Gas leak detection device and gas leak detection method

Abstract

To provide a gas leak detector and a gas leak detection method, capable of early detecting leak of an insulating gas sealed in a pressure vessel with high accuracy by a low cost pressure sensor.SOLUTION: A gas leak detector 100 includes: a pressure sensor 2 for measuring an internal pressure of a pressure vessel 1; a temperature sensor 3 for measuring a temperature of the pressure vessel 1; a recording part 5; a calculation part (6) for correcting the pressure measured by the pressure sensor 2 with a correction function to calculate a correction pressure; and an examination part (6) for examining existence of leak of the insulating gas on the basis of a time change in the correction pressure. When setting an extreme value temperature measured for a predetermined period before examining the leak to Ts, the correction function is represented by a polynomial including a product Pm(T / Ts) n of power of the measured pressure P and power of a temperature ratio T / Ts of a measurement temperature T to the extreme value temperature Ts as a variable. A gas leak detection method comprises the steps of: correcting the pressure measured by the pressure sensor 2 with the correction function to calculate the correction pressure; and examining the existence of leak of the insulating gas on the basis of the time change in correction pressure.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a gas leak detection device and a gas leak detection method for detecting leakage of insulating gas sealed in gas-insulated equipment or the like. [Background technology]

[0002] Electric power facilities such as substations and power plants are equipped with gas-insulated equipment such as gas insulated switchgear (GIS), gas-insulated transformers, and gas-insulated transmission lines. Gas-insulated equipment is constructed with conductors housed in a pressure vessel. The pressure vessel is filled with insulating gas to ensure electrical insulation and current interruption. Known insulating gases include sulfur hexafluoride (SF6) and carbon dioxide (CO2).

[0003] SF6 is a type of greenhouse gas with a high Global Warming Potential (GWP). The GWP of SF6 is 23,900. SF6 and CO2 have a negative impact on the environment when they leak from gas-insulated equipment, so their emission amounts are regulated. Electrical standards for GIS set the management standard for SF6 slow leaks at 0.5% per year. Because the pressure vessels of gas-insulated equipment have a large capacity, even small, slow leaks must be detected early.

[0004] A common device for detecting gas leaks is a gas density switch, which monitors the pressure and density of the gas. However, gas density switches have a low detection sensitivity, making it difficult to detect minute leaks, and it takes time for an alarm to sound to warn of a leak. By the time the alarm sounds, a considerable amount of insulating gas has already leaked, so improvements are needed from the perspective of environmental conservation.

[0005] Patent Document 1 describes a gas leak detection method using a gas pressure sensor. Gas pressure values ​​fluctuate significantly depending on the ambient temperature. Furthermore, when measured outdoors, they are affected by sunlight, temperature, weather, and other factors. Therefore, this detection method converts the measurement data of the gas pressure sensor into gas pressure at a predetermined reference temperature based on measurement data from a temperature sensor measured at the same time. The temperature sensor that minimizes the standard deviation of the conversion results is determined, and a gas leak is determined based on the change over time in the temperature-corrected gas pressure corrected for that measurement temperature.

[0006] Generally, there are various types of pressure sensors, such as metal strain gauge sensors and semiconductor pressure sensors that use the piezoresistance effect. Metal strain gauge sensors have high detection accuracy but the disadvantage of low measurement sensitivity. On the other hand, semiconductor pressure sensors that use the piezoresistance effect have high measurement sensitivity, with a gauge factor that is more than 50 times that of metal strain gauge sensors. Semiconductor pressure sensors that use the piezoresistance effect are used in many fields due to their high measurement sensitivity, high temperature resistance, and wide operating temperature range. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] Japanese Patent Application Laid-Open No. 2007-263584 Summary of the Invention [Problem to be solved by the invention]

[0008] Generally, gas-insulated equipment such as GIS is constructed with multiple pressure vessels connected together. Therefore, when monitoring for insulating gas leaks in gas-insulated equipment, it is desirable to measure the pressure of each pressure vessel. From the perspective of reducing equipment costs, an appropriate solution is to install only the minimum number of inexpensive sensors required. However, the commonly used metal strain gauge sensors are relatively expensive, which poses a cost issue.

[0009] On the other hand, semiconductor pressure sensors that use the piezoresistance effect are cheaper than metal strain gauge sensors, but they have issues with nonlinearity and a large temperature coefficient. In semiconductor pressure sensors that use the piezoresistance effect, the relationship between stress and resistance exhibits temperature-dependent nonlinearity. They may also exhibit temperature-dependent zero-point shifts. Because the magnitude of the strain that generates electrical resistance is strongly affected by temperature, there are problems with nonlinearity and temperature errors.

[0010] Generally, the pressure of an insulating gas measured by a pressure sensor is converted to pressure at a reference temperature, such as 20°C, as in Patent Document 1. Converting to pressure at a reference temperature reduces the effects of sunlight, weather, Joule heating of the conductor, and the like. However, even with this processing, semiconductor pressure sensors that utilize the piezoresistance effect have large nonlinearity and temperature coefficients that can cause errors. Because nonlinearity and temperature errors cannot be eliminated, it is difficult to detect small, slow leaks at an early stage.

[0011] Therefore, an object of the present invention is to provide a gas leak detection device and a gas leak detection method that can detect leaks of insulating gas sealed in a pressure vessel early and with high accuracy using a low-cost pressure sensor. [Means for solving the problem]

[0012] In order to solve the above problems, a gas leak detection device according to the present invention is a gas leak detection device that detects leakage of an insulating gas in a pressure vessel filled with the insulating gas, and includes: a pressure sensor that measures the pressure inside the pressure vessel; a temperature sensor that measures the temperature on the surface or inside the pressure vessel; a recording unit that records the pressure measurements measured by the pressure sensor and the temperature measurements measured by the temperature sensor; a calculation unit that calculates a corrected pressure by correcting the pressure measured by the pressure sensor with a correction function; and a testing unit that tests for the presence or absence of a leak of the insulating gas based on a change in the corrected pressure over time, wherein the correction function is expressed as the product P of a power of the measured pressure P and a power of a temperature ratio T / Ts of the measured temperature T and the extreme temperature Ts, where P is the measured pressure of the insulating gas, T is the measured temperature of the insulating gas, and Ts is an extreme temperature that is a minimum or maximum value of the measured temperature measured during a predetermined period before the leak test m (T / Ts) n (where m and n are numbers equal to or greater than 0) as variables.

[0013] Further, a gas leak detection method according to the present invention is a gas leak detection method for detecting leakage of an insulating gas in a pressure vessel filled with the insulating gas, and includes the steps of: measuring a pressure inside the pressure vessel and a temperature on or inside the pressure vessel; correcting the pressure measured by the pressure sensor with a correction function to obtain a corrected pressure; and verifying the presence or absence of leakage of the insulating gas based on a change over time in the corrected pressure, wherein the correction function is a product P of a power of the measured pressure P and a power of a temperature ratio T / Ts of the measured temperature T and the extreme temperature Ts, where P is a measured pressure measured as the pressure of the insulating gas, T is a measured temperature measured as the temperature of the insulating gas, and Ts is an extreme temperature which is a minimum or maximum value of the measured temperature measured during a predetermined period before the leak verification. m (T / Ts) n (where m and n are numbers equal to or greater than 0) as variables. [Effects of the Invention]

[0014] According to the present invention, it is possible to provide a gas leak detection device and a gas leak detection method that can detect leakage of insulating gas sealed in a pressure vessel early and with high accuracy using a low-cost pressure sensor. [Brief explanation of the drawings]

