Detection method and device of air insulation switch cabinet, storage medium and electronic equipment

Abstract

The invention discloses a detection method and device of an air insulation switch cabinet, a storage medium and electronic equipment, and relates to the technical field of air insulation switch cabinet state detection. The method comprises the following steps: detecting whether characteristic gas exists in the air-insulated switch cabinet according to type information of gas in the air-insulated switch cabinet; if the characteristic gas exists in the air-insulated switch cabinet, determining that a target defect exists in the air-insulated switch cabinet, and acquiring gas concentration information in the air-insulated switch cabinet; and inputting the gas concentration information in the air-insulated switch cabinet into the target evaluation model for processing to obtain the defect type and defect degree of the target defect in the air-insulated switch cabinet. According to the invention, the problem of low accuracy of detecting whether defects such as partial discharge and partial overheating exist in the air insulation switch cabinet in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the technical field of state detection of air-insulated switchgear, and in particular to a detection method and device, storage medium and electronic equipment of an air-insulated switchgear. Background Art

[0002] Air-insulated switchgear is one of the most widely used equipment in the power transmission and distribution link. During the production and long-term operation, various defects will appear inside the air-insulated switchgear, such as partial discharge, local overheating, etc. Moreover, in order to ensure the normal operation of the equipment, it is necessary to detect the operating status of the air-insulated switchgear and promptly detect the insulation defects inside the air-insulated switchgear to prevent them from developing into faults.

[0003] In the related art, the pulse current method is generally used to detect whether there are insulation defects such as partial discharge and local overheating inside the air-insulated switchgear. Moreover, the pulse current method used in the related art generally includes the following steps: first, a series of pulse currents are applied to the inside of the switchgear through a pulse current generator, and these currents will generate partial discharge at the insulation defects; then, the current and voltage inside the switchgear are monitored and recorded using sensors or detection equipment. When insulation defects occur, abnormal current and voltage signals will be generated; then, based on the monitored current and voltage signals, it can be determined whether there are insulation defects in the air-insulated switchgear, and further analysis and positioning can be performed on them.

[0004] However, in the related art, when the pulse current method is used to detect insulation defects inside an air-insulated switchgear, electromagnetic interference may exist in the environment inside the air-insulated switchgear, and the electromagnetic interference factor will affect the propagation of the pulse current and the detection results, thereby affecting the accuracy of the detection.

[0005] Currently, no effective solution has been proposed to the problem of low accuracy in detecting whether there are defects such as local discharge and local overheating inside air-insulated switchgear in related technologies. Summary of the invention

[0006] The main purpose of the present application is to provide a detection method and device, storage medium and electronic equipment for an air-insulated switchgear, so as to solve the problem of low accuracy in detecting whether there are defects such as local discharge and local overheating inside the air-insulated switchgear in the related art.

[0007] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a detection method for an air-insulated switchgear is provided. The method comprises: detecting whether there is a characteristic gas in the air-insulated switchgear according to the type information of the gas in the air-insulated switchgear; if there is a characteristic gas in the air-insulated switchgear, determining that there is a target defect in the air-insulated switchgear, and obtaining the gas concentration information in the air-insulated switchgear, wherein the gas concentration information comprises at least one of the following: the concentration value of the characteristic gas and the concentration ratio between each characteristic gas; inputting the gas concentration information in the air-insulated switchgear into the target evaluation model for processing, and obtaining the defect type and defect degree of the target defect in the air-insulated switchgear, wherein the target evaluation model is a model constructed based on the Bayesian optimization algorithm and the support vector machine algorithm.

[0008] Furthermore, the target evaluation model is obtained in the following manner: a training data set is obtained based on multiple concentration values ​​of characteristic gases in the air-insulated switchgear, wherein the multiple concentration values ​​include at least: concentration values ​​of characteristic gases in the air-insulated switchgear under different defect types and different defect degrees, and concentration values ​​of characteristic gases at different detection positions in the air-insulated switchgear; based on the Bayesian optimization algorithm and the support vector machine algorithm, an original evaluation model for evaluating defects in the air-insulated switchgear is constructed; and the training data set is used to learn and train the original evaluation model to obtain the target evaluation model.

[0009] Furthermore, based on multiple concentration values ​​of characteristic gases in the air-insulated switchgear, obtaining a training data set includes: performing feature expansion processing on the multiple concentration values ​​of characteristic gases in the air-insulated switchgear to obtain processed gas concentration information, wherein the processed gas concentration information at least includes: the concentration value of the characteristic gas and the concentration ratio between each characteristic gas; performing feature selection processing on the processed gas concentration information to obtain multiple concentration information, wherein the multiple concentration information is concentration information selected from the processed gas concentration information with an importance higher than a preset importance; based on the multiple concentration information, obtaining the training data set.

