Compensation for gas fuel flow rate measurement and improved turbomachine control

Abstract

A compensation for the lag error in a measurement (QFlowmeter), by a flow meter, of the flow rate of a gaseous fuel in a turbomachine pipe is proposed. The flow meter has a transfer function (H(p)). A flow rate measurement (Qm) derived from a pressure ratio of the pressures across the terminals of the metering member is obtained by measuring the pressures upstream and downstream of the metering member, and then by applying the Barré Saint Venant relationship to the pressures. A dynamic measurement error (ΔQm) of the flow meter is calculated using the transfer function and this flow rate measurement (Qm) derived from the pressure ratio. The dynamic error is then added to the flow rate measurement from the flow meter to obtain a corrected measurement (QFlowCorr).

Description

[0001]DESCRIPTION TITLE: GAS FUEL FLOW MEASUREMENT COMPENSATION AND IMPROVED TURBOMACHINE REGULATION Technical Field The present invention relates to the field of aircraft turbomachinery, and more particularly to the regulation of the gas supply to the combustion chamber of an aircraft turbomachine, as well as an aircraft turbomachine operating with gaseous fuel and equipped with a fuel system providing such supply regulation. Prior Arts Climate change is a major concern for many legislative and regulatory bodies worldwide. Indeed, various restrictions on carbon emissions have been, are being, or will be adopted by various states. In particular, an ambitious standard applies to both new types of aircraft and those currently in service, requiring the implementation of technological solutions to bring them into compliance with current regulations.Civil aviation has been actively contributing to the fight against climate change for several years now. Technological research efforts have already led to significant improvements in the environmental performance of aircraft. The Applicant takes into account factors impacting all phases of design and development to obtain aeronautical components and products that are less energy-intensive, more environmentally friendly, and whose integration and use in civil aviation have moderate environmental impacts, with the aim of improving aircraft energy efficiency.Consequently, the Applicant is constantly working to reduce its climate impact by employing environmentally sound methods and processes for development and manufacturing that minimize greenhouse gas emissions to the absolute minimum possible, thereby reducing the environmental footprint of its activities. This sustained research and development work focuses on new generations of aircraft engines, aircraft weight reduction, particularly through the use of lighter materials and onboard equipment, the development of electric propulsion technologies, and finally, aviation fuels. Within this framework, the use of cryogenic fuel to power the combustion chamber of an aircraft turbomachine is well-established. Such cryogenic fuels include, for example, liquefied natural gas (LNG) or liquid hydrogen (LH2).While cryogenic fuels are less expensive than conventional kerosene, their main advantage lies in their ability to significantly reduce CO2 emissions. LNG emits 25% less CO2 per unit of energy, while LH2 emits no CO2 at all. Hydrogen-powered aircraft propulsion is therefore a major factor in the decarbonization of the aviation sector. Cryogenic fuel is stored in liquid form, generally at low pressure (2-3 bar) and low temperature (113K or -160°C for LNG, 20K or -253°C for LH2) to minimize its transport volume. However, it is injected into the combustion chamber in a gaseous state. The "fuel system" encompasses all the components, from the storage tank onward, that distribute and meter the fuel into the combustion chamber to meet specific engine performance requirements. Figure 1 schematically illustrates such a fuel system100.This system includes a storage tank 110 for cryogenic fuel in its liquid state, connected to injectors 117 located in the combustion chamber by a supply line 120. Mounted on the line 120 from the tank 110 are a pressurization device 111 maintaining injection pressure, a heating device 112 raising the temperature of the cryogenic fuel to allow its evaporation into gaseous fuel, a gaseous fuel accumulator 113 acting as a buffer, and an injection system 114 controlling the injection of gaseous fuel from the accumulator 113 into the combustion chamber. The injection system 114 includes a metering device 115 that meters the mass flow rate of the gaseous fuel entering the combustion chamber, an optional shut-off valve 116, and the injector(s) 117.These various components are controlled based on measurements taken via sensors, such as a temperature sensor 130 upstream of the metering unit 115, a pressure sensor 131 also upstream of the metering unit 115, and a flow sensor 132 located as close as possible to the injectors 117. The thrust of a turbojet