An Adaptive Test and Calibration Method for an SF6 Gas Sensor

Abstract

This invention discloses an adaptive testing and calibration method for SF6 gas sensors, belonging to the field of gas sensor technology. The method includes establishing a sealed high-pressure tank testing system, connecting multiple sensors under test and a reference sensor, and purging with low-humidity gas; determining the proportion of sensors exceeding the allowable moisture content; if a preset value is reached, repurging; pressurizing to a preset pressure value; circulating gas and sampling moisture data; after determining system stability, reading the pressure, temperature, and moisture errors of each sensor and the reference data; if qualified, the test is passed; otherwise, the humidity is increased to a second operating condition for retesting; extracting historical data from unqualified sensors to calculate pressure, temperature, and moisture compensation parameters and writing them into local storage; iterative calibration until qualified or reaching the maximum number of calibrations. This invention solves the technical problem of batch accurate testing and automatic calibration of SF6 gas sensors under simulated real operating conditions. It achieves batch automated testing, improving product consistency and factory reliability.

Description

Technical Field

[0001] This invention belongs to the field of gas sensor technology, and particularly relates to an adaptive testing and calibration method for an SF6 gas sensor. Background Technology

[0002] SF6 moisture density sensors (hereinafter referred to as SF6 gas sensors) are widely used in GIS switchgear, circuit breakers, and other gas-insulated equipment to monitor SF6 gas state parameters in real time. Based on physical parameters such as pressure, temperature, and humidity, they calculate gas density and moisture content to determine the equipment's insulation performance and operational safety. Typically, the operating pressure of SF6 gas inside GIS equipment is approximately 5–6 Bar (relative pressure at 20°C), and the moisture content is generally required to be ≤150 ppm. Therefore, high requirements are placed on the measurement accuracy, stability, and factory consistency of SF6 gas sensors.

[0003] In existing technologies, SF6 gas sensors typically undergo only single-point testing at atmospheric pressure, manual sampling, or simple ventilation calibration after production. They lack simulation testing under actual operating pressure and humidity conditions, leading to discrepancies between test results and real-world usage. Furthermore, during transportation, storage, and assembly, the sensor's gas path and parts in contact with the outside air easily absorb water vapor. For example, at an ambient temperature of 20°C and relative humidity of 50%, the moisture content in the air can reach approximately 11,800 ppm. In coastal and other high-humidity areas, where relative humidity is typically above 80%, the moisture absorption is even more pronounced. If installed directly into GIS equipment without sufficient purging and calibration, residual humid air inside the sensor may mix with the drying SF6 gas within the equipment, causing falsely high moisture readings in the initial commissioning phase. This can lead to false alarms, misjudgments of insulation status, and even unnecessary maintenance shutdowns.

[0004] Invention data content The purpose of this invention is to provide an adaptive testing and calibration method for SF6 gas sensors, which solves the technical problem of performing batch accurate testing and automatic calibration of SF6 gas sensors under simulated real working conditions.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: An adaptive test and calibration method for an SF6 gas sensor includes the following steps: S1. Establish a closed high-pressure tank test system, connect multiple SF6 gas sensors to multiple test ports of the high-pressure tank, and set up a reference sensor for collecting reference data. Purge the high-pressure tank with low humidity gas. S2. Read the micro-moisture data of each SF6 gas sensor to be tested. When the proportion of the number of SF6 gas sensors with micro-moisture values ​​greater than the first preset threshold to the total number of SF6 gas sensors to be tested reaches a preset proportion, repeat step S1; otherwise, repeat step S3. S3. Fill the high-pressure tank with carrier gas and pressurize it to the preset pressure value; S4. Start the circulation pump to circulate the gas in the high-pressure tank, and perform sliding window sampling on the micro-water data collected by the reference sensor. When the sampling result meets the first preset humidity threshold and the continuous rate of change is less than the stability threshold, the system is determined to be stable. S5. After the system stabilizes, read the pressure, temperature and moisture values ​​output by each SF6 gas sensor under test and compare them with the reference data. When all errors are less than the preset allowable error threshold, the SF6 gas sensor under test is deemed qualified; otherwise, it is deemed unqualified. S6. Increase the humidity inside the high-pressure tank to the second preset humidity threshold, and repeat steps S4 to S5. S7. Extract historical sampling data and corresponding baseline data from the SF6 gas sensor that is determined to be unqualified, calculate pressure compensation parameters, temperature compensation parameters and micro-water compensation parameters, and write them into the local storage of the unqualified SF6 gas sensor. S8. Repeat steps S4, S5 and S7 for the SF6 gas sensors that are determined to be unqualified, until all the SF6 gas sensors to be tested reach the qualified state, or the unqualified SF6 gas sensors reach the preset calibration limit. S9. Replace the next batch of SF6 gas sensors to be tested and repeat steps S1 to S8 above.

