetic or ultrasonic waves. Some approaches offer non-intrusive measurement of the fuel level, notably using bulk or guided ultrasonic waves. The disadvantage of the bulk ultrasonic wave approach is that it is necessary to be able to capture the wave after reflection at the fuel / gas interface. Since the liquid is rarely at rest in the tank, the wave is diffracted rather than being reflected specularly, which makes the backscattered wave difficult to detect by the sensor,since it is of low energy. The guided ultrasonic wave approach overcomes this drawback but does not allow for the influence of temperature on the estimation error of its level to be taken into account, unless the fuel properties are known. A guided ultrasonic wave approach exploits the conversion of an antisymmetric Lamb mode A0 into a Quasi-Scholte (QS) mode during its propagation in the tank wall (Lingyu Yu and Zhenhua Tian, ​​Case study of guided wave propagation in a one-side water-immersed steel plate, Case Studies in Nondestructive Testing and Evaluation, vol. 3, pp. 1-8, 2015). It is shown that the amplitude and flight time of the wave vary with the fuel level, due to the conversion, giving access to a much more accurate and robust measurement of the fuel level. However, no solution is proposed to improve the robustness of the measurement with respect to environmental conditions, in particular temperature. Therefore,at a given temperature, the level measurement as presented in this approach is relative. In other words, the fuel level is measured relatively between two level measurements, but not absolutely with respect to the dimensions of the tank. In practice, knowing the temperature, for example through a thermocouple, is not sufficient. This requires having a very precise measurement of the temperature at different locations in the tank, as thermal equilibrium is generally not reached during the use of an aircraft. This implies having a large number of temperature measurement points. At a minimum, a temperature measurement is required at the fuel level and another at the gas level, in the tank. Knowledge of the temperature therefore requires additional instrumentation, and a fusion of two technologies within the same system,which increases the installation and maintenance constraints as well as the probabilities of system failure. Consequently, there is a need for a simplified system for determining a level and / or a mass of liquid in a container and / or a conduit, robust to experimental conditions and compatible with the instrumentation constraints on an aircraft. SUMMARY OF THE INVENTION The invention offers a solution to the problems mentioned above, by proposing a method and a system allowing the measurement of the different quantities necessary for determining the level of a liquid in a container and / or a conduit from a single set of sensors. A first aspect of the invention relates to a method for determining a level of a liquid in a container and / or a conduit, the container and / or the conduit comprising: - a wall, in particular with an internal face and an external face,the wall delimiting an internal volume of the container and / or of the conduit comprising the liquid, and - a plurality of ultrasonic sensors comprising at least a first sensor positioned above the liquid level and at least a second sensor positioned below the liquid level,the liquid level in the container and / or the conduit being determined from: - The propagation speed of a first propagation mode determined by propagation of a guided wave in the wall above the liquid level and captured by the first sensor; - The propagation speed of a second propagation mode determined by propagation of a guided wave in the wall below the liquid level and captured by the second sensor; and - The propagation time of a guided wave generated by the first and / or the second sensor propagating in the wall according to the first propagation mode above the liquid level and according to the second propagation mode below the liquid level. Thanks to the invention, it is possible to determine a liquid height by taking into account the environmental conditions in which the container and / or the conduit containing it is located. Indeed,the method allows self-calibration of the sensors for measuring the liquid level, making it possible to overcome the influence of experimental conditions, including temperature. The level is therefore estimated robustly to these experimental conditions. The method is, moreover, simple to implement since it only requires the emission of ultrasonic waves, in particular in the wall. It is, moreover, advantageously possible to implement the method by generating a single ultrasonic guided wave, for example by the first sensor, to determine both the propagation speeds of the first and second modes and the liquid level in the container and / or conduit. Furthermore, the method does not require multiplying the measuring instruments to carry out the measurements. Indeed, a system comprising only ultrasonic sensors is sufficient to determine the different quantities involved. It is therefore, for example,no need to add temperature sensors on the tank to compensate for the effect of temperature on the propagation of guided waves. Above all, the method does not require intrusive measurement to be implemented. The method is therefore compatible with an application under ATEX regulations and without risk of chemical attack for the sensors due to the nature of the liquid, particularly in the case of fuel. The method therefore allows rapid instrumentation in terms of preparation and installation time, with low maintenance costs. Since any ultrasonic sensor technology can be used without losing the advantages of the method, it is possible to use sensor technologies that are compact and easy to instrument, facilitating access to acquisition systems and sensors for maintenance. Finally,the method is compatible for any type of container and / or conduit and any application subject to ATEX type instrumentation constraints and / or chemical attack. The method is therefore compatible for any aeronautical application, in particular during flight, including for semi-rigid or flexible tanks such as those on board a helicopter. Other applications of the method are possible, such as in the automotive, naval or storage fields, for example for drinking water, chemical products, agri-food products, etc. Although described here for liquids (of different physicochemical natures: viscosity, density, etc.), the proposed approach is also compatible for powdery or granular compounds. In addition to the characteristics which have just been mentioned in the previous paragraph, the method according to the first aspect of the invention may have one or more additional characteristics among the following,considered individually or in all technically possible combinations. In one embodiment, the liquid is a fuel. In one embodiment: - The first propagation mode is the antisymmetric mode A0 of a guided Lamb wave; - The second propagation mode is the Quasi-Scholte mode of a guided elastic wave. The determination of the liquid level is therefore based on the propagation of guided modes which are simple to generate and do not require specific instrumentation other than an ultrasonic sensor generating guided waves. In addition, the A0 mode, converting into the Quasi-Scholte mode, is sensitive to the presence of the liquid, thus allowing the liquid level to be calculated accurately. In one embodiment: - The propagation speed ^^, ^^0of the first propagation mode is estimated by generation then capture of a wave guided by the first sensor after reflection on a first reflector located on the wall, in particular the internal face, and above the liquid level, such that ^^ ^^0 = 2ℎ 1^ ^ 1, where ℎ1 is the distance between the first sensor and the first reflector and ^^1 is the round-trip propagation time of the first propagation mode between the first sensor and the first reflector; - The propagation speed ^^ ^^^^ of the second propagation mode is estimated by generation then capture of a wave guided by the second sensor after reflection on a second reflector located on the wall, in particular the internal face, and below the liquid level, such that where ℎ2 is the distance between the second sensor and the second reflector and ^^2 is the round-trip propagation time of the second propagation mode between the second sensor and the second reflector; and - The height ^^ liquid level relative to the second sensor is determined by ^^ 12 − ^^ 12 = ^^ ^ ^^ 12 − ^^ 12 ^^1 ^^ ^0 2ℎ 11 1 − = ^^ 2 ^^0 − ^^ 1, ^ ^ ^^^^ ^^ 2ℎ 2 2ℎ 1where ^^ 12 is the propagation time of the guided wave propagating between the first and second sensors according to the first propagation mode above the liquid level and according to the second propagation mode below the liquid level, and ^^ 12is the distance between the first sensor and the second sensor. In one embodiment, the first reflector is at least a portion of the upper surface of the container and / or the conduit and / or the second reflector is at least a portion of the lower surface of the container and / or the conduit. In one embodiment, the plurality of sensors further comprises a third sensor disposed below the liquid level and a fourth sensor disposed above the liquid level, the first and fourth sensors being positioned at substantially the same height, and the second and third sensors being positioned at substantially the same height, and wherein: - The propagation speed ^^ ^^0 of the first propagation mode is estimated by capturing a guided wave transmitted between the first and fourth sensors, such that where ^^ 14 is the distance between the first and fourth sensors, and ^^ 14is the propagation time of the first mode of the transmitted guided wave to travel the distance ^^ 14 ; - The speed of propagation ^^ ^^^^ of the second propagation mode is estimated by capturing a guided wave transmitted between the second and third sensors, such that where ^^ 23 is the distance between the second and third sensors, and ^^ 23 is the propagation time of the second mode of the transmitted guided wave to travel the distance ^^ 23 ; and - The height ^^ liquid level relative to the second sensor is determined by where ^^ 12 is the propagation time of the guided wave propagating between the first and second sensors according to the first propagation mode above the liquid level and according to the second propagation mode below the liquid level, and ^^ 12is the distance between the first sensor and the second sensor. Thanks to these embodiments, the determination of the liquid height to take into account the experimental conditions is simple and quick to implement. In one embodiment, the method further comprises: - Determining the density of the liquid from: o An estimate of the wall temperature; o A propagation speed of the second propagation mode; and o A dispersion curve of the second propagation mode; or o A measurement of the electrical impedance of one of the sensors of the plurality of sensors, said sensor being positioned below the liquid level; - Determining the mass of liquid from the density of the liquid and the level of the liquid. It is thus possible to determine the mass of fuel on board from the level measurement carried out,without adding or modifying the instrumentation of the sensors on the container and / or conduit. In one embodiment, the temperature of the wall is estimated by means of a guided wave generated by one of the first or second sensors and captured by the other of the first or second sensors and propagating in the wall according to a third propagation mode. In one embodiment, the temperature of the wall is estimated by inverting a dispersion curve of a third propagation mode as a function of the propagation speed of said third propagation mode in the wall. In one embodiment, the third propagation mode is the symmetric mode S0 of a guided Lamb wave. In one embodiment, the temperature of the wall is estimated by means of a measurement via a temperature sensor, in particular a thermocouple, which can be positioned outside the container and / or conduit, in particular on the external face. In one embodiment,the density of the liquid is determined by estimating the propagation speed of the second propagation mode in the wall and by inverting the dispersion curve of the second propagation mode for the estimated wall temperature. In one embodiment, the first propagation mode and / or the second propagation mode are generated using a broadband excitation signal. In one embodiment, the third propagation mode is generated using a broadband excitation signal. The advantage of generating the guided wave using the broadband excitation signal is to excite the first, second and third propagation modes over a controlled frequency range. This makes it possible to determine the propagation time of each mode by averaging the propagation times measured for each frequency of each mode, which