Method for determining amount of gas contained in insulated switchgear
Through embedded sensors and corrected gas temperature model, the accuracy problem of small leakage detection in gas insulated switchgear tanks is solved, and higher-precision gas volume calculation and leakage detection are achieved.
Patent Information
- Application Number
- CN202410559830.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-05-10
- Filing Date
- 2024-05-08
- Publication Date
- 2024-11-12
AI Technical Summary
It is difficult for the prior art to accurately detect small leakage in the tank of the gas insulated switchgear, and due to uneven gas temperature and environmental influences, it is difficult for traditional methods to achieve high-precision gas quantity calculation.
An embedded gas temperature sensor and pressure sensor are used to obtain multiple sets of sample data through the calibration stage, and a corrected gas temperature model is established to improve the accuracy of gas volume calculation, and the model is used for leakage detection during the measurement stage.
It improves the accuracy of gas volume calculation, can detect small leaks more accurately, reduces misjudgment, and ensures the safe operation of electrical equipment.
Smart Images

Figure CN118937146A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to medium voltage power distribution systems, and more particularly to a method for determining the amount of gas contained in a tank of a gas insulated switchgear. In addition, the method for determining the amount of gas can serve as a basis for a tank leakage detection method. Background Art
[0002] Gas-insulated switchgear comprises a connecting / disconnecting element for one or more phases of a medium voltage network. The connecting / disconnecting element is installed in a tank filled with gas. The dielectric properties of the gas facilitate arc interruption or provide electrical insulation between different phases of the network. The tank and all its interfaces are sealed, so that refilling of the tank is not required during the entire service life of the equipment. The gas contained in the tank can be, for example, an inert gas such as sulfur hexafluoride SF 6 In other applications, the gas may also be pressurized air.
[0003] It is known to monitor the amount of gas in a tank by measuring the temperature and pressure of the gas in the tank. For this purpose, a sensor having a pressure sensing probe and a temperature sensing probe is fixed in an interface plug, which is fixed to the tank panel. The sensing element is thus in contact with the gas in the tank. Thus, during the operation of the switchgear, the amount of gas can be calculated and monitored based on the pressure and temperature measurements. The correct operation of the electrical equipment can be checked by this amount monitoring. For example, a drop in the amount of gas may indicate a premature leak in the tank, which may require maintenance operations to correct it.
[0004] Detection of small leaks can be difficult. Small leaks by their very nature result in a slow reduction in the amount of gas. For example, a pressure drop from 1.2 bar to 1.1 bar may take weeks or months and still require detection. In addition, the pressure inside the tank is affected by the gas temperature, which itself is affected by the ambient temperature and the intensity of the current in the electrical conductor. Therefore, the internal pressure and temperature of the tank can be affected by the operating conditions of the electrical equipment. Since the gas pressure and temperature sensors are located on the outer wall of the tank, the measured temperature may not represent the true gas temperature. In addition, the temperature of the gas contained in the tank may not be uniform and temperature gradients may exist. Therefore, the accuracy of gas volume calculation required to detect low-level leaks may be difficult to achieve.
[0005] The present disclosure proposes a method for estimating the amount of gas in a gas filling tank with higher accuracy and also proposes a leak detection method using the method for estimating the amount of gas. Summary of the invention
[0006] To this end, a method is proposed for determining the amount of gas contained in a tank of a gas-insulated switchgear, the tank comprising an embedded gas temperature sensor and an embedded gas pressure sensor,
[0007] The method includes the following steps during the calibration phase:
[0008] (i) obtaining a first plurality of consecutive sets of calibration samples, each set of calibration samples comprising a gas pressure measured by an embedded gas pressure sensor, a gas temperature measured by an embedded gas temperature sensor, and an ambient temperature measured by an ambient temperature sensor external to the tank,
[0009] (ii) for a second plurality of sets of calibration samples included in the first plurality of sets of calibration samples, determining the amount of gas contained in the tank from a gas equation of state based on the acquired gas pressure and based on a corrected gas temperature model based on the acquired gas temperature, the acquired gas pressure and the acquired ambient temperature,
[0010] (iii) updating the correction gas temperature model so as to minimize the difference between the gas quantities determined for each set of calibration samples,
[0011] The method also includes the following steps during the measurement phase:
[0012] (iv) obtaining the gas temperature measured by the embedded gas temperature sensor, the gas pressure measured by the embedded pressure temperature sensor, and the ambient temperature measured by the ambient temperature sensor outside the tank,
[0013] (v) determining a corrected gas temperature from a corrected gas temperature model and from the acquired gas temperature, the acquired gas pressure and the acquired ambient temperature,
[0014] (vi) Determine the gas quantity from the gas state equation and the obtained gas pressure and the determined corrected gas temperature.
[0015] During the calibration phase (whose duration is negligible compared to the expected service life of the switchgear), the tank is considered to be completely sealed and therefore the amount of gas contained in the tank is assumed to be constant. Possible differences between the gas quantities determined for consecutive calibration samples do not come from any actual leakage, but from the fact that the measured gas temperature is not fully representative of the actual gas temperature inside the tank. Therefore, the correction gas temperature can be determined by applying the constraint that the determined gas quantity must be the same for all sets of samples acquired during the calibration phase. The correction gas temperature is the temperature at which the gas quantity predicted by the gas state equation is the same for all sets of calibration samples, or at least the temperature at which the differences between the gas quantities are minimal. Once the model for determining this correction gas temperature has been determined, the same model is also applied during the measurement phase. The use of this gas temperature model improves the accuracy of the gas quantity calculation compared to methods based solely on raw measurements from different sensors. Since the model is extracted from a limited number of experimental data, it is easy to calibrate and can be quickly adapted to any gas tank / temperature sensor configuration. The method does not require the installation of any additional sensors during the calibration phase, so the electrical equipment of the integrated tank can remain in its original configuration without any modification, even temporary. The calibration phase, also called the learning phase, can be carried out when the GIS is new and commissioned. It can also be repeated after every major maintenance operation on the equipment.
[0016] The following features can optionally be implemented individually or in combination with each other:
[0017] The calibration phase is followed by the measurement phase. In other words, the calibration phase is performed before the measurement phase is performed.
[0018] The measurement phase is a monitoring phase that can be performed throughout the entire service life of the switchgear.
[0019] The measurement phase can be performed continuously.
[0020] Alternatively, the measurement phases may be performed at different time intervals.
[0021] The calibration phase includes a sub-step of checking the canister tightness.