[0006] Preferably, the reference data includes collected pressure data, temperature data, and micro-moisture data; The sealed high-pressure tank testing system includes a gas circulation channel, an air inlet channel, a carrier gas channel, an exhaust channel, and a sensor channel. The gas circulation channel consists of a high-pressure tank, a circulation pump, and a one-way valve forming a closed loop. The air intake channel is used to inject low-humidity gas into the high-pressure tank; The carrier gas channel is used to inject carrier gas into the high-pressure tank. The exhaust channel is used for venting the high-pressure tank. The sensor channel is used to connect to the SF6 gas sensor under test.

[0007] Preferably, when performing step S1, the low-humidity gas in step S1 is filtered dry air.

[0008] Preferably, when performing step S2, the first preset threshold is less than the second preset threshold.

[0009] Preferably, when performing step S4, the sliding window sampling result is: ; where m represents the length of the sliding window, in units of the number of sampling points; j represents the traversal sequence number of the data within the sliding window, j = 0, 1, 2... m - 1; H1(t - j) is the micro - water value collected by the micro - water sensor at the (t - j)th moment; When judging the system stability, first take the last k smoothed values. If any adjacent points meet the following judgment conditions and meet them continuously for k times, it is determined to be stable. The specific formula is as follows: ; where stability is the stability threshold.

[0010] Preferably, the allowable error threshold includes a pressure allowable error threshold, a temperature allowable error threshold, and a micro - water allowable error threshold. The conditions for judging whether the待测 SF6 gas sensor is qualified are as follows: When {SGi P (t) - P1(t)} / P1(t) < Error P and {SGi T (t) - T1(t)} / T1(t) < Error T and {SGi H (t) - H1(t)} / H1(t) < Error H then it is judged that the待测 SF6 gas sensor is qualified; otherwise, the待测 SF6 gas sensor is unqualified; where SGi P (t) is the pressure value collected by the ith SF6 gas sensor at the tth moment; SGi T [[ID=3)]] (t) is the temperature value collected by the ith SF6 gas sensor at the tth moment; SGi H [[ID=)]] (t) is the micro - water value collected by the ith SF6 gas sensor at the tth moment; Preferably, when calculating the pressure compensation parameter, temperature compensation parameter, and micro - water compensation parameter in step S7, the specific algorithm is as follows: P′ = P + C Pi ; T′ = T + C Ti ; H′ = H + C Hi ; where P′, T′, and H′ are the pressure value, temperature value, and micro - water value respectively output by the SF6 gas sensor after final compensation; P, T, and H are the pressure value, temperature value, and micro - water value respectively actually measured by the SF6 gas sensor; CPi C Ti C Hi These are pressure compensation parameters, temperature compensation parameters, and micro-water compensation parameters, respectively. The pressure compensation parameters, temperature compensation parameters, and micro-water compensation parameters are all obtained by using an algorithm that calculates the average of the differences between historical sampled values ​​and corresponding baseline data.

[0011] Preferably, the upper limit of the preset calibration times in step S8 is q times. If an SF6 gas sensor is still deemed unqualified after q consecutive calibrations, it is marked as an SF6 sensor fault and calibration is stopped.