[0010] Furthermore, obtaining the gas concentration information in the air-insulated switchgear includes: determining a target detection position, and collecting characteristic gases in the air-insulated switchgear based on the target detection position; determining a concentration signal of the characteristic gases in the air-insulated switchgear based on the characteristic gases collected from the air-insulated switchgear; converting the concentration signal of the characteristic gases in the air-insulated switchgear into an electrical signal, and obtaining the gas concentration information in the air-insulated switchgear based on the electrical signal.

[0011] Furthermore, determining the target detection position includes: obtaining the concentration values ​​of characteristic gases at different detection positions in the air-insulated switchgear; determining the detection area with the most types of characteristic gases and the largest concentration of characteristic gases based on the concentration values ​​of characteristic gases at different detection positions in the air-insulated switchgear; determining the target detection position based on the detection area with the most types of characteristic gases and the largest concentration of characteristic gases.

[0012] Furthermore, after inputting the gas concentration information in the air-insulated switchgear into the target assessment model for processing and obtaining the defect type and defect degree of the target defect in the air-insulated switchgear, the method also includes: determining a target strategy for performing maintenance processing on the air-insulated switchgear based on the defect type and defect degree of the target defect in the air-insulated switchgear; sending the target strategy to the target object, and prompting the target object to perform maintenance processing on the air-insulated switchgear based on the target strategy.

[0013] Furthermore, after inputting the gas concentration information in the air-insulated switchgear into the target assessment model for processing and obtaining the defect type and defect degree of the target defect in the air-insulated switchgear, the method also includes: judging whether the defect degree is higher than the preset defect degree; if the defect degree is higher than the preset defect degree, sending an alarm prompt information to the target object, wherein the alarm prompt information is used to prompt the target object to perform maintenance on the air-insulated switchgear; if the defect degree is not higher than the preset defect degree, continuing to detect whether the target defect exists in the air-insulated switchgear.

[0014] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a detection device for an air-insulated switchgear is provided. The device comprises: a first detection unit, which is used to detect whether there is a characteristic gas in the air-insulated switchgear according to the type information of the gas in the air-insulated switchgear; a first processing unit, which is used to determine whether there is a target defect in the air-insulated switchgear if there is a characteristic gas in the air-insulated switchgear, and obtain the gas concentration information in the air-insulated switchgear, wherein the gas concentration information includes at least one of the following: the concentration value of the characteristic gas and the concentration ratio between each characteristic gas; a second processing unit, which is used to input the gas concentration information in the air-insulated switchgear into a target evaluation model for processing, and obtain the defect type and defect degree of the target defect in the air-insulated switchgear, wherein the target evaluation model is a model constructed based on the Bayesian optimization algorithm and the support vector machine algorithm.

[0015] Furthermore, the target evaluation model is obtained through the following units: a first determination unit, used to obtain a training data set based on multiple concentration values ​​of characteristic gases in the air-insulated switchgear, wherein the multiple concentration values ​​include at least: concentration values ​​of characteristic gases in the air-insulated switchgear under different defect types and different defect degrees, and concentration values ​​of characteristic gases at different detection positions in the air-insulated switchgear; a first construction unit, used to construct an original evaluation model for evaluating defects in air-insulated switchgear based on a Bayesian optimization algorithm and a support vector machine algorithm; a first training unit, used to use the training data set to learn and train the original evaluation model to obtain the target evaluation model.

[0016] Furthermore, the first determination unit includes: a first processing module, used to perform feature expansion processing on multiple concentration values ​​of characteristic gases in the air-insulated switchgear to obtain processed gas concentration information, wherein the processed gas concentration information at least includes: the concentration value of the characteristic gas and the concentration ratio between each characteristic gas; a second processing module, used to perform feature selection processing on the processed gas concentration information to obtain multiple concentration information, wherein the multiple concentration information is concentration information selected from the processed gas concentration information with an importance higher than a preset importance; a first determination module, used to obtain the training data set based on the multiple concentration information.

[0017] Furthermore, the first processing unit includes: a third processing module, used to determine the target detection position, and based on the target detection position, collect characteristic gases in the air-insulated switchgear; a second determination module, used to determine the concentration signal of the characteristic gases in the air-insulated switchgear based on the characteristic gases collected from the air-insulated switchgear; and a fourth processing module, used to convert the concentration signal of the characteristic gases in the air-insulated switchgear into an electrical signal, and based on the electrical signal, obtain the gas concentration information in the air-insulated switchgear.