[0022] Each time a set of samples is acquired, the gas pressure measured by the embedded gas pressure sensor, the gas temperature measured by the embedded gas temperature sensor, and the ambient temperature measured by the ambient temperature sensor external to the tank are acquired simultaneously.
[0023] The volume of the tank is constant.
[0024] The amount of gas includes a closely related physical parameter, which can be the number of moles of gas or the mass of gas. It also includes the gas density, since the internal volume of the tank remains constant.
[0025] In step (iii), the model may be updated so that the amount of gas determined for each set of samples is the same.
[0026] The first plurality of consecutive sets of calibration samples includes at least three samples.
[0027] In an embodiment, the calibration phase comprises the following steps:
[0028] (iii-a1) if the difference between the gas quantities determined for each set of calibration samples in step (iii) is below a first predetermined threshold, then terminating the calibration phase,
[0029] (iii-a2) If the difference between the gas amounts determined for each set of calibration samples in step (iii) is greater than or equal to a first predetermined threshold, an additional set of calibration samples is added to the second plurality of sets of calibration samples to obtain an updated second plurality of sets of calibration samples, and steps (ii) and (iii) are repeated with the updated second plurality of sets of calibration samples.
[0030] When the expected accuracy is reached, the calibration phase ends. As long as the expected accuracy has not been obtained, some calibration samples are added to the second plurality of sets of samples, which means using more samples to calibrate the model. As long as this target accuracy is not reached, more calibration data is added to improve the model.
[0031] In an embodiment, the method comprises the steps of:
[0032] (iii-a3) If the difference between the gas quantities determined for the first plurality of calibration samples in step (iii-a2) is greater than or equal to a first predetermined threshold, repeating step (i) to obtain at least one additional set of calibration samples and obtaining an updated first plurality of consecutive sets of calibration samples, and repeating steps (ii) and (iii) to update the corrected gas temperature model.
[0033] In other words, samples are added to the second plurality of sets of samples until all calibration samples of the first plurality of sets of samples are integrated, which means that the second plurality of sets of samples are identical to the first plurality of sets of samples. If, after all calibration samples have been used to calibrate the model, the determined difference is still above or equal to a predetermined acceptance threshold, some additional sets of samples are acquired to continue the calibration process with more sets of samples in order to achieve the desired accuracy. Alternatively, the calibration process can be started from scratch with a completely new set of samples.
[0034] In an embodiment, the method comprises a validation phase of the correction gas temperature model between the calibration phase and the measurement phase, the validation phase comprising the following steps:
[0035] (iii-b1) for a third plurality of sets of calibration samples included in the first plurality of sets of calibration samples, determining the amount of gas contained in the tank from a gas state equation and from a corrected gas temperature model based on the acquired gas temperature, the acquired gas pressure, and the acquired ambient temperature,
[0036] (iii-b2) determining a difference between the gas quantities determined for each set of samples of the third plurality of sets of calibration samples,
[0037] (iii-b3) If the determined difference is below a second predetermined threshold, entry into the measurement phase is authorized.
[0038] Once the calibration acceptance criteria have been reached, the calibration phase ends and an intermediate verification phase can be performed before entering the measurement phase. For this purpose, some sample sets of initially acquired data can be used to check whether the relevant determined gas quantities are close enough to the expected results.
[0039] The third plurality of sets of calibration samples may have no common elements with the second plurality of sets of calibration samples.
[0040] In other words, the model may be calibrated using certain sample sets and validated using different sample sets. The same set of samples is not used for calibration and validation of the gas-corrected temperature model.
[0041] In an embodiment, the method comprises the following steps:
[0042] (iii-b4) If the determined difference is above or equal to a second predetermined threshold, repeating step (i) to obtain at least one additional set of calibration samples, and repeating steps (ii) and (iii) to update the correction gas temperature model.
[0043] If, after all available data initially acquired have been used, the accuracy criteria for accepting the model are still not met, some additional data acquisition is performed. To this end, step (i) is repeated to add at least one more set of samples, and the gas-corrected temperature model M is updated with the additional set of samples. The second predetermined threshold may be higher than the first predetermined threshold. The second predetermined threshold may be equal to the first predetermined threshold.
[0044] According to an embodiment of the proposed method, the gas state equation is the ideal gas law.
[0045] For the ideal gas model, the equation of state is:
[0046] P*V=n*r*T (Equation 1)
[0047] According to another embodiment of the proposed method, the gas state equation is a real gas law. For example, the van der Waals state equation may be used. The Virial state equation may also be used. Other different real gas laws may also be used.
[0048] In an embodiment of the method, the corrected temperature model is a multiple linear regression model.
[0049] For example, the corrected gas temperature is a linear combination of the acquired gas pressure, the acquired gas temperature, and the acquired ambient temperature.
[0050] The calibration gas temperature satisfies:
[0051] Tcor=K 0 +K 1 *P+K 2 *T+K 3 *TA (Equation 2)
[0052] Tcor represents the calibration gas temperature, P represents the acquired gas pressure, T represents the acquired gas temperature, TA represents the acquired ambient temperature, and K 0 , K 1 , K 2 , K 3 is a constant coefficient.
[0053] When the number of acquired samples is less than the number of measured variables, the parameters of such an underdetermined linear system can be derived due to the minimum L2 norm solution using, for example, the Lagrange multiplier method. On the contrary, if the number of samples is greater than the number of measured variables, a classical regression method such as the least squares regression method can be chosen.
[0054] In another embodiment of the method, the corrected temperature model is a perceptron model.
[0055] The acquired gas pressure, acquired gas temperature and acquired ambient temperature are the inputs of the sensor. The corrected temperature is the output of the sensor.
[0056] In yet another embodiment of the method, the corrected temperature model is support vector regression.
[0057] In another embodiment of the method, the correction temperature model is a gradient boosting regression.
[0058] The time interval between two consecutive sets of calibration samples of the first plurality of consecutive sets of calibration samples may be greater than one minute and less than one hour.
[0059] In an embodiment of the method, the step (i) of acquiring a first plurality of consecutive sets of calibration samples may comprise the sub-steps of:
[0060] - After taking one sample and before taking the next sample, modifying the current intensity in the switchgear and thus the temperature of the gas contained in the tank.
[0061] By extending the values of this calibration sample set, the robustness of the model for determining the correction gas temperature can be improved, since the measurement is less affected by noise and environmental conditions.