engine is controlled via the metering unit 115, coupled to the flow sensor 132 – typically a Coriolis flow meter – to enable the metering unit 115 to supply the desired quantity of gaseous fuel to the injectors 117, corresponding to a thrust command. More generally, the regulation of a turbomachine requires precise and rapid measurement of the fuel flow rate to meet various engine constraints, including the pumping protection requirements of the high-pressure compressor.However, flow measurement sensors or probes, such as the Coriolis flow meter, exhibit inertia and latency in the delivery of measurements, particularly during flow transients. These transients induce a measurement lag effect, detrimental to the proper control of the turbomachine, and even to its integrity, by delaying fuel metering regulation. Consequently, there is a need to improve the regulation of gaseous fuel flow in fuel systems, and in particular to improve the accuracy of flow measurement (mass or volumetric) of gaseous fuel in a turbomachine piping. Description of the Invention: The invention aims to overcome at least some of the aforementioned drawbacks and to provide an improved fuel system, specifically a more responsive fuel system capable of more precisely regulating the quantity of gaseous fuel injected into the combustion chamber.Such a fuel system would be more durable over time. It would therefore promote the development of environmentally friendly turbomachines operating on cryogenic fuel. To this end, the invention relates firstly to a method for measuring the flow rate of a regulated gaseous fuel in a turbomachine pipe by a metering device, the method comprising the following steps: obtaining, from a flow meter, a flow measurement in the pipe, the flow meter having a transfer function; determining a dynamic measurement error of the flow meter using the transfer function and a flow measurement derived from a ratio of pressures across the meter terminals; and adding the determined dynamic error to the flow measurement of the flow meter to obtain a corrected flow measurement of the gaseous fuel in the turbomachine pipe. Indeed, the flow measurement derived from the pressure ratio provides good representation of the relative deviations (i.e.(variations) in flow rates during flow transients, i.e., in dynamic regimes, because pressures are easily recovered instantaneously. The dynamic error, representing the lag of measurements by the flow meter, can thus be calculated in real time based on the modeling of this lag, through the flow meter's transfer function. This error is then added to the flow meter measurement value, compensating for the flow meter lag. The method thus offers a hybrid of a reliable measurement in static (or steady-state) regime, namely the flow meter measurement, with a reliable measurement in dynamic (or transient) regime, namely the measurement derived from the pressure ratio, in order to correct in real time the dynamic error of the reliable static measurement.This method is particularly well-suited to an aeronautical environment, as Coriolis flow meters are already qualified for such environments, as are easily integrated pressure sensors. The invention also relates to a method for regulating the flow rate of gaseous fuel in a turbomachine, comprising the following steps: defining a flow control setpoint at the inlet of a gaseous fuel metering device; measuring the gaseous fuel flow rate in real time in a turbomachine pipe in order to control the metering device; and a method in which the measurement of the gaseous fuel flow rate conforms to the aforementioned measurement method.The invention also relates to a system for measuring the flow rate of a regulated gaseous fuel in a turbomachine line controlled by a metering device, comprising: a flow meter configured to obtain a flow rate measurement in the line, the flow meter having a transfer function, and a controller configured to: - determine a dynamic measurement error of the flow meter using the transfer function and a flow rate measurement derived from a ratio of pressure drops across the metering device, and - add the determined dynamic error to the flow rate measurement of the flow meter in order to obtain a corrected flow rate measurement of the gaseous fuel in the turbomachine line. A system for regulating the flow rate of a gaseous fuel in a turbomachine can therefore use such a measurement system to facilitate the control of the metering device. Optional features of embodiments are defined in the attached claims.Some of these characteristics are explained below with reference to a process, while they can be transposed into system characteristics. In one embodiment, determining the dynamic error involves: obtaining the flow measurement from the pressure ratio across the metering device, applying the flowmeter's transfer function to the obtained measurement to obtain a filtered measurement, and calculating the difference between the obtained