[0018] Furthermore, the third processing module includes: a first acquisition submodule, used to acquire the concentration value of the characteristic gas at different detection positions in the air-insulated switchgear; a first determination submodule, used to determine the detection area with the most types of characteristic gases and the largest characteristic gas concentration based on the concentration values ​​of the characteristic gases at different detection positions in the air-insulated switchgear; and a second determination submodule, used to determine the target detection position based on the detection area with the most types of characteristic gases and the largest characteristic gas concentration.

[0019] Furthermore, the device also includes: a second determination unit, which is used to input the gas concentration information in the air-insulated switchgear into the target assessment model for processing, and after obtaining the defect type and defect degree of the target defect in the air-insulated switchgear, determine the target strategy for repairing the air-insulated switchgear according to the defect type and defect degree of the target defect in the air-insulated switchgear; and a first sending unit, which is used to send the target strategy to the target object, and prompt the target object to repair the air-insulated switchgear according to the target strategy.

[0020] Furthermore, the device also includes: a first judgment unit, used to input the gas concentration information in the air-insulated switchgear into the target evaluation model for processing, and after obtaining the defect type and defect degree of the target defect in the air-insulated switchgear, judge whether the defect degree is higher than a preset defect degree; a second sending unit, used to send an alarm prompt information to the target object if the defect degree is higher than the preset defect degree, wherein the alarm prompt information is used to prompt the target object to perform maintenance on the air-insulated switchgear; a second detection unit, used to continue to detect whether the target defect exists in the air-insulated switchgear if the defect degree is not higher than the preset defect degree.

[0021] In order to achieve the above-mentioned object, according to another aspect of the present application, a computer-readable storage medium is provided, wherein the storage medium stores a program, wherein the program executes any one of the above-mentioned methods for detecting an air-insulated switchgear.

[0022] In order to achieve the above-mentioned purpose, according to another aspect of the present application, an electronic device is provided, which includes one or more processors and a memory, and the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement any one of the above-mentioned air-insulated switchgear detection methods.

[0023] Through the present application, the following steps are adopted: based on the type information of the gas in the air-insulated switchgear, detecting whether there is a characteristic gas in the air-insulated switchgear; if there is a characteristic gas in the air-insulated switchgear, determining that there is a target defect in the air-insulated switchgear, and obtaining the gas concentration information in the air-insulated switchgear, wherein the gas concentration information includes at least one of the following: the concentration value of the characteristic gas and the concentration ratio between each characteristic gas; the gas concentration information in the air-insulated switchgear is input into the target evaluation model for processing to obtain the defect type and defect degree of the target defect in the air-insulated switchgear, wherein the target evaluation model is a model constructed based on the Bayesian optimization algorithm and the support vector machine algorithm, which solves the problem of low accuracy in detecting whether there are defects such as local discharge and local overheating inside the air-insulated switchgear in the related art. By detecting the characteristic gas in the air-insulated switchgear and inputting the concentration information of the detected characteristic components in the air-insulated switchgear into the state assessment model for processing, the defect type and defect degree of the insulation defects inside the air-insulated switchgear are intelligently output. The state assessment model is a model constructed based on two intelligent algorithms, the Bayesian optimization algorithm and the support vector machine algorithm. This can avoid the situation in which the pulse current method used in the related technology to detect defects inside the air-insulated switchgear is easily affected by electromagnetic interference factors, resulting in low detection accuracy, thereby achieving the effect of improving the accuracy of detecting whether there are defects such as local discharge and local overheating inside the air-insulated switchgear. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0025] Figure 1 is a flow chart of a detection method for an air-insulated switchgear provided in an embodiment of the present application;

[0026] Figure 2 is a flow chart of an optional air-insulated switchgear detection method provided according to an embodiment of the present application;

[0027] Figure 3 Schematic diagram of the best detection position for state detection of the air-insulated switchgear in this embodiment;

[0028] Figure 4 is a schematic diagram of an optional air-insulated switchgear state detection device provided according to an embodiment of the present application;

[0029] Figure 5 is a schematic diagram of a detection device for an air-insulated switchgear provided according to an embodiment of the present application;

[0030] Figure 6 is a schematic diagram of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0032] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0034] It should be noted that the relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties. For example, an interface is set between this system and the relevant user or organization. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0035] The present invention is described below in conjunction with preferred implementation steps. Figure 1 is a flow chart of a detection method for an air-insulated switchgear provided in an embodiment of the present application, such as Figure 1 As shown, the method comprises the following steps:

[0036] Step S101, based on the type information of the gas in the air-insulated switch cabinet, detecting whether there is a characteristic gas in the air-insulated switch cabinet.

