Online monitoring device and method for SF6 gas decomposer

Abstract

According to the online monitoring device and method for the SF6 gas decomposer, in the online monitoring device for the SF6 gas decomposer, a sensing unit comprises a gas sensor used for obtaining components of the SF6 gas decomposer and a temperature and humidity sensor used for obtaining temperature and humidity data; the acquisition unit is connected with the sensing unit, the acquisition unit comprises a voltage division circuit, an analog-to-digital conversion circuit and a single-chip microcomputer, data acquired by the sensing unit is converted into voltage signals through the voltage division circuit and converted into digital signals through the analog-to-digital conversion circuit, and the data are input into the single-chip microcomputer; the communication unit is connected with the acquisition unit to transmit data; and the diagnosis unit is connected with the communication unit so as to identify the components and the concentration of the SF6 gas decomposer based on the data.

Description

Technical Field

[0001] The present invention relates to the technical field of gas detection, and in particular to an online monitoring device and method for SF6 gas decomposition products. Background Art

[0002] Sulfur hexafluoride (SF6) is a commonly used insulating medium in high-voltage electrical equipment, such as GIS, due to its excellent electrical insulation properties and stable chemical properties. However, when electrical equipment experiences faults such as partial discharge or arcing, SF6 gas may decompose, producing low-fluorinated sulfides such as SF2, SF3, and SF4. These react with air and moisture inside the equipment to produce harmful compounds such as SO2 and H2S, posing a potential threat to equipment safety and the environment.

[0003] Existing research shows that the component concentrations of SF6 gas decomposition products can reflect the operating status of electrical equipment. Therefore, by monitoring the composition of the gas within the equipment, the operating status of the SF6 equipment can be effectively assessed and possible failures can be warned in advance. Existing gas detection technologies mainly include gas chromatography, infrared spectroscopy, and photoacoustic spectroscopy. However, gas chromatography requires multiple steps such as sample collection, pretreatment, and analysis, resulting in a long detection time and inability to achieve real-time monitoring. Infrared spectroscopy may be interfered with by other gases or environmental factors, resulting in reduced detection accuracy. Although photoacoustic spectroscopy has high sensitivity, it may also be affected by gas interference, and the equipment is relatively complex and requires high operating requirements. Therefore, these existing detection technologies generally have problems such as cumbersome operation and high cost, making it difficult to meet the needs of online monitoring.

[0004] The information disclosed in this Background section is only for enhancement of understanding of the background of the invention and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the Invention

[0005] In view of the shortcomings or defects of the above-mentioned prior art, an online monitoring device and method for SF6 gas decomposition products are provided to overcome the above-mentioned defects, with high real-time detection accuracy, convenient operation and low cost.

[0006] The purpose of the present invention is achieved through the following technical solutions.

[0007] An online monitoring device for SF6 gas decomposition products includes:

[0008] A sensing unit comprising a gas sensor for obtaining components of SF6 gas decomposition products;

[0009] an acquisition unit connected to the sensing unit, comprising a voltage divider circuit, an analog-to-digital conversion circuit, and a single-chip microcomputer. The acquisition unit converts the data collected by the sensing unit into a voltage signal via the voltage divider circuit and into a digital signal via the analog-to-digital conversion circuit, and inputs the data into the single-chip microcomputer. The single-chip microcomputer calculates the resistance value of the gas sensor based on the input voltage signal and the resistance value of the matching resistor in the voltage divider circuit;

[0010] a communication unit connected to the acquisition unit to transmit the resistance value, and sending the gas sensor resistance value output by the single chip microcomputer to the diagnosis unit via the Modbus protocol;

[0011] A diagnosis unit is connected to the communication unit to identify the components and concentrations of SF6 gas decomposition products based on the resistance value and a gas identification algorithm.

[0012] The online monitoring device for SF6 gas decomposition products also includes a gas testing platform for screening gas sensors, the gas testing platform includes:

[0013] A mass flow controller that adjusts the flow ratio of SF6 gas as background gas and standard gas to configure a predetermined concentration of the gas to be measured;

[0014] a sensor array connected to the mass flow controller to introduce a gas to be measured to obtain sensor data, the sensor array comprising a plurality of metal oxide semiconductor gas sensors;

[0015] Waste treatment, which is connected to a closed gas chamber, passes the experimental waste gas through a gas washing bottle filled with calcium hydroxide solution for harmless treatment;

[0016] a sampling board connected to the sensor array to collect the resistance value of each gas sensor;

[0017] The host computer is connected to the sampling board to calculate the response value, response recovery time and repeatability based on the resistance value of the sensor, and at least one metal oxide semiconductor gas sensor is screened out from a plurality of metal oxide semiconductor gas sensors based on the response value, response recovery time and repeatability as a gas sensor suitable for detecting SF6 gas decomposition products in an oxygen-free environment.