improves the accuracy of the estimation of the propagation time of said mode. In one embodiment,the method comprises a preliminary step of determining the thickness of the wall by calibration with a known fluid or by direct measurement. In one embodiment, the plurality of sensors further comprises an intermediate sensor arranged at a distance from the wall or on the wall, in particular on the external face, and positioned between the first and second sensors, and the method comprises determining a gradient of the propagation speed of the first and / or second propagation modes from the propagation time of the first and / or second propagation modes, respectively, by means of the intermediate sensor. In one embodiment, the method comprises determining a gradient of the propagation speed of the third propagation mode from the propagation time of the third propagation mode by means of the intermediate sensor. It is thus possible to take into account the inhomogeneity of the properties of the liquid,in particular of its density which can vary with the depth in the container and / or conduit. A second aspect of the invention relates to a system for determining a liquid level in a container and / or a conduit, the container and / or the conduit comprising a wall with an internal face and an external face, the internal face of the wall delimiting an internal volume of the container and / or the conduit comprising the liquid, the system comprising: - A plurality of ultrasonic sensors arranged at a distance from the wall and / or on the wall, comprising: o A first sensor, positioned above the liquid level; and o A second sensor,positioned below the liquid level; - An acquisition module connected to the plurality of ultrasonic sensors and configured to control the plurality of ultrasonic sensors so as to generate a guided wave and / or to acquire samples of a guided wave captured by one of the sensors of the plurality of ultrasonic sensors; and - A processing module connected to the acquisition module and configured to implement the method according to the first aspect from the acquired samples. Thanks to this second aspect, the system for determining the liquid level is simple and quick to instrument on the container and / or conduit to implement the method according to the first aspect of the invention. A third aspect of the invention relates to a computer program product comprising instructions which, when the program is executed on a computer,lead the latter to implement the steps of the method according to the first aspect. A fourth aspect of the invention relates to a computer-readable recording medium comprising instructions which, when executed by a computer, lead the latter to implement the steps of the method according to the first aspect. A fifth aspect of the invention relates to a method for determining a mass of a liquid in a container and / or a conduit comprising: - a wall, in particular with an internal face and an external face, the wall delimiting an internal volume of the container and / or the conduit comprising the liquid, and - at least one ultrasonic sensor arranged at a distance from the wall and / or on the wall, the mass of the liquid in the container and / or the conduit being determined from the density of the liquid and the level of the liquid in the container and / or the conduit,the density being determined from: - An estimate of the wall temperature; - The propagation speed of a second propagation mode determined by propagation of a guided wave in the wall below the liquid level and captured by the sensor; and - A dispersion curve of the second propagation mode. In addition, by using sensors placed outside the container and / or the conduit, it may be possible to determine a mass of liquid contained in a container and / or the conduit in a non-intrusive manner because the method is then based on non-intrusive sensor technologies. The method is therefore compatible with an application under ATEX regulations and without risk of chemical attack for the sensors due to the nature of the liquid, particularly in the case of a fuel. The method therefore allows rapid instrumentation in terms of preparation and installation time, with a low maintenance cost. The method is, in addition,simple to implement since it only requires the emission of ultrasonic waves in the wall. It is, moreover, advantageously possible to implement the method by generating a single ultrasonic guided wave, for example via a first sensor, to determine both the temperature of the wall and the density of the liquid. Furthermore, the method does not require multiplying the measuring instruments to carry out the measurements. Indeed, a system comprising only ultrasonic sensors is sufficient to determine the different quantities involved. It is therefore, for example, not obligatory to add temperature sensors on the tank in order to compensate for the effect of temperature on the propagation of the guided waves. Since any ultrasonic sensor technology can be used without losing the advantages of the method, it is possible to use sensor technologies that are compact and easy to instrument,facilitating access to acquisition systems and sensors for maintenance. Finally, the method is compatible for any type of container and / or conduit and any application subject to instrumentation constraints such as ATEX and / or chemical attack. The method is therefore compatible for any aeronautical application, particularly during flight, including for semi-rigid or flexible tanks such as those on board a helicopter. Other applications of the method are possible, such as in the automotive, naval or storage fields, for example for drinking water, chemicals, agri-food products, etc. Although described here for liquids (of different physicochemical natures: viscosity, density, etc.), the proposed approach is also compatible for powdery or granular compounds. In addition to the characteristics just mentioned in the previous paragraph,the method according to the fifth aspect of the invention may have one or more complementary characteristics among the following, considered individually or according to all technically possible combinations. In one embodiment, the liquid is a fuel. In one embodiment, the second propagation mode is the Quasi-Scholte mode of a guided elastic wave. The determination of the mass of liquid is based on the propagation of a guided mode which is simple to generate and does not require specific instrumentation other than an ultrasonic sensor generating guided waves. In addition, the Quasi-Scholte mode is sensitive to the presence of the liquid, thus improving the accuracy of the calculation of the density. In one embodiment, the method according to the fifth aspect comprises at least one first sensor and at least one second sensor, and in which the temperature of the wall is estimated by means of a guided wave generated by the first,respectively the second, sensor and captured by the second, respectively the first, sensor and propagating in the wall according to a third propagation mode. In one embodiment, the third propagation mode is the symmetrical mode S, 0of a guided Lamb wave. In one embodiment, the wall temperature is estimated by inverting a dispersion curve of the third propagation mode as a function of the propagation speed of said third propagation mode in the wall. It is thus possible to quickly and easily determine the wall temperature without using any equipment other than the two sensors mentioned. It is therefore not useful to add additional equipment to evaluate this temperature. Furthermore, the determination of the wall temperature is based on the propagation of a guided mode which is simple to generate and does not require specific instrumentation other than an ultrasonic sensor generating guided waves. In addition, the S0 mode is little or not sensitive to the presence of liquid in the container and / or conduit, allowing for accurate temperature measurement.In one embodiment, the wall temperature is estimated by means of a measurement via a temperature sensor, for example a thermocouple, in particular positioned outside the container and / or conduit, on the wall. It is therefore alternatively possible to use another measuring instrument to evaluate the wall temperature. In one embodiment, the density of the liquid is determined by estimating the propagation speed of the second propagation mode in the wall and by inverting the dispersion curve of the second propagation mode for the estimated wall temperature. In one embodiment, the method comprises at least one first sensor positioned above the liquid level and at least one second sensor positioned below the liquid level, and wherein the liquid level is determined from: - The propagation speed ^^. ^^0of a first propagation mode is estimated by generation then capture of a wave guided by the first sensor after reflection on a first reflector located on the internal face and above the liquid level, such that 2ℎ ^^1 ^^0 = ^^1, where ℎ1 is the distance between the first sensor and the first reflector and ^^1 is the round-trip propagation time of the first propagation mode between the first sensor and the first reflector; - The propagation speed ^^ ^^^^ of the second propagation mode is estimated by generation then capture of a wave guided by the second sensor after reflection on a second reflector located on the internal face and below the liquid level, such that where ℎ2 is the distance between the second sensor and the second reflector and ^^2 is the round-trip propagation time of the second propagation mode between the second sensor and the second reflector; the height ^^ of the liquid level relative to the second sensor being determined by where ^^ 12 is the propagation time of the guided wave propagating between the first and second sensors according to the first propagation mode above the liquid level and according to the second propagation mode below the liquid level, and ^^ 12is the distance between the first sensor and the second sensor. In one embodiment, the first reflector is at least a portion of the upper surface of the container and / or the conduit and / or the second reflector is at least a portion of the lower surface of the container and / or the conduit. In one embodiment, the method according to the fifth aspect comprises at least a first sensor, a second sensor, a third sensor and a fourth sensor, - the first sensor and the fourth sensor being positioned above the liquid level, in particular at substantially the same height, and - the second sensor and the third sensor being positioned below the liquid level, in particular at substantially the same height, and wherein the liquid level is determined from: - The propagation speed ^^ ^^0of the first propagation mode is estimated by capturing a guided wave transmitted between the first and fourth sensors, such that where ^^ 14 is the distance between the first and fourth sensors, and ^^ 14 is the propagation time of the first mode of the transmitted guided wave to travel the distance ^^ 14 ; - The speed of propagation ^^ ^^^^ of the second propagation mode is estimated by capturing a guided wave transmitted between the second and third sensors, such that where ^^ 23 is the distance between the second and third sensors, and ^^ 23 is the propagation time of the second mode of the transmitted guided wave to travel the distance ^^ 23 ; the height ^^ of the liquid level relative to the second sensor being determined by where ^^ 12is the propagation time of the guided wave propagating between the first and second sensors according to the first propagation mode above the liquid level and according to the second propagation mode below the liquid level, and ^^ 12is the distance between the first sensor and the second sensor. In one embodiment, the method according to the fifth aspect comprises at least one first sensor and at least one second sensor, the first sensor and the second sensors being in particular positioned substantially at the same height, and in which the liquid level is a determined point level such that: - The liquid level is above or at the same height as the first and second sensors when a wave emitted by the first and / or second sensors and picked up by the second and / or first sensors, respectively, propagates according to the second propagation mode; - The liquid level is below the height of the first and second sensors when a wave emitted by the first and / or second sensors and picked up by the second and / or first sensors, respectively, propagates according to the first propagation mode.With these embodiments, determining the liquid height to take into account the experimental conditions is simple and quick to implement. It is not necessary to add additional measuring instruments since the ultrasonic sensors are sufficient in themselves to measure and determine all the data necessary for calculating the liquid level in the container and / or conduit. In one embodiment, the first