[0037] For example, the characteristic gases mentioned above may be four gases, namely, NO2 (nitrogen dioxide), CO (carbon monoxide), O3 (ozone) and NO (nitric oxide). For example, it may be detected that the air-insulated switchgear contains multiple gases, such as CO2 (carbon dioxide), NO2 (nitrogen dioxide), CO (carbon monoxide), O3 (ozone) and NO (nitric oxide), and then it may be determined whether the multiple gases detected in the air-insulated switchgear contain any one of the characteristic gases, namely, NO2 (nitrogen dioxide), CO (carbon monoxide), O3 (ozone) and NO (nitric oxide).

[0038] Step S102: If there is a characteristic gas in the air-insulated switchgear, it is determined that there is a target defect in the air-insulated switchgear, and the gas concentration information in the air-insulated switchgear is obtained, wherein the gas concentration information includes at least one of the following: a concentration value of the characteristic gas and a concentration ratio between each characteristic gas.

[0039] For example, if it is determined that the multiple gases in the air-insulated switchgear do not contain any characteristic gas among NO2 (nitrogen dioxide), CO (carbon monoxide), O3 (ozone) and NO (nitric oxide), it can be determined that there are no insulation defects such as local discharge and local overheating in the air-insulated switchgear (corresponding to the above-mentioned target defects); on the contrary, if it is determined that the multiple gases in the air-insulated switchgear contain any one or more characteristic gases among NO2 (nitrogen dioxide), CO (carbon monoxide), O3 (ozone) and NO (nitric oxide), it can be determined that there are insulation defects such as local discharge and local overheating in the air-insulated switchgear (corresponding to the above-mentioned target defects), and then the component concentration information in the air-insulated switchgear can be detected (corresponding to the above-mentioned gas concentration information).

[0040] For example, if it is determined that the multiple gases in the air-insulated switchgear contain only the characteristic gas NO2 (nitrogen dioxide), but do not contain the three characteristic gases CO (carbon monoxide), O3 (ozone) and NO (nitric oxide), then the concentration value of the characteristic gas NO2 (nitrogen dioxide) in the air-insulated switchgear (corresponding to the above-mentioned gas concentration information) and / or NO2 / SUM (the concentration ratio of nitrogen dioxide to the total gas content in the air-insulated switchgear) can be detected; if it is determined that the multiple gases in the air-insulated switchgear contain NO2 (nitrogen dioxide), CO (carbon monoxide), O3 (ozone) and NO (nitric oxide), then the concentration value of the characteristic gas NO2 (nitrogen dioxide) in the air-insulated switchgear (corresponding to the above-mentioned gas concentration information) and / or NO2 / SUM (the concentration ratio of nitrogen dioxide to the total gas content in the air-insulated switchgear) can be detected. O (carbon monoxide) and O3 (ozone) three characteristic gases, but not NO (nitric oxide) characteristic gas, then the NO2 / O3 (concentration ratio of nitrogen dioxide and ozone), NO2 (concentration value of nitrogen dioxide), NO2 / SUM (concentration ratio of nitrogen dioxide and the total gas content in the air-insulated switchgear), CO / SUM (concentration ratio of carbon monoxide and the total gas content in the air-insulated switchgear) and NO2 / CO (concentration ratio of nitrogen dioxide and carbon monoxide) (corresponding to the above-mentioned gas concentration information) in the air-insulated switchgear can be detected.

[0041] Step S103, inputting the gas concentration information in the air-insulated switchgear into the target evaluation model for processing to obtain the defect type and defect degree of the target defect in the air-insulated switchgear, wherein the target evaluation model is a model constructed based on the Bayesian optimization algorithm and the support vector machine algorithm.

[0042] For example, based on the content and changing rules of characteristic components (corresponding to the characteristic gases mentioned above) under different insulation defect types and defect degrees, as well as the concentration information of characteristic components (corresponding to the characteristic gases mentioned above) at different detection positions in the air-insulated switchgear, a state assessment model (corresponding to the target assessment model mentioned above) for identifying the defect types and defect degrees of air-insulated switchgear can be constructed by using the Bayesian optimization algorithm and the support vector machine algorithm.

[0043] For example, the detected NO2 / O3 (concentration ratio of nitrogen dioxide and ozone), NO2 (concentration value of nitrogen dioxide), NO2 / SUM (concentration ratio of nitrogen dioxide and total gas content in the air-insulated switchgear), CO / SUM (concentration ratio of carbon monoxide and total gas content in the air-insulated switchgear) and NO2 / CO (concentration ratio of nitrogen dioxide and carbon monoxide) (corresponding to the above-mentioned gas concentration information) in the air-insulated switchgear can be used as input parameters of the state assessment model (corresponding to the above-mentioned target assessment model). That is, the detected air-insulated switchgear can be used as the input parameters of the state assessment model (corresponding to the above-mentioned target assessment model). The NO2 / O3 (concentration ratio of nitrogen dioxide and ozone), NO2 (concentration value of nitrogen dioxide), NO2 / SUM (concentration ratio of nitrogen dioxide and the total gas content in the air-insulated switchgear), CO / SUM (concentration ratio of carbon monoxide and the total gas content in the air-insulated switchgear) and NO2 / CO (concentration ratio of nitrogen dioxide and carbon monoxide) in the air-insulated switchgear are input into the state assessment model (corresponding to the above-mentioned target assessment model) for assessment and processing, and the defect type and defect degree of the insulation defect (corresponding to the above-mentioned target defect) in the air-insulated switchgear are output.