[0015] [Figure 1] 1 is a diagram showing a schematic configuration of a gas leak detection device according to an embodiment of the present invention. [Figure 2] FIG. 1 is a diagram showing the relationship between temperature and pressure change in a semiconductor pressure sensor utilizing the piezoresistance effect. [Figure 3] FIG. 10 is a diagram showing a zero point shift of a pressure sensor. [Figure 4] 1 is a flowchart showing the flow of a gas leak detection method according to a first embodiment of the present invention. [Figure 5] 10A and 10B are diagrams illustrating a process of converting an initial pressure for each gas amount at a reference temperature into a sample pressure for each sample temperature. [Figure 6] 10A and 10B are diagrams illustrating a process of converting a measurement pressure measured in a real field into a converted corrected pressure. [Figure 7] 3 is a diagram showing a process of inspecting the presence or absence of leakage of insulating gas in the gas leak detection method according to the first embodiment. FIG. [Figure 8] 10 is a flowchart showing the flow of a gas leak detection method according to a second embodiment of the present invention. [Figure 9] 10 is a diagram showing a process of inspecting the presence or absence of leakage of insulating gas in the gas leak detection method according to the second embodiment. FIG. DETAILED DESCRIPTION OF THE INVENTION

[0016] A gas leak detection device and a gas leak detection method according to an embodiment of the present invention will be described below with reference to the drawings. In the following drawings, common components are designated by the same reference numerals and duplicated descriptions will be omitted.

[0017] FIG. 1 is a diagram showing a schematic configuration of a gas leak detection device according to an embodiment of the present invention. 1, the gas leak detection device 100 is provided in association with a pressure vessel 1 of gas-insulated equipment. The gas leak detection device 100 includes a pressure sensor 2, a temperature sensor 3, an A / D converter 4, a recording unit 5, a processing unit 6, and a display unit 7.

[0018] The gas leak detection device 100 is a device that detects leakage of insulating gas in a pressure vessel 1 of gas-insulated equipment. The gas leak detection device 100 corrects the pressure measured by the pressure sensor 2 using a correction function to obtain a corrected pressure, and verifies the presence or absence of an insulating gas leak based on the change over time in the corrected pressure.

[0019] In the gas leak detection device 100, by using a predetermined correction function, nonlinear errors caused by the nonlinearity of the temperature dependence of the pressure sensor 2 and temperature errors caused by the zero point shift of the temperature dependence of the pressure sensor 2 are reduced, thereby performing highly accurate detection.

[0020] Gas-insulated equipment includes a sealed pressure vessel 1. The pressure vessel 1 may be provided in a structure in which multiple vessels are connected together, or in a structure in which the vessel is partitioned into multiple compartments. Examples of gas-insulated equipment include gas-insulated switchgear (GIS), gas-insulated transformers, and gas-insulated power lines. The pressure vessel 1 of the gas-insulated equipment contains conductors such as bus bars, and electrical equipment such as circuit breakers, disconnecting switches, earthing switches, current transformers, and voltage transformers.

[0021] An insulating gas is sealed inside the pressure vessel 1 to achieve a predetermined initial pressure. Examples of insulating gases include sulfur hexafluoride (SF6). By sealing the insulating gas inside the pressure vessel 1, electrical insulation and current interruption properties are ensured for the conductors and electrical equipment housed in the pressure vessel 1.

[0022] A pressure sensor 2 and a temperature sensor 3 are installed in the pressure vessel 1. The pressure sensor 2 and the temperature sensor 3 are connected to an A / D converter 4 via a communication line 8. The communication line 8 may be configured as a wired or wireless communication line. The A / D converter 4, recording unit 5, processing unit 6, and display unit 7 are connected in this order via signal lines.

[0023] The pressure sensor 2 measures the pressure inside the pressure vessel 1 as the pressure of the insulating gas. The pressure sensor 2 can be installed in the pressure vessel 1 via piping and a valve. The pressure inside the pressure vessel 1 can be considered as the pressure of the insulating gas sealed in the pressure vessel 1. If the pressure vessel 1 is provided in a connected structure or a separated chamber structure, the pressure sensor 2 can be installed in each space to measure the pressure in each space.

[0024] As the pressure sensor 2, any appropriate sensor can be used, such as a metal strain gauge sensor, a semiconductor pressure sensor utilizing the piezoresistance effect, a gas density pressure sensor, a capacitance pressure sensor, etc. From the viewpoint of equipment cost and the effectiveness of correction using a predetermined correction function, a pressure sensor that exhibits temperature-dependent nonlinearity or zero-point shift is preferred as the pressure sensor 2, and a semiconductor pressure sensor utilizing the piezoresistance effect is particularly preferred.

[0025] The temperature sensor 3 measures the surface or internal temperature of the pressure vessel 1 as the temperature of the insulating gas. The temperature sensor 3 is installed on the outer surface of the pressure vessel 1 or inside the pressure vessel 1. The surface temperature or internal temperature of the pressure vessel 1 can be considered to be the temperature of the insulating gas sealed in the pressure vessel 1. If the pressure vessel 1 has a connected structure or a divided chamber structure, the temperature sensor 3 can be installed in each space to measure the temperature of each space.

[0026] As the temperature sensor 3, for example, a suitable sensor such as a resistance temperature detector or a thermocouple can be used. As the resistance temperature detector, a suitable sensor such as a platinum-based or nickel-based sensor can be used as long as the measurement temperature range is appropriate. The temperature sensor 3 can also be installed integrally with the pressure sensor 2.

[0027] When measuring the temperature on the surface of the pressure vessel 1, the temperature sensor 3 is preferably placed on the surface of the lower half of the pressure vessel 1, and more preferably placed below the bottom of the pressure vessel 1. It is also preferable to perform the measurement at a specific time during the night. When measuring the temperature inside the pressure vessel 1, it is preferably placed in the lower half of the interior of the pressure vessel 1, and more preferably placed at the bottom inside the pressure vessel 1. In this configuration, the temperature is less susceptible to the effects of sunlight, weather, temperature, etc., and the error between the measured value and the actual gas temperature can be reduced.

[0028] The A / D converter 4 receives the analog signal of the pressure measured by the pressure sensor 2 and the analog signal of the temperature measured by the temperature sensor 3 and converts these analog signals into digital signals. Measurements by the pressure sensor 2 and the temperature sensor 3 are made periodically at any time interval when monitoring the pressure vessel 1 for leaks of insulating gas.

[0029] The recording unit 5 records the pressure measurement values ​​measured by the pressure sensor 2 and the temperature measurement values ​​measured by the temperature sensor 3. The recording unit 5 is composed of a storage device such as an internal memory or a hard disk. The measurement data of the pressure measured by the pressure sensor 2 and the measurement data of the temperature measured by the temperature sensor 3 are recorded periodically at any time interval when monitoring for insulating gas leaks in the pressure vessel 1.

[0030] The processing unit 6 performs calculations based on the pressure measurement data measured by the pressure sensor 2 and the temperature measurement data measured by the temperature sensor 3. The processing unit 6 is composed of a CPU (Central Processing Unit), RAM (Random Access Memory), ROM (Read Only Memory), a timer, an input unit, an output unit, etc.

[0031] The processing unit 6 constitutes an analysis unit that performs fitting using multivariate regression analysis to determine a correction function. During the learning period, the processing unit 6 performs fitting using a predetermined matrix equation to derive the correction function based on data on the pressure of each insulating gas amount. Then, the data resulting from the fitting is output to the recording unit 5 for recording so that the correction function can be used during the diagnosis period.

[0032] The processing unit 6 also includes a calculation unit that corrects the pressure measured by the pressure sensor 2 using a correction function to obtain a corrected pressure, and a testing unit that tests for the presence or absence of an insulating gas leak based on the change in the corrected pressure over time. The processing unit 6 corrects the pressure data measured by the pressure sensor 2 using the correction function during the diagnosis period to generate corrected pressure data. The processing unit 6 then compares the data on the change in the corrected pressure over time with a predetermined threshold value to test for the presence or absence of an insulating gas leak. The processing unit 6 outputs the test result data to the display unit 7.