[0018] In the online monitoring device for SF6 gas decomposition products, the standard gases include H2S, CO, SO2, SOF2 and SO2F2.

[0019] In the online monitoring device for SF6 gas decomposition products, the response value is:

[0020]

[0021] Among them, R a is the resistance of the metal oxide semiconductor gas sensor in air; Rg is the resistance of the metal oxide semiconductor gas sensor in the gas to be measured,

[0022] Response time τ res The recovery time τ is the time required for the metal oxide semiconductor gas sensor to reach 90% of the maximum change from contacting the gas to be measured. rec It is defined as the time required for a metal oxide semiconductor gas sensor to reach 10% of the maximum change from being separated from the gas to be measured.

[0023] Repeatability is the consistency of the response curve obtained by a metal oxide semiconductor gas sensor when repeatedly measuring the same gas sample under the same conditions. Repeatability is described by introducing the coefficient of variation (CV). The smaller the coefficient of variation, the better the repeatability of the sensor. The coefficient of variation (CV) is:

[0024]

[0025] Among them, SD S is the standard deviation of multiple response values; Mean S is the average of multiple response values.

[0026] In the online monitoring device for SF6 gas decomposition products, the voltage divider circuit includes a heating resistor R H and matching resistor R L .

[0027] In the online monitoring device for SF6 gas decomposition products, the communication unit includes an RS-485 transceiver. The communication unit receives instructions sent by the diagnostic unit through the RS-485 transceiver based on the Modbus communication protocol, and replies to the diagnostic unit with specified data according to the function code.

[0028] In the online monitoring device for SF6 gas decomposition products, the diagnostic unit includes a host computer embedded with a gas identification algorithm module, and the gas identification algorithm is an integrated learning algorithm module integrating K-nearest neighbor, support vector machine, random forest, extreme tree and multi-layer perceptron.

[0029] In the online monitoring device for SF6 gas decomposition products, the host computer also includes an interface for display.

[0030] In the online monitoring device for SF6 gas decomposition products, the acquisition unit also includes a power supply module connected to the single-chip microcomputer and the digital-to-analog conversion circuit.

[0031] The online monitoring method of the online monitoring device for SF6 gas decomposition products includes the following steps:

[0032] S100: The decomposition products of the SF6 gas to be measured are introduced into the sensing unit;

[0033] S200: The acquisition unit acquires the gas signal of the gas decomposition product to be measured and converts it into a digital signal and transmits it to the single chip microcomputer;

[0034] S300: The communication unit transmits the data to the diagnosis unit;

[0035] S400: The diagnostic unit identifies the components and concentrations of the SF6 gas decomposition products to be tested based on the data collected by the sensor unit, and displays them in real time on the host computer interface.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] The present invention detects gas based on a semiconductor gas sensor, which has the advantages of low cost, small size and low power consumption. It combines a gas identification algorithm to identify the component concentrations of SF6 gas decomposition products and displays them in real time on a host computer, thereby realizing online monitoring of SF6 gas decomposition products. It can be further extended to online monitoring of the operating status and fault diagnosis of SF6 gas-insulated equipment, and has broad application prospects.

[0038] The above description is only an overview of the technical solution of the present invention. In order to make the technical means of the present invention clearer and easier to understand, so that those skilled in the art can implement it according to the contents of the description, and in order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are illustrated below. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Various other advantages and benefits of the present invention will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiments below. The accompanying drawings are intended only to illustrate preferred embodiments and are not to be construed as limiting the present invention. It should be understood that the drawings described below are merely examples of the present invention, and that those skilled in the art will be able to derive other drawings from these drawings without inventive effort. Throughout the drawings, identical reference numerals are used to denote identical components.