[0012] This invention discloses an adaptive testing and calibration method for SF6 gas sensors, which solves the technical problem of performing accurate batch testing and automatic calibration of SF6 gas sensors under simulated real-world operating conditions. The invention constructs a sealed high-pressure tank testing system to simultaneously perform purging, pressurization, cyclic steady-state testing, and dual-humidity condition testing on multiple SF6 gas sensors under near-actual operating pressure conditions. The system automatically determines the sensor performance status based on reference sensor data. For sensors that fail the test, compensation parameters are calculated based on historical sampling data and written into the sensor, achieving online adaptive calibration, realizing automated batch testing, reducing labor costs, and improving product consistency and factory reliability. Attached Figure Description

[0013] Figure 1 This is the main flowchart of the present invention; Figure 2 This is a flowchart of the system setup and purging process of the present invention; Figure 3 This is a flowchart of the purging effect judgment of the present invention; Figure 4 This is a flowchart of the gas circulation and system stability determination of the present invention; Figure 5 This is a flowchart of the sensor performance testing and qualification determination process of the present invention; Figure 6 This is a flowchart of the second humidity condition test of the present invention; Figure 7 This is a flowchart of steps S7 to S9 of the present invention; Figure 8 This is a schematic diagram of the system architecture of the sealed high-pressure tank testing system of the present invention; In the diagram: 1. Air compressor; 2. Gas flow meter; 3. Check valve; 4. Check valve; 5. Air valve; 6. Carrier gas transmitter; 7. Check valve; 8. Air valve; 9. SF6 gas sensor; 10. High-pressure tank; 11. Pressure sensor; 12. Micro-water sensor; 13. Air valve; 14. Circulating pump; 15. Pressure relief valve; 16. Air valve; 17. Silencer. Detailed Implementation

[0014] Depend on Figures 1-8 An adaptive test and calibration method for an SF6 gas sensor 9, as shown, includes the following steps: S1. Establish a closed high-pressure tank test system, connect multiple SF6 gas sensors 9 to multiple test ports of the high-pressure tank 10, and set up a reference sensor for collecting reference data. Purge the high-pressure tank 10 with low-humidity gas.

[0015] In this embodiment, the sealed high-pressure tank testing system includes a gas circulation channel, an inlet channel, a carrier gas channel, an outlet channel, a sensor channel, and a reference sensor; The gas circulation channel includes a high-pressure tank 10 (used to contain gas and control the internal environment), a circulation pump 14, and a one-way valve 7. The outlet of the high-pressure tank 10 is connected to the inlet of the circulation pump 14. The outlet of the circulation pump 14 is connected to the inlet of the one-way valve 7 through a gas valve 13. The outlet of the one-way valve 7 is then connected to the inlet of the high-pressure tank 10, forming a gas circulation loop.

[0016] The air intake channel includes an air compressor 1 (for regulating gas humidity) and a gas flow meter 2. The outlet of the air compressor 1 is connected to the inlet of a one-way valve 3 through the gas flow meter 2. The outlet of the one-way valve 3 is connected to the inlet of the high-pressure tank 10 to control the speed and pressure of gas flowing into the high-pressure tank 10.

[0017] The carrier gas channel includes a carrier gas emitter 6, which is connected to the inlet of a one-way valve 4 via a gas valve 5. The outlet of the one-way valve 4 is connected to the inlet of the high-pressure tank 10 to provide dry gas or a suitable carrier gas.

[0018] The exhaust channel includes a pressure relief valve 15, a gas valve 16, and a silencer 17. The exhaust port of the high-pressure tank 10 is connected to the silencer 17 through the pressure relief valve 15 and the gas valve 16. After being processed by the silencer 17, the gas is released to the outside through the exhaust port. In this embodiment, the silencer 17 is used to stabilize the airflow in order to reduce noise during gas release.

[0019] The sensor channel includes multiple valves Fi and multiple SF6 gas sensors 9 to be tested. The valves Fi include F1 to Fn, labeled Fi.

[0020] The SF6 gas sensor 9 is denoted as SGi. The high-pressure tank 10 has multiple test ports, each of which is connected to a valve Fi, where i is the number.

[0021] Before the test calibration begins, connect each SGi to be tested to a valve Fi. For example, connect valve F1 to SG1, valve F2 to SG2, and so on.