[0033] The display unit 7 displays the test result indicating whether or not there is a leak of insulating gas in the pressure vessel 1. The display unit 7 is configured by, for example, a liquid crystal display, an organic EL display, a cathode ray tube, etc. The display unit 7 displays the test result indicating whether or not there is a leak of insulating gas as an image together with an alarm or the like to warn of the leak.

[0034] The display of the test results can include, for example, a display showing the time when the insulating gas leak occurred, a display showing the rate of leakage of the insulating gas, a regression line, a graph, or a table showing the relationship between the pressure measured by the pressure sensor 2 and time, or a graph or a table showing the relationship between the temperature measured by the temperature sensor 3 and time. The test results can be displayed for each pressure vessel 1 of the gas-insulated equipment.

[0035] FIG. 2 is a diagram showing the relationship between temperature and pressure change in a semiconductor pressure sensor that utilizes the piezoresistance effect. Semiconductor pressure sensors that use the piezoresistance effect exhibit nonlinearity that depends on temperature and pressure, as shown in Figure 2. In this type of pressure sensor, when strain due to pressure occurs, the electrical resistance of the semiconductor changes, and pressure is measured based on the change in output.

[0036] However, the electrical resistance of a piezoresistance element changes nonlinearly depending on the amount of strain. Furthermore, the stress of a piezoresistance element changes nonlinearly depending on temperature. Therefore, pressure measurements taken by semiconductor pressure sensors that utilize the piezoresistance effect have a temperature-dependent nonlinear error. The temperature-dependent nonlinear error increases as the pressure applied to the sensor diaphragm increases, along with the amount of strain.

[0037] FIG. 3 is a diagram showing the zero point shift of the pressure sensor. As shown in Figure 3, metal strain gauge sensors and semiconductor pressure sensors that utilize the piezoresistance effect can experience temperature-dependent zero point shift. Zero point shift is a phenomenon in which the zero point of the measured value shifts in a certain direction as measurements are repeated. The lower the temperature of the pressure sensor 2, the smaller the zero point of the pressure measurement value tends to become.

[0038] Such nonlinearity and zero-point shift of pressure sensors make it difficult to accurately measure the pressure of insulating gas when detecting leaks. Because gas-insulated equipment is often installed outdoors, the temperature of the insulating gas sealed in the pressure vessel is easily affected by factors such as sunlight, air temperature, and weather. The pressure of the insulating gas is significantly affected by these temperature fluctuations. Pressure sensors also have a detection limit, making them unable to measure minute pressures.

[0039] When there is a large nonlinear error due to the nonlinearity of the temperature dependency or a large temperature error due to a zero-point shift in the temperature dependency, accurate pressure measurement is not possible, making it difficult to detect slow leaks and other problems early on. Pressure changes cannot be detected correctly, particularly near the detection limit. If a small, slow leak continues, the insulating gas will continue to leak undetected, leading to the release of a large amount of insulating gas.

[0040] Therefore, in the gas leak detection device and gas leak detection method according to this embodiment, the pressure measured by the pressure sensor 2 is corrected using a predetermined correction function that includes as a variable the temperature ratio T / Ts between the measured temperature T, measured as the temperature of the insulating gas, and the extreme temperature Ts, to obtain a corrected pressure Ra(P,T).The presence or absence of an insulating gas leak is then determined based on the change over time in the corrected pressure Ra(P,T).

[0041] The correction function for correcting the pressure measured by the pressure sensor 2 can be generalized and expressed by a matrix formula. The matrix formula for deriving the correction function is expressed by the following formula (1).

[0042]

number

[0043] [In formula (1), Ra(P,T) is the corrected pressure corrected for temperature dependency, P is the measured pressure measured as the pressure of the insulating gas, T is the measured temperature measured as the temperature of the insulating gas, Ts is the extreme temperature, which is the minimum or maximum value of the measured temperature measured during a specified period before the leak test, K 11 ~K mn are n × m independent correction coefficients, and m and n are numbers greater than or equal to 0 that can be used to calculate the matrix equation.

[0044] The measured pressure P is the pressure of the insulating gas measured by a pressure sensor 2 provided in the pressure vessel 1 of the gas-insulated equipment, or a pressure sensor that measures the pressure inside a test vessel that simulates the pressure vessel 1.

[0045] The measured temperature T is the temperature measured as the temperature of the insulating gas by a temperature sensor 3 provided in the pressure vessel 1 of the gas-insulated equipment, or a temperature sensor that measures the temperature on the surface or inside of a test vessel simulating the pressure vessel 1.

[0046] The extreme temperature Ts refers to the extreme value during a predetermined period of the measured temperatures T measured during that period prior to the leak detection, i.e., the minimum or maximum value during that period. The extreme temperature Ts can be determined from the measured temperatures T measured during the predetermined period involving the most recent temperature fluctuation when correcting the pressure measured by the pressure sensor 2. Any appropriate period can be used as the predetermined period for determining the extreme temperature Ts, depending on factors such as the leak detection cycle, as long as it includes the reference minimum or maximum value.

[0047] As the extreme temperature Ts, it is particularly preferable to use the minimum or maximum value of the measured temperature T measured within the 24 hours immediately prior to the leak test. Normally, the temperature of the insulating gas sealed in the pressure vessel of gas-insulated equipment changes in a 24-hour cycle due to the influence of sunlight, etc. The temperature of the insulating gas tends to be highest during the day and lowest at night. Therefore, using 24 hours as the specified period allows the measured temperature T to be properly normalized when monitoring leaks in units of time less than 24 hours.

[0048] As the extreme temperature Ts, it is more preferable to use the minimum value of the measured temperature T than the maximum value of the measured temperature T. Using the minimum value increases the sensitivity of the temperature ratio T / Ts included in the matrix equation expressed by formula (1). Therefore, a correction function that enables more accurate correction can be derived from the matrix equation expressed by formula (1).

[0049] If multiple minimum or maximum values ​​occur within a predetermined period, the extreme temperature Ts may be an extreme value that occurs with relatively high periodicity or the same extreme value as the most recently used extreme value.If no minimum or maximum value is included within the predetermined period, the extreme temperature Ts may be the same extreme value as the most recently used extreme value.

[0050] Correction coefficient K(K 11 ~K mn ) is a correction coefficient that corrects for the influence of factors within the system of the gas leak detection device 100. The correction coefficient K indicates an arbitrary number. The factors within the system may be unknown phenomena or known phenomena. Factors within the system include influence factors derived from the sensor. By using the correction coefficient K, it is possible to calibrate errors due to factors within the system in addition to nonlinear errors and temperature errors of the pressure sensor 2.

[0051] The matrix equation expressed by Equation (1) can be expanded into a polynomial correction function. The correction function expanded from the matrix equation expressed by Equation (1) is the product P of the power of the measured pressure P and the power of the temperature ratio T / Ts of the measured temperature T and the extreme temperature Ts, where P is the measured pressure of the insulating gas, T is the measured temperature of the insulating gas, and Ts is the extreme temperature, which is the minimum or maximum value of the measured temperature measured during a predetermined period before the leak test. m (T / Ts) n (where m and n are numbers equal to or greater than 0) as variables.

[0052] The correction function, which is an expansion of the matrix equation expressed by formula (1), is used to correct for the influence of factors within the device system. It is calculated by multiplying the measured pressure P by a power, the temperature ratio T / Ts by a power, and the correction coefficient K (K 11 ~K mn ) The matrix of the correction coefficient K may be input as a unit matrix or the like depending on the required detection accuracy of the insulating gas leak.