[0040] In the attached figure:

[0041] Figure 1 This is a schematic structural diagram of an online monitoring device for SF6 gas decomposition products provided by one embodiment of the present disclosure;

[0042] Figure 2 is a schematic diagram of a high-throughput gas-sensing testing platform provided in one embodiment of the present disclosure;

[0043] Figure 3 This is a dynamic characteristic curve of a WO3 sensor provided by one embodiment of the present disclosure when detecting SO2 gas with a concentration of 50 ppm;

[0044] Figure 4 This is a repeatability curve of a WO3 sensor provided by one embodiment of the present disclosure when detecting SO2 gas with a concentration of 50 ppm;

[0045] Figure 5 is a schematic diagram of a voltage divider circuit in a collection unit provided in one embodiment of the present disclosure;

[0046] Figure 6 is a flow chart of an integrated learning algorithm provided by one embodiment of the present disclosure;

[0047] Figure 7 Schematic diagram of a GIS discharge fault simulation platform provided by one embodiment of the present disclosure;

[0048] Figure 8 This is a schematic diagram of a host computer providing real-time monitoring of the concentration of SF6 gas decomposition products according to an embodiment of the present disclosure;

[0049] Figure 9 is a response curve diagram of a gas sensor under two discharges provided by an embodiment of the present disclosure;

[0050] Figure 10 This is a test result diagram of the SF6 decomposition product tester provided in one embodiment of the present disclosure.

[0051] The present invention will be further explained below with reference to the accompanying drawings and embodiments. DETAILED DESCRIPTION

[0052] Specific embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although specific embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0053] It should be noted that certain words are used in the specification and claims to refer to specific components. Those skilled in the art should understand that technicians may use different nouns to refer to the same component. This specification and claims do not use the difference in nouns as a way to distinguish components, but use the difference in the functions of the components as the criterion for distinction. As mentioned throughout the specification and claims, "including" or "comprising" is an open term, so it should be interpreted as "including but not limited to". The subsequent description of the specification is a preferred embodiment of the present invention, but the description is based on the general principles of the specification and is not intended to limit the scope of the invention. The scope of protection of the present invention shall be as defined in the attached claims.

[0054] To facilitate understanding of the embodiments of the present invention, several specific embodiments will be further explained below with reference to the accompanying drawings. However, the accompanying drawings do not limit the embodiments of the present invention.

[0055] For better understanding, Figures 1 to 10 As shown, an online monitoring device for SF6 gas decomposition products includes:

[0056] A sensing unit comprising a gas sensor for obtaining components of SF6 gas decomposition products;

[0057] an acquisition unit connected to the sensing unit, comprising a voltage divider circuit, an analog-to-digital conversion circuit, and a single-chip microcomputer. The acquisition unit converts the data collected by the sensing unit into a voltage signal via the voltage divider circuit and into a digital signal via the analog-to-digital conversion circuit, and inputs the data into the single-chip microcomputer. The single-chip microcomputer calculates the resistance value of the gas sensor based on the input voltage signal and the resistance value of the matching resistor in the voltage divider circuit;

[0058] a communication unit connected to the acquisition unit to transmit data and sending the gas sensor resistance value output by the single chip microcomputer to the diagnosis unit via the Modbus protocol;

[0059] A diagnosis unit is connected to the communication unit to identify the components and concentrations of SF6 gas decomposition products based on the data and a gas identification algorithm.

[0060] In a preferred embodiment of the online monitoring device for SF6 gas decomposition products, a gas testing platform for screening gas sensors is further included. The gas testing platform includes:

[0061] A mass flow controller (MFC) adjusts the flow ratio of SF6 gas (background gas) and the standard gas to configure a predetermined concentration of the gas to be measured;

[0062] a sensor array connected to the mass flow controller to introduce a gas to be measured to obtain sensor data, the sensor array comprising a plurality of metal oxide semiconductor gas sensors;

[0063] Waste treatment, which is connected to a closed gas chamber, passes the experimental waste gas through a gas washing bottle filled with calcium hydroxide solution for harmless treatment;

[0064] a sampling board connected to the sensor array to collect the resistance value of each gas sensor;

[0065] The host computer is connected to the sampling board to calculate the response value, response recovery time and repeatability based on the sensor data, and selects at least one metal oxide semiconductor gas sensor from a plurality of metal oxide semiconductor gas sensors as a gas sensor suitable for detecting SF6 gas decomposition products in an oxygen-free environment based on the response value, response recovery time and repeatability.

[0066] In a preferred embodiment of the online monitoring device for SF6 gas decomposition products, the standard gases include H2S, CO, SO2, SOF2 and SO2F2.