[0022] The carrier gas launcher 6 provides dry gas, the air compressor 1 is used to regulate the gas temperature, the flow meter 2 regulates the gas flow rate, the circulation pump 14 ensures uniform circulation of gas within the system, and the sensor channel connects to different SF6 gas sensors 9 to provide real-time feedback on the gas status.

[0023] These SF6 gas sensors 9 are denoted as SGi. Each SF6 gas sensor 9 outputs the gas parameters generated by the gas detected in the high-pressure tank 10, which are used as sensor parameters. For SGi, its sensor parameters include pressure data SGi. P Temperature data SGi T and micro water data SGi H .

[0024] The reference sensor includes a pressure sensor 11, a temperature sensor, and a micro-water sensor 12. The reference sensor is used to measure the gas parameters generated by the gas in the high-pressure tank 10 and use them as reference parameters. The reference parameters include reference pressure P1(t), reference temperature T1(t), and reference micro-water data H1(t), where t represents the sampling time.

[0025] The high-pressure tank 10 is connected to various channels through multiple pipes, forming a closed-loop gas flow system. During testing and calibration, the gas circulates within the high-pressure tank 10. A reference sensor provides reference parameters, and each SF6 gas sensor 9 under test collects and outputs its detected sensor parameters, which are then uploaded to the detection and control system. The detection and control system compares the sensor parameters of each SF6 gas sensor 9 with the reference parameters, calibrates each SF6 gas sensor 9 accordingly, and completes the automatic testing and calibration tasks.

[0026] In this embodiment, the detection and control system consists of an industrial computer and a digital signal expansion board. The industrial computer communicates with the SF6 gas sensor 9, the reference sensor and the digital signal expansion board through at least one RS485 bus. Each device is distinguished and interacts with data through a preset communication address.

[0027] In normal use, the industrial control computer communicates with multiple SF6 gas sensors 9 through the first RS485 bus, and communicates with the reference sensor and the switch expansion board through the second RS485 bus respectively.

[0028] The SF6 gas sensor 9 communicates with the industrial control computer via an RS485 bus, and the pressure sensor 11, temperature sensor, and micro-water sensor 12 also communicate with the industrial control computer via an RS485 bus.

[0029] In this embodiment, the pressure sensor 11 and the temperature sensor are the same composite pressure sensor, and its output includes pressure value and temperature value.

[0030] The industrial control computer and the switch expansion board communicate via an RS485 bus. The switch expansion board is used to output multiple switch signals to drive and control the air compressor 1, gas flow meter 2, check valve 3, carrier gas generator Q5, air valve 5, check valve 4, check valve 7, air valve 13, circulating pump 14, air valves F1 to Fn, pressure relief valve 15, and air valve 16.

[0031] In this embodiment, check valve 3, gas valve 5, check valve 4, check valve 7, gas valve 13, gas valves F1 to Fn, pressure relief valve 15, and gas valve 16 are all solenoid valves. The SF6 gas sensor to be tested 9 is a composite SF6 gas sensor 9, which includes at least one micro-water sensor, one SF6 gas concentration sensor, one temperature sensor, and one pressure sensor.

[0032] The drive control of air compressor 1, gas flow meter 2, carrier gas generator Q5 and circulating pump 14 is achieved by controlling their power supply circuit.

[0033] After connecting all n SF6 gas sensors 9 to be tested (the total number is determined by the number of valves Fi; if there are N valves Fi, i.e. valves F1 to Fn, then the total number of sensors in this batch is n), the sealed high-pressure tank test system is turned on and enters the purging program. Gas valves F1 to Fn and gas valve 5 are opened, and the output of the carrier gas generator is adjusted to make the carrier gas pressure 1.5 Bar. The flow rate of the gas injected by the air compressor 1 is controlled by the flow meter Q2 to regulate the air entering the high-pressure tank 10, thereby regulating the humidity of the gas inside the high-pressure tank 10. Open valve 16 to purge the gas and replace the high-humidity gas in the tank with low-humidity air.

[0034] In this embodiment, low-humidity gas (filtered high-purity air) is used instead of the humid air inside the tank to ensure that the equipment is in a dry environment before testing. The purging time is generally 5 minutes during use.