[0062] The present disclosure also relates to a method for detecting gas leakage in a tank of a gas insulated switchgear, comprising the following steps:
[0063] - determining the amount of gas contained in the tank by the method described above,
[0064] - Detection of gas leaks based on the evolution of the gas volume over time.
[0065] The gas leak detection method may include the following steps:
[0066] - determine the initial amount of gas contained in the tank at the end of the calibration phase,
[0067] - determine the updated amount of gas contained in the tank during the measurement phase,
[0068] - determining a difference between the determined initial amount and the determined updated amount,
[0069] - if the determined difference is above a predetermined threshold, the presence of a leak is detected,
[0070] If the determined difference is lower than or equal to a predetermined threshold, it is detected that the can is sealed.
[0071] In an embodiment, the determined difference is an absolute difference.
[0072] In this case, the predetermined threshold is an absolute threshold.
[0073] In another embodiment, the determined difference may be a relative difference.
[0074] In this case, the predetermined threshold is a relative threshold.
[0075] The method may include the following steps:
[0076] - If a leak is detected, a warning signal is issued.
[0077] The present disclosure also relates to an electrical equipment, which includes a switchgear, a tank configured to accommodate the switchgear, and an electronic control unit configured to implement the aforementioned gas amount determination method or the aforementioned gas leakage detection method. BRIEF DESCRIPTION OF THE DRAWINGS
[0078] Additional features, details and advantages will be seen in the following detailed description and accompanying drawings, in which:
[0079] Figure 1is a schematic diagram of an electrical installation containing a gas-insulated switchgear in a tank,
[0080] Figure 2 is a time curve showing the evolution of the internal temperature of the gas contained in the tank,
[0081] Figure 3 is a time curve showing the evolution of the pressure of the gas contained in the tank,
[0082] Figure 4 is a time graph showing the evolution of the corrected gas temperature determined for the gas contained in the tank,
[0083] Figure 5 is a time curve showing the evolution of a determined amount of gas contained in the tank when the tank is fluid-tight,
[0084] Figure 6 is another time curve showing the evolution of a determined amount of gas contained in a tank when a tank leak occurs,
[0085] Figure 7 is a block diagram of the different steps of an embodiment of the proposed method,
[0086] Figure 8 is a time plot illustrating some aspects of the proposed method. DETAILED DESCRIPTION
[0087] In order to make the drawings easier to read, the various elements are not necessarily represented to scale. In these figures, the same elements receive the same reference numerals. Certain elements or parameters can be indexed, that is to say, designated by, for example, "first element" or "second element", or first parameter and second parameter, etc. The purpose of such indexing is to distinguish between similar but not identical elements or parameters. Such indexing does not mean that one element or one parameter takes precedence over another, and their names can be interchanged. When it is mentioned that a subsystem includes a given element, it does not exclude the presence of other elements in the subsystem.
[0088] Figure 1 The electrical equipment 100 is schematically shown, which includes a switchgear 2, a tank 1 configured to accommodate the switchgear 2, and an electronic control unit 15. The electronic control unit 15 is configured to implement a gas quantity determination method to be described below. The electronic control unit 15 is also configured to implement a gas leak detection method, which will also be described in detail below.
[0089] The tank 1 defines an enclosure configured to house the switchgear 2. The tank 1 comprises a number of panels which are assembled to form a sealed enclosure defining a sealed volume. For example, the panels of the tank 1 are welded together. The panels may be fixed to a subframe which ensures sufficient rigidity to resist the pressure difference between the inside and the outside of the tank 1 without significant deformation. Thus, the volume of the tank 1 is constant throughout its service life.
[0090] The tank 1 comprises a sealed interface 14 allowing access to three electrical conductors 11, 12, 13 connected to the switchgear 2. Each of the three electrical conductors 11, 12, 13 corresponds to a different phase of the circuit. Each phase of the circuit can be interrupted by a circuit breaker 21, 22, 23, respectively. Figure 1 In FIG. 2 , the switchgear 2 is shown in the open position of the circuit breakers 21 , 22 , 23 , in which the current is interrupted.
[0091] When the tank 1 is in normal operation, the internal volume of the tank 1 is filled with pressurized gas G. When the switchgear 2 is operated to interrupt the current in the circuit, the pressurized gas G prevents the generation of arcs. The outside of the tank 1 is exposed to the temperature and pressure of the ambient air A in the room where the tank 1 is installed. The gas pressure inside the tank 1 is higher than the ambient pressure outside the tank 1, i.e. the local atmospheric pressure. The tank 1 can be filled with an inert gas, such as sulfur hexafluoride (SF 6 ). Tank 1 may also be filled with pressurized air.
[0092] Although the tank 1 is constructed to be fluid-tight, small leaks may occur during the life of the electrical equipment 100, and the amount of gas G contained in the internal volume of the tank 1 may slowly decrease. Therefore, the dielectric properties slowly degrade and the risk of arcing during equipment operation increases. To this end, the amount of gas N contained in the tank 1 is monitored so that the operator of the electrical equipment 100 can take appropriate corrective measures when the amount of gas N is insufficient to safely operate the electrical equipment 100. This gas amount monitoring is based on gas temperature and gas pressure measurements. As mentioned above, the gas amount N can be a gas mass or a gas mole number. It can also be a gas density because the internal volume of the tank remains constant. The conversion of one parameter selected from mass, mole number, density to another of these selected parameters can be performed by a multiplication factor. Therefore, these three parameters can be considered to be equivalent physical parameters.
[0093] The tank 1 is equipped with an embedded gas temperature sensor 3. The embedded temperature sensor 3 includes a temperature sensitive element 6, which is configured to transmit an electrical signal representing the temperature of the sensitive element 6. The temperature sensitive element 6 of the embedded temperature sensor 3 is located inside the tank 1 and is in contact with the gas G contained in the tank 1. The temperature sensing element 6 of the embedded gas temperature sensor 3 may include a thermistor, such as a negative coefficient thermistor. In a variant, the temperature sensing element 6 may include a thermocouple.
[0094] The embedded temperature sensor 3 includes a sensor body 7 in contact with the ambient air A outside the tank 1. The sensor body 7 protrudes from the outside of the panel of the tank 1. The implementation of the sensor is easy because it is arranged outside the tank 1 and its connection lines are kept outside the internal volume of the tank 1, so that the sealing of the tank 1 is not disturbed. The embedded temperature sensor 3 is attached to an interface plug 9, which is fixed to the panel of the tank 1. The interface plug 9 is inserted into a hole formed in the panel and is welded to the panel, for example, along the periphery of the interface plug. The interface plug 9 defines an opening 10 in the panel of the tank 1, through which the sensitive element 6 can contact the gas G inside the tank 1. The interface plug 9 may include an internal thread for fixing the sensor body 7. When the gas temperature sensor 3 is installed, the opening 10 is closed by the body 7 of the gas temperature sensor 3. A seal (not shown) may be provided between the interface plug 9 and the body 7 of the temperature sensor to ensure the fluid tightness of the tank 1.