[0067] In a preferred embodiment of the online monitoring device for SF6 gas decomposition products, the response value is:

[0068]

[0069] Among them, R a is the resistance of the metal oxide semiconductor gas sensor in air; R g is the resistance of the metal oxide semiconductor gas sensor in the gas to be measured,

[0070] Response time τ res The recovery time τ is the time required for the metal oxide semiconductor gas sensor to reach 90% of the maximum change from contacting the gas to be measured. rec It is defined as the time required for a metal oxide semiconductor gas sensor to reach 10% of the maximum change from being separated from the gas to be measured.

[0071] Repeatability is the consistency of the response curve obtained by a metal oxide semiconductor gas sensor when repeatedly measuring the same gas sample under the same conditions. Repeatability is described by introducing the coefficient of variation (CV). The smaller the coefficient of variation, the better the repeatability of the sensor. The coefficient of variation (CV) is:

[0072]

[0073] Among them, SD S is the standard deviation of multiple response values; Mean S is the average of multiple response values.

[0074] In a preferred embodiment of the online monitoring device for SF6 gas decomposition products, the response value, response time and recovery time of the sensor are obtained by conducting dynamic characteristic experiments; the repeatability of the sensor is obtained by conducting repeatability experiments.

[0075] In a preferred embodiment of the online monitoring device for SF6 gas decomposition products, the sensor needs to be screened based on the response value, response recovery time and repeatability, and the sensor should be selected to have a high response value, a short response recovery time and good repeatability.

[0076] In a preferred embodiment of the online monitoring device for SF6 gas decomposition products, the voltage divider circuit includes a heating resistor R H and matching resistor R L .

[0077] In a preferred embodiment of the online monitoring device for SF6 gas decomposition products, the communication unit includes an RS-485 transceiver. The communication unit receives instructions sent by the diagnostic unit through the RS-485 transceiver based on the Modbus communication protocol, and replies to the diagnostic unit with specified data according to the function code.

[0078] In a preferred embodiment of the online monitoring device for SF6 gas decomposition products, the diagnostic unit includes a host computer embedded with a gas identification algorithm module, and the gas identification algorithm is an integrated learning algorithm module integrating K-nearest neighbor, support vector machine, random forest, extreme tree and multi-layer perceptron.

[0079] In a preferred embodiment of the online monitoring device for SF6 gas decomposition products, the host computer further includes an interface for display.

[0080] In a preferred embodiment of the online monitoring device for SF6 gas decomposition products, the acquisition unit further includes a power supply module connected to the single chip microcomputer and the digital-to-analog conversion circuit.

[0081] The online monitoring method of the online monitoring device for SF6 gas decomposition products includes the following steps:

[0082] S100: The decomposition products of the SF6 gas to be measured are introduced into the sensing unit;

[0083] S200: The acquisition unit acquires the gas signal of the gas decomposition product to be measured and converts it into a digital signal and transmits it to the single chip microcomputer;

[0084] S300: The communication unit transmits the data to the diagnosis unit;

[0085] S400: The diagnostic unit identifies the components and concentrations of the SF6 gas decomposition products to be tested based on the data collected by the sensor unit, and displays them in real time on the host computer interface.

[0086] In one embodiment, an online monitoring device for SF6 gas decomposition products includes:

[0087] A sensing unit comprising a gas sensor for obtaining components of SF6 gas decomposition products;

[0088] an acquisition unit connected to the sensing unit, comprising a voltage divider circuit, an analog-to-digital conversion circuit, and a single-chip microcomputer. The acquisition unit converts the data collected by the sensing unit into a voltage signal via the voltage divider circuit and into a digital signal via the analog-to-digital conversion circuit, and inputs the data into the single-chip microcomputer. The single-chip microcomputer calculates the resistance value of the gas sensor based on the input voltage signal and the resistance value of the matching resistor in the voltage divider circuit;

[0089] a communication unit connected to the acquisition unit to transmit data and sending the gas sensor resistance value output by the single chip microcomputer to the diagnosis unit via the Modbus protocol;

[0090] A diagnosis unit is connected to the communication unit to identify the components and concentrations of SF6 gas decomposition products based on the data and a gas identification algorithm.

[0091] In another embodiment, in the sensing unit, the selection of semiconductor gas sensors suitable for detecting SF6 gas decomposition products in an oxygen-free environment needs to rely on a high-throughput gas-sensitive testing platform. The schematic diagram of the platform is as follows: Figure 2 The selection of sensors mainly depends on three indicators: response value, response recovery time and repeatability.