[0044] Through the above steps S101 to S103, by detecting the characteristic gas in the air-insulated switchgear, and inputting the concentration information of the detected characteristic components in the air-insulated switchgear into the state assessment model for processing, the defect type and defect degree of the insulation defects inside the air-insulated switchgear are intelligently output. The state assessment model is a model constructed based on two intelligent algorithms, namely the Bayesian optimization algorithm and the support vector machine algorithm. This can avoid the situation in which the pulse current method used in the related technology to detect defects inside the air-insulated switchgear is easily affected by electromagnetic interference factors, resulting in low detection accuracy, thereby achieving the effect of improving the accuracy of detecting whether there are defects such as local discharge and local overheating inside the air-insulated switchgear.

[0045] Optionally, in the detection method of air-insulated switchgear provided in an embodiment of the present application, a target evaluation model is obtained in the following manner: a training data set is obtained based on multiple concentration values ​​of characteristic gases in the air-insulated switchgear, wherein the multiple concentration values ​​include at least: concentration values ​​of characteristic gases in the air-insulated switchgear under different defect types and different defect degrees, and concentration values ​​of characteristic gases at different detection positions in the air-insulated switchgear; based on the Bayesian optimization algorithm and the support vector machine algorithm, an original evaluation model for evaluating defects in the air-insulated switchgear is constructed; and the training data set is used to learn and train the original evaluation model to obtain a target evaluation model.

[0046] For example, the content and change law of characteristic components (corresponding to the above-mentioned characteristic gases) under different defect types and different defect degrees can be obtained first, and the concentration information of characteristic components (corresponding to the above-mentioned characteristic gases) under different detection positions in the air-insulated switchgear can be obtained. Then, a training data set can be constructed based on the obtained content and change law of characteristic components (corresponding to the above-mentioned characteristic gases) under different defect types and different defect degrees, and the concentration information of characteristic components (corresponding to the above-mentioned characteristic gases) under different detection positions in the air-insulated switchgear; then, the Bayesian optimization algorithm and the support vector machine algorithm can be used to construct an original evaluation model for evaluating defects in the air-insulated switchgear; the original evaluation model is then trained using the constructed training data set, and the recognition accuracy can be improved by changing the model parameters and the number of iterations, and when the recognition accuracy reaches more than 95%, this evaluation model can be considered to be a trained state evaluation model (corresponding to the above-mentioned target evaluation model).

[0047] Through the above scheme, the original condition assessment model can be easily learned and trained, thereby improving the accuracy of identifying the defect type and degree of the air-insulated switchgear through the condition assessment model.

[0048] Optionally, in the detection method for air-insulated switchgear provided in an embodiment of the present application, obtaining a training data set based on multiple concentration values ​​of characteristic gases in the air-insulated switchgear includes: performing feature expansion processing on the multiple concentration values ​​of characteristic gases in the air-insulated switchgear to obtain processed gas concentration information, wherein the processed gas concentration information at least includes: the concentration value of the characteristic gas and the concentration ratio between each characteristic gas; performing feature selection processing on the processed gas concentration information to obtain multiple concentration information, wherein the multiple concentration information is concentration information selected from the processed gas concentration information with an importance higher than a preset importance; and obtaining a training data set based on the multiple concentration information.

[0049] For example, when constructing a training data set based on the acquired characteristic component (corresponding to the above-mentioned characteristic gas) content and change law under different defect types and different defect degrees, as well as the characteristic component (corresponding to the above-mentioned characteristic gas) concentration information at different detection positions in the air-insulated switchgear, the detected characteristic component (corresponding to the above-mentioned characteristic gas) content can be first used as the original data, and feature expansion can be performed. The expanded characteristic parameters may include the content value of each characteristic component (corresponding to the above-mentioned characteristic gas), the ratio of the content between each characteristic component (corresponding to the above-mentioned characteristic gas), and the gas production rate (corresponding to the above-mentioned processed gas concentration information). Then, the top five characteristic parameters (corresponding to the above-mentioned multiple concentration information) that contribute the most to the recognition result can be selected through a feature selection method, and then the top five characteristic parameters (corresponding to the above-mentioned multiple concentration information) that contribute the most to the recognition result can be used to construct the training data set. For example, the top five feature parameters (corresponding to the above-mentioned multiple concentration information) that contribute most to the recognition result can be NO2 / O3 (concentration ratio of nitrogen dioxide and ozone), NO2 (concentration value of nitrogen dioxide), NO2 / SUM (concentration ratio of nitrogen dioxide and the total gas content in the air-insulated switchgear), CO / SUM (concentration ratio of carbon monoxide and the total gas content in the air-insulated switchgear) and NO2 / CO (concentration ratio of nitrogen dioxide and carbon monoxide), and then the original state assessment model can be trained using a large number of data sets (corresponding to the above-mentioned training data sets).