[0095] The tank 1 is also equipped with an embedded pressure sensor 4. The embedded pressure sensor 4 includes a pressure sensitive element 8, which is configured to transmit an electrical signal representing the pressure applied to the sensitive element. The sensitive element 8 of the embedded pressure sensor 4 is located inside the tank 1 and is in contact with the gas G contained in the tank 1. The pressure sensing element 8 and the temperature sensing element 6 can be accommodated in the same sensor body 7, as is the case in the example shown. A single interface plug 9 is used to mount both sensors and obtain gas temperature measurements and gas pressure measurements. Therefore, the sensor is a multifunctional sensor combining pressure measurement and temperature measurement.
[0096] According to a design variant not shown, the embedded pressure sensor 4 can be a dedicated sensor, separate and independent from the gas temperature sensor 3. In this case, the pressure sensor 4 is fixed to another interface plug defining another opening in the panel of the tank 1.
[0097] The electrical equipment 100 includes an ambient temperature sensor 5. The ambient temperature sensor 5 includes a sensitive element configured to transmit an electrical signal representing the temperature of the sensitive element. The sensitive element of the ambient temperature sensor 5 is located outside the tank 1 and is in contact with the ambient air outside the tank 1. The ambient temperature sensor 5 is installed, for example, on a control panel 20 of a switchgear 2. The ambient temperature sensor 5 and the outer surface of the tank 1 are spaced apart from each other. The distance between the ambient temperature sensor 5 and the outer surface of the tank 1 is preferably greater than 30 centimeters. Therefore, the ambient temperature measurement is not affected by the heat emitted by the tank 1. The measurement value of the ambient temperature sensor 5 is considered to fully represent the ambient temperature of the room where the tank 1 is installed. The ambient temperature sensor 5 may include a thermistor, such as a negative coefficient thermistor. The ambient temperature sensor 5 may include a thermocouple.
[0098] Figure 2 The evolution of the internal temperature of the gas contained in the tank measured by different methods is shown.
[0099] Curve C1 shows the temperature measured by the embedded temperature sensor 3. Curve C2 shows the average temperature measured by a group of sensors installed in the tank 1. For example, one sensor is installed near the top of the tank 1, another sensor is installed in the middle of the tank, and another sensor is installed near the bottom of the tank 1. These sensors are installed only for experiments, and when the electrical equipment 100 is operating normally, these sensors are not part of the electrical equipment 100.
[0100] The two curves C1 and C2 are offset, indicating that the average temperature of the gas measured at different locations of the tank 1 is different from the temperature measured by the sensor mounted to the panel of the tank 1. One reason is that the electrical conductors 11, 12, 13 dissipate heat while no forced gas circulation occurs. Therefore, there are some temperature gradients within the tank 1.
[0101] Furthermore, the temperature of the gas contained in the tank 1 varies with the power demanded by the grid, since the heat dissipated by the current circulating in the housing of the tank 1 varies. Figure 2 In the example, the current intensity increases at 1400 minutes and decreases at 3000 minutes. These current intensity changes cause the temperature to increase and then decrease.
[0102] Figure 3 The evolution of the pressure of the gas contained in the tank measured by the embedded pressure sensor 4 is shown. Due to the change in temperature, the pressure of the gas also changes over time. The temperature curve and the pressure curve have similar shapes. The pressure of the gas contained in the tank 1 can be considered to be uniform. The pressure gradients between different positions of the tank 1 are negligible.
[0103] The proposed method is used to determine the amount N of gas G contained in a tank 1 of a gas insulated switchgear 2, the tank 1 comprising an embedded gas temperature sensor 3 and an embedded gas pressure sensor 4. The method comprises the following steps in a calibration phase:
[0104] (i) Obtaining a first plurality of n consecutive sets of calibration samples S 1 , S 2 ,…,S n , each set of calibration samples S 1 , S 2 ,…,S n The gas pressure P measured by the embedded gas pressure sensor 4 1 , P 2 ,…,P n , the gas temperature T measured by the embedded gas temperature sensor 3 1 , T2 ,…,T n and the ambient temperature TA measured by the ambient temperature sensor 5 outside the tank 1 1 TA 2 ,…,TA n ,
[0105] (ii) For the calibration samples S included in the first plurality of n groups 1 , S 2 ,…,S n The second plurality of p sets of calibration samples S 1 , S 2 ,…,S p , based on the obtained gas pressure P 1 , P 2 ,…,P p and based on the obtained gas temperature T 1 , T 2 ,…,T p , the obtained gas pressure P 1 , P 2 ,…,P p and the obtained ambient temperature TA 1 TA 2 ,…,TA p The corrected gas temperature model M is used to determine the amount of gas N contained in tank 1 from the gas state equation E 1 、N 2 ,…,N p ,
[0106] (iii) Update the calibration gas temperature model M so as to minimize 1 , S 2 ,…,S p Determine the amount of gas N 1 、N 2 ,…,N p The difference between.
[0107] The method also includes the following steps during the measurement phase:
[0108] (iv) obtaining the gas temperature Tsens measured by the embedded gas temperature sensor 3, the gas pressure Psens measured by the embedded pressure temperature sensor 4, and the ambient temperature Tamb measured by the ambient temperature sensor 5 outside the tank 1,
[0109] (v) determining the correction gas temperature Tcor from the correction gas temperature model M and from the acquired gas temperature Tsens, the acquired gas pressure Psens and the acquired ambient temperature Tamb,
[0110] (vi) The gas quantity N is determined from the gas state equation E and from the acquired gas pressure Psens and the determined corrected gas temperature Tcor.
[0111] The duration of the calibration phase is considered to be negligible compared to the expected service life of the switchgear 2. In practice, the duration of the calibration phase is of the order of a few hours or days, whereas the service life of the switchgear 2 is of the order of more than 10 years. During the calibration phase, the tank 1 is considered to be completely sealed. Therefore, the amount of gas N contained in the tank 1 can be considered constant. All sets of samples S 1 , S 2 ,…,S n Theoretically, this should provide the same result for the determination of the amount of gas contained in tank 1, in other words, for all quantities N 1 、N 2 ,…,N n Should be equal.