[0092] In this embodiment, the response value reflects the sensitivity of the device to the gas. For a specific concentration of gas, the higher the response value, the higher the sensitivity of the device to the gas and the better the performance. The specific definition is as follows:

[0093]

[0094] Among them, R a is the resistance of the sensor in air; R g is the resistance of the sensor in the gas to be measured.

[0095] Response time (τ res ) is defined as the time required for the sensor to reach 90% of the maximum change from contact with the gas to be measured, and the recovery time (τ rec ) is defined as the time it takes for the sensor to reach 10% of its maximum change after being separated from the gas being measured. Generally speaking, the shorter the response time and recovery time, the better.

[0096] Repeatability is another important indicator of gas sensors. Repeatability refers to the consistency of the response curve obtained by a gas sensor when repeatedly measuring the same gas sample under the same conditions. Repeatability can be described by introducing the coefficient of variation (CV). The smaller the coefficient of variation, the better the repeatability of the sensor. The specific definition is as follows:

[0097]

[0098] Among them, SD S is the standard deviation of multiple response values; MeanS is the average of multiple response values.

[0099] In this embodiment, the response value, response time, and recovery time of the gas sensor are obtained by conducting a dynamic characteristic experiment. The specific experimental process is as follows:

[0100] 1) First, the background gas (SF6) is introduced into the gas chamber at a steady flow rate for flushing. After a period of time, the temperature and humidity in the gas chamber can be maintained stable;

[0101] 2) Use a mass flow controller (MFC) to control the flow ratio of background gas and standard gas, thereby configuring a certain concentration of the gas to be tested, which is then introduced into the gas chamber to cause the sensor array to respond;

[0102] 3) After a full response, the background gas is introduced again to flush the gas chamber to restore the sensor, and the experimental exhaust gas is rendered harmless by passing through a gas washing bottle filled with calcium hydroxide solution.

[0103] Through the dynamic characteristic experiment, the dynamic characteristic curve of the gas sensor detecting a certain concentration of SF6 gas decomposition products in an oxygen-free environment can be obtained, and the response value and response recovery time can be calculated. Figure 3 The dynamic characteristic curve of the WO3 sensor when detecting SO2 gas with a concentration of 50ppm is given. According to the formula above, the response value S is calculated to be 44.819%, and the response time τ is res is 247s, and the recovery time τ rec It is 866s.

[0104] In this embodiment, the repeatability of the gas sensor is obtained by conducting a repeatability experiment. The specific experimental process is as follows:

[0105] 1) First, the background gas (SF6) is introduced into the gas chamber at a steady flow rate for flushing. After a period of time, the temperature and humidity in the gas chamber can be maintained stable;

[0106] 2) Use a mass flow controller (MFC) to control the flow ratio of the background gas and the standard gas, thereby configuring a certain concentration of the gas to be tested, which is then introduced into the gas chamber to cause the sensor array to respond;

[0107] 3) After a full response, the background gas is introduced again to flush the gas chamber, and the next response is performed after the resistance of the sensor returns to its initial state;

[0108] 4) After the gas sensor repeats the response recovery process four times, background gas is introduced to flush the gas chamber to restore the sensor. The experimental exhaust gas is then rendered harmless by passing it through a gas washing bottle filled with calcium hydroxide solution.

[0109] Through repeatability experiments, the repeatability curve of the gas sensor detecting a certain concentration of SF6 gas decomposition products in an oxygen-free environment can be obtained, and the coefficient of variation CV can be calculated to quantitatively analyze the quality of the sensor repeatability. Figure 4 The repeatability curve of the WO3 sensor when detecting SO2 gas with a concentration of 50ppm is given, and the coefficient of variation CV can be calculated as 2.12% based on the formula above.

[0110] In this embodiment, based on the sensor performance indicators obtained from the gas sensing experiment results, gas sensors with higher response values, shorter response recovery times and smaller coefficients of variation are selected for detecting SF6 gas decomposition products in an oxygen-free environment.

[0111] In another embodiment, in the acquisition unit, the voltage divider circuit is composed of a heating resistor (R H ) and matching resistor (R L ) composition, such as Figure 5 When the sensor is working, a voltage is applied across the heating resistor to heat the gas-sensitive material to a suitable working temperature. When the gas sensor is exposed to different target gases, the resistance of the gas-sensitive material (R S ) will change accordingly, which will be reflected in the voltage change on the matching resistor. The resistance of the gas-sensitive material is calculated by the following formula:

[0112]

[0113] Among them, V C To measure the circuit voltage; V OUT is the matching resistor R L The voltage on.