[0035] After purging is complete, close gas valve 16 and carrier gas channel valve V7, and set gas flow meter 2 to zero to bring the system into a static state.

[0036] S2. Read the micro-moisture data of each SF6 gas sensor 9 to be tested. When the proportion of the number of SF6 gas sensors 9 with micro-moisture values ​​greater than the first preset threshold reaches the preset proportion of the total number of SF6 gas sensors 9 to be tested, repeat step S1; otherwise, repeat step S3.

[0037] In this embodiment, step S2 specifically includes reading the micro-water data collected by each SF6 gas sensor 9, denoted as SGi. HIf the number of SF6 gas sensors 9 with a moisture content greater than 200 ppm accounts for 50% or more of the total number n, it is determined that the residual moisture inside the system is too high, and S1 is executed again to continue purging; otherwise, step S3 is executed. In this embodiment, the industrial control computer reads the micro-water data SGi collected by each SF6 gas sensor 9 one by one via the RS485 bus. H And judge according to the following judgment formula: N faulty / N≥50%; N faulty For SGi H The number of SF6 gas sensors 9 with a concentration >200ppm; N is the total number n.

[0038] S3. Fill the high-pressure tank 10 with carrier gas (which is actually high-purity air output from the carrier gas generator) and pressurize it to the preset pressure value. In this embodiment, when performing step S3, the specific steps include adjusting the output of the carrier gas generator Q5 to adjust the carrier gas pressure to 5 Bar, pressurizing the high-pressure tank 10 to simulate the actual operating conditions of the SF6 gas sensor 9; and stopping the gas intake when the pressure inside the high-pressure tank 10 reaches the set value of 5 Bar.

[0039] In this embodiment, the industrial control computer determines whether the pressure value collected by the pressure sensor has reached 5 Bar: if yes, it stops the power supply to the carrier gas generator Q5 and stops the air intake; otherwise, it keeps the power supply on and continues to intake air.

[0040] S4. Start the circulation pump to circulate the gas in the high-pressure tank 10, and perform sliding window sampling on the micro-water data collected by the reference sensor. When the sampling result meets the first preset humidity threshold and the continuous rate of change is less than the stability threshold, the system is determined to be stable. In step S4, the specific steps include starting the circulation pump 14 and the gas valve 13 to fully circulate the gas in the high-pressure tank 10; and performing sliding window sampling on the baseline micro-water data H1(t) collected by the micro-water sensor to obtain the sliding window sampling result. ; ; Where m represents the length of the sliding window, in units of the number of sampling points; j represents the traversal sequence number of the data within the sliding window, j=0,1,2...m-1; H1(tj) is the micro-water value at time tj.

[0041] In this embodiment, m is at least 20, meaning the window length includes at least 20 sampling points.