[0112] In fact, at different times t 1 , t 2 ,…,t k Determine the amount of various gases N 1 , N 2 ,…,N k These differences arise from the fact that the temperature measured by the embedded gas temperature sensor 3 is not fully representative of the actual temperature of the gas contained in the tank 1. Therefore, the simulated gas temperature, also called the corrected gas temperature Tcor, can be determined by applying a calculation constraint that is calculated for different sets of samples S taken during the calibration phase. 1 , S 2 ,…,S k , the determined gas quantity must be the same. The corrected gas temperature Tcor is the gas quantity predicted by the gas state equation E that is closest to each group of samples S 1 , S 2 ,…,S ktemperature. When the fit of the correction gas temperature model is not ideal, corresponding to most actual situations, some residual differences may remain between the gas quantity predictions obtained from each individual set of samples. Therefore, the gas correction temperature model M is updated iteratively until the best fit is obtained. When the fit of the correction gas temperature model is ideal, the gas quantities predicted by different samples are the same. In this case, the difference between the predictions for each sample is zero. Once the model M for determining the correction gas temperature Tcor is determined, the same model M is applied during the measurement phase. The model M using the correction temperature Tcor improves the accuracy of the gas quantity calculation compared to methods that use only the raw measurements of different sensors, and in particular, only the raw gas temperature. Since the model M is extracted from a limited amount of experimental data, it is easy to calibrate and can be quickly adapted to any gas tank / sensor configuration. No additional sensors need to be installed during the calibration phase. The electrical equipment 100 using the tank 1 does not need to be modified in any way.
[0113] The calibration phase is followed by the measurement phase. In other words, the calibration phase is performed before the measurement phase is performed.
[0114] The measurement phase is a monitoring phase that can be performed throughout the service life of the switchgear. The measurement phase can be performed continuously. The measurement phase can also be performed at different time intervals. For example, the measurement phase can be performed once an hour, or once a day. Any time period can be used depending on the available data acquisition resources and data processing resources, and on the expected detection time for ensuring the safe operation of the electrical equipment.
[0115] The calibration phase may include a sub-step of checking the tightness of the tank 1. This sub-step is done before taking measurements. The purpose of this preliminary check is to ensure that the tank is properly sealed and that the assumption that the amount of gas does not change during the calibration phase is correct.
[0116] The first set of samples S 1 Corresponding to the first moment t 1 The second group of samples S 2 Corresponding to the second moment t 2 The nth group of samples S n Corresponding to the nth time t n The first moment t 1 At the second moment t 2 Before, the second moment t 2 At the third moment t 3 Before, and so on. The first moment t 1 , the second time t 2 and the third moment t 3 Appear successively in chronological order. The n-1th moment and the nth moment appear successively in chronological order.
[0117] The first plurality of n consecutive sets of calibration samples S 1 , S 2 ,…,S n In the first plurality of n consecutive groups of calibration samples S 1 , S 2 ,…,S n There is no maximum limit on the number of elements in .
[0118] Each time a set of samples is acquired, the acquired set of samples includes the gas pressure P measured by the embedded gas pressure sensor 4, the gas temperature T measured by the embedded gas temperature sensor 3, and the ambient temperature TA measured by the ambient temperature sensor 5 outside the tank 1. Optionally, other parameters are also acquired. Each time a set of samples is acquired, different samples are acquired simultaneously.
[0119] Simultaneously means that a gas temperature sample, a gas pressure sample and an ambient temperature sample of a given set of samples are taken within the same time interval of duration less than a predetermined threshold. The predetermined threshold is, for example, 0.5 seconds. A shorter threshold corresponding to a faster sampling frequency can be adopted, but this is not mandatory, since the time evolution of the different parameters is relatively slow. Some tests carried out by the applicant have shown that the time constant of the temperature evolution is several hours, for example between 8 and 10 hours depending on the internal volume of the tank 1.
[0120] Figure 8 The acquisition of a set of samples during the calibration phase is schematically shown. The evolution of the gas pressure P, the gas temperature T and the ambient temperature TA over time is plotted. 1 ,t 2 ,…,t n Each corresponds to the acquisition of a set of samples, namely S 1 , S 2 ,…,S n Each set of samples includes a gas pressure sample, a gas temperature sample and an ambient temperature sample. Symbol P1 refers to the first plurality of consecutive sets of calibration samples S 1 , S 2 ,…,S n The symbol P2 refers to the second plurality of p groups of calibration samples S 1 , S 2 ,…,S p In the example shown, the second plurality of groups of samples includes 4 groups of samples.
[0121] The second plurality of p groups of calibration samples S 1 , S 2 …、S p The first plurality of n sets of calibration samples S are included 1 , S 2 …、S nThis means that initially n sets of samples may be acquired, but the gas correction temperature model M is defined from a smaller number of sample sets. For example, initially n=30 sets of samples may be acquired, but only 10 sets of samples may be used for the definition and update of the gas correction temperature model M.
[0122] In an embodiment, the calibration phase comprises the following steps:
[0123] (iii-a1) If in step (iii) for each set of calibration samples S 1 , S 2 ,…,S p Determine the amount of gas N 1 、N 2 ,…,N p If the difference between is less than the predetermined threshold value Th, the calibration phase ends.
[0124] (iii-a2) If in step (iii) for each set of calibration samples S 1 , S 2 ,…,S p Determine the amount of gas N 1 、N 2 ,…,N p If the difference between the two is greater than or equal to a predetermined threshold, an additional set of calibration samples S p+1 Add to the second plurality p sets of calibration samples S 1 , S 2 ,…,S p To obtain the updated second plurality of p+1 sets of calibration samples S 1 , S 2 ,…,S p , S p+1 , and use the updated second plurality of p+1 sets of calibration samples S 1 , S 2 Repeat steps (ii) and (iii).
[0125] The calibration phase ends when the expected accuracy is achieved, i.e. when the differences between the determined quantities of the various gases are below an acceptance threshold. As long as the expected accuracy has not yet been achieved, calibration samples are added to the second plurality of samples, thereby using more samples to calibrate the model M, which helps to improve its accuracy. As long as the acceptance threshold has not been reached, further sets of calibration samples are added to the second plurality of samples in order to improve the model M.