propagation mode, the second propagation mode and / or the third propagation mode are generated using a broadband excitation signal. The advantage of generating the guided wave using the broadband excitation signal is to excite the first, second and third propagation modes over a controlled frequency range.This makes it possible to determine the propagation time of each mode by averaging the propagation times measured for each frequency of each mode, which improves the accuracy of the estimation of the propagation time of said mode. In one embodiment, the method further comprises a preliminary step of determining the thickness of the wall by calibration with a known fluid or by direct measurement. In one embodiment, the method according to the fifth aspect further comprises an intermediate sensor arranged at a distance from the wall or on the wall, and positioned between the first and second sensors, and the method comprising determining a gradient of the propagation speed of the first, second and / or third propagation modes from the propagation time of said first, second and / or third propagation modes by means of the intermediate sensor.It is thus possible to take into account the inhomogeneity of the properties of the liquid, in particular its density which can vary with the depth in the container and / or conduit.A sixth aspect of the invention relates to a system for determining a mass of liquid in a container and / or a conduit, the container and / or the conduit comprising a wall with an internal face and an external face, the wall delimiting an internal volume of the container and / or the conduit comprising the liquid, the system comprising: - At least one ultrasonic sensor arranged at a distance from the wall and / or on the wall; - An acquisition module connected to the ultrasonic sensor and configured to control the ultrasonic sensor so as to generate a guided wave and / or to acquire samples of a guided wave captured by the ultrasonic sensor; and - A processing module connected to the acquisition module and configured to implement the method according to the fifth aspect from the acquired samples.Thanks to this sixth aspect, the gauging system is simple and quick to instrument on the container and / or conduit to implement the method according to the fifth aspect of the invention. A seventh aspect of the invention relates to a computer program product comprising instructions which, when the program is executed on a computer, cause the latter to implement the steps of the method according to the fifth aspect. An eighth aspect of the invention relates to a computer-readable recording medium comprising instructions which, when executed by a computer, cause the latter to implement the steps of the method according to the fifth aspect. The invention and its various applications will be better understood upon reading the following description and examining the accompanying figures. BRIEF DESCRIPTION OF THE FIGURES The figures and examples described are presented for information purposes only and in no way limit the invention.- Figure 1 shows a schematic representation of the gauging system according to the invention implemented on a tank containing a fuel, according to one embodiment. - Figure 2 is a block diagram illustrating the sequence of steps of a gauging method according to the invention, according to one embodiment. - Figure 3 is a schematic representation of the container and / or conduit according to Figure 1 detailing the physical quantities used by the method according to Figure 2 according to a first embodiment. - Figure 4 is a schematic representation of the container and / or conduit according to Figure 1 detailing the physical quantities used by the method according to Figure 2 according to a second embodiment. - Figure 5 is a schematic representation of an embodiment of the instrumentation of the container and / or conduit.- Figure 6 is a set of experimental signals acquired using the system according to Figure 1, according to one embodiment. - Figure 7 illustrates the variations in the propagation speed of a guided wave following the S0 mode in an aluminum plate, as a function of temperature variations. - Figure 8 is a schematic representation of the positioning of sensors for determining a point level in a container and / or a conduit, according to one embodiment. - Figure 9 is a schematic representation of the propagation of a guided wave in the wall of a container and / or a conduit, according to one embodiment. DETAILED DESCRIPTION Unless otherwise specified, the same element appearing in different figures has a single reference. The invention relates to an ultrasonic gauging method and system for determining a level and / or a mass of liquid in a container and / or a conduit.The application proposed below relates to an aircraft tank. The invention can however be used for any other container and / or conduit containing a liquid in an environment subject to instrumentation constraints and safety requirements, in particular linked to an environment under explosive atmosphere (ATEX) and to risks of chemical attack. The invention is based on the generation of guided ultrasonic waves in the wall of the container and / or conduit, here the tank, to determine the level and the density of the liquid, here the fuel, in the tank, in order to deduce the on-board mass of liquid. In particular, three guided wave modes are excited: the antisymmetric mode A0 of a Lamb wave, the symmetric mode S0 of a Lamb wave and the Quasi-Scholte mode. In Figure 1 is illustrated the gauging system 10 implemented on a container and / or conduit 1, the container and / or conduit is here a tank 1.The tank 1 comprises a wall 2, which comprises an inner face 2i and an outer face 2e. The wall, in particular its inner face 2i, delimits an internal volume of the container and / or conduit 1. The internal volume contains the liquid. The inner face 2i is therefore in contact with the liquid. The outer face 2e is exposed to the surrounding environment in which the container and / or conduit 1 is located. In the first embodiment presented, the tank 1 is on board an aircraft, such as an airplane or a helicopter, and the tank 1 is located in a storage space dedicated to said tank 1. The tank 1 also comprises an upper surface 3 and a lower surface 4. The internal volume is therefore also delimited by the upper 3 and lower 4 surfaces.The tank 1 may also comprise other elements, such as equipment used for circulating fuel to the engine to supply it, such as a pump device and a connection circuit, and equipment used to convey fuel into the tank for storage therein, such as a supply circuit. The term "liquid level" means the height of the interface formed by the upper surface of the liquid with a gas in the container and / or conduit 1 relative to a fixed reference such as the lower surface 4. The dimension of the liquid level 5 is preferably a distance, for example in millimeters, centimeters, meters, etc. The liquid level 5 thus delimits an upper part 1a and a lower part 1b in said container and / or conduit 1 such that the lower part 1b corresponds to the part of the container and / or conduit where the liquid is located and the upper part 1a corresponds to the part of the container and / or conduit where the gas is located.The gas is a chemical compound mainly composed, for example, of fuel vapors and air or nitrogen-enriched air. The wall then comprises an upper section 2a and a lower section 2b, corresponding to the upper part 1a and the lower part 1b, respectively, of the container and / or conduit 1. The geometry and dimensions of the container and / or conduit 1 are predefined. That is to say, the geometry and dimension of the internal volume, the upper surface 3, the lower surface 4 and the wall 2 are predefined. In certain embodiments, the thickness of the wall is not predefined and will have to be determined. “Predefined” is what is perfectly known at the time of implementation of the method and is not a characteristic for which a determination must be made. In particular, the properties of the container and / or conduit and its elements are predefined since they are the subject of a controlled design and manufacturing process.The characteristics and positions of the elements of the gauging system are also predefined because they are controlled and known to the operator. The gauging system 10 comprises a plurality of ultrasonic sensors 11, an acquisition module 12 and a processing module 13. The plurality of ultrasonic sensors 11 can be arranged on the wall 2, in particular on the external face 2e of the wall 2. In other words, each sensor 11 of the plurality of sensors 11 is positioned on the container and / or conduit 1, in contact with the wall 2. The gauging of the liquid is therefore implemented in a non-intrusive manner. In certain alternatives, depending on the sensor technology used and the intended application, the sensors 11 can be non-contact sensors. By "non-contact" is meant sensor technologies making it possible to generate an ultrasonic wave in a part without the generation of the wave requiring contact of the source with the part.For example, it may be a sensor technology very close to the wall (such as electromagnetic acoustic sensors, or "EMAT" for "Electromagnetic Acoustic Transducer" in English) or even at a distance (such as one or more laser sources). In such a case, the sensors 11 may be positioned outside the container and / or conduit, close to and / or at a distance from the wall 2. In this context, "close" means that the distance between the wall and the sensor is between 1 µm and 50 mm, or even between 100 µm and 10 mm, or even between 1 mm and 5 mm. Also in this context, "at a distance" means that the distance between the wall and the sensor is greater than 50 mm, or even greater than 100 mm, or even greater than 500 mm. In still other alternatives, the sensors may be placed inside the container and / or conduit 1, for example on the wall 2 or at a distance therefrom.The ultrasonic sensors 11 are sensors allowing the generation and capture of ultrasonic waves propagating in the wall 2 of the container and / or conduit 1. In particular, these sensors 11 serve to generate and capture one or more guided waves in the wall 2, that is to say in the thickness of said wall 2. For example, an illustration of such propagation is provided in Figure 9, where, in particular in Figure 9(a), a guided wave ^^ propagates in the wall 2 of the container and / or conduit 1, seen from the front, between two sensors 11 positioned one above the other, and where, in particular in Figure 9(b), a guided wave ^^ propagates in the wall 2 of the container and / or conduit 1, seen from above, between two sensors 11 positioned at a substantially similar height on said wall 2.Similarly, typically when the container and / or the conduit 1 is of a substantially cylindrical shape in a horizontal direction, in particular in Figure 9(c), a guided wave ^^ propagates in the wall 2 of the container and / or conduit 1 between two sensors 11 positioned one above the other, and where, in particular in Figure 9(d), a guided wave ^^ propagates in the wall 2 of the container and / or conduit 1 between two sensors 11 positioned at a substantially similar height on said wall 2. For the sake of simplification, the thickness of the wall 2 is not shown in the figures other than Figure 9. Preferably, these sensors 11 are configured to excite the anti-symmetric A0 and symmetric S0 propagation modes of a Lamb and / or Quasi-Scholte QS wave.Any technology can be used for these sensors, such as ultrasonic transducers exploiting the piezoelectric effect which converts a mechanical displacement into a voltage, and vice versa. The sensors can then be single-element transducers, otherwise called pellets, or linear or matrix multi-element transducers. These sensors are, in the first embodiment, placed in contact with the wall 2, but they are also compatible for non-contact applications. In the latter case, a couplant can be used to minimize the impedance break between the wall and the medium surrounding the container and / or conduit. Alternatively, certain applications can use measurement technologies not based on the piezoelectric effect, such as optoacoustic technologies, for example using one or more laser vibrometers or magnetostrictive transducers.The sensors 11 of the plurality of sensors 11 may be based, in whole or in part, on different technologies and / or combine different sensor technologies. The sensors are therefore configured to suit the application (geometry, dimensions, emission frequency, emitted wavelength, emission energy, etc.). The position of each sensor 11 of the plurality of sensors 11 remotely or on the container and / or conduit 2 is predefined. The plurality of sensors 11 comprises a first sensor 11 and a second sensor 11. The first sensor 11 is arranged to be positioned on the wall 2 above the liquid level 5. The second sensor 11 is, for its