[0114] In this embodiment, the device uses a high-precision analog-to-digital converter, supports multiple differential analog signal input channels, has a high sampling rate and low nonlinear error, can meet the needs of high-precision measurement, its gain error and offset error are kept in a low range, the noise level is also low, and the accuracy of the measurement is effectively improved. The reference voltage is provided by an external voltage source, which supports a wide voltage range to adapt to different application scenarios. The device uses a low-noise operational amplifier to build a voltage follower circuit, and inputs the voltage signal into the analog-to-digital converter after filtering to ensure signal stability and accuracy. The device uses a single-chip microcomputer to coordinate and manage various functional modules. The single-chip microcomputer receives the digital signal output by the analog-to-digital converter through the SPI interface, and transmits the real-time data of the gas sensor to the 485 communication module through serial communication to achieve efficient data transmission.

[0115] In another embodiment, the communication unit utilizes an isolated RS-485 transceiver, enabling single-chip isolation of half-duplex or full-duplex RS-485 interfaces without the need for an external isolated power supply. This simplifies power supply design and optimizes circuit layout. The chip integrates an isolated power supply, digital isolation channels, and RS-485 transceiver modules. Compared to traditional optocoupler isolation solutions, this significantly reduces PCB area and component count, while improving system reliability and design simplicity.

[0116] In this embodiment, the device supports data transmission based on the Modbus protocol, enabling communication of sensor data via function code 03. The output data includes key sensor parameters such as voltage, real-time temperature and humidity. By optimizing the RS-485 isolation design and Modbus protocol implementation, the device maintains excellent anti-interference performance in high-noise environments, ensuring stable and reliable data transmission, making it suitable for online monitoring of SF6 gas decomposition products.

[0117] In another embodiment, the diagnostic unit includes a host computer embedded with a gas identification algorithm. The gas identification algorithm is an ensemble learning algorithm module that integrates K-nearest neighbor, support vector machine, random forest, extreme tree, and multilayer perceptron. The K-nearest neighbor algorithm is simple and efficient, suitable for processing small sample data; support vector machine performs well in high-dimensional data classification; random forest and extreme tree have strong feature selection capabilities and robustness, suitable for multivariate and nonlinear problems; and multilayer perceptron, through its multi-layer neural network structure, can automatically extract latent features from complex data, especially effective in large-scale data processing and multi-category gas identification. Furthermore, ensemble learning methods can be used to combine these models, thereby integrating the advantages of different algorithms, better addressing noise, missing values, and complex patterns in the data, and improving overall recognition performance.

[0118] This embodiment builds an integrated learning algorithm based on K-nearest neighbor (KNN), support vector machine (SVM), random forest (RF), extra tree (ExtraTrees) and multi-layer perceptron (MLP). The process is as follows: Figure 6 As shown in Figure 2. During the training phase, each algorithm's training dataset is randomly sampled from the original training dataset, and each algorithm's training process is independent of the others. During the testing phase, ensemble learning combines the predictions of all algorithms through hard voting (i.e., majority voting) to generate a final prediction. Ensemble learning combines the predictions of multiple models through hard voting, reducing the risk of overfitting of a single model.

[0119] In this embodiment, the specific parameters of the five algorithms are shown in Table 1. Among them, n_neighbors is the number of neighbors used in the K-nearest neighbor algorithm to determine the reference for classification or regression, which directly affects the complexity and generalization ability of the model; kernel is the function used in the support vector machine to map data to a high-dimensional space, which determines the shape of the model's decision boundary in the high-dimensional space; n_estimators is the number of decision trees in random forests and extreme trees. More trees will improve the stability and accuracy of the model, but will also increase the computational cost; hidden_layer_sizes is the structure of the hidden layer in the multilayer perceptron. More hidden layers and neurons can enhance the expressive power of the model, but too many neurons and layers may lead to overfitting, especially when the amount of data is small, and the training time will increase significantly, resulting in higher computational costs.

[0120] Table 1 Algorithm parameter settings

[0121] Recognition algorithm Parameter settings KNN n_neighbors:1 Support Vector Machine kernel:linear Random Forest n_estimators: 100 Extra Tree n_estimators: 100 MLP hidden_layer_sizes: (100, 100, 100)

[0122] In this example, based on the results of the gas sensing experiment, a decomposition product dataset containing five SF6 decomposition products of H2S, CO, SO2, SOF2, and SO2F2 at different concentrations was constructed. The detailed information of the dataset is shown in Table 2.