[0053] The matrix equation expressed by Equation (1) can be fitted by multivariate regression analysis based on data collected before leak testing. Multivariate regression analysis involves the use of powers of the measured pressure P, powers of the temperature ratio T / Ts, and a correction coefficient K (K 11 ~K mn ) is used as an explanatory variable, and the sample pressure Ra(P,T) corresponding to the corrected pressure Ra(P,T) is used as a response variable.

[0054] By fitting using multivariate regression analysis, the power exponent m of the measured pressure P, the power exponent n of the temperature ratio T / Ts, and the correction coefficient K 11 ~K mn can be obtained for each pressure sensor 2. The correction function obtained by expanding the matrix equation expressed by the fitted formula (1) is used to correct the pressure measured by the pressure sensor 2 when monitoring for leakage of insulating gas.

[0055] Next, a gas leak detection method performed using the gas leak detection device according to this embodiment will be described in more detail.

[0056] First Embodiment FIG. 4 is a flowchart showing the flow of the gas leak detection method according to the first embodiment of the present invention. As shown in Fig. 4, the matrix equation expressed by Equation (1) can be fitted by multivariate regression analysis based on data collected in a preliminary test conducted before leak monitoring. The gas leak detection method shown in Fig. 4 consists of a learning period (steps S1 to S4) and a diagnosis period (steps S5 to S8).

[0057] The learning period (steps S1 to S4) is carried out in a preliminary test simulating an insulating gas leak using a test vessel simulating the pressure vessel 1 of the gas-insulated equipment. During the learning period (steps S1 to S4), pressure data for known amounts of insulating gas are collected and fitted to the matrix equation expressed by Equation (1).

[0058] The diagnostic period (steps S5 to S8) is carried out when monitoring the pressure vessel 1 of the gas-insulated equipment in the actual field for leakage of insulating gas. During the diagnostic period (steps S5 to S8), a correction function obtained by expanding the matrix operation formula expressed by the fitted formula (1) is used to detect leakage of insulating gas sealed in the pressure vessel 1 of the gas-insulated equipment.

[0059] The preliminary test is conducted by simulating an insulating gas leak in the field and sealing a predetermined amount of insulating gas in a test vessel. The test vessel may be the same pressure vessel as that used for the gas-insulated equipment, or a simulated pressure vessel. The preliminary test can be conducted at any temperature, but it is preferable to conduct it at a standard temperature such as 20°C.

[0060] As a temperature sensor for measuring the temperature of the insulating gas sealed in the test vessel, any appropriate sensor such as a resistance temperature detector, a thermocouple, etc. As a pressure sensor for measuring the pressure of the insulating gas sealed in the test vessel, any appropriate sensor can be used, but it is preferable to use a standard device with high measurement accuracy.

[0061] (Step S1) First, in a preliminary test, the temperature and pressure of an insulating gas at a given gas amount sealed in a test vessel are measured. The temperature and pressure measurements are performed by adjusting the amount of insulating gas sealed in the test vessel to different gas amounts. By measuring the pressure for each gas amount, multiple data points for the initial pressure Pa for each gas amount at a predetermined reference temperature Ta are obtained.

[0062] In step S1, the amount of insulating gas is adjusted to simulate an insulating gas leak at a predetermined leak rate [% / t]. The amount of insulating gas is adjusted to a range above the detection limit. It is preferable to adjust the amount of insulating gas to the target detection amount [% / t] that is the target for detection. By adjusting the gas amount in this way, data that can be used to handle a small, slow leak with a slow leak rate can be collected.

[0063] An example of the reference temperature Ta is 20°C, which corresponds to room temperature. The pressure of the insulating gas depends on the temperature. By obtaining data on the initial pressure Pa for each gas amount at the reference temperature Ta, pressure errors due to the influence of temperature can be eliminated. By collecting data on the reference initial pressure Pa for each gas amount, it becomes possible to fit the matrix equation expressed by Equation (1) more accurately.

[0064] Data on the initial pressure Pa for each gas amount at the reference temperature Ta may be collected by adjusting the insulating gas to the reference temperature Ta, or may be collected by adjusting the insulating gas to any temperature other than the reference temperature Ta. The initial pressure Pa for each gas amount at the reference temperature Ta can be converted using the gas state equation based on the pressure measurement at any temperature.

[0065] The equation of state for a gas can be, for example, the Beattie and Bridgman real gas equation, which is expressed as the following equation (2):

[0066]

number

[0067] [In formula (2), P is the pressure of the gas [Pa], V is the molar volume of the gas [m 3 / mol], R is the gas constant [8.31 J / mol K], T is the temperature of the gas [K], and A0, B0, a, b, and c are constants for each type of gas.

[0068] In formula (2), A and B are A0 = 1.578 and a = 0.1062 × 10 when the insulating gas is sulfur hexafluoride (SF6). -3 , B0=0.366×10 -3 , b=0.1236×10 -3 It can be calculated as follows. At room temperature, ε can be approximated to 0.

[0069] Temperature and pressure measurements for each gas amount can be performed in one test vessel or in multiple test vessels. Assuming a situation in which a slow leak continues at a constant rate, one test vessel can be prepared and data collection can be repeated while the gas amount is gradually increased to achieve a predetermined leak rate [% / t]. Alternatively, multiple test vessels with different gas amounts can be prepared and data can be collected for each vessel.

[0070] Temperature and pressure measurements are performed for multiple different gas amounts. The number of gas amount types and the number of measurements can be any number. However, the more measurements are performed with different gas amounts, the more data can be obtained, enabling more accurate fitting. For example, in the initial state corresponding to before a leak, an initial pressure Pa0 can be measured. Also, in the leak state corresponding to after a leak, initial pressures Pa1, Pa2, Pa3, etc. corresponding to different gas amounts can be measured.

[0071] (Step S2) Next, multiple data of the initial pressure Pa for each gas amount at a predetermined reference temperature Ta are converted into multiple data of sample pressure Ra(P, T) for each gas amount at different sample temperatures T measured by a temperature sensor, and based on the gas state equation.

[0072] In step S2, the data of the initial pressure Pa for each gas amount at the reference temperature Ta is converted into data of the sample pressure Ra(P,T) for each gas amount at different sample temperatures T, thereby creating a data set for fitting the matrix equation expressed by Equation (1). As the gas state equation, the Beatty-Bridgeman real gas state equation expressed by Equation (2) can be used.

[0073] 5 is a diagram illustrating the process of converting the initial pressure for each gas amount at the reference temperature into the sample pressure for each sample temperature. The left diagram shows the initial pressure Pa for each gas amount at the reference temperature Ta before conversion using the gas equation of state. The right diagram shows the sample pressure Ra(P, T) for each gas amount at the sample temperature T (T1, T2, T3) after conversion using the gas equation of state. As shown in FIG. 5, in step S2, the initial pressure Pa for each gas amount at the reference temperature Ta is converted into a sample pressure Ra(P, T) for each gas amount at an arbitrary sample temperature T, thereby creating a data set showing the pressure for each temperature and each gas amount.

[0074] The sample temperature T can be a temperature that is expected to be measured in the pressure vessel 1 of the gas-insulated equipment in the field. It is preferable to use a plurality of different temperatures as the sample temperatures T. The number of types of sample temperatures T can be any number. However, the more types of sample temperatures T there are, the more data can be obtained, enabling more accurate fitting.

[0075] For example, different sample temperatures T0, T1, T2, T3... within the temperature range expected to be measured in pressure vessel 1 of gas-insulated equipment in the field can be used to create data as the sample temperature T. The initial pressures Pa0, Pa1, Pa2, Pa3... for each gas volume at reference temperature Ta can be converted into sample pressures Ra0(Pa0, T0), Ra1(Pa1, T1), Ra2(Pa2, T2), Ra3(Pa3, T3)... at sample temperatures T0, T1, T2, T3... using the gas state equation.