[0042] Sampling results of the sliding window Each micro-water value in the data is judged, when If all trace moisture values ​​are less than the preset trace moisture threshold of 200 ppm (in this case, all SF6 gas sensors 9 are tested under the condition of trace moisture value of 200 ppm), then the system stability judgment is executed; otherwise, S1 is executed. The system stability judgment is specifically as follows: when the following conditions are met... If the change of 20 consecutive sampling points is less than 2%, the system is considered stable. At this time, it continues to run for 10 minutes, then the gas valve 13 is closed and the circulation pump 14 is stopped (in this embodiment, the circulation pump 14 is de-energized), and S5 is executed; otherwise, S4 is executed to continue the cycle. At the same time, if the cycle time exceeds 30 minutes, a fault alarm is generated and the process ends. In this embodiment, when determining system stability, the most recent 20 smoothed values ​​are first taken, for example, the value is: If any two adjacent points satisfy the following condition for 20 consecutive times, then the point is considered stable. The specific formula is as follows: ; S5. After the system stabilizes, read the pressure, temperature, and moisture values ​​output by each SF6 gas sensor 9 under test and compare them with the reference data. If all errors are less than the preset allowable error threshold, the SF6 gas sensor 9 under test is deemed qualified; otherwise, it is deemed unqualified. In this embodiment, after the system stabilizes, it is allowed to stand for another 2 minutes. Then, the industrial control computer reads the data collected by each SF6 gas sensor 9 and the reference sensor in real time and caches it as historical data. After the preset acquisition time threshold is reached, the industrial control computer reads the sensor parameters collected by each SF6 gas sensor 9 at time t from the historical data one by one. The sensor parameters include pressure value, temperature value and trace moisture value; and constructs the data sequence SGi for each SF6 gas sensor 9. GP (t): SGi GP (t) = (SGi) P (t), SGi T (t), SGi H (t)); Among them, SGi P (t), SGi T (t), SGi H (t) represents the pressure, temperature, and trace moisture values ​​collected by the i-th SF6 gas sensor 9 at time t, respectively; Simultaneously, the reference parameter E(t) acquired by the reference sensor is read from historical data: E(t)=(P1(t), T1(t), H1(t)); Where P1(t) is the reference pressure; T1(t) is the reference temperature; and H1(t) is the reference micro-water data. Compare the data sequence SGi GP (t) with the reference parameter E(t). When {SGi P (t) - P1(t)} / P1(t) < 1% and {SGi T (t) - T1(t)} / T1(t) < 1% and {SGi H (t) - H1(t)} / H1(t) < 5%, it is determined that SGi is qualified; otherwise, SGi is unqualified; [[ID=IO]]S6. Increase the humidity in the high-pressure tank 10 to the second preset humidity threshold, and repeat steps S4 to S5; When performing step S6, it specifically includes starting the gas flowmeter 2 to input 10 SCCM of high-humidity air, so that the micro-water value in the high-pressure tank 10 gradually increases; during the micro-water climbing process, steps S4 to S5 are periodically repeated to detect each SF6 gas sensor 9; when the micro-water value H1(t) collected by the micro-water sensor > 1000 ppm, stop the humidity climbing and end this round of humidity condition testing; Repeat the method of steps S4 to S5, but set the preset micro-water threshold to 1000 ppm, and judge the qualified status of all SF6 gas sensors 9 under the condition of a micro-water value of 1000 ppm.

[0043] If there is an unqualified SF6 gas sensor 9, it is recorded as SGi faulty , and enter the calibration process of S7; otherwise, it is determined that all SF6 gas sensors 9 in this batch are qualified, and S9 is executed.

[0044] In this embodiment, the SF6 gas sensor 9 must be qualified in all test conditions (the two conditions of 200 ppm and 1000 ppm) to be finally qualified.

[0045] S7. Extract the historical sampling data and its corresponding reference data of the determined unqualified SF6 gas sensor 9, calculate the pressure compensation parameter, temperature compensation parameter and micro-water compensation parameter, and write them into the local storage of the unqualified SF6 gas sensor 9.

[0046] When performing step S7, it specifically includes reading the unique ID (96-bit address) of the unqualified SGi faulty , and extracting the historical data X(t) of SGi faulty at multiple sampling moments according to the unique ID, and simultaneously extracting the reference data E(t) at multiple sampling moments accordingly: X(t) = (SGi faultyP (t), SGi faultyT (t), SGi faultyH (t)); E(t)=(P1(t),T1(t),H1(t)); Among them, SGi faultyP (t) represents SGi at time t. faulty Measured pressure value; SGi faultyT (t) represents SGi at time t. faulty Measured temperature value; SGi faultyH (t) represents SGi at time t. faulty The measured micro-moisture value; Calculate the average error values ​​for pressure, temperature, and micro-water content, respectively, and use them as compensation parameters: Micro-water compensation parameter C Hi The calculation method is as follows: ; Pressure compensation parameter C Pi The calculation method is as follows: ; Temperature compensation parameter C Ti The calculation method is as follows: ; Where N is the total number of historical sampling points. In this embodiment, the value of N is determined by the number of sampling times extracted. The more sampling times extracted, the more accurate the final calibration result, but the time consumption increases. In actual use, the default value of N is 40. The industrial control computer will calculate the micro-water compensation parameter C. Hi Pressure compensation parameter C Pi and temperature compensation parameter C Ti All sent to SGi faulty ,SGi faulty Perform local storage; and in subsequent measurements, execute the compensation algorithm locally. The compensation algorithm is as follows: P′=P+C Pi ; T′=T+C Ti ; H′=H+C Hi ; Wherein, P′, T′, and H′ are respectively SGi faulty The final output values ​​are pressure, temperature, and moisture content; P, T, and H are respectively the values ​​of SGa. faulty The actual measured pressure, temperature, and moisture content values.