flow measurement and the filtered measurement, this difference corresponding to the dynamic error. The drag compensation principle proposed here is based on estimating this drag by filtering the flow measurement from the pressure ratio with the flowmeter's transfer function. By calculating the difference between this measurement and its equivalent filtered by the transfer function, the drag of the flow measurement by the flowmeter is deduced in real time.In one embodiment, obtaining the flow rate measurement from the pressure ratio across the metering device comprises: obtaining a pressure upstream of the metering device and a pressure downstream of the metering device, and applying the Saint Venant equation to the upstream and downstream pressures to obtain a flow rate. In one embodiment, the transfer function is of the form of a low-pass filter with pure delay. In another embodiment, the transfer function is of the form , with Tr a pure delay, and τ a time constant. Brief description of the drawings: The invention will be better understood upon a detailed study of two embodiments taken by way of non-limiting examples and illustrated by the accompanying drawings, in which: [Fig 1] represents a known fuel system; [Fig 2] represents an example of an embodiment of a turbomachine structure, here a double-flow, double-acting turbomachine. corps ;[Fig 3] schematically illustrates a method for measuring the flow rate of a regulated gaseous fuel in a turbomachine pipeline according to embodiments; [Fig 4] represents a detailed embodiment of an injection system for implementing the method, according to embodiments; [Fig 5] shows a numerical implementation of the transfer function ^^^^^^ ൌ ^^ି ^்.^ ^^ାఛ.^ ; et [Fig 6] It highlights, through flow curves, the benefits of the proposed measurement method. For clarity, the same elements are designated by the same references in the different figures. Furthermore, the various figures are not drawn to scale, as is usual in schematic representations. D escription détailléeReferring to Figure 1, the tank 110 can be any type of tank for cryogenic fuel in liquid form, such as liquefied natural gas (LNG) or liquid hydrogen (LH2). Cryogenic fuel is typically stored at low pressure (2-3 bar) and low temperature (113 K or -160°C for LNG, 20 K or -253°C for LH2). The pressurization device 111 allows the cryogenic fuel in liquid form to be pressurized in the supply line 120 in order to maintain an injection pressure higher than that of the turbomachine's combustion chamber.Specifically, at a given injection flow rate, the difference between the pressure at the outlet of the pressurization unit 111 and the pressure in the combustion chamber is maintained greater than the sum of the pressure losses of the various equipment or components along the fuel supply line 120 of the fuel system 100 up to the injectors 117. For illustrative purposes only, the pressurization unit 111 may be one or more high-pressure pumps, for example, centrifugal pumps in series, controlled by a controller (not shown in the figure, 'CTRL' in Figure 4), such as the electronic control unit or ECU (for "Engine Control Unit"). The pressure in the injection chamber for aeronautical applications can be several tens of bar at maximum takeoff speed, typically between 30 and 50 bar for hydrogen and a few bar for LPG.The heating element 112 raises the temperature of the cryogenic fuel to allow its evaporation into gaseous fuel, specifically within a temperature range permissible for its injection into the combustion chamber and thus for its combustion. For example, the permissible temperature range, and therefore the target temperature at the outlet of the heating element 112, can be set at 300 K + / - 15 K (27°C + / - 15°C) for dihydrogen and at 323 K + / - 15 K (50°C + / - 10°C) for LPG. The heating element 112 can be of any type, including those described in publication FR3110938A1. For illustrative purposes only, this is a closed-loop heat exchanger with a heat transfer fluid that recovers heat from the exhaust gases exiting nozzle 220 (Figure 2) and transfers it to the cryogenic fuel within the exchanger. The heating element 112 is controlled by the controller, for example, to the midpoint of the range during steady-state operation.The accumulator 113 is any container suitable for storing a gas such as LPG or H2 in its gaseous state. The accumulator 113 acts as a buffer zone for gaseous fuel, that is, an intermediate reserve ensuring the fuel supply to the injectors 117 under all circumstances – particularly during sudden fuel demands – without the pressure in the supply line 120 dropping under the injection pressure. The accumulated gas is stored at the pressure set by the pressurization device 111, taking into account any pressure losses (from the heating device 112 and in the accumulator 113). The injection system 114 controls the injection of the gaseous fuel from the accumulator 113 into the combustion chamber. It