[0126] exist Figure 8 In the figure, symbol P2 shows the initial definition of the second plurality of groups of samples for generating the corrected gas temperature model M, and includes 4 samples S 1 , S 2 , S 3 and S 4Assume that for each set of calibration samples S 1 , S 2 , S 3 , S 4 Determine the amount of gas N 1 、N 2 、N 3 、N 4 The difference between them is greater than or equal to the predetermined threshold, the fifth group of samples S 5 is integrated into the second plurality of group samples. Symbol "P2" shows an updated second plurality of group samples, into which the fifth sample is integrated.
[0127] In an embodiment, the method comprises the steps of:
[0128] (iii-a3) If in step (iii-a2) the first plurality of n sets of calibration samples S 1 , S 2 ,…,S n Determine the amount of gas N 1 、N 2 ,…,N n If the difference between is greater than or equal to a predetermined threshold, then repeat step (i) to obtain at least one additional set of calibration samples S n+1 And obtain the updated first plurality of n+1 consecutive sets of calibration samples S 1 , S 2 ,…,S n , S n+1 , and repeat steps (ii) and (iii) to update the corrected gas temperature model M.
[0129] As long as the accuracy acceptance threshold is not achieved, additional sets of samples are added to the second plurality of set samples until all set calibration samples of the first plurality of set samples are integrated. The second plurality and the first plurality of set samples are identical. If, after all calibration samples have been used to calibrate the model M, the determined difference is still above or equal to the predefined acceptance threshold, some additional sets of samples are acquired to continue the calibration process with more data in order to reach the expected accuracy. Alternatively, the calibration process can be started from scratch with a completely new set of samples. In this case, the method is applied in the same way as before.
[0130] Once the expected accuracy is achieved, the calibration phase ends. However, a verification can be performed before actually entering the measurement phase.
[0131] To this end, the method may include a validation phase of the correction gas temperature model M between the calibration phase and the measurement phase. The validation phase includes the following steps:
[0132] (iii-b1) For the calibration samples S included in the first plurality of n groups 1 , S 2,…,S n The third plurality of q sets of calibration samples S k ,…,S k+q ,, determine the amount of gas N contained in tank 1 from the gas state equation E and the corrected gas temperature model M k ,…,N k+q , the gas temperature model is corrected based on the obtained gas temperature T k ,…,T k+q , the obtained gas pressure P k ,…,P k+q and the obtained ambient temperature TA k ,…,TA k+q ,
[0133] (iii-b2) is determined as the third plurality of q groups of calibration samples S k ,…,S k+q The amount of gas N determined for each set of samples k ,…,N k+q The difference between
[0134] (iii-b3) If the determined difference is below a second predetermined threshold Th, entry into the measurement phase is authorized.
[0135] Some sample groups of the initially acquired data can be used to check whether the relevant determined gas quantities are close enough to the expected results, i.e. for validating the corrected gas temperature model. If the validation is satisfactory, meaning that the empirical differences between the predictions of the different sample groups are below the acceptance threshold, the start of the measurement phase is authorized.
[0136] The third plurality of q groups of calibration samples S k ,…,S k+q A second plurality of p groups of calibration samples S may be 1 , S 2 ,…,S p There are no common elements. In other words, a certain set of samples may be used to calibrate the model, and a different set of samples may be used to validate the model. The same set of samples is not used for calibration and validation of the gas-corrected temperature model. Figure 8 In , the symbol P3 refers to a third plurality of sets of samples, used to validate the model M and authorize entry into the measurement phase. It is also possible that the third plurality of sets of calibration samples and the second plurality of sets of calibration samples overlap each other and have some common elements.
[0137] In an example not shown, the data initially acquired includes 30 sets of samples. For example, the first 12 sets of samples are used to calibrate the model M. The next 8 sets of samples are then used to validate the constructed model. If the difference between the gas quantities calculated for the 8 sets of samples is low enough, the model M is valid and the measurement phase can be entered. In this example, the last 10 sets of samples of the initial batch of 30 sets of samples are not used because the validation phase gives satisfactory results without using all the available data.
[0138] In an embodiment, the method comprises the steps of:
[0139] (iii-b4) If the determined difference is higher than or equal to the second predetermined threshold value Th2, repeat step (i) to obtain at least one additional set of calibration samples S n+1 , and repeat steps (ii) and (iii) to update the corrected gas temperature model M.
[0140] The second predetermined threshold Th2 may be higher than the first predetermined threshold Th1. In this case, the verification phase allows for more differences between the measured gas quantities than the calibration phase. The second predetermined threshold Th2 may also be equal to the first predetermined threshold Th1.
[0141] Figure 4 The sequence of the calibration phase, the verification phase and the measurement phase is shown. Starting from t=0 to t=t c The end of the time period corresponds to the calibration phase, denoted by the symbol CP. c From start to t = t v The end time period corresponds to the verification phase, denoted by the symbol VP. v The initial time period corresponds to the measurement phase, which is indicated by the symbol MP. In order to simplify the figure, only three sets of calibration samples are shown. 1 , S 2 , S 3 The time t 1 ,t 2 and t 3 At time t = t m The acquired sample sets are used for measurement and leak detection. Curve C5 shows the evolution of the corrected gas temperature Tcor when the electrical equipment 100 is running. The temperature can increase or decrease depending on the variation of the electrical power and the ambient temperature.
[0142] The volume of the tank 1 is constant. The amount N of gas G includes closely related physical parameters, which can be the number of moles of gas or the mass of gas. It also includes the gas density, since the internal volume of the tank remains constant.
[0143] In step iii, the model M can be updated so that for each set of samples S 1 , S2 ,…,S p Determine the amount of different gases N 1 、N 2 ,…,N p are the same. In other words, for each set of samples S 1 , S 2 ,…,S p Measured gas volume N 1 、N 2 ,…,N p The difference between may be zero.
[0144] According to an embodiment of the proposed method, the gas state equation E is the ideal gas law.
[0145] For the ideal gas model, the equation of state is:
[0146] P*V=N*r*Tcor(Equation 1)
[0147] Where P is the gas pressure, V is the gas volume, i.e. the internal volume of the tank, N is the amount of gas, r is the ideal gas constant 8.32 J / mol / K, and Tcor is the corrected gas temperature in Kelvin. This model is simple to calculate and is sufficiently accurate for conditions using electrical equipment with relatively low pressures and temperatures.
[0148] According to another embodiment of the proposed method, the gas state equation E is a real gas law. Using a real gas model can improve accuracy. Many different real gas laws are available. For example, the van der Waals state equation or the Virial state equation can be used.