[0050] Through the above scheme, the data set used to train the state evaluation model can be determined quickly and accurately.

[0051] Optionally, in the detection method of the air-insulated switchgear provided in the embodiment of the present application, obtaining the gas concentration information in the air-insulated switchgear includes: determining the target detection position, and collecting characteristic gases in the air-insulated switchgear based on the target detection position; determining the concentration signal of the characteristic gases in the air-insulated switchgear based on the characteristic gases collected from the air-insulated switchgear; converting the concentration signal of the characteristic gases in the air-insulated switchgear into an electrical signal, and obtaining the gas concentration information in the air-insulated switchgear based on the electrical signal.

[0052] For example, the above-mentioned target detection position may be the best detection position for actually detecting defects inside the air-insulated switchgear. For example, characteristic gas may be extracted from the air-insulated switchgear according to the best detection position (corresponding to the above-mentioned target detection position) determined for the actual detection of the switchgear, and the gas component (corresponding to the above-mentioned characteristic gas) signal (corresponding to the above-mentioned concentration signal) in the air-insulated switchgear may be obtained based on the characteristic gas extracted from the air-insulated switchgear and the gas sensor; then the gas component signal (corresponding to the above-mentioned concentration signal) may be converted into an electrical signal, thereby converting the electrical signal into a concentration value and a concentration ratio between each component (corresponding to each characteristic gas) (corresponding to the above-mentioned gas concentration information).

[0053] In summary, through the conversion between signals, the concentration value in the air-insulated switchgear and the concentration ratio between the components can be obtained quickly and accurately.

[0054] Optionally, in the detection method of the air-insulated switchgear provided in the embodiment of the present application, determining the target detection position includes: obtaining the concentration values ​​of the characteristic gas at different detection positions in the air-insulated switchgear; determining the detection area with the most types of characteristic gases and the highest concentration of characteristic gases based on the concentration values ​​of the characteristic gases at different detection positions in the air-insulated switchgear; determining the target detection position based on the detection area with the most types of characteristic gases and the highest concentration of characteristic gases.

[0055] For example, the diffusion process and distribution law of the characteristic component (corresponding to the above-mentioned characteristic gas) inside the switch cabinet can be obtained through fluid simulation, and the optimal detection position of the physical switch cabinet (corresponding to the above-mentioned target detection position) can be obtained in combination with experimental detection. For example, different defects can be set at different positions inside the switch cabinet, and different detection points can be set at different positions on the surface of the switch cabinet, and then the characteristic component content (corresponding to the above-mentioned characteristic gas concentration) and type measured at each detection position within a certain period of time are statistically analyzed to obtain the detection area with the most types of characteristic components (corresponding to the above-mentioned characteristic gases) and the largest content (corresponding to the above-mentioned characteristic gas concentration), and the detection area with the most types of characteristic components (corresponding to the above-mentioned characteristic gases) and the largest content (corresponding to the above-mentioned characteristic gas concentration) is used as the above-mentioned target detection position.

[0056] Through the above scheme, the best detection position for detecting internal defects of the air-insulated switchgear can be determined quickly and accurately.

[0057] Optionally, in the detection method of air-insulated switchgear provided in an embodiment of the present application, after the gas concentration information in the air-insulated switchgear is input into the target evaluation model for processing to obtain the defect type and defect degree of the target defect in the air-insulated switchgear, the method also includes: determining a target strategy for performing maintenance processing on the air-insulated switchgear based on the defect type and defect degree of the target defect in the air-insulated switchgear; sending the target strategy to the target object, and prompting the target object to perform maintenance processing on the air-insulated switchgear based on the target strategy.

[0058] For example, the target objects may be operators and field staff who operate the air-insulated switchgear. For example, according to the defect type and defect degree of the insulation defect (corresponding to the target defect) in the identified air-insulated switchgear, the corresponding operation and maintenance strategy (corresponding to the target strategy) may be intelligently recommended to guide the operators and field staff (corresponding to the target objects) to perform operation and processing.