includes a metering unit 115, an optional shut-off valve 116 and the injector(s) 117 opening into the combustion chamber.The metering unit 115 is, for example, a pressure regulator coupled to an adjustable throat or a metering valve, configured to meter the mass flow rate of the gaseous fuel entering the combustion chamber. It is controlled by the controller according to control commands. These various components, typically the pressurization unit 111, the heating unit 112, the internal heating unit 113, and the injection system 114 (including its internal components), are controlled by one or more controllers (hereinafter referred to as "the controller") which receive measurements taken on the supply line 120. In particular, a temperature sensor 130 provides the controller with temperature measurements of the fuel in the line 120. In one embodiment, the temperature sensor 130, for example, a temperature probe or thermocouple sensor, is positioned upstream of the metering unit 115, after the accumulator 113.Alternatively, the temperature sensor 130 is an internal temperature sensor within the accumulator. A pressure sensor 131 provides fuel pressure measurements in line 120 to the controller. The pressure sensor 131 can also be positioned upstream of the metering unit 115, after the accumulator 113. In this way, the controller receives the temperature and pressure of the gaseous fuel at the inlet line of the injection system 114. Finally, a flow sensor or flow meter 132 provides fuel flow measurements in line 120 to the controller. For example, a mass flow meter can be placed as close as possible to the injectors 117 opening into the combustion chamber. The controller can also receive external control commands, typically a command to vary the fuel flow rate to meet a need to modify the turbomachine's thrust (engine acceleration or deceleration).A sensor acquisition frequency on the order of hundredths to tenths of a second allows for dynamic control of all components. For example, the pressurization pump or device 111 is activated by the controller when the pressure measured by sensor 131 falls below a low trigger threshold, higher than the pressure in the combustion chamber, and is deactivated by the controller when the measured pressure reaches a high shut-off threshold. Figure 2 schematically illustrates the structure of a twin-spool, twin-flow turbomachine. The twin-flow turbomachine 2 comprises, successively in the direction of airflow, i.e., from upstream (left in the figure) to downstream (right in the figure), an air inlet 20 and a fan 21, which delivers air to a primary duct 22 and a secondary duct 23. The term "duct" refers to the volume through which an airflow passes.The airflow circulating in the primary channel 22 successively passes through a low-pressure compressor 24a, a high-pressure compressor 24b, a combustion chamber 25, a high-pressure turbine 26a, and a low-pressure turbine 26b, before being ejected through a primary flow nozzle 220. Furthermore, the secondary airflow circulating in the secondary channel 23 is ejected separately through a secondary flow nozzle 230, after passing through a series of guide vanes 231. Figure 3 schematically illustrates a method for measuring the flow rate of a regulated gaseous fuel in the pipeline 120, according to various embodiments. Figure 4 illustrates a detailed embodiment of the injection system 114 for implementing the method. The method is implemented by a computing unit, typically the CTRL controller (Figure 4). The process includes obtaining a flow measurement of 300 in the 120 pipe.This measurement, denoted QFlowmeter, is performed by the flowmeter 132. The QFlowmeter can be a mass or volumetric flow rate, depending on the flowmeter typically used. In one embodiment, a Coriolis mass flowmeter is used. In particular, H2 mass flowmeters are commercially available for hydrogen and LNG flowmeters for LNG-type fuels. Coriolis mass flow meters typically provide QFlowmeter measurements at a frequency of 10 to 100 Hz. Of course, other flow sensors, for example vortex, ultrasonic or electromagnetic, can be used. The flow meter 132 presents a real transfer function, that is to say a mathematical model of the relationship between the input (the actual flow QReel in the pipe 120) and the output (measured flow QFlowmeter), a model most often considered to be invariant: QFlowmeter=QReal*H(p).We will hereafter call a transfer function 'H(p)' such a transfer function representing the flowmeter 132 providing the measurement Qflowmeter, regardless of the model used. 'p' is the Laplace variable. It is therefore possible to have a mathematical model of the flowmeter. The transfer function H(p) can be provided by the manufacturer of the flowmeter 132 or determined on a test bench. In one embodiment, a parameterizable reference transfer function is used on the test bench, only its parameters being determined by the tests. For example, the transfer function is of the form of a low-pass filter, of any order, in particular order 1 or 2. It can also include a delay characterizing the inertia and latency of the flowmeter. In a particular embodiment, a transfer function of the type of a first-order low-pass filter is used, i.e., having the form ^^^^ ^^^. ் ^ ^ ൌ ି ^ .