part, arranged to be positioned on the wall 2 below the liquid level 5. These first and second sensors 11 may alternatively be based, for one or the other, on a contactless sensor technology.The first and / or second sensors are then positioned so as to generate a guided wave in the wall 2 respectively from a first and / or second source point, these source points being located above and / or below the liquid level 5, respectively. These source points correspond, for example, to the respective positions of the first and / or second sensors, respectively, if these are in contact with the wall 2. The position of the first sensor 11 and the position of the second sensor 11 on the wall are dependent on the instrumentation conditions and constraints linked to the environment as well as to the geometry, accessibility and size of the container and / or conduit 1.In particular, other elements of the container and / or conduit 1, such as the supply and feed circuits, as well as the chassis of the aircraft and the presence of other systems around the container and / or conduit 1 such as other containers and / or conduits, may constrain the positioning of the sensors 11. Furthermore, the internal volume of the container and / or conduit, in particular for a fuel tank, may comprise an upper zone, called bubble B, and a lower zone, called pool F. Bubble B serves as a buffer to allow sufficient time for the operator filling the container and / or conduit to stop filling so as not to exceed the capacity of the internal volume of the container and / or conduit. Pool F serves as a fuel reserve guaranteeing the pilot a minimum flight autonomy to complete his flight and land the aircraft.According to an exemplary embodiment, the first sensor and the second sensor may be arranged on the external face 2e so as to be above the low interface Bb of the bubble B and below the high interface Fh of the puddle F, respectively. The interest is to ensure that the first and second sensors are always above and below, respectively, the fuel level that it is desired to measure outside the inert volume. The relative positioning of the first sensor with respect to the second sensor does not depend on specific conditions other than those previously mentioned. It is however preferable that the first and second sensors are arranged on the same side of the container and / or conduit in order to promote the transmission of guided waves from one to the other.Similarly, it is preferable that the propagation path between the first and second sensors does not include a mechanical interruption, such as an impedance break, for example in the form of an attenuating polymer seal or a geometric discontinuity. When the geometry of the container and / or conduit 1 allows it, it is preferable that the first sensor and the second sensor are substantially vertically aligned with each other, along a vertical axis of the container and / or conduit, so that the transmission of a wave is made along a direct propagation path between these sensors. The vertical axis of the container and / or conduit is orthogonal to the plane defined by the surface of the liquid at its interface with the gas when the apparatus is on the ground and is not in motion. In other words, when the container and / or conduit is a cylinder, the first and second sensors are positioned so as to be substantially in a plane tangent to the wall 2.This alignment makes it possible to simplify the calculations of the gauging method described later. The acquisition module 12 is connected to the plurality of sensors 11. The acquisition module 12 is configured to control the plurality of ultrasonic sensors so as to generate a guided wave in the wall 2. Preferably, the acquisition module is connected to each sensor of the plurality of sensors. By "connected" is meant that a connection intermediary, for example a wired or wireless connection, connects the elements in question and allows the transfer of data from the elements between them. The generation of the guided wave is implemented by the excitation of the sensor by means of an excitation signal. The excitation signal is, optionally, digitized by the acquisition module 12. In this case, the excitation signal comprises a plurality of successive samples digitized at a predetermined sampling frequency.The sampling frequency is generally greater than or equal to twice the highest frequency of the excitation signal, so as to comply with the Shannon-Nyquist criterion. The excitation signal can be any type of signal allowing the generation of the guided wave. The excitation signal therefore depends on the application in question. For example, the excitation signal is a pulse, a single-frequency sine wave train, a multi-frequency sine wave train, a signal modulated around a central frequency (in English, "chirp"), a broadband signal, etc. The excitation signal is preferably of finite time support. In the first embodiment, the excitation signal is a single-frequency sine wave train of given excitation frequency and digitized at the given sampling frequency, and preferably greater than 1 MHz.The excitation frequency is the frequency of the sinusoid of the sinusoidal wave train. The values ​​of these different frequencies are dependent on the material of the wall of the container and / or the conduit. For example, for a 3 mm thick aluminum wall, the excitation frequency is 50 kHz for the first propagation mode and 100 kHz for the third propagation mode. The sampling frequency is then at least equal to 500 kHz, or even greater than 1 MHz for these two propagation modes. The processing module is also configured to acquire a signal of a guided wave, in particular propagating in the wall, and captured by one of the sensors of the plurality of sensors. The acquired signal comprises a plurality of samples acquired at an acquisition frequency.The acquisition frequency is generally greater than or equal to the maximum frequency of the propagation mode of the wave captured by the sensor that is to be acquired, so as to comply with the Shannon-Nyquist criterion. The acquisition frequency may be equal to the sampling frequency. The acquisition module 12 therefore comprises a processor and a memory. The memory comprises instructions which, when executed by the processor, allow the digitization of the excitation signal, the generation of the guided wave, in particular in the wall, by means of a sensor of the plurality of ultrasonic sensors and / or the acquisition of the signal of a guided wave, in particular propagating in the wall. The processing module 13 is connected to the acquisition module 12. The processing module comprises a memory and a processor and is configured to implement the gauging method.The processing module 13 therefore includes, in its memory, instructions which, when executed by the processor of the processing module 13, allow the implementation of the gauging method. The processing module is, for example, a computer. The gauging method 100, as illustrated in FIG. 2, includes three main steps numbered from 110 to 130. The method 100 may also include a preliminary step 105 for determining a thickness of the wall 2, if it is not predefined. The physical quantities used by the method 100 are shown in FIG. 3. It will appear that the first step 110 and the second 120 are independent and can be implemented in any order or concurrently with one another. More details are given below. In the first embodiment, the assumption is made that the properties of the liquid are homogeneous in the internal volume of the container and / or conduit 1.The first step 110 is a step of determining the liquid level 5 in the container and / or conduit 1. This step therefore aims to determine the liquid height in the container and / or conduit. The first step 110 comprises three sub-steps, numbered 111 to 113. The proposed approach is based on measuring the propagation time of a guided wave propagating between the sensors 11 in order to deduce the liquid height therefrom. The liquid height is here determined relative to the position of the second sensor. It is necessary to use a propagation mode sensitive to the presence of the liquid. In particular, the guided wave propagates between the sensors according to a first mode in the upper section 2a of the wall 2, above the liquid level 5 and propagates according to a second propagation mode, sensitive to the presence of the liquid, in the lower section 2b of the wall 2, below the liquid level 5.The second propagation mode may require the presence of a liquid to propagate and therefore may not exist in the upper section 2a. Preferably, the first propagation mode is the anti-symmetric mode A0 of a Lamb guided wave and the second propagation mode is the Quasi-Scholte mode QS. The QS mode can be generated in the lower part 2b directly by the second sensor 11 or by conversion of the mode A. 0at the passage of the liquid / gas interface after emission by the first sensor 11 (FB Cegla, P. Cawley, and MJS Lowe, Material property measurement using the quasi-Scholte mode - A waveguide sensor, JASA, vol.117, 1098, 2005; FB Cegla, P. Cawley, and MJS Lowe, Fluid bulk velocity and attenuation measurements in non-Newtonian liquids using a dipstick sensor, Meas. Sci. Technol., vol.17(2), 264, 2006). The advantage of the QS mode is that it allows better measurement accuracy to be achieved than the other modes conventionally used. The excitation signal is preferably adapted to solicit a first mode, a second mode and / or a third propagation mode. It is possible to use different excitation signals to solicit each of the first, second and / or third propagation modes independently.It is noted that the excitation signal adapted to solicit the first mode, in the case of the A0 mode, is also adapted to solicit the second mode, in the case of the QS mode, and vice versa, due to the conversion of the A0 mode into the QS mode. The sensors are therefore dimensioned (diameter, emission wavelength, energy, etc.) to solicit the first, second and / or third mode. The implementation of sub-steps 111 and 112 makes it possible to determine the propagation speed of the first and second propagation modes. Advantageously, the approach proposed for determining these propagation speeds is self-calibrated, that is to say it makes it possible to take into account the experimental conditions at the time of measurement, in particular the temperature in the internal volume, and more particularly the temperature of the liquid. Sub-steps 111 and 112 are independent of each other and can be implemented in any order or concurrently.The first sub-step 111 is a step of determining the propagation speed of the first propagation mode, in this case the A0 mode, propagating in the upper section 2a of the wall 2. The excitation signal is therefore chosen to promote the propagation of the A0 mode. The determination is based on the “pulse-echo” technique, i.e. the generation of a wave and its capture by the same sensor after reflection on a first reflector. The wave therefore makes a round trip between the sensor and the first reflector. Preferably, the first reflector is an element of the container and / or conduit in contact with the upper section 2a and not requiring the addition of an additional element on the internal face 2i. Alternatively, the first reflector is an additional element calibrated and positioned at any position on the internal face 2i so as to be above the liquid level 5. The position of the first reflector is predefined.Preferably, the position of the first reflector is such that the transmission path between the first reflector and the first sensor is a direct path. The wave is therefore generated by the first sensor so that it propagates according to the first propagation mode, i.e. mode A0. The first reflector is here the upper surface 3 or a part thereof. Preferably, the first sensor is positioned so as to be sufficiently far from the upper surface 3 to guarantee an accurate measurement of the propagation time of the first propagation mode, for example at a distance ℎ1 equal to at least two wavelengths, or even at least equal to three wavelengths, or even at least five wavelengths or even ten wavelengths of the first propagation mode at a given frequency, for example the excitation frequency.The distance ℎ1 between the first sensor and the first reflector, here the upper surface 3, being predefined, the distance traveled by the wave propagating according to the mode A0 and reflected by the first reflector is also predefined. The propagation time ^^1 of the first mode along the path ℎ1 is determined by calculating a difference between the instant when the echo of the guided wave is captured, after reflection on the first reflector, and the instant of emission of this wave. In this case, the propagation speed ^^. ^^0 of mode A0 is then The second sub-step 112 is a step of determining the propagation speed of the second propagation mode, in this case the QS mode, propagating in the lower