[0076] (Step S3) Next, using multiple data of the initial pressure Pa for each gas amount at a predetermined reference temperature Ta and multiple data of the sample pressure Ra(P,T) for each gas amount at different sample temperatures T, a multivariate regression analysis is performed using the matrix equation represented by Equation (1) as a model. By fitting using the multivariate regression analysis, the power exponent m of the measured pressure P, the power exponent n of the temperature ratio T / Ts, and the correction coefficient K(K 11 ~Kmn ) is determined.

[0077] The multivariate regression analysis was performed using the power of the initial pressure Pa for each gas amount at the reference temperature Ta, the power of the temperature ratio T / Ts, and the correction coefficient K (K 11 ~K mn ) is used as an explanatory variable, and the sample pressure Ra(P,T) for each gas amount at the sample temperature T is used as the response variable. Fitting is performed using multivariate regression analysis to identify the matrix equation expressed by Equation (1).

[0078] The exponent m of the measured pressure P, the exponent n of the temperature ratio T / Ts, and the correction coefficient K in the matrix equation (1) are 11 ~K mn is the coefficient of determination R calculated in multivariate regression analysis 2 The coefficient of determination R 2 is the sample value y i ,y j , where the mean value of the sample values ​​is y-(bar) and the estimated value is y^(hat), it can be expressed as the following equation (3).

[0079]

number

[0080] Coefficient of determination R 2 The larger the coefficient of determination, R, the more the dependent variable is explained by the explanatory variables. 2 The combination that maximizes the coefficient of determination R 2 The combinations for which the coefficient of determination R is equal to or greater than the standard value can be adopted. 2 The reference value of is not particularly limited, but is preferably 0.95 or more, for example. Usually, the exponent m of the measurement pressure P and the exponent n of the temperature ratio T / Ts can be fitted to m=3 or less and n=3 or less.

[0081] (Step S4) Next, the data of the matrix equation represented by Equation (1) fitted by multivariate regression analysis is used to calculate the power exponent m of the measured pressure P, the power exponent n of the temperature ratio T / Ts, and the correction coefficient K 11 ~K mn The data is stored in the recording unit 5 of the gas leak detection device 100 together with the above data.

[0082] The fitted matrix equation (1) is the product P of the power of the measured pressure P and the power of the temperature ratio T / Ts between the measured temperature T and the extreme temperature Ts. m (T / Ts) n (where m and n are numbers equal to or greater than 0.) The data of the correction function may be stored in the recording unit 5.

[0083] (Step S5) When monitoring for insulating gas leaks in the field, the pressure sensor 2 measures the pressure inside the pressure vessel 1 at a temperature corresponding to the pressure of the insulating gas. The temperature sensor 3 measures the temperature on the surface or inside the pressure vessel 1 corresponding to the temperature of the insulating gas. The data on the measured pressure P and the measured temperature T are converted into digital signals and recorded in the recording unit 5.

[0084] Pressure measurement by the pressure sensor 2 and temperature measurement by the temperature sensor 3 are performed periodically at predetermined time intervals. By performing the periodic measurements, data on the measured temperature T for each measurement time and data on the measured pressure P for each measurement time at the measured temperature T are collected. The data on the measured temperature T and the data on the measured pressure P are associated with data on the measurement time and stored in the recording unit 5 as time-series data.

[0085] (Step S6) Next, each time the measurement temperature T and the measurement pressure P are measured, the processing unit 6 of the gas leak detection device 100 reads out the data of the measurement temperature T, the data of the measured pressure P for each measurement time at the measurement temperature T, and the data of the correction function obtained by expanding the matrix calculation formula expressed in formula (1), and corrects the measured pressure P measured by the pressure sensor 2 with the correction function obtained by expanding the matrix calculation formula expressed in formula (1) to obtain the corrected pressure Ra'(P, T).

[0086] In step S6, the measured pressure P for each measurement time at the latest measurement temperature T measured by the pressure sensor 2 is input, and a calculation corresponding to the matrix calculation formula expressed by the fitted formula (1) is performed. By this calculation, the corrected pressure Ra'(P, T) for each measurement time at the measurement temperature T is obtained.

[0087] (Step S7) Next, the processing unit 6 converts the data of the corrected pressure Ra'(P,T) for each measurement time at the measurement temperature T into a converted corrected pressure Ra' for each measurement time at a predetermined reference temperature Ta based on the gas state equation. Ta Convert to (P,T) data. Conversion corrected pressure Ra' Ta The (P, T) data is associated with the data of the measurement time and stored in the recording unit 5 as time-series data.

[0088] In step S7, the data of the corrected pressure Ra'(P,T) for each measurement time at the measurement temperature T is converted into the converted corrected pressure Ra' for each measurement time at a predetermined reference temperature Ta. Ta By converting the data into (P,T), the pressure error due to the influence of temperature can be eliminated. The Beatty-Bridgeman real gas equation of state, expressed as equation (2), can be used as the gas equation.

[0089] The reference temperature Ta may be, for example, 20° C., which corresponds to room temperature. However, the reference temperature Ta may be the same or different in steps S1, S2, and S7.

[0090] 6 is a diagram illustrating the process of converting the measured pressure measured in the actual field into a converted corrected pressure. The left diagram shows the measured pressure P for each measurement time at the measurement temperature T before conversion using the correction function and the gas state equation. The right diagram shows the converted corrected pressure Ra' for each measurement time at the reference temperature Ta after conversion using the correction function and the gas state equation. Ta FIG. 10 is a diagram showing (P, T). As shown in FIG. 6, in steps S6 and S7, the measured pressure P at each measurement time at the measurement temperature T is converted into a converted corrected pressure Ra' at each measurement time at the reference temperature Ta. Ta By converting it into (P, T), data calibrated by the temperature ratio T / Ts between the measurement temperature T and the extreme temperature Ts can be created.

[0091] As shown in the left diagram of Figure 6, if the measured pressure P of the insulating gas is not corrected by the correction function obtained by expanding the matrix equation expressed by Equation (1), it is difficult to determine whether the decrease in pressure inside the pressure vessel 1 is due to an increase in the amount of leakage or a decrease in temperature. If the change in pressure at each measurement time is within the range caused by temperature fluctuations, the change in pressure will not be detected as an insulating gas leak.

[0092] In contrast, as shown in the right diagram of Figure 6, when correction is made using a correction function that expands the matrix operation formula expressed in equation (1), the conversion is based on the temperature ratio T / Ts between the measured temperature T and the extreme temperature Ts, so the change in pressure due to temperature fluctuations becomes smaller. This makes it easier to distinguish pressure changes due to an increase in the amount of leakage, enabling highly accurate detection of insulating gas leaks.

[0093] (Step S8) Next, the processing unit 6 calculates the converted corrected pressure Ra' for each measurement time at a predetermined reference temperature Ta. Ta The time change rate of (P, T) is calculated, and the converted corrected pressure Ra' is calculated for each measurement time at the reference temperature Ta. Ta Based on the time rate of change of (P, T), the presence or absence of a leak of the insulating gas sealed in the pressure vessel 1 is verified.

[0094] The test for the presence or absence of an insulating gas leak is carried out by measuring the converted corrected pressure Ra' at the reference temperature Ta for each measurement time. Ta This can be done by determining whether the rate of change of (P, T) over time is equal to or greater than a predetermined threshold value.

[0095] Conversion corrected pressure Ra' Ta The time rate of change of (P, T) can be calculated for a suitable time range. For example, it can be calculated for the time range from the start of leak monitoring to the current measurement when the latest pressure is measured, or for the time range from the previous measurement to the current measurement. The processing unit 6 calculates the converted corrected pressure Ra' for each measurement time calculated for a predetermined time range. Ta A plurality of data points (P, T) are read out, and the rate of change over time within the time range is calculated.