^^ାఛ.^, with Tr a pure delay, and τ a time constant. The tests allow Tr and τ to be determined. These two time parameters take values ​​ranging from a few tens of milliseconds (e.g., 20 or 30 ms) to a few seconds (e.g., 3 to 5 s). Of course, transfer functions other than the low-pass filter can be used. Similarly, instead of using a parameterized transfer function, it is possible to obtain a set of test data (inputs, outputs) and use this set to select a transfer function (from a library) that is most suitable, or even to use this set in a generative artificial intelligence tool to generate a mathematical model H(p). This transfer function H(p) characterizes the entire flowmeter 132; it therefore models both the static component of the measurement and the dynamic component of the measurement.For a flowmeter with good static accuracy but insufficient dynamic performance, as is the case with flowmeters generally used in aeronautics, the transfer function H(p), after subtracting its static part, is useful for modeling the dynamic or transient error of the flowmeter. It is this characteristic that is used, as described below, to correct the flowmeter measurement QFlowmeter. Following the acquisition of the flowmeter measurement QFlowmeter, the process continues with the determination 310 of this dynamic measurement error of the flowmeter. The determination 310 is carried out using the transfer function H(p) and a flow measurement, denoted Qm, derived from a ratio of pressures across the metering unit 115. For this purpose, a pressure sensor, denoted P, is provided upstream of the metering unit 115. a m , for example sensor 131, and a pressure sensor 133 downstream of the dosing unit 115, denoted P avFor example, in the case of a pressure-reducing type metering device coupled to a throat, the pressure sensors can be positioned on either side of the throat, preferably as close as possible to the throat. Similarly, the flow meter 132 is placed on the pipe 120 as close as possible to where the pressure ratio is measured (i.e., to the pressure sensors and the metering device 115) so that both devices detect the flow rate simultaneously. Conventional pressure sensors capable of providing pressure measurements at a frequency of 50 to 1000 Hz are used. The upstream pressure is denoted P. a m while the downstream pressure is denoted P a v The Barré-Saint-Venant equation, known to those skilled in the art, allows the flow rate through an orifice to be calculated, in this case the neck of the dosing device 115. This equation provides a flow rate measurement as a function of the pressure ratio P av / P a mAt the terminals of the dosing device, the calculation formula differs depending on whether the flow rate is sonic (Pav / Pa m ≥ a threshold) or not (Pav / Pa m < the threshold). Obtaining a flow rate measurement using this relationship has the advantage of providing rapid measurements (and therefore reliable measurements of the dynamic flow rate variation) due to a very short time constant (at the pressure sensor level), regardless of the measurement quality in static conditions. The Barré-Saint-Venant equation for calculating mass flow rate is as follows: where C d is the discharge coefficient, for example of the neck, A is the cross-sectional area (in m²) 2 ) in the neck, ^ a m is the density of the upstream gas (in kg.m-3) and ^ is a constant of the gas considered, equal to C p / C v (C p etc v (being the specific heat capacities of the gas). The parameters C dand A are correlated to the pilot (thrust) control of the metering unit 115. Indeed, the flow rate is regulated by the variation in the opening of the passage section A. The relationship between the pilot control and the product Cd.A can be determined on a test bench, notably using the flow meter 132 in static / stabilized operation, or can be provided by the metering unit manufacturer. A sensor of the RVDT (Rotary Variable Differential Transformer) or LVDT (Linear Variable Differential Transformer) type, known to those skilled in the art, can be used to determine the exact positioning 'PositionDoser' of the metering unit 115 at different test points in stabilized operation. The LVDT type sensor is used when the control element of the metering unit 115 has a linear displacement, typically a cylinder.The RVDT type sensor is used when the control element of the dosing unit 115 exhibits angular displacement. At the various test points in steady state, the flowmeter 132 is used to determine the actual flow rate QFlowmeter=QActual in the pipe 120 (to within the static accuracy of the flowmeter), while the pressures P. a m and P a vare also obtained. By inverting the