section 2b of the wall 2. The excitation signal is therefore chosen to promote the propagation of the QS mode. The “pulse-echo” technique is also used. A second reflector, different from the first reflector, is used. Preferably, the second reflector is an element of the container and / or conduit arranged on the lower section 2b and not requiring the addition of an additional element on the internal face 2i. Alternatively, the second reflector is an additional element calibrated and positioned at any position on the internal face 2i so as to be below the liquid level 5. The position of the second reflector is predefined.Preferably, the position of the second reflector is such that the transmission path between the second reflector and the second sensor is a direct path. The wave is therefore generated by the second sensor so that it propagates according to the second propagation mode, i.e. the QS mode. The second reflector is here the lower surface 4 or a part thereof. Preferably, the second sensor is positioned so as to be sufficiently far from the lower surface 4 to guarantee an accurate measurement of the propagation time of the first propagation mode, for example at a distance ℎ2 equal to at least two wavelengths, or even at least three wavelengths, or even at least five wavelengths or even ten wavelengths of the second propagation mode at a given frequency, for example the excitation frequency.The distance ℎ2 between the second sensor and the second reflector, here the lower surface 4, being predefined, the distance traveled by the wave propagating according to the QS mode and reflected by the second reflector is also predefined. The propagation time ^^2 of the second propagation mode along the path ℎ2 is determined by calculating a difference between the instant when the echo of the guided wave is captured, after reflection on the second reflector, and the instant of emission of this wave. In this case, the propagation speed ^^. ^^^^ of the QS mode is then 2ℎ ^^2 ^^^^= ^^2. The third sub-step 113 is then a step of determining the liquid level 5. In particular, the height ^^ of liquid relative to the position of the second sensor is determined. To do this, a guided wave is transmitted between the sensors 11 of the plurality of sensors so that the path of the wave crosses the liquid / gas interface. The wave therefore propagates according to the first propagation mode above the liquid level 5 and according to the second propagation mode below the liquid level 5, after conversion of the first mode into the second propagation mode upon passing the liquid / gas interface. In this first embodiment, this guided wave is generated by the first sensor and is picked up by the second sensor. The positions of the sensors being predefined, the distance ^^ 12 between the first sensor and the second sensor is predefined. The height ^^ of liquid is determined by where ^^ 12is the propagation time of the guided wave propagating between the first and second sensors 11 according to the first propagation mode above the liquid level 5 and according to the second propagation mode below the liquid level 5. The propagation time ^^ 12 is determined by calculating a difference between the instant of capture of the second propagation mode of the guided wave by the second sensor and the instant of generation of the guided wave by the first sensor. Alternatively, the wave can be generated by the second sensor and be captured by the first sensor. In this case, the first propagation mode is generated by conversion of the second propagation mode upon passing the liquid / gas interface. The height ^^ is determined in the same way as before and the propagation time ^^ 12is determined by calculating a difference between the instant of capture of the first propagation mode of the guided wave by the first sensor and the instant of generation of the second propagation mode of the guided wave by the second sensor. The liquid level 5 relative to the lower surface 4 of the container and / or conduit 1 is therefore given by the summation of the height ^^ of liquid measured relative to the second sensor with the distance ℎ2 from the second sensor relative to the lower surface 4. For illustration purposes, Figure 6 shows a set of signals acquired by capture of the ultrasonic guided wave by the second sensor after generation by the first sensor. The guided wave therefore crosses the interface of the liquid level during its propagation. Each line corresponds to one of these signals, each acquired for a different height ^^ of liquid between a maximum level ^^ ^^^^^^ and a minimal level ^^ ^^^^^^. It is observed that the propagation time of the S0 mode is invariant with the variation of the liquid height, while that of the A0 mode, converted into QS mode at the interface, decreases as the liquid height decreases. The A0 mode propagates here at 125 kHz. The second step 120 is a step of determining the density of the liquid. The density can be determined from the knowledge of the propagation speed of the second propagation mode, in this case the QS mode. To do this, it is necessary to know the temperature of the wall and to have a dispersion curve of the second propagation mode as a function of this temperature. For this purpose, the second step 120 comprises at least a first and a third sub-step, numbered 121 and 123 and, possibly, a second sub-step 122.As is known, the calculation of a theoretical dispersion curve requires knowing a certain number of parameters of the propagation medium such as, for example, the wall thickness, the Young and Poisson moduli and its density. It is also required to know the temperature of the wall and the material properties of the liquid such as its isostatic modulus of elasticity and its density. With the exception of the temperature of the wall and the density of the liquid, all these parameters are predefined. The calculation of the theoretical dispersion curves is used to obtain the velocity curve of the second mode, here the QS mode, at a given frequency or over a frequency range around a central emission frequency. The first sub-step 121 is a step of determining the temperature of the wall.The proposed approach is based on measuring the propagation time of a guided wave propagating between the sensors 11 according to the third propagation mode which is insensitive to the presence of the liquid but sensitive to the temperature. The guided wave propagates according to the third mode throughout the wall 2. Preferably, the third propagation mode is the symmetrical mode S0 of a Lamb guided wave. The guided wave propagating according to the mode S0 is generated by the first sensor and captured by the second sensor or is generated by the second sensor and is captured by the first sensor. For illustration purposes, variations in the propagation speed of the mode S0, at different temperatures and in a 3 mm thick aluminum plate are shown in Figure 7. One of the faces of said plate is in contact with the air, the other is in contact with air or water.It can be appreciated that the presence of water has a second-order influence on the velocity of the S0 mode, unlike its influence on the temperature. Particularly for this reason, the S0 propagation mode is favored here to estimate the plate temperature, regardless of the water level. The wall temperature is determined by inverting a dispersion curve of the third propagation mode as a function of the propagation velocity of the third propagation mode of the guided wave. The propagation velocity ^^. ^^0 of the third mode of propagation is given by The propagation time ^^ ^^0of the third propagation mode is determined by calculating a difference between the instant of capture of the third propagation mode of the guided wave by the sensor which captures the wave and the instant of generation of the third propagation mode of the guided wave by the sensor which emits the wave. The dispersion curve is a theoretical curve, for example calculated by a numerical model or obtained by an analytical function, for example following a low frequency approximation, or an experimental curve obtained via an acquisition system, and makes it possible to determine the propagation speed of the third propagation mode at different frequencies at a given wall temperature. The inversion consists, knowing the excitation frequency and the propagation speed of the third propagation mode of the guided wave, in finding the corresponding dispersion curve to deduce the temperature.The inversion can be implemented using known inversion techniques, such as the inversion of the analytical equation of the dispersion curve, the inversion of the numerical model used to calculate the dispersion curve, or the search among a pre-constructed database of dispersion curves obtained analytically, experimentally and / or by the numerical model. The advantage of using an inversion based on searching a database is that its implementation, once the database has been constructed, is much faster than other approaches, and is therefore more suitable for an in situ implementation of the method 100.At a given frequency, the database can be constructed by defining variation ranges for the wall temperature and for the excitation frequency, and by determining the associated propagation speed by reading a calculated dispersion curve, taking into account the temperature / excitation frequency pair in question. This database can be stored in the memory of the processing module 13. It is then sufficient, during sub-step 121, to search in the database for the pair corresponding to the propagation speed of the third propagation mode and to the related excitation frequency, to determine the wall temperature. The inversion therefore consists of finding the dispersion curve whose wall temperature from which it is calculated generates the propagation of the third propagation mode at the generation frequency with the propagation speed ^^. ^^0determined. The mechanical properties of the wall necessary for calculating the dispersion curve are predefined. The second sub-step 122 is a step of determining the propagation speed of the second propagation mode, here the QS mode. This second sub-step 122 is implemented only if this propagation speed has not been determined in sub-step 112, in particular if the second step 120 is implemented before the first step 110. In this case, it is not necessary to implement the second sub-step 112. The propagation speed of the second mode is then determined, during the second sub-step 122, in the same way as in the second sub-step 112. The second sub-step 122 is therefore independent of the first sub-step 121 and can be implemented before, after or concurrently with it.The third sub-step 123 is then a step of determining the density of the liquid, knowing the temperature of the wall and the propagation speed of the second propagation mode, here QS. The density of the liquid is determined by knowing (through modeling or calibration measurements on known liquids) the dispersion curve of the second guided wave propagation mode and the temperature of the wall. The density is, for example, obtained by inverting the equation of a dispersion curve of the propagation speed of the second propagation mode taking into account the determined temperature of the wall. The inversion consists, knowing the excitation frequency, the propagation speed of the second propagation mode and the temperature of the wall, in finding the corresponding dispersion curve to deduce the density of the liquid.The inversion can be implemented using known inversion techniques, such as the inversion of an analytical equation of the dispersion curve, the inversion of the numerical model used to determine the dispersion curve, or the search among a pre-constructed database of dispersion curves obtained analytically, experimentally and / or by the numerical model. The advantage of using an inversion based on searching a database is that its implementation, once the database has been constructed, is much faster than other approaches, and is therefore more suitable for an in situ implementation of the method 100.At a given frequency, the database can be constructed by defining variation ranges for the wall temperature and for the density of the liquid, and by determining the associated propagation speed by reading a calculated dispersion curve, taking into account the temperature / density pair in question. This database can be stored in the memory of the processing module 13. It is then sufficient, during sub-step 123, to search in the database for the pair corresponding to the propagation speed of the second propagation mode, determined in sub-step 112 or in sub-step 122, and to the wall temperature determined in sub-step 121 to determine the density of the liquid.The inversion therefore consists of finding the dispersion curve whose density of the liquid, from which it is calculated, generates the propagation of the third propagation mode at the generation frequency with