[0096] FIG. 7 is a diagram showing a process of inspecting the presence or absence of leakage of insulating gas in the gas leak detection method according to the first embodiment. As shown in FIG. 7, the converted corrected pressure Ra' at each measurement time at a predetermined reference temperature Ta Ta The relationship between (P, T) and time can be linearly approximated using the least squares method.

[0097] Conversion corrected pressure Ra' for each measurement time at reference temperature Ta Ta (P, T) is the elapsed time calculated from the start of leak monitoring, t, and the converted corrected pressure, Ra' Ta The time rate of change of (P,T) is Ra' Ta (P,T) / dt=α, converted correction pressure Ra' Ta When the initial value of (P, T) is β, it can be expressed by the following equation (4). Ra' Ta (P,T)=αt+β (4)

[0098] In step S7, the processing unit 6 calculates the converted corrected pressure Ra' Ta Calculate the time change rate α of (P, T) and calculate the converted corrected pressure Ra' Ta The time change rate α of (P, T) is compared with a preset threshold value to obtain the converted corrected pressure Ra'Ta It is possible to determine whether the time change rate α of (P, T) is equal to or greater than a predetermined threshold value that has been set in advance.

[0099] The threshold value can be, for example, a numerical value based on the standard deviation σ. The pressure measurement value measured by the pressure sensor 2 varies due to the influence of sunlight, temperature, weather, etc. If it is assumed that the variation in the pressure measurement value is a normal distribution, it can be determined that there is an insulating gas leak when the variation deviates from a predetermined fluctuation range based on the standard deviation σ.

[0100] For example, the ±3σ range includes approximately 99.73% of the data. When high-precision detection of insulating gas leaks is required, the converted corrected pressure Ra' Ta It can be determined whether the time rate of change α of (P, T) satisfies the following formula (5). |α|>6σ / t (5)

[0101] In step S7, the converted corrected pressure Ra' Ta When the time rate of change α of (P, T) is equal to or greater than a predetermined threshold, it can be determined that there is an insulating gas leak. If a leak is determined to exist, the detection result indicating the existence of a leak is displayed, and an alarm is issued to warn of the leak.

[0102] On the other hand, in step S7, the converted corrected pressure Ra' Ta When the time rate of change α of (P, T) is less than a predetermined threshold, it can be determined that there is no leakage of insulating gas. If it is determined that there is no leakage, the temperature and pressure continue to be measured and the leakage of insulating gas continues to be monitored.

[0103] In this gas leak detection device and gas leak detection method, the measured pressure P measured by the pressure sensor 2 is corrected using a correction function derived from the matrix equation expressed by Equation (1). Therefore, the measured pressure P measured by the pressure sensor 2 is calibrated using the temperature ratio T / Ts between the measured temperature T and the extreme temperature Ts. This reduces nonlinear errors due to temperature-dependent nonlinearity and temperature errors due to temperature-dependent zero-point shifts. Because calibration is performed based on temperature changes relative to the extreme temperature Ts, nonlinear errors can be canceled out between data. Furthermore, using the measured temperature T during a predetermined period immediately prior to the leak detection as the extreme temperature Ts reduces the influence of accumulated temperature-dependent factors, thereby eliminating errors due to zero-point shifts. Therefore, even when using a semiconductor pressure sensor that generates nonlinear errors or a pressure sensor that generates temperature errors, it is possible to detect slow, minute leaks. This enables early, highly accurate detection of insulating gas leaks sealed in a pressure vessel using a low-cost pressure sensor.

[0104] Second Embodiment FIG. 8 is a flowchart showing the flow of a gas leak detection method according to the second embodiment of the present invention. As shown in Fig. 8, the matrix equation expressed by Equation (1) can also be fitted by multivariate regression analysis based on data collected during leak monitoring, instead of data collected in a preliminary test performed before leak monitoring. The gas leak detection method shown in Fig. 8 consists of a learning period (steps S11 to S14) and a diagnosis period (steps S15 to S18).

[0105] The learning period (steps S11 to S14) is performed while monitoring for insulating gas leaks in the pressure vessel 1 of the gas-insulated equipment in the field. During the learning period (steps S11 to S14), pressure data at the leak rate of the target detection amount [% / t] is collected, and fitting is performed to the matrix calculation formula expressed by formula (1).

[0106] The diagnostic period (steps S15 to S18) is performed when monitoring for insulating gas leaks from the pressure vessel 1 of the gas-insulated equipment in the actual field. During the diagnostic period (steps S15 to S18), a correction function obtained by expanding the matrix operation formula expressed by the fitted formula (1) is used to detect leaks of the insulating gas sealed in the pressure vessel 1 of the gas-insulated equipment.

[0107] (Step S11) First, in the actual field, the pressure sensor 2 measures the pressure inside the pressure vessel 1 at a temperature corresponding to the pressure of the insulating gas. The temperature sensor 3 measures the temperature on the surface or inside the pressure vessel 1 corresponding to the temperature of the insulating gas. The data on the measured pressure P and the measured temperature T are converted into digital signals and recorded in the recording unit 5.

[0108] The pressure measurement by the pressure sensor 2 and the temperature measurement by the temperature sensor 3 are repeated over time. By repeating the measurements, multiple data sets of the measured temperature T for each measurement time and multiple data sets of the measured pressure P for each measurement time at the measured temperature T are collected. The data sets of the measured temperature T and the measured pressure P are associated with the data sets of the measurement times and stored in the recording unit 5 as time-series data.

[0109] In step S11, assuming that there is no insulating gas leak from the pressure vessel 1, a change in pressure corresponding to a minute leak that is equal to or less than the target detection amount [% / t] is measured over time over a predetermined learning period. The learning period may be, for example, the entire period from the start of leak monitoring to the current measurement when the most recent pressure is measured, or a part of the period from the start of leak monitoring to the current measurement.

[0110] For example, in an actual pressure vessel 1 in the field, in the initial state corresponding to before a leak, it is possible to measure the measurement temperature T0 and the initial pressure P0 at the measurement temperature T0, assuming that there is no leakage of insulating gas. Also, after time has passed with the leakage falling below the target detection amount [% / t], it is possible to measure the measurement temperatures T1, T2, T3... and the measurement pressures P1, P2, P3... at the measurement temperatures T1, T2, T3...

[0111] (Step S12) Next, after a predetermined learning period has elapsed, the processing unit 6 converts multiple data of measured pressure P for each measurement time at different measurement temperatures T into multiple data of sample pressure Rb(P, T) for each measurement time at a predetermined reference temperature Tb based on the gas state equation.

[0112] In step S12, the data of the measured pressure P at each measurement time at the measurement temperature T is converted into data of the sample pressure Rb(P,T) at each measurement time at a predetermined reference temperature Tb, thereby creating a data set for fitting the matrix equation expressed by Equation (1). As the equation of state of the gas, the Beatty-Bridgman equation of state for real gases expressed by Equation (2) can be used.

[0113] For example, the measured pressure P (P0, P1, P2, P3...) at each measurement time at the measurement temperature T (T0, T1, T2, T3...) measured assuming no leakage of insulating gas can be converted into the sample pressure Rb(P,T) at each measurement time at the reference temperature Tb (Rb0(P0, T0), Rb1(P1, T1), Rb2(P2, T2), Rb3(P3, T3)...) using the gas equation of state.

[0114] (Step S13) Next, using multiple data of the measured pressure P at each measurement time at different measurement temperatures T and multiple data of the sample pressure Rb(P,T) at each measurement time at a predetermined reference temperature Tb, a multivariate regression analysis is performed using a matrix equation represented by Equation (1) as a model. By fitting using the multivariate regression analysis, the power exponent m of the measured pressure P, the power exponent n of the temperature ratio T / Ts, and the correction coefficient K(K 11 ~K mn ) is determined.