[0149] In an embodiment of the method, the corrected temperature model is a multiple linear regression model.
[0150] For example, the corrected gas temperature Tcor is a linear combination of the acquired gas pressure P, the acquired gas temperature T, and the acquired ambient temperature TA.
[0151] Tcor=K 0 +K 1 *P+K 2 *T+K 3 *TA (Equation 2)
[0152] Where K 0 , K 1 , K 2 , K 3 is a constant coefficient. By iteratively adjusting the four coefficients K 0 , K 1 , K 2 , K 3 , so that the difference between the gas amounts predicted by different samples is minimized.
[0153] When the number of samples is less than the number of measured variables, the parameters of such an underdetermined linear system can be derived by using a minimum L2 norm solution such as the Lagrange multiplier method. On the contrary, if the number of samples is greater than the number of measured variables, a classical regression method such as the least squares regression method can be selected.
[0154] In another embodiment of the method, the corrected temperature model is a perceptron model.
[0155] For example, the corrected gas temperature Tcor is a function of a real-valued vector X containing the acquired gas pressure P, the acquired gas temperature T, and the acquired ambient temperature TA, a real-valued weight vector, and a bias value. The weights and biases are adjusted using an iterative method such as stochastic gradient descent. The perceptron model can be extended to a multi-layer neural network, where back propagation is used to adjust the weights and biases.
[0156] The acquired gas pressure P, the acquired gas temperature T and the acquired ambient temperature TA are the inputs of the sensor, and the corrected temperature Tcor is the output of the sensor.
[0157] In yet another embodiment of the method, the corrected temperature model is support vector regression.
[0158] For example, the correction Tcor is obtained by mapping the input variables (the acquired gas pressure P, the acquired gas temperature T, and the acquired ambient temperature TA) into a high-dimensional space, finding the hyperplane that best fits the data, and then mapping the predicted output back to the original space by calculating the dot product between the input vector and the support vector consisting of a subset of the training data points.
[0159] In another embodiment of the method, the correction temperature model is a gradient boosting regression.
[0160] For example, the correction Tcor is obtained by training an ensemble of decision trees for prediction. In this combination of decision trees, each decision tree is added to correct the mistakes made by the previous decision tree. The process is repeated until the algorithm is no longer able to improve the model about the predicted output compared to the actual output.
[0161] The first plurality of n consecutive sets of calibration samples S 1 , S 2 ,…,S n Two consecutive sets of calibration samples S k , S k+1 The time interval between.
[0162] At the k+1th time t k+1 and the kth time t k The durations are offset from each other by longer than one minute but shorter than one hour.
[0163] In an embodiment of the method, a first plurality of n consecutive sets of calibration samples S are obtained. 1 , S 2 ,…,S n The step (i) may comprise the sub-steps of:
[0164] - After taking one calibration sample and before taking the next sample, the current intensity in the switchgear 2 is modified, thereby modifying the temperature of the gas contained in the tank 1. By spreading the values of different sets of samples, the robustness of the model M for determining the corrected gas temperature can be improved, since the measurement is less affected by noise.
[0165] exist Figure 2 , curve C3 shows the corrected gas temperature Tcor determined from the model M and from the acquired gas temperature Tsens, the acquired gas pressure Psens and the acquired ambient temperature Tamb used as inputs to the model M. It can be observed that during the time interval from 1500 seconds to 3000 seconds, the corrected gas temperature represented by curve C3 is higher than curve C2, which is the average of different measurements at different positions of the tank, which itself is higher than curve C1, which is the measurement of the embedded sensor.
[0166] The present disclosure also relates to a method for detecting gas leakage in a tank 1 of a gas insulated switchgear 2. The method comprises the following steps:
[0167] - determining the quantity N of gas G contained in the tank 1 by the method described above,
[0168] - Detection of gas leaks based on the evolution of the gas quantity N over time.
[0169] The gas leak detection method may include the following steps:
[0170] - determining the initial quantity N0 of gas G contained in tank 1 at the end of the calibration phase,
[0171] - determining the updated quantity N of gas G contained in the tank 1,
[0172] - determining the difference D between the determined initial amount N0 and the determined updated amount N,
[0173] - if the determined difference D is above a predetermined threshold value Smax, the presence of a leak is detected,
[0174] If the determined difference D is lower than or equal to a predetermined threshold value Smax, it is detected that the tank 1 is leaktight.
[0175] The determined difference D may be an absolute difference. In this case, the predetermined threshold value Smax is also an absolute threshold value. The predetermined threshold value Smax is a constant value determined relative to the internal volume of the tank, since the initial amount of gas filling the tank is proportional to the volume of the tank.
[0176] In another embodiment, the determined difference D may be a relative difference. In this case, the predetermined threshold value Smax is a relative threshold value. The predetermined threshold value Smax is, for example, between 8% and 12%.
[0177] Figure 5 and 6 The evolution of the gas quantity N over time t is shown. The initial gas quantity determined at time t=0 is denoted by N0 on the vertical axis. Figure 5 Corresponds to an airtight tank, i.e. no leakage. Figure 6 Corresponds to an initially sealed tank that starts to leak. Curves C6 and C7 represent the evolution of the gas quantity N over time, continuously updated. Figure 5 In , the estimated gas amount N remains constant over time. Figure 6 The estimated gas volume N is approximately constant until time t = t f From this moment on, the estimated gas amount N represented by the curve C7 gradually decreases and keeps deviating from the initial value N0 because of the presence of the leak. The arrow represented by the symbol D represents the deviation between the initial gas amount N0 represented by the horizontal line and the updated gas amount N represented by the curve C7. Until the moment t d , the difference is small enough that no leakage is detected. d , the difference D becomes higher than a defined threshold Smax and thus a leak is detected. The threshold Smax is chosen large enough to avoid false positives, i.e. faults are detected when no real fault exists. It is also chosen small enough to ensure a short response time in case of a real fault.
[0178] The gas leak detection method may include the following steps:
[0179] - If a leak is detected, a warning signal is issued.
[0180] Warning signals give the operator of the electrical equipment an opportunity to investigate and correct the detected fault.