the propagation speed ^^. ^^^^ determined. The mechanical properties of the wall required to calculate the dispersion curve are predefined. For example, at 20°C for a 3 mm thick aluminum tank filled with a liquid, reading the dispersion curves for an excitation frequency of 50 kHz provides the propagation velocity of the QS mode for different density values ​​^^, according to the table below. ^^ (g / cm 3 ) VQS (m / s) 0.7 1931 0.75 1924 0.8 1918 0.85 1911 0.9 1905 0.95 1900 1 1894 At 15 °C, the different density values ​​^^ are provided in the following table. ^^ (g / cm 3) VQS (m / s) 0.7 1925 0.75 1918 0.8 1912 0.85 1906 0.9 1900 0.95 1894 1 1888 Over this range of variation in density, we can estimate a linear law of variation of the speed as a function of the density ^^. This law can then be used to obtain the density once the measurement has been made ^^ ^^^^ performed. Statistical calculations show that, in this case, a measurement of the speed ^^ ^^^^to within 1%, and knowledge of the temperature to within 5°C, lead to the estimation of the liquid density to within approximately 13%. This step can be repeated at several different frequencies and the result averaged in order to improve the accuracy of the estimation of the density. Alternatively, a numerical model of the dispersion curve (linking the speed of the second mode to the frequency) can be stored in the memory of the processing module 13 to accurately estimate the propagation speed of the second mode over a whole frequency range by minimizing the difference between the experimental dispersion curve and the dispersion curve obtained by an analytical or numerical model. Indeed, the accuracy of the measurement of this speed directly conditions the accuracy of the estimation of the density in fine. The third step 130 is a step of determining the mass of liquid on board.The mass of liquid is determined from the density determined in the second step 120, from the volume of liquid in the container and / or conduit 1, more particularly by multiplying these two quantities. The volume of liquid is determined by applying a height / volume law taking into account the liquid level 5 determined in the first step 110 and knowing the geometry of the tank as well as the attitude of the aircraft (yaw, pitch, roll, etc.). When the geometry of the tank is simple, such as a cylinder, the volume of liquid is given by the analytical equation of the volume of the geometry in question, one of the dimensions of which is the liquid level 5. When the geometry of the tank is complex, the volume of liquid can be determined by means of analytical or numerical formulations via a numerical volume calculation model.It is advantageously observed that the determination of the propagation speed of the first propagation mode, the determination of the propagation speed of the second propagation mode and the propagation speed of the third propagation mode can be carried out by generating a single signal by one of the first or second sensors. The excitation signal is then adapted to generate these three modes simultaneously. The on-board mass is then transmitted to an operator, such as the pilot, or to an external system, for example for storing the information on a memory or in order to maintain the center of gravity of the aircraft relative to the center of thrust and not to unbalance the aircraft by an excess or a deficiency of fuel mass. In a second embodiment, it is not possible to use reflectors to implement the sub-steps 111 and 112 of determining the propagation speeds of the first and second propagation modes.In this case, the plurality of sensors 11 comprises a third sensor and a fourth sensor. The third sensor is arranged so as to be positioned on the wall 2 below the liquid level 5. The fourth sensor 11 is, for its part, arranged so as to be positioned on the wall 2 above the liquid level 5. In this second embodiment, it is also assumed that the properties of the liquid are homogeneous throughout the container and / or conduit 1. In this embodiment, it is therefore not necessary for the first sensor to be positioned so as to be sufficiently far from the upper surface 3, nor for the second sensor to be positioned sufficiently far from the lower surface 4.The position of the third sensor 11 and the position of the fourth sensor 11 on the wall 2 are dependent on the instrumentation conditions and constraints linked to the environment as well as to the geometry, accessibility and size of the container and / or conduit 1, as for the first and second sensors. Similarly, it is preferable that the propagation path between the third and fourth sensors does not include a mechanical interruption, such as an impedance break, for example in the form of an attenuating polymer seal or a geometric discontinuity. For example, it is possible to position the third and fourth sensors on the same side, typically on the same face, of the tank as the first and second sensors, in particular in order to promote the transmission and detection of the guided waves in the wall.Preferably, the fourth sensor and the third sensor are arranged on the wall 2 so as to be above the low interface Bb, the bubble B, and below the high interface Fh, the puddle F, respectively. Preferably again, the fourth sensor is substantially horizontally aligned with the first sensor, for example in a plane parallel to the liquid level when the device is on the ground and is not in motion, so that the transmission of a wave is made along a direct propagation path between these sensors. Similarly, the third sensor is substantially horizontally aligned with the second sensor, for example in a plane parallel to the liquid level when the device is on the ground and is not in motion, so that the transmission of a wave is made along a direct propagation path between these sensors.In the example of Figure 4, the tank is cylindrical and the fourth sensor is positioned substantially at the same height as the first sensor. Similarly, the third sensor is positioned substantially at the same height as the second sensor. This alignment simplifies the calculation of the liquid level 5. It is preferable that the propagation paths between the first and fourth sensors and between the second and third sensors do not include a mechanical interruption, such as an impedance break, for example in the form of an attenuating polymer seal or a geometric discontinuity. The relative positioning of the fourth and third sensors does not depend on specific conditions other than those previously discussed. However, it is preferable that the third and fourth sensors are arranged on the same side of the container and / or conduit in order to promote the transmission of a guided wave from one to the other.When the geometry of the container and / or conduit 1 allows it, it is preferable for the third sensor and the fourth sensor to be substantially vertically aligned with each other, along the vertical axis of the container and / or conduit, so that the transmission of a wave is made along a direct propagation path between these sensors. This alignment makes it possible to simplify the calculation of the liquid level 5. In this second embodiment, the determination of the propagation speeds of the first mode and the second mode, in this case of the modes A0 and QS, is not based on a “pulse-echo” type measurement using known reflectors, but it is based on a direct measurement in transmission of a wave guided in the wall between the first and fourth sensors and between the second and third sensors, respectively. Thus, in sub-step 111, the propagation speed of the first propagation mode, i.e. mode A, is determined. 140 rmined by ^^ ^^0 = ^^ 14 , where ^^ 14 is the distance between the first sensor and the fourth sensor and ^^ 14 is the propagation time of the first propagation mode between the first and fourth sensors. The propagation time ^^ 14 is determined by calculating a difference between the instant of capture of the first propagation mode of the guided wave by the sensor that captures the wave, i.e. the first or the fourth sensor, and the instant of generation of said first propagation mode of the guided wave by the sensor that emits the wave, i.e. the fourth or the first sensor, respectively. Similarly, in sub-step 112, the propagation speed of the second propagation mode, i.e. the QS mode, is determined by where ^^ 23 is the distance between the second sensor and the third sensor and ^^ 23is the propagation time of the second propagation mode between the second and third sensors and is determined by calculating a difference between the instant of capture of the second propagation mode of the guided wave by the sensor that captures the wave, i.e. the second or third sensor, and the instant of generation of said second propagation mode of the guided wave by the sensor that emits the wave, i.e. the third or second sensor, respectively. In sub-step 113, the height ^^ of liquid is then determined by This equation is given in the case where the guided wave is transmitted between the first and second sensors. Alternatively, the wave can be generated by the second sensor and be picked up by the first sensor. The height ^^ is determined according to the same equation and the propagation time ^^ 12is determined by calculating a difference between the instant of capture of the first propagation mode of the guided wave by the first sensor and the instant of generation of the second propagation mode of the guided wave by the second sensor. In sub-step 113, the guided wave can, alternatively, be transmitted between the fourth and third sensors. The height ^^ of liquid is then determined by where ^^ 34 is the distance between the fourth sensor and the third sensor and ^^ 34 is the propagation time of the guided wave propagating between the third and fourth sensors 11 according to the first propagation mode above the liquid level 5 and according to the second propagation mode below the liquid level 5. The propagation time ^^ 34is determined by calculating a difference between the instant of capture of the second propagation mode of the guided wave by the third sensor and the instant of generation of the guided wave by the fourth sensor. Alternatively, the wave can be generated by the third sensor and be captured by the fourth sensor. In this case, the first propagation mode is generated by conversion of the second propagation mode upon passing the liquid / gas interface. The height ^^ is determined in the same way as before and the propagation time ^^ 34is determined by calculating a difference between the instant of capture of the first propagation mode of the guided wave by the fourth sensor and the instant of generation of the second propagation mode of the guided wave by the third sensor. In a third embodiment, compatible with the preceding embodiments, the measurements carried out at the different steps of the method 100 may be based on the capture by several sensors of the generated waves. In this case, the plurality of sensors comprises one or more intermediate sensors 11i, arranged on the wall 2 so as to be positioned between the first and second sensors, as illustrated in FIG. 5. The intermediate sensors 11i may be regularly spaced according to a pitch, given in units of length. The pitch may be constant or may vary depending on the position along the wall 2.Depending on the sensor technology used, these intermediate sensors can be arranged at a distance from or in contact with the wall 2 and can be placed in or outside the container and / or conduit. The advantage is then to capture the guided wave generated throughout its path between the first and second sensors by the intermediate sensors 11i, which is advantageous if the container and / or conduit is large. In addition, these intermediate sensors allow redundancy of measurements in the event of failure of certain sensors, in particular the first and second sensors and / or the third and fourth sensors, if applicable. This also makes it possible to detect a variation in the propagation speed of the wave and the modes, by measuring the associated propagation times, according to which it propagates, during its propagation. There is therefore a stratification of the propagation speeds.It is then possible to determine a gradient of the propagation speed for one or more of the first, second or third modes of propagation of the guided wave in the wall 2. These gradients can be obtained by interpolation by any interpolation method, for example by means of a linear function, of the propagation speed values ​​determined by the measurement of the sensors 11 of the plurality of sensors 11, in particular the intermediate sensors 11i. A possible temperature gradient is therefore taken into account when determining the propagation speeds. The propagation speed gradient can thus be taken into account for the first, second and third modes in the calculation of the height ^^ of liquid, thus improving the accuracy and reliability of the