[0047] S8. Repeat steps S4, S5 and S7 for the SF6 gas sensors 9 that are determined to be unqualified, until all the SF6 gas sensors 9 to be tested reach the qualified state, or the unqualified SF6 gas sensors 9 reach the preset calibration limit. When performing step S8, it specifically includes repeating S4-S7 until all the unqualified SF6 gas sensors 9 have been calibrated. During this process, if a certain SF6 gas sensor 9 fails to meet the standard after three consecutive calibrations, it is marked as an SF6 gas sensor fault and calibration is stopped. When all SF6 gas sensors 9 have reached the qualified state or have been marked as faulty, step S9 is executed. S9. Replace the next batch of SF6 gas sensors 9 to be tested and repeat steps S1 to S8 above.

[0048] When performing step S9, the specific steps include closing valves F1-Fn, replacing a new batch of SF6 gas sensors 9 to be tested, and retesting or calibrating according to the methods of steps S1 to S8.

[0049] In this embodiment, after all the SF6 gas sensors 9 to be tested have been tested and calibrated, the control module shuts off all valves and equipment power, opens gas valve 16, releases the gas pressure in the tank to 1 Bar (adjustable), and then closes V5.

[0050] This invention discloses an adaptive testing and calibration method for SF6 gas sensors, which solves the technical problem of performing accurate batch testing and automatic calibration of SF6 gas sensors under simulated real-world operating conditions. The invention constructs a sealed high-pressure tank testing system to simultaneously perform purging, pressurization, cyclic steady-state testing, and dual-humidity condition testing on multiple SF6 gas sensors under near-actual operating pressure conditions. The system automatically determines the sensor performance status based on reference sensor data. For sensors that fail the test, compensation parameters are calculated based on historical sampling data and written into the sensor, achieving online adaptive calibration, realizing automated batch testing, reducing labor costs, and improving product consistency and factory reliability.

Claims

1. An adaptive test and calibration method for an SF6 gas sensor, characterized in that: Includes the following steps: S1. Establish a closed high-pressure tank test system, connect multiple SF6 gas sensors (9) to multiple test ports of the high-pressure tank (10), and set up a reference sensor for collecting reference data. Purge the high-pressure tank (10) with low humidity gas. S2. Read the micro-moisture data of each SF6 gas sensor (9) to be tested. When the proportion of the number of SF6 gas sensors (9) with micro-moisture values ​​greater than the first preset threshold reaches the preset proportion of the total number of SF6 gas sensors (9) to be tested, repeat step S1; otherwise, execute step S3. S3. Fill the high-pressure tank (10) with carrier gas and pressurize it to the preset pressure value; S4. Start the circulation pump to circulate the gas in the high-pressure tank (10), and perform sliding window sampling on the micro-water data collected by the reference sensor. When the sampling result meets the first preset humidity threshold and the continuous change rate is less than the stable threshold, the system is determined to be stable. S5. After the system stabilizes, read the pressure value, temperature value and micro-moisture value output by each SF6 gas sensor (9) under test, and compare them with the reference data. When all errors are less than the preset allowable error threshold, the SF6 gas sensor (9) under test is deemed qualified; otherwise, it is deemed unqualified. S6. Increase the humidity inside the high-pressure tank (10) to the second preset humidity threshold, and repeat steps S4 to S5. S7. Extract historical sampling data and corresponding baseline data from the SF6 gas sensor (9) that is determined to be unqualified, calculate pressure compensation parameters, temperature compensation parameters and micro-water compensation parameters, and write them into the local storage of the unqualified SF6 gas sensor (9). S8. Repeat steps S4, S5 and S7 for the SF6 gas sensors (9) that are determined to be unqualified, until all the SF6 gas sensors (9) to be tested reach the qualified state, or the unqualified SF6 gas sensors (9) reach the preset calibration limit. S9. Replace the next batch of SF6 gas sensors to be tested (9) and repeat steps S1 to S8 above.