formulas of the Barré-Saint-Venant relation (the other parameters being known), knowing that QFlowmeter = Qm thanks to the stabilized regime, it is possible to obtain the values ​​of the product A.Cd as a function of each position of the metering unit 115 at the test points, and thus establish the relation A.Cd = f(PositionMeter). This relation can be modeled, and the resulting mathematical model (for example, a function) can be used subsequently. Alternatively, a lookup table can be used that links several positions PositionMeter to the corresponding values ​​of the product AC. d Step 310 then consists of determining the dynamic error of the flow meter 132 using the Saint Venant Barré equation, given that the AC equation is known. d = f(DosingPosition). In operation, a flow measurement is derived from the pressure ratio P a v / P a mThe pressure across the terminals of the metering unit 115 is first determined. This includes obtaining the pressure P a m upstream of the dosing unit 115 and the pressure P a v downstream of the metering device, then applying the Saint-Venant law to the upstream and downstream pressures to obtain the flow rate Qm. The application of the Saint-Venant law also takes into account the 'PositionDoseur' position of the metering device 115 to determine AC d The 'PositionPosition' is determined by the RVDT or LVDT type sensor. Once the flow rate measurement Qm is obtained, the flow meter's transfer function is applied. Substep 312 yields a filtered Qm measurement. Figure 5 illustrates a numerical implementation of the transfer function. .A person skilled in the art can develop numerical implementations for any other type of transfer function. The numerical implementation in the figure first comprises a sub-block 3120 for calculating the pure delay component, taking as inputs the pure delay parameter Tr and the flow rate measurement Qm. It then comprises a first-order low-pass filter sub-block 3122, taking as inputs the time constant τ, the flow rate measurement Qm, and the output of sub-block 3120. The SampleTime parameter ST represents the calculation period of the algorithm. The time parameters (Tr and τ) are therefore converted into algorithm units using the SampleTime parameter. Returning to Figure 3, once the filtered flow rate measurement QmFiltered is obtained, the difference between the flow rate measurement Qm and this filtered measurement QmFiltered is calculated.Substep 314 provides an estimate of the dynamic error ΔQm, that is, an estimate of the measurement drift error by flowmeter 132. This dynamic error ΔQm is added to the flowmeter's measurement QFlowmeter. Step 320 provides a corrected measurement QFFCorriged of the gaseous fuel flow rate in the turbomachine line 120. The concept described above aims to compensate for the drift of a gas flow measurement probe. The compensation is based on estimating the probe's drift by filtering a reliable measurement Qm under dynamic conditions with the probe's transfer function H(p). By calculating the difference between the measurement Qm and its equivalent filtered by H(p), the probe's measurement drift is deduced in real time. This estimate of the dynamic error thus calculated is then added to the value of the measurement QFlowmeter of the probe, which makes it possible to compensate for the drag induced by H(p).This results in a more precise gas flow measurement (QDebCorrigé) in both static and dynamic conditions. Figure 6 illustrates the more precise gas flow measurement achieved through the proposed hybridization. The figure represents the flow rates QDebmètre, Qm, and QDebCorrigé as a function of time. The figure shows the slow dynamics of the QDebmètre measurement from the flow meter, and conversely, the rapid evolution of the Qm measurement derived from the pressure ratio Pav / Pam. However, the Qm measurement is unreliable in static conditions (hence the y-axis shift between the two curves). The QDebCorrigé result of hybridizing the two measurements using the measurement method described above allows us to recover the dynamics of the Qm measurement with the absolute accuracy of the QDebmètre measurement. This drag compensation is advantageously used in regulating the gaseous fuel flow in turbomachine 2.The CTRL controller can define a flow control setpoint at the inlet of the metering unit 115, for example, based on a desired thrust. To control the metering unit, it is useful to measure the gaseous fuel flow rate in the pipeline 120 in real time, typically as close as possible to the metering unit 115. This real-time measurement is carried out according to the measurement method described above, incorporating measurement lag compensation by adding the dynamic measurement error obtained using H(p) and the flow rate measurement Qm derived from the pressure ratios across the metering unit 115. Of course, the present invention is not limited to the embodiments described above by way of example; it extends to other variations.