[0123] Table 2 Dataset information

[0124] Gas type Concentration (ppm) Number of samples <![CDATA[H2S]]> 2,5,20,50,100 100 <![CDATA[SO2]]> 10,20,30,40,50 100 CO 10,20,30,40,50 100 <![CDATA[SOF2]]> 10,20,30,40,50 100 <![CDATA[SO2F2]]> 10,20,30,40,50 100

[0125] To verify the effectiveness of the gas identification algorithm, the decomposition product dataset was randomly divided into a training set and a test set in a 4:1 ratio. An ensemble learning algorithm was trained on the training set using 5-fold cross-validation. After obtaining the average cross-validation accuracy, the algorithm was tested on the test set to obtain the test set accuracy. Cross-validation was used to train the model because it allows the training set to be divided into five subsets, with four of these subsets used in rotation to train the model and the remaining subset used to verify model performance. This ensures model stability across different data subsets and avoids overfitting. The recognition accuracy of the ensemble learning algorithm on the decomposition product dataset is shown in Table 3. As can be seen, the ensemble learning algorithm achieved high average cross-validation accuracy and test set accuracy for all five SF6 decomposition products. This result demonstrates the effectiveness of the algorithm, which can accurately identify SF6 decomposition products and is suitable for online monitoring of SF6 gas decomposition.

[0126] Table 3 Recognition accuracy of ensemble learning algorithm

[0127] Gas type Average accuracy (%) Test set accuracy (%) <![CDATA[H2S]]> 90.50 89.00 <![CDATA[SO2]]> 91.75 90.00 CO 89.25 89.00 <![CDATA[SOF2]]> 93.75 90.00 <![CDATA[SO2F2]]> 92.50 92.00

[0128] In another embodiment, the present disclosure further provides an online monitoring method for SF6 gas decomposition products, comprising the following steps:

[0129] S100: The decomposition products of the SF6 gas to be measured are introduced into the sensing unit;

[0130] S200: The acquisition unit acquires the gas signal of the gas decomposition product to be measured and converts it into a digital signal and transmits it to the single chip microcomputer;

[0131] S300: The communication unit transmits data to the diagnostic unit via the RS-485 transceiver using the Modbus protocol;

[0132] S400: The diagnostic unit identifies the components and concentrations of the SF6 gas decomposition products to be tested based on the data collected by the sensor unit, and displays them in real time on the host computer interface.

[0133] In this embodiment, the following Figure 7 The GIS discharge fault simulation platform shown in Figure 1 was used to verify the effectiveness of the proposed method. An SF6 gas decomposition product online monitoring device was installed on the fault simulation platform, and a needle electrode was added to the GIS cavity, while high-pressure SF6 gas was injected. After applying a high voltage of 15kV to the needle electrode, the electrode broke down and discharged. The response curve of the sensor is shown in Figure 2. Figure 8 As shown in the figure, it can be observed that after the breakdown discharge occurs, the device successfully detects the decomposition gas within 5 minutes, and the sensor module shows an obvious response; after 12 hours, the second breakdown discharge is carried out, and the device shows an obvious response again within 5 minutes. The response curve is shown in Figure 9 As shown. The precision gas analyzer also detected the composition of SF6 decomposition gas, the results are as follows Figure 10 As shown, the effectiveness of the online monitoring device and method for SF6 gas decomposition products is further verified.

[0134] The basic principles of the present application have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, and effects mentioned in this application are merely illustrative and not restrictive, and it should not be assumed that these advantages, strengths, and effects are required of each embodiment of this application. In addition, the specific details disclosed above are merely illustrative and facilitating understanding, and are not restrictive. The above details do not limit this application to necessarily being implemented using the above specific details.