[0059] Through the above scheme, according to the defect type and defect degree of the insulation defect in the air-insulated switchgear identified, the corresponding operation and maintenance inspection strategy can be automatically recommended to the operator and the on-site staff.

[0060] Optionally, in the detection method of air-insulated switchgear provided in an embodiment of the present application, after the gas concentration information in the air-insulated switchgear is input into the target evaluation model for processing to obtain the defect type and defect degree of the target defect in the air-insulated switchgear, the method also includes: judging whether the defect degree is higher than a preset defect degree; if the defect degree is higher than the preset defect degree, sending an alarm prompt information to the target object, wherein the alarm prompt information is used to prompt the target object to perform maintenance on the air-insulated switchgear; if the defect degree is not higher than the preset defect degree, continuing to detect whether there is a target defect in the air-insulated switchgear.

[0061] For example, when the component concentration (corresponding to the above-mentioned gas concentration information) reaches the warning value, that is, the defect type and defect degree of the identified insulation defect (corresponding to the above-mentioned target defect) in the air-insulated switchgear are higher than the preset defect degree, an alarm can be issued through sound, light and other signals to prompt the operator (corresponding to the above-mentioned target object) to perform maintenance on the air-insulated switchgear; when the component concentration (corresponding to the above-mentioned gas concentration information) does not reach the warning value, that is, the defect type and defect degree of the identified insulation defect (corresponding to the above-mentioned target defect) in the air-insulated switchgear are not higher than the preset defect degree, the defect type and defect degree of the insulation defect (corresponding to the above-mentioned target defect) in the air-insulated switchgear can continue to be detected.

[0062] Through the above scheme, according to the sound, light and other signals of the alarm prompt, the operator can be promptly prompted to repair the insulation defects in the air-insulated switchgear.

[0063] For example, the method provided in this embodiment belongs to the technical field of air-insulated switchgear state detection, and in particular, relates to a method and device for air-insulated switchgear state detection based on characteristic components.

[0064] Moreover, when partial discharge and other faults occur inside the air-insulated switchgear, natural air will produce complex chemical reactions under discharge or overheating, generating a series of decomposition products. The solid insulating material coated on the conductor inside the switchgear will also be damaged under discharge and overheating, generating gases such as CO and CO2. Since the structure of the air-insulated switchgear is not completely airtight, the temperature and humidity of the air inside it will change due to the influence of the operating environment, and the temperature and humidity have a significant impact on the gas decomposition products. In addition, there are obvious differences in the PD (Partial Discharge) characteristics caused by different types of insulation defects, resulting in different decomposition characteristics of the air. At the same time, due to the large volume of the air-insulated switchgear, as the defect time continues, the distribution of the characteristic gas generated at the defect position inside the switchgear is different, and the characteristic gas content measured at different measurement positions is also different, which affects the accuracy of insulation diagnosis. Therefore, through experiments and simulations, the generation characteristics of characteristic products in the air during partial discharge and overheating in the switchgear under different temperatures and humidities are obtained, and experiments are carried out in the real switchgear to obtain the distribution differences of characteristic gases at different measurement positions, providing theoretical and practical support for the application of gas component analysis in the field of switchgear state detection.

[0065] For example, the purpose of this embodiment is to provide an air-insulated switchgear state detection method and device based on characteristic gas analysis, and to diagnose the state of the air-insulated switchgear by the concentration and type of characteristic components (corresponding to the above-mentioned characteristic gases) for the types of characteristic gases generated by the air-insulated switchgear under different insulation defects. Specifically, it includes obtaining the types and change laws of characteristic components under different conditions through reaction molecular dynamics simulation, determining the target detection characteristic components (corresponding to the above-mentioned characteristic gases), further determining the content and change law of characteristic components (corresponding to the above-mentioned characteristic gases) under different insulation defect types and degrees through experiments, obtaining the diffusion process and distribution law of characteristic components (corresponding to the above-mentioned characteristic gases) inside the switchgear through fluid simulation, and combining experimental detection to obtain the best detection position of the physical switchgear (corresponding to the above-mentioned target detection position), and then constructing an intelligent algorithm evaluation model to identify the defect type and degree of the air-insulated switchgear, and designing a set of characteristic component detection devices suitable for switchgear state detection.

[0066] For example, the detection method of the air-insulated switchgear provided in the embodiment of the present application includes the following contents:

[0067] First, the characteristic gas composition and change law of air under different conditions are obtained through reaction molecular dynamics simulation. In order to determine the decomposition and recombination mechanism of air and the change law of the generated products under partial discharge / overheating conditions, molecular dynamics simulation calculations are performed on the air system and the air and insulating material system with the help of chemical calculation software. Among them, the air system mainly considers the air discharge situation that does not involve solid insulating materials, and the air and insulating material system focuses on the defect types involving solid insulating materials, such as surface discharge and local overheating defects. By counting the number of various components in the simulation results and observing the change law of each component, the gas with a large content, relatively stable change and detectable content is selected as the characteristic component (corresponding to the above-mentioned characteristic gas) for detection.