Claims

CLAIMS 1. A method for measuring the flow rate of a regulated gaseous fuel in a turbomachine (2) pipeline (120) by a metering device (115), the method comprising the following steps: obtaining (300) a flow rate measurement (QFlowmeter) from a flowmeter (132) in the pipeline, the flowmeter having a transfer function (H(p)); determining (310) a dynamic error (ΔQm) in the flowmeter (132) measurement using the transfer function and a flow rate measurement (Qm) derived from a ratio of pressures across the metering device (115); and adding (320) the determined dynamic error to the flow rate measurement (QFlowmeter) of the flowmeter to obtain a corrected flow rate measurement (QCorrectedFlow) of the gaseous fuel in the turbomachine pipeline (120). 2.A method according to claim 1, wherein determining (310) the dynamic error comprises: obtaining the flow measurement (Qm) from the ratio of pressures across the metering device (115), applying (312) the transfer function (H(p)) of the flowmeter (132) to the measurement obtained to obtain a filtered measurement (QmFiltered), and calculating (314) the difference (ΔQm) between the flow measurement obtained and the filtered measurement, said difference corresponding to the dynamic error.

3. A method according to claim 2, wherein obtaining the flow measurement (Qm) from the ratio of pressures across the metering device (115) comprises: obtaining a pressure (P. a m ) upstream of the dosing unit and a pressure (P a v) downstream of the metering device, and apply the Saint Venant Barré equation to the upstream and downstream pressures to obtain a flow rate.

4. A method according to any one of claims 1 to 3, wherein the transfer function is of the form a low-pass filter with pure delay.

5. A method according to any one of claims 1 to 4, wherein the transfer function is of the form , with Tr a pure delay, and τ a time constant.

6. Method for regulating the flow of a gaseous fuel in a turbomachine (2), comprising the following steps: defining a flow control setpoint at the inlet of a metering device (115) of the gaseous fuel, measuring in real time a flow rate (QDébCorrigé) of gaseous fuel in a turbomachine pipe (120), in order to control the metering device, method in which the measurement of the gaseous fuel flow rate conforms to the measurement method according to any one of claims 1 to 5. 7.A system for measuring the flow rate of a regulated gaseous fuel in a turbomachine (2) pipeline (120) by a metering device (115), comprising: a flowmeter (132) configured to obtain a flow rate measurement (QFlowmeter) in the pipeline, the flowmeter having a transfer function (H(p)), and a controller (CTRL) configured to: - determine a dynamic error (ΔQm) in the flowmeter (132) measurement using the transfer function and a flow rate measurement (Qm) derived from a ratio of pressures across the metering device (115), and - add the determined dynamic error to the flow rate measurement of the flowmeter in order to obtain a corrected flow rate measurement (QCorrectedFlow) of the gaseous fuel in the turbomachine pipeline (120). 8.System according to claim 7, wherein the controller (CTRL) is configured to: obtain the flow measurement (Qm) from the ratio of pressures across the metering unit (115), apply (312) the transfer function (H(p)) of the flowmeter (132) to the measurement obtained to obtain a filtered measurement (QmFiltered), and calculate (314) the difference (ΔQm) between the flow measurement obtained and the filtered measurement, said difference corresponding to the dynamic error.

9. System according to claim 8, wherein the controller (CTRL) is configured to: obtain a pressure (P. a m ) upstream of the dosing unit and a pressure (P a v ) downstream of the dosing unit, and Apply the Saint Venant Barré relation to the upstream and downstream pressures to obtain the flow measurement (Qm) from the pressure ratio.

10. System according to any one of claims 7 to 9, wherein the transfer function is of the form ^^^^^^ ൌ ^^ି ^்.^ ^^ାఛ.^, with Tr a pure delay, and τ a time constant.

Images