[0135] The above description has been provided for the purpose of illustration and description. Furthermore, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although a number of example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. An online monitoring device for SF6 gas decomposition products, characterized in that: These include, A sensing unit comprising a gas sensor for obtaining components of SF6 gas decomposition products; an acquisition unit connected to the sensing unit, comprising a voltage divider circuit, an analog-to-digital conversion circuit, and a single-chip microcomputer. The acquisition unit converts the data collected by the sensing unit into a voltage signal via the voltage divider circuit and into a digital signal via the analog-to-digital conversion circuit, and inputs the data into the single-chip microcomputer. The single-chip microcomputer calculates the resistance value of the gas sensor based on the input voltage signal and the resistance value of the matching resistor in the voltage divider circuit; a communication unit connected to the acquisition unit to transmit the resistance value, and sending the gas sensor resistance value output by the single chip microcomputer to the diagnosis unit via the Modbus protocol; A diagnosis unit is connected to the communication unit to identify the components and concentrations of SF6 gas decomposition products based on the resistance value and a gas identification algorithm.

2. The online monitoring device for SF6 gas decomposition products according to claim 1, characterized in that: Preferably, it also includes a gas testing platform for screening gas sensors, the gas testing platform including: A mass flow controller that adjusts the flow ratio of SF6 gas as background gas and standard gas to configure a predetermined concentration of the gas to be measured; a sensor array connected to the mass flow controller to introduce a gas to be measured to obtain sensor data, the sensor array comprising a plurality of metal oxide semiconductor gas sensors; Waste treatment, which is connected to a closed gas chamber, passes the experimental waste gas through a gas washing bottle filled with calcium hydroxide solution for harmless treatment; a sampling board connected to the sensor array to collect the resistance value of each gas sensor; The host computer is connected to the sampling board to calculate the response value, response recovery time and repeatability based on the resistance value of the sensor, and at least one metal oxide semiconductor gas sensor is screened out from a plurality of metal oxide semiconductor gas sensors based on the response value, response recovery time and repeatability as a gas sensor suitable for detecting SF6 gas decomposition products in an oxygen-free environment.

3. The online monitoring device for SF6 gas decomposition products according to claim 2, characterized in that: The standard gases include H2S, CO, SO2, SOF2 and SO2F2.

4. The online monitoring device for SF6 gas decomposition products according to claim 2, characterized in that: The response value is: ， Among them, R a is the resistance of the metal oxide semiconductor gas sensor in air; R g is the resistance of the metal oxide semiconductor gas sensor in the gas to be measured, Response time τ res The recovery time τ is the time required for the metal oxide semiconductor gas sensor to reach 90% of the maximum change from contacting the gas to be measured. rec It is defined as the time required for a metal oxide semiconductor gas sensor to reach 10% of the maximum change from being separated from the gas to be measured. Repeatability is the consistency of the response curve obtained by a metal oxide semiconductor gas sensor when repeatedly measuring the same gas sample under the same conditions. Repeatability is described by introducing the coefficient of variation (CV). The smaller the coefficient of variation, the better the repeatability of the sensor. The coefficient of variation (CV) is: ， Among them, SD S is the standard deviation of multiple response values; Mean S is the average of multiple response values.

5. The online monitoring device for SF6 gas decomposition products according to claim 1, characterized in that: The voltage divider circuit includes a heating resistor R H and matching resistor R L .

6. The online monitoring device for SF6 gas decomposition products according to claim 1, characterized in that: The communication unit includes an RS-485 transceiver. The communication unit receives instructions sent by the diagnosis unit through the RS-485 transceiver based on the Modbus communication protocol, and replies specified data to the diagnosis unit according to the function code.

7. The online monitoring device for SF6 gas decomposition products according to claim 1, characterized in that: The diagnostic unit includes a host computer embedded with a gas identification algorithm module, and the gas identification algorithm is an integrated learning algorithm module integrating K-nearest neighbor, support vector machine, random forest, extreme tree and multi-layer perceptron.

8. The online monitoring device for SF6 gas decomposition products according to claim 7, characterized in that: The host computer also includes an interface for display.

9. The online monitoring device for SF6 gas decomposition products according to claim 1, characterized in that: The acquisition unit also includes a power supply module connected to the single chip microcomputer and the digital-to-analog conversion circuit.

10. The online monitoring method of the online monitoring device for SF6 gas decomposition products according to any one of claims 1 to 9, characterized in that: It includes The following steps, S100: The decomposition products of the SF6 gas to be measured are introduced into the sensing unit; S200: The acquisition unit acquires the gas signal of the gas decomposition product to be measured and converts it into a digital signal and transmits it to the single chip microcomputer; S300: The communication unit transmits the data to the diagnosis unit; S400: The diagnostic unit identifies the components and concentrations of the SF6 gas decomposition products to be tested based on the data collected by the sensor unit, and displays them in real time on the host computer interface.

Images