2. The adaptive test and calibration method for an SF6 gas sensor as described in claim 1, characterized in that: The baseline data includes collected pressure data, temperature data, and micro-moisture data; The sealed high-pressure tank test system includes a gas circulation channel, an air inlet channel, a carrier gas channel, an exhaust channel, and a sensor channel. The gas circulation channel consists of a high-pressure tank (10), a circulation pump, and a one-way valve forming a closed loop. The air intake channel is used to inject low-humidity gas into the high-pressure tank (10); The carrier gas channel is used to inject carrier gas into the high-pressure tank (10); The exhaust channel is used for exhausting the high-pressure tank (10); The sensor channel is used to connect to the SF6 gas sensor (9) under test.

3. The adaptive testing and calibration method for an SF6 gas sensor as described in claim 1, characterized in that: When performing step S1, the low-humidity gas mentioned in step S1 is filtered dry air.

4. The adaptive test and calibration method for an SF6 gas sensor as described in claim 1, characterized in that: When performing step S2, the first preset threshold is less than the second preset threshold.

5. The adaptive test and calibration method for an SF6 gas sensor as described in claim 1, characterized in that: During step S4, the sliding window sampling result for: ； Where m represents the length of the sliding window, in units of the number of sampling points; j represents the traversal sequence number of the data within the sliding window, j=0,1,2...m-1; H1(tj) is the micro-water value collected by the micro-water sensor at time tj; When determining system stability, the k most recent smoothed values ​​are first selected. If any adjacent points satisfy the following condition for k consecutive times, the system is considered stable. The specific formula is as follows: ； Where stability is the stability threshold.

6. The adaptive test and calibration method for an SF6 gas sensor as described in claim 1, characterized in that: The allowable error thresholds include pressure allowable error threshold, temperature allowable error threshold and micro-moisture allowable error threshold. The conditions for determining whether the SF6 gas sensor (9) under test is qualified are as follows: When {SGi} is satisfied P (t)-P1(t)} / P1(t)<Error P And {SGi T (t)-T1(t)} / T1(t)<Error T And {SGi H (t)-H1(t)} / H1(t)<Error H If the test result is positive, the SF6 gas sensor (9) to be tested is deemed qualified; otherwise, the SF6 gas sensor (9) to be tested is deemed unqualified. Among them, SGi P (t) represents the pressure value collected by the i-th SF6 gas sensor (9) at time t; SGi T (t) represents the temperature value collected by the i-th SF6 gas sensor (9) at time t; SGi H (t) represents the micro-water value collected by the i-th SF6 gas sensor (9) at time t.

7. The adaptive test and calibration method for an SF6 gas sensor as described in claim 1, characterized in that: The specific algorithm for calculating the pressure compensation parameters, temperature compensation parameters, and micro-water compensation parameters in step S7 is as follows: P′=P+C Pi ; T′=T+C Ti ; H′=H+C Hi ; Wherein, P′, T′, and H′ represent the pressure, temperature, and micro-moisture values ​​output by the SF6 gas sensor (9) after final compensation, respectively; P, T, and H represent the pressure, temperature, and micro-moisture values ​​actually measured by the SF6 gas sensor (9), respectively; C Pi C Ti C Hi These are pressure compensation parameters, temperature compensation parameters, and micro-water compensation parameters, respectively. The pressure compensation parameters, temperature compensation parameters, and micro-water compensation parameters are all obtained by using an algorithm that calculates the average of the differences between historical sampled values ​​and corresponding baseline data.

8. The adaptive test and calibration method for an SF6 gas sensor as described in claim 1, characterized in that: The preset calibration count limit in step S8 is q times. When an SF6 gas sensor (9) is still deemed unqualified after q consecutive calibrations, it is marked as an SF6 sensor fault and calibration is stopped.

Images