calculation.Furthermore, having propagation velocity gradients is particularly advantageous for applications where the properties of the liquid in the container and / or conduit 1 are not homogeneous. For example, during its various stopovers, an aircraft refuels at different refueling points for which the properties of the fuel refueled are different. This is particularly the case for the density of the fuel. The use of intermediate sensors 11i to determine the propagation velocity gradient of the first, second and / or third modes makes it possible to determine a gradient of the density of the fuel in the container and / or conduit. Consequently, it is possible to determine with great precision the mass of fuel on board the container and / or conduit by applying the height / volume law taking into account the density gradient.It is also possible to arrange intermediate sensors 11i between the fourth and third sensors on the wall 2. This makes it possible to form pairs of sensors associating one of the intermediate sensors 11i arranged between the first and second sensors with one of the intermediate sensors 11i arranged between the fourth and third sensors. Each pair is such that the sensors of said pair are substantially placed at the same height and / or on the same horizontal plane, for example parallel to the liquid level when the device is on the ground and is not in motion. The advantage is to be able to evaluate with great precision the gradient of the propagation speeds of the first and second modes calculated according to the sub-steps 111 and 112.In a fourth embodiment, compatible with the preceding embodiments, the method 100 comprises the preliminary step 105 which is a step of determining the thickness of the wall 2 of the container and / or conduit 1. The thickness is the distance separating the external face 2e and the internal face 2i of the wall 2. The thickness is determined by calibration using a known fluid. The thickness is, for example, obtained by inverting the equation of a dispersion curve of the propagation speed of the first, second and / or third propagation mode while the container and / or conduit contains a known fluid. The known fluid is a fluid whose properties are predefined, in particular its density. The temperature of the wall and in the volume of the container and / or conduit, including the known fluid, is predefined.Preferably, this determination is implemented by propagation of the first propagation mode in the wall, i.e. the A0 mode, which is the most sensitive to the thickness, while the container and / or the conduit are filled with a gas, in particular air. The inversion consists, knowing the excitation frequency and the propagation speed of the propagation mode concerned, preferably the first propagation mode, in finding the corresponding dispersion curve to deduce the thickness of the wall. The inversion can be implemented by means of known inversion techniques, such as the inversion of an analytical equation of the dispersion curve, the inversion of the numerical model making it possible to determine the dispersion curve, or the search among a pre-constructed database of dispersion curves obtained analytically, experimentally and / or by the numerical model.The advantage of using an inversion based on searching in a database is that its implementation, once the database has been constructed, is much faster than other approaches, and is therefore more suitable for in situ implementation of the method 100. At a given frequency, the database can be constructed by defining variation ranges for the wall thickness and for the excitation frequency, and by determining the associated propagation speed by reading a calculated dispersion curve, taking into account the thickness / excitation frequency pair in question. This database can be stored in the memory of the processing module 13. It is then sufficient, during the preliminary step 105, to search in the database for the pair corresponding to the propagation speed of the propagation mode concerned, here the first propagation mode, and to the related excitation frequency, to determine the wall thickness.The inversion therefore consists of finding the dispersion curve whose wall thickness from which it is calculated generates the propagation of the third propagation mode at the generation frequency with the propagation speed ^^. ^^0determined. The mechanical properties of the wall necessary for calculating the dispersion curve are predefined. Alternatively, the thickness is determined via direct measurement, for example using an ultrasonic probe. Advantageously, this preliminary step only needs to be implemented once, preferably before the first implementation of steps 110 to 130 of the method 100. It is possible to combine this embodiment with the embodiment comprising the intermediate sensors 11i so as to evaluate the thickness of the wall 2 over several sections, each section being delimited by two successive intermediate sensors 11i. This variant makes it possible to precisely determine the thickness of the wall when it is not constant between the first and second sensors and / or between the fourth and third sensors.In a fifth embodiment, compatible with the previous embodiments, the excitation signal is a broadband signal. This makes it possible to excite the first, second and third propagation modes according to several excitation frequencies. It is then possible to average the measurement of the different propagation times involved and thus improve the resolution of the propagation times determined and necessary for the calculations of the different propagation speeds involved. It is possible to use bandpass filtering, following known approaches, in order to isolate the excitation frequency or frequencies of each of the modes in order to calculate the respective propagation speeds. In a sixth embodiment, compatible with embodiments two to five, one or more point levels are determined. These point levels indicate whether there is a presence of liquid at the sensor or not.The determination of the point levels is implemented by generating a wave transmitted between two sensors positioned substantially at the same height, for example between the first and fourth sensors and / or between the second and third sensors and / or between the sensors of a pair of intermediate sensors 11i. A variation in the propagation speed of the A0 mode, due to its conversion to QS mode, thus makes it possible to detect the presence of the liquid. In a seventh embodiment, compatible with the previous embodiments, the method further comprises a step (not shown) of determining the liquid level in the bubble B and / or in the puddle F. These levels are determined by means of “pulse-echo” measurement using a wave generated by the first or second sensors and reflected by the first or second reflector, respectively. In the bubble, the liquid level is then given by. with ^^ 1^^the propagation time of the guided wave propagating between the first sensor and the first reflector according to the second mode, here QS, between the first sensor and the liquid level, and according to the first mode, here A0, between the liquid level and the first reflector. In the puddle, the liquid level is then given by with ^^ 2^^the propagation time of the guided wave propagating between the second sensor and the second reflector according to the first mode, here A0, between the second sensor and the liquid level, and according to the second mode, here QS, between the liquid level and the second reflector. In an alternative embodiment, the density of the liquid is determined by means of a measurement of the electrical impedance of one of the piezoelectric sensors positioned below the liquid level, for example the second sensor. The electrical impedance measurement is carried out at the terminals of the sensor over a frequency range, possibly only at a given frequency, for example at a resonance frequency of the sensor. Such an impedance measurement varies with the transfer of mechanical energy to the liquid, which depends in particular on the ratio between the acoustic impedance of the mechanical wall and that of the liquid, the acoustic impedance of the liquid being given by the product ^^ × ^^,with ^^ the density of the liquid and ^^ the speed of sound propagation in the liquid. A one-dimensional model, as presented in I. Perrissin-Fabert Et al., "Nondestructive evaluation of materials using an inserted piezoelectric sensor- correlation with hardness measurements" IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, vol. 42(4), pp. 641-648, 1995, can then be used to invert the density of the liquid. Other more complex models, such as possibly a three-dimensional numerical model, can also be used. This variant is particularly advantageous for a container and / or conduit with a flexible and thin wall, for example for a helicopter tank, which reduces the impedance break between the wall and the liquid contained in the tank and makes the method more sensitive to the density of said liquid.In an alternative embodiment, the temperature of the wall is directly measured by means of one or more thermocouples, or one or more sensors from another temperature measurement technology, positioned outside the container and / or conducted on the wall 2. For an aeronautical application, it is possible to duplicate the plurality of sensors of the gauging system in order to take into account the attitude of the aircraft in the calculation of the volume of fuel on board. For example, a first plurality of sensors and a second plurality of sensors, both comprising the sensors mentioned above, are positioned on different sides of the tank. Alternatively, these sensors are positioned on the same side, typically on the same face, of the tank, so as to promote the transmission and detection of the guided waves in the wall.This second plurality of sensors also allows for backup acquisition if some of the sensors in the first plurality of sensors are defective. In some embodiments, since step 110 is independent of steps 120 and 130, the latter are not implemented and only the determination, step 110, of the liquid level in the container and / or conduit is implemented. In this case, method 100 is a method for determining the liquid level in the container and / or conduit, which only comprises the implementation of step 110, as well as the embodiments and alternatives relating to the calculation of the liquid level. In some embodiments, to determine the onboard liquid mass, the liquid level is determined or otherwise known via a similar or different approach than that described in step 110.In this case, it is possible to implement only steps 120 and 130 of the method 100 for gauging the mass of the liquid. In such a case, there is no need to use the first and second sensors 11. Indeed, a single sensor may be sufficient to implement steps 120 and 130 of the method 100 for gauging the mass of the liquid. The plurality of sensors is then at least one sensor, which has no specific positional constraints, except those required to implement steps 120 and 130. By way of example, this at least one sensor may be the second sensor. Also by way of example, when necessary, the at least one sensor may comprise the first and second sensors, and sometimes even the third and fourth sensors.In an alternative to step 110, the first sensor 11 and the second sensor 11 are not necessarily arranged so as to be positioned above and below the liquid level 5, respectively. Rather, it is simply required that these two sensors are positioned on the wall 2, or at a distance, depending on the sensor technology used. The position of the first and second sensors are also dependent on the instrumentation conditions and constraints and they may be arranged so as to be above the low interface Bb of the bubble B and below the high interface Fh of the puddle F, respectively. It is also possible that the first and second sensors are arranged on the same side of the container and / or conduit and that the path between these sensors does not include a mechanical interruption.These two sensors can be substantially aligned horizontally with each other, along a horizontal axis of the container and / or conduit, so that the transmission of a wave is made along a direct propagation path between these sensors, in the wall. The horizontal axis of the container and / or conduit is parallel to the plane defined by the surface of the liquid at its interface with the gas when the device is on the ground and is not moving. In other words, when the container and / or conduit is a cylinder, the first and second sensors are positioned so as to be substantially in a plane orthogonal to the wall 2. This alignment also makes it possible to simplify the calculations of the gauging method. For example, these sensors make it possible to determine a point level. The liquid level is then alternatively determined via a step of estimating the point level of liquid in the