[0115] The multivariate regression analysis is performed using the power of the measured pressure P at each measurement time at the measurement temperature T, the power of the temperature ratio T / Ts, and the correction coefficient K (K 11 ~K mn ) is used as an explanatory variable, and the sample pressure Rb(P,T) for each measurement time at the reference temperature Tb is used as the objective variable. Fitting is performed using multivariate regression analysis, and the matrix equation expressed by Equation (1) is identified, as in step S3.

[0116] (Steps S14 to S16) Next, similarly to steps S4 to S6, data of the matrix equation represented by equation (1) fitted by multivariate regression analysis is saved (step S14), the measured temperature T and measured pressure P in the actual field are measured (step S15), and correction is performed using a correction function obtained by expanding the matrix equation represented by equation (1) (step S16). In step S16, the corrected pressure Rb'(P, T) for each measurement time at the measured temperature T is calculated.

[0117] (Step S17) Next, the processing unit 6 converts the data of the corrected pressure Rb'(P, T) for each measurement time at the measurement temperature T into a converted corrected pressure Rb' for each measurement time at a predetermined reference temperature Tb based on the gas state equation. Tb Convert to (P,T) data. Corrected pressure Rb' Tb The (P, T) data is associated with the data of the measurement time and stored in the recording unit 5 as time-series data.

[0118] The reference temperature Tb may be, for example, 20° C., which corresponds to room temperature. However, the reference temperature Tn may be the same or different in steps S11, S12, and S17.

[0119] (Step S18) Next, the processing unit 6 calculates a converted corrected pressure Rb' for each measurement time at a predetermined reference temperature Tb. Tb The time change rate of (P, T) is calculated, and the converted corrected pressure Rb' is calculated for each measurement time at the reference temperature Tb. Tb Based on the time rate of change of (P, T), the presence or absence of a leak of the insulating gas sealed in the pressure vessel 1 is verified.

[0120] The test for the presence or absence of an insulating gas leak is carried out in the same manner as in step S8, by measuring the converted corrected pressure Rb' at the reference temperature Tb for each measurement time. Tb This can be done by determining whether the rate of change of (P, T) over time is equal to or greater than a predetermined threshold value.

[0121] FIG. 9 is a diagram showing a process of inspecting the presence or absence of leakage of insulating gas in the gas leak detection method according to the second embodiment. As shown in FIG. 9, the converted corrected pressure Rb' at each measurement time at a predetermined reference temperature Tb Tb The relationship between (P, T) and time can be linearly approximated. Assuming that there is no leakage of insulating gas during a certain measurement time t=0 to t1, the converted corrected pressure Rb' Tb The rate of change of (P, T) over time is calculated, and the leakage of insulating gas can be detected at a certain measurement time t=t1 to t2.

[0122] In step S13, only the data of the measured pressure P measured by the pressure sensor 2 is used in fitting the matrix calculation formula represented by equation (1), but in addition to the data of the measured pressure P measured by the pressure sensor 2, the data of the measured pressure P obtained by calculation may also be used.

[0123] For example, let us assume that there is no leakage of insulating gas during a certain measurement time t=0 to t1, that the pressure change ΔP of the insulating gas is constant during this time range, and that fluctuations exceeding the constant pressure change ΔP are the detection target. When pressure is measured periodically at a predetermined measurement interval Δt during measurement time t=0 to t1, if the initial pressure when there is no leakage of insulating gas is P0 and the detection target fluctuation amount per measurement is ΔP / Δt, the measured pressure P(q) measured in the qth measurement can be expressed by the following formula (6). P(q)=P0×(1-q×ΔP / Δt) (6)

[0124] The measured pressure P(q) measured in this q-th measurement can be calculated based on the measured pressure P measured by the pressure sensor 2. By calculating a portion of the data on the measured pressure P, it is possible to obtain a large amount of data on the sample pressure Rb(P, T) required for fitting without actually measuring the pressure multiple times using the pressure sensor 2. Even if it is difficult to secure a time range in which it can be assumed that there is no leakage of insulating gas, it is possible to collect appropriate data required for fitting.

[0125] Conversion corrected pressure Rb' for each measurement time at reference temperature Tb Tb (P, T) is the elapsed time calculated from the start of leak monitoring, t, and the converted corrected pressure, Rb' Tb The time rate of change of (P,T) is Rb' Tb (P,T) / dt=α', converted correction pressure Rb' Tb When the value of (P, T) at time t1 is β', it can be expressed by the following equation (7). Rb' Tb (P,T)=α'(t2-t1)+β' (7)

[0126] In step S17, the processing unit 6 calculates the converted corrected pressure Rb' in the same manner as in step S7. Tb The time rate of change α' of (P, T) is calculated and the converted corrected pressure Rb' is calculated. Tb The time change rate α' of (P, T) is compared with a preset threshold value to obtain the converted corrected pressure Rb' TbIt is possible to determine whether the time rate of change α′ of (P, T) is equal to or greater than a predetermined threshold value that has been set in advance.

[0127] For example, the ±3σ range includes approximately 99.73% of the data. When high-precision detection of insulating gas leaks is required, the converted corrected pressure Rb' Tb It can be determined whether the time rate of change α′ of (P, T) satisfies the following formula (8). |α'|-ΔP / t>6σ / t (8)

[0128] In step S17, the converted corrected pressure Rb' Tb When the time rate of change α' of (P, T) is equal to or greater than a predetermined threshold, it can be determined that there is an insulating gas leak. If a leak is determined to exist, the detection result indicating the existence of a leak is displayed, and an alarm is issued to warn of the leak.

[0129] On the other hand, in step S17, the converted corrected pressure Rb' Tb When the time rate of change α' of (P, T) is less than a predetermined threshold, it can be determined that there is no leakage of insulating gas. If it is determined that there is no leakage, the temperature and pressure continue to be measured and the leakage of insulating gas continues to be monitored.

[0130] According to this gas leak detection device and gas leak detection method, the matrix calculation formula expressed by formula (1) is corrected using a correction function expanded, so similar to the above-mentioned gas leak detection device and gas leak detection method, leaks of insulating gas sealed in a pressure vessel can be detected early and with high accuracy using a low-cost pressure sensor without conducting a preliminary test in advance. Since the matrix calculation formula expressed by formula (1) is fitted with data obtained in the actual field, more accurate detection is possible.

[0131] <Third embodiment> The matrix equation expressed by Equation (1) can be transformed into other forms depending on the type of pressure sensor 2 and the range of environmental temperature. If the pressure sensor 2 does not produce a temperature-dependent zero point shift, the matrix equation for deriving the correction function can also be replaced with the equation expressed by the following Equation (9).

[0132]

number

[0133] [In equation (9), Ra(P,T) is the corrected pressure corrected for temperature dependency, P is the measured pressure measured as the pressure of the insulating gas, T is the measured temperature measured as the temperature of the insulating gas, K 11 ~K mn are n × m independent correction coefficients, and m and n are numbers greater than or equal to 0 that can be used to calculate the matrix equation.

[0134] The matrix equation expressed by Equation (9) corresponds to the case where the extreme temperature Ts=1 in the matrix equation expressed by Equation (1) and the measured temperature T is not normalized. In the gas leak detection device and gas leak detection method described above, the extreme temperature Ts is a number other than 1, but if the pressure sensor 2 does not produce a temperature-dependent zero point shift, the measured temperature T does not need to be normalized.

[0135] When monitoring for insulating gas leaks from a pressure vessel 1 of gas-insulated equipment in the field, the matrix calculation formula expressed by formula (1) and the matrix calculation formula expressed by formula (9) can be used together depending on the type of pressure sensor 2 and the range of environmental temperature. For sensors with large temperature characteristics and large fluctuations in pressure measurements relative to temperature changes, it is preferable to use the matrix calculation formula expressed by formula (1), while for sensors with small temperature characteristics and small fluctuations in pressure measurements relative to temperature changes, it is preferable to use the matrix calculation formula expressed by formula (9).