Claims
1. A method for determining the amount (N) of gas (G) contained in a tank (1) of a gas-insulated switchgear (2), the tank (1) comprising an embedded gas temperature sensor (3) and an embedded gas pressure sensor (4), the method comprising the following steps during a calibration phase: (i) obtaining a first plurality (n) of consecutive sets of calibration samples (S1, S2, ..., S n ), each set of calibration samples (S1, S2, …, S n ) includes gas pressures (P1, P2, ..., P n ), the gas temperature (T1, T2, ..., T n ) and the ambient temperature (TA1, TA2, ..., TA2) measured by the ambient temperature sensor (5) outside the tank (1) n ), (ii) for the calibration samples (S1, S2, ..., S n ) in the second plurality (p) of calibration samples (S1, S2, ..., S p ), based on the obtained gas pressure (P1, P2, ..., P p ) and determining the amount of gas (N1, N2 . . . , N2) contained in the tank (1) from the gas state equation (E) based on the corrected gas temperature model (M) n ), the gas temperature model (M) is corrected based on the acquired gas temperatures (T1, T2, ..., T p ), the obtained gas pressure (P1, P2, ..., P p ) and the acquired ambient temperature (TA1, TA2, ..., TA n ), (iii) updating the calibration gas temperature model (M) so as to minimize the second plurality of sets of calibration samples (S1, S2, ..., S p ) in each of the determined gas quantities (N1, N2, ..., N p ), The method further comprises the following steps during the measurement phase: (iv) obtaining the gas temperature (Tsens) measured by the embedded gas temperature sensor (3), the gas pressure (Psens) measured by the embedded pressure temperature sensor (4), and the ambient temperature (Tamb) measured by the ambient temperature sensor (5) outside the tank (1), (v) determining a correction gas temperature (Tcor) from a correction gas temperature model (M) and from the acquired gas temperature (Tsens), the acquired gas pressure (Psens) and the acquired ambient temperature (Tamb), (vi) Determine the gas quantity (N) from the gas state equation (E) and from the acquired gas pressure (Psens) and the determined corrected gas temperature (Tcor).
2. The method according to claim 1, wherein the calibration phase comprises the following steps: (iii-a1) If in step (iii) for each set of calibration samples (S1, S2, ..., S p ) to determine the gas quantity (N1, N2, ..., N p ) is less than a first predetermined threshold value (Th1), then the calibration phase is terminated. (iii-a2) If in step (iii) for each set of calibration samples (S1, S2, ..., S p ) to determine the gas quantity (N1, N2, ..., N p ) is greater than or equal to a first predetermined threshold (Th1), an additional set of calibration samples (S p+1 ) are added to the second plurality (p) of calibration samples (S1, S2, ..., S p ) to obtain updated second plurality (p+1) groups of calibration samples (S1, S2, ..., S p , S p+1 ), and repeat steps (ii) and (iii) with an updated second plurality (p+1) of calibration samples.
3. The method according to claim 2, comprising the steps of: (iii-a3) If in step (iii-a2) the first plurality (n) of calibration samples (S1, S2, ..., S n ) to determine the gas quantity (N1, N2, ..., N n ) is greater than or equal to a first predetermined threshold (Th1), repeat step (i) to obtain at least one additional set of calibration samples (S n+1 ) and obtain the updated first plurality (n+1) of consecutive sets of calibration samples (S1, S2, ..., S n , S n+1 ), and repeat steps (ii) and (iii) to update the corrected gas temperature model (M).
4. The method according to any of the preceding claims, comprising a validation phase of the correction gas temperature model (M) between the calibration phase and the measurement phase, the validation phase comprising the following steps: (iii-b1) For the calibration samples (S1, S2, ..., S n ) in the third plurality (q) of calibration samples (S k ,…,S k+q ), the amount of gas (N) contained in the tank (1) is determined from the gas state equation (E) and from the corrected gas temperature model (M) k ,…,N k+q ), the gas temperature model (M) is corrected based on the obtained gas temperature (T k ,…,T k+q ), the obtained gas pressure (P k ,…,P k+q ) and the acquired ambient temperature (TA k ,…,TA k+q ), (iii-b2) is determined as the third plurality (q) of calibration samples (S k ,…,S k+q ) for each determined gas quantity (N k ,…,N k+q ), (iii-b3) If the determined difference is below a second predetermined threshold (Th2), entry into the measurement phase is authorized.
5. The method according to claim 4, comprising the steps of: (iii-b4) If the determined difference is higher than or equal to a second predetermined threshold value (Th2), repeating step (i) to obtain at least one additional set of calibration samples (S n+1 ), and repeat steps (ii) and (iii) to update the corrected gas temperature model (M).
6. A method according to any one of the preceding claims, wherein the gas state equation (E) is the ideal gas law.
7. The method according to any one of claims 1 to 6, wherein the corrected gas temperature model (M) is a multiple linear regression model.
8. Method according to any one of claims 1 to 6, wherein the corrected gas temperature model (M) is a perceptron model.
9. The method according to any one of claims 1 to 6, wherein the corrected gas temperature model (M) is a support vector regression or a gradient boosting regression.
10. The method according to any one of the preceding claims, wherein a first plurality (n) of consecutive groups of calibration samples (S1, S2, ..., S n ) of two consecutive sets of calibration samples (S k , S k+1 ) is greater than one minute and less than one hour.
11. The method according to any one of the preceding claims, wherein a first plurality (n) of consecutive sets of calibration samples (S1, S2, ..., S n ) comprises the following sub-steps: After taking one calibration sample and before taking the next calibration sample, the current intensity in the switchgear (2) is modified, thereby modifying the temperature of the gas contained in the tank (1).
12. A method for detecting gas leakage in a tank (1) of a gas insulated switchgear (2), comprising the following steps: - determining the amount (N) of gas (G) contained in the tank (1) by a method according to any one of the preceding claims, - Detection of gas leaks based on the evolution of the gas quantity (N) over time.
13. The method according to claim 12, comprising the steps of: - determining the initial quantity (N0) of gas (G) contained in said tank (1) at the end of the calibration phase, - determining, during a measurement phase, the updated quantity (N) of gas (G) contained in said tank (1), - determining the difference (D) between the determined initial amount (N0) and the determined updated amount (N), - if the determined difference (D) is above a maximum threshold (Smax), the presence of a leak is detected, If the determined difference (D) is lower than or equal to a maximum threshold value (Smax), it is detected that the tank (1) is sealed.
14. The method according to claim 12 or 13, comprising the steps of: If a leak is detected, a warning signal is issued.
15. An electrical device (100), comprising a switchgear (2), a tank (1) configured to accommodate the switchgear (2), and an electronic control unit (15) configured to implement the gas quantity determination method according to any one of claims 1 to 11 or the gas leakage detection method according to any one of claims 12 to 14.