container and / or conduit.By "point level" is meant an estimation of the liquid level relative to a threshold above or below which the liquid level in the tank is located. It is therefore a binary estimation and less precise than the determination of the level proposed in step 110. This point level is determined via the emission by the first sensor and / or the second sensor of a wave, which is picked up by the second sensor and / or the first sensor, respectively. In this case, the first sensor and the second sensor are positioned so as to be substantially aligned along a horizontal line, to form the position of the threshold of the point level. Thus, when the wave propagates while the liquid level is above the position of the sensors, this wave propagates according to the second propagation mode. On the contrary, when the liquid level is below the position of the sensors, this wave propagates according to the first propagation mode.Since these two propagation modes do not have the same propagation speed, it is possible to discriminate in a binary way whether the liquid level is above or below the position of the sensors, since the first propagation mode is generated when there is no liquid at the height of the propagation path, while the second mode is generated only when there is liquid at the height of this path. To do this, it is sufficient to determine the propagation time of the first or second propagation mode according to which the wave propagates between the sensors, to determine a propagation speed associated with this propagation, then to compare this determined speed with a reference speed. This comparison then makes it possible to discriminate whether the determined speed corresponds to that of the first propagation mode or to that of the second propagation mode.For illustration, as shown in Figure 8, consider that the first sensor 11 and the second sensor 11 are aligned horizontally and are separated by a distance ^^. 12 known. These two sensors are positioned at a height ℎ from the lower surface 4. In the example of Figure 8 a), the wave propagates according to the second propagation mode, while in the example of Figure 8 b), the wave propagates according to the first propagation mode. The propagation speed ^^ 12 to discriminate for the wave propagating between the two sensors is determined by ^^ ^^ 12 12 = ^^ 12 , where ^^ 12is the propagation time of said wave between the two sensors, one way. To make the comparison, the reference speed can be a predetermined speed of the propagation speed of the first or second propagation modes of the wave in the wall. This predetermined speed can be estimated theoretically via an analytical or numerical model, knowing the properties of the wall 2 and of the liquid, or be evaluated experimentally, for example according to the methods of implementation of steps 111 and / or 112 previously described. Thus, when a distance between propagation speed ^^ 12 and the reference speed is lower than a predefined threshold, it is deduced that this propagation speed ^^ 12 corresponds to the reference propagation speed (which can be that of the first or second mode). Conversely, when this distance is greater than the predefined threshold, it is deduced that the propagation speed ^^12 does not correspond to the reference propagation speed (that of the first or second mode) but corresponds to the propagation speed of the other propagation mode (that of the second or first mode, respectively). It is observed that it is possible to predetermine the propagation speeds of the first and second propagation modes, following the aforementioned modalities. In this case, the propagation speed ^^ 12corresponds to the propagation speed of the first or second mode with respect to which a calculated distance is the smallest, or less than said predetermined threshold. Alternatively, the reference speed is a speed whose value is between the propagation speed of the first propagation mode and the propagation speed of the second propagation mode. For example, it is a speed sufficiently far from those of the two modes in question so that it cannot be confused with the respective speeds of these modes. For example again, this reference speed is the central value of the interval defined by the respective speeds of the two modes. Thus, when the propagation speed ^^ 12 is lower than the reference speed, it is considered that the propagation speed ^^ 12corresponds to that of the second propagation mode, since this mode propagates more slowly than the first propagation mode. Conversely, when the propagation speed ^^ 12 is greater than the reference speed, it is considered that the propagation speed ^^ 12 corresponds to that of the first propagation mode. Consequently, when the propagation speed ^^ 12 corresponds to the propagation speed of the first propagation mode, the liquid level is lower than the point level defined by the height ℎ of the sensors 11 with the lower surface, and when the propagation speed ^^ 12corresponds to the propagation speed of the second propagation mode, the liquid level is higher than the point level defined by the height ℎ. Subsequently, the point level indication makes it possible to determine that the onboard liquid level is at least higher than or equal to that of the liquid when the latter has a level above or equal to the point level (second propagation mode identified), or that the onboard liquid level is at least lower than that of the liquid when the latter is below the point level (first propagation mode identified). In certain embodiments, the system comprises an arrangement, in the form of an intermediate waveguide. This intermediate waveguide is used to transmit the different waves, according to the first, second and / or third propagation mode, from the outside of the container and / or conduit 1 into the wall 2 thereof.

Claims

CLAIMS

1. Method (100) for determining a level of a liquid in a container and / or a conduit (1) comprising: - a wall (2), in particular with an internal face (2i) and an external face (2e), the wall (2) delimiting an internal volume of the container and / or the conduit (1) comprising the liquid, and - a plurality of ultrasonic sensors (11) comprising at least one first sensor (11) positioned above the liquid level (5) and at least one second sensor (11) positioned below the liquid level (5), the liquid level (5) in the container and / or the conduit (1) being determined (110) from: - The propagation speed of a first propagation mode determined (111) by propagation of a guided wave in the wall (2) above the liquid level (5) and captured by the first sensor (11);- The propagation speed of a second propagation mode determined (112) by propagation of a guided wave in the wall (2) below the liquid level (5) and captured by the second sensor (11); and - The propagation time of a guided wave generated (113) by the first and / or the second sensor (11) propagating in the wall (2) according to the first propagation mode above the liquid level (5) and according to the second propagation mode below the liquid level (5).

2. Method (100) according to the preceding claim, wherein the liquid is a fuel.

3. Method (100) according to one of the preceding claims, wherein: - The first propagation mode is the antisymmetric mode A0 of a guided Lamb wave; - The second propagation mode is the Quasi-Scholte mode of a guided elastic wave.

4. Method (100) according to one of the preceding claims, in which: - The propagation speed of the first propagation mode is estimated (111) by generation then capture of a wave guided by the first sensor (11) after reflection on a first reflector located on the wall (2), in particular the internal face (2i), and above the liquid level, such that where ℎ1 is the distance between the first sensor (11) and the first reflector and ^^1 is the round-trip propagation time of the first propagation mode between the first sensor (11) and the first reflector; - The propagation speed ^^ ^^^^of the second propagation mode is estimated (112) by generation then capture of a wave guided by the second sensor (11) after reflection on a second reflector located on the wall (2), in particular the internal face (2i), and below the liquid level, such that 2ℎ ^^ =2 ^^^^ ^^2where ℎ2is the distance between the second sensor (11) and the second reflector and ^^2is the round trip propagation time of the second propagation mode between the second sensor (11) and the second reflector; and - The height ^^ liquid level (5) relative to the second sensor (11) is determined (113) by ^^ 12 − ^^ 12 ^ ^ ^^ ^^ 12 − ^^ 12 ^^1 ^^ = 0 2ℎ 11 1 = ^^ 2 ^^ 1, ^ ^ ^^^^ − ^ ^ ^^0 2ℎ 2− 2 ℎ 1where ^^ 12is the propagation time of the guided wave propagating between the first and second sensors (11) according to the first propagation mode above the liquid level (5) and according to the second propagation mode below the liquid level (5), and ^^ 12 is the distance between the first sensor (11) and the second sensor (11).

5. Method (100) according to claim 4, wherein the first reflector is at least a portion of the upper surface of the container and / or the conduit (1) and / or the second reflector is at least a portion of the lower surface of the container and / or the conduit (1).

6. Method (100) according to one of claims 1 to 3, wherein the plurality of sensors (11) further comprises a third sensor (11) arranged below the liquid level (5) and a fourth sensor (11) arranged above the liquid level, the first and fourth sensors (11) being positioned substantially at the same height, and the second and third sensors (11) being positioned substantially at the same height, and wherein: - The propagation speed ^^ ^^0 of the first propagation mode is estimated (111) by capturing a guided wave transmitted between the first and fourth sensors (11), such that where ^^14 is the distance between the first and fourth sensors (11), and ^^ 14 is the propagation time of the first mode of the transmitted guided wave to travel the distance ^^ 14 ; - The speed of propagation ^^ ^^^^ of the second propagation mode is estimated (112) by capturing a guided wave transmitted between the second and third sensors (11), such that where ^^ 23 is the distance between the second and third sensors (11), and ^^ 23 is the propagation time of the second mode of the transmitted guided wave to travel the distance ^^ 23 ; and - The height ^^ liquid level (5) relative to the second sensor (11) is determined (113) by where ^^ 12is the propagation time of the guided wave propagating between the first and second sensors (11) according to the first propagation mode above the liquid level (5) and according to the second propagation mode en- below the liquid level (5), and ^^ 12is the distance between the first sensor (11) and the second sensor (11).

7. Method (100) according to one of the preceding claims, wherein the first propagation mode and / or the second propagation mode are generated by means of a broadband excitation signal.

8. Method (100) according to one of the preceding claims, wherein the plurality of sensors (11) further comprises an intermediate sensor (11i) arranged at a distance from the wall (2) or on the wall (2), in particular on the external face (2e), and positioned between the first and second sensors (11), and the method (100) comprising determining a gradient of the propagation speed of the first and / or the second propagation modes from the propagation time of the first and / or the second propagation modes, respectively, by means of the intermediate sensor (11i).

9. System for determining a liquid level in a container and / or a conduit (1), the container and / or the conduit (1) comprising a wall (2) with an inner face (2i) and an outer face (2e), the inner face (2i) of the wall (2) delimiting an internal volume of the container and / or the conduit (1) comprising the liquid, the system comprising: - A plurality of ultrasonic sensors (11) arranged at a distance from the wall (2) and / or on the wall (2), comprising: o A first sensor (11), positioned above the liquid level (5); and o A second sensor (11), positioned below the liquid level (5); - An acquisition module connected to the plurality of ultrasonic sensors (11) and configured to control the plurality of ultrasonic sensors (11) so as to generate a guided wave and / or to acquire samples of a guided wave captured by one of the sensors (11) of the plurality of ultrasonic sensors (11); and. - A processing module connected to the acquisition module and configured to implement the method (100) according to one of claims 1 to 8 from the acquired samples.

10. Computer program product comprising instructions which, when the program is executed on a computer, cause the latter to implement the steps of the method according to one of claims 1 to 8.

11. Computer-readable recording medium comprising instructions which, when executed by a computer, cause the latter to implement the steps of the method according to one of claims 1 to 8.

Images