[0136] Using the matrix equation expressed by Equation (9) makes it easier to fit the matrix equation. When the sensor diaphragm is made of a material with poor temperature characteristics, such as stainless steel, leaks of insulating gas sealed in a pressure vessel can be detected early and with high accuracy using the easily calculated correction function.

[0137] <Fourth embodiment> The matrix calculation formula expressed by Equation (1) can be transformed into other forms depending on the influence of factors outside the system of the gas leak detection device 100. When correcting the influence of factors outside the system of the device, the matrix calculation formula for deriving the correction function can also be replaced with the formula expressed by the following Equation (10).

[0138]

number

[0139] [In equation (10), Ra(P,T) is the corrected pressure corrected for temperature dependency, P is the measured pressure measured as the pressure of the insulating gas, T is the measured temperature measured as the temperature of the insulating gas, Ts is the extreme temperature, which is the minimum or maximum value of the measured temperature measured during a predetermined period before the leak detection, F is an input variable due to factors outside the device system, K 11 ~K mn are n × m independent correction coefficients, and l, m, and n are numbers greater than or equal to 0 that can be used to calculate the matrix equation.

[0140] The input variable F is a variable that corrects the influence of factors outside the system of the gas leak detection device 100. The factors outside the system may be unknown phenomena or known phenomena. Examples of factors outside the system include mechanical vibrations and electromagnetic vibrations applied to the pressure sensor 2. The vibration frequency of the mechanical vibrations, the vibration frequency of the electromagnetic vibrations, etc. can be input as the input variable F.

[0141] By using the matrix operation formula expressed by Equation (10), it is possible to properly correct the pressure measurement value measured by the pressure sensor 2 in the pressure vessel 1 of gas-insulated equipment in the field, even when the pressure sensor 2 is affected by factors outside the equipment system, such as mechanical vibrations and electromagnetic vibrations. In cases where the sensor diaphragm exhibits regular erroneous displacement, leaks of the insulating gas sealed in the pressure vessel can be detected early and with high accuracy.

[0142] Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments and includes various modifications without departing from the technical scope. For example, the above-described embodiments are not necessarily limited to those including all of the configurations described above. Furthermore, it is possible to replace part of the configuration of an embodiment with another configuration, or to add another configuration to the configuration of an embodiment. Furthermore, it is also possible to add other configurations to, delete configurations from, or replace configurations with respect to part of the configuration of an embodiment.

[0143] For example, the gas leak detection device and gas leak detection method described above address the nonlinearity of the temperature dependence of pressure sensors and zero-point shift, but they can also be similarly applied to pressure sensors with small temperature dependence, such as gas density pressure sensors. The processing unit 6 of the gas leak detection device 100 can be configured as appropriate as long as it has the same functions as described above, and may be realized by one piece of hardware or by multiple pieces of hardware. [Explanation of symbols]

[0144] 1. Pressure vessels 2 Pressure Sensors 3 Temperature Sensor 4 A / D converter 5 Recording section 6 Processing section (calculation section, testing section) 7 Display section 100 Gas leak detection device

Claims

1. A gas leak detection device for detecting leakage of an insulating gas in a pressure vessel in which the insulating gas is sealed, comprising: a pressure sensor for measuring the pressure inside the pressure vessel; a temperature sensor for measuring the temperature on or inside the pressure vessel; a recording unit that records pressure measurements measured by the pressure sensor and temperature measurements measured by the temperature sensor; a calculation unit that calculates a corrected pressure by correcting the pressure measured by the pressure sensor using a correction function; a test unit that tests whether or not there is a leak of the insulating gas based on a change over time of the corrected pressure, The correction function is a product P of the power of the measured pressure P and the power of the temperature ratio T / Ts of the measured temperature T and the extreme temperature Ts, where P is the measured pressure measured as the pressure of the insulating gas, T is the measured temperature measured as the temperature of the insulating gas, and Ts is the extreme temperature which is the minimum or maximum value of the measured temperature measured during a predetermined period before the leak inspection. ｍ (T / Ts) ｎ (where m and n are numbers equal to or greater than 0) as variables.

2. The gas leak detection device according to claim 1, A gas leak detection device in which the correction function is expressed as a polynomial which is the sum of the products of a power of the measured pressure P, a power of the temperature ratio T / Ts, and a correction coefficient which corrects for the influence of factors within the device system.

3. The gas leak detection device according to claim 2, A gas leak detection device in which the exponent m of the measured pressure P, the exponent n of the temperature ratio T / Ts, and the correction coefficient are determined based on the magnitude of a coefficient of determination calculated in multivariate regression analysis.

4. The gas leak detection device according to claim 3, the multivariate regression analysis is performed based on data collected in a preliminary test, using a power of the measured pressure P, a power of the temperature ratio T / Ts, and the correction coefficient as explanatory variables, and using a plurality of sample pressures as response variables; A gas leak detection device in which the sample pressure is a pressure obtained by converting the pressure for each gas amount at a predetermined reference temperature obtained in the preliminary test into pressures at multiple different sample temperatures measured by the temperature sensor.

5. The gas leak detection device according to claim 4, the preliminary test is a test for determining a pressure of the insulating gas sealed in a test vessel simulating the pressure vessel for each gas amount at the reference temperature, In the preliminary test, the insulating gas is adjusted to different gas amounts, and the pressure for each gas amount at the reference temperature is determined; A gas leak detection device in which the pressure for each gas amount at the reference temperature is converted into the pressure for each gas amount at the sample temperature based on a plurality of different sample temperatures measured by the temperature sensor and the gas's equation of state.

6. The gas leak detection device according to claim 3, the multivariate regression analysis is performed based on data collected during leak monitoring, using a power of the measured pressure P, a power of the temperature ratio T / Ts, and the correction coefficient as explanatory variables, and using a plurality of sample pressures as response variables; The sample pressure is a pressure measured by the pressure sensor at each measurement time during leak monitoring, converted to a pressure at each measurement time at a predetermined reference temperature based on a plurality of different measurement temperatures measured by the temperature sensor.

7. The gas leak detection device according to claim 6, During the leak monitoring, the pressure of the insulating gas sealed in the pressure vessel is measured at the measurement temperature for each measurement time; A gas leak detection device in which the pressure at each measurement time at the measurement temperature is converted into the pressure at each measurement time at the reference temperature based on the gas state equation.

8. The gas leak detection device according to claim 1, A gas leak detection device in which the correction function is expressed by a polynomial which is the sum of products of a power of the measured pressure P, a power of the temperature ratio T / Ts, a correction coefficient which corrects the influence of factors within the device system, and an input variable which corrects the influence of factors outside the device system.

9. The gas leak detection device according to claim 1, The extreme temperature Ts is the minimum or maximum value of the temperature measured within 24 hours immediately before the leak detection.

10. A gas leak detection method for detecting a leak of an insulating gas in a pressure vessel filled with the insulating gas, comprising: measuring the pressure inside the pressure vessel and the temperature on or inside the pressure vessel; a step of correcting the pressure measured by the pressure sensor using a correction function to obtain a corrected pressure; and a step of verifying whether or not there is a leak of the insulating gas based on a change over time of the corrected pressure, The correction function is a product P of the power of the measured pressure P and the power of the temperature ratio T / Ts of the measured temperature T and the extreme temperature Ts, where P is the measured pressure measured as the pressure of the insulating gas, T is the measured temperature measured as the temperature of the insulating gas, and Ts is the extreme temperature which is the minimum or maximum value of the measured temperature measured during a predetermined period before the leak inspection. ｍ (T / Ts) ｎ (where m and n are numbers equal to or greater than 0) as variables.

Images