[0068] Furthermore, the characteristic component content and variation law under different defect types and degrees are obtained. Different defect types are simulated, including metal protrusion defects, insulator metal contamination defects, insulator air gap defects and local overheating defects, and the characteristic component content and variation law within 24 hours are statistically analyzed, and the test is conducted every 4 hours. In the simulation test, the air chamber volume is small, the characteristic components are evenly diffused, and the characteristic component content in each area is basically the same, so there is no need to test the components in different areas.

[0069] Furthermore, in order to obtain the diffusion law of characteristic components inside the physical switch cabinet, the diffusion characteristics of characteristic components in the switch cabinet are obtained through fluid mechanics simulation. A three-dimensional model of the physical switch cabinet is built, and by setting the fault area and the gas production rate of the characteristic component, the component volume fraction cloud map of different simulation times is statistically analyzed to observe the diffusion process of different characteristic components in the switch cabinet. The component volume fractions at different detection positions are statistically analyzed, and the content differences at different measuring points are compared, so as to propose the key detection areas that should be paid attention to in the switch cabinet (corresponding to the above-mentioned target detection positions). In order to verify and obtain the distribution of characteristic components in the actual situation of the physical switch cabinet, different types of defects are set inside the physical switch cabinet, and the defects are placed at different positions to simulate the defects occurring at different positions. Different gas extraction detection positions are set, and the characteristic component content at each detection position is tested every 8 hours, for a total of 72 hours. Compare the differences in component content at different measurement positions, and obtain the area with the highest content of each characteristic component as the recommended detection position (corresponding to the above-mentioned target detection position).

[0070] Furthermore, a characteristic gas atlas library under insulation defects of switch cabinet is constructed, and a state evaluation model is constructed. The characteristic component contents corresponding to different insulation defect types and degrees are collected and counted to form a characteristic gas atlas library. Through characteristic parameter expansion and feature selection, the five characteristic parameters that contribute most to the recognition accuracy are selected as evaluation parameter inputs, and a state evaluation model based on BO-SVM (a support vector machine method based on Bayesian optimization) is constructed to realize the discrimination of defect types and the division of defect degrees. Finally, a portable air switch cabinet state detection device based on characteristic gas component analysis is proposed. And the device mainly includes a gas detection module, an identification and diagnosis module, an operation and maintenance strategy recommendation module and a human-computer interaction interface. The gas detection module can detect and output the volume fraction of characteristic components in real time through a sampling probe, a built-in air pump, a gas sensor and a data processing unit, and the gas sensor can be an electrochemical gas sensor and an ultraviolet gas sensor. It is also equipped with a temperature and humidity measurement unit, which can test the ambient temperature and relative humidity to assist in the determination of the results. The identification and diagnosis module includes a trained BO-SVM (a support vector machine method based on Bayesian optimization) model, data input and output units, which can pre-process the detected characteristic gas content, obtain the characteristic parameter values ​​required for the evaluation model and output the diagnosis results. The operation and maintenance strategy recommendation module can intelligently recommend corresponding operation and maintenance strategies based on the results given by the identification and diagnosis module to guide operators and on-site staff to perform operations. The human-computer interaction interface can set display mode, detection mode, equipment self-test and other functions.

[0071] Compared with the prior art, the detection method of the air-insulated switchgear provided by the embodiment of the present application has the following beneficial effects: the state of the air-insulated switchgear can be systematically judged based on the characteristic component content and the change law, and an evaluation method and device for the air-insulated switchgear are proposed. The structure of the air-insulated switchgear itself is not completely airtight, so changes in atmospheric conditions will cause changes in the properties such as the proportion and composition of the air components in the cabinet. At present, the research basis is mostly laboratory exploration, mainly based on simulation experiments. However, due to the large internal volume and complex structure of the switchgear, and the large number of locations where defects may occur, the diffusion of gas in the compartment causes different gas contents in different parts, resulting in differences in gas concentrations measured at different detection positions, which will have a great impact on the diagnostic accuracy of the threshold method and defect identification method driven by gas content data, and there are currently few studies on the distribution of characteristic components inside the physical switchgear. In this embodiment, the characteristic gas of the target to be detected is first determined by a molecular dynamics simulation method, and the characteristic component concentrations of different types of defects and different detection positions in the physical switchgear are analyzed to propose the best detection position, data analysis method and gas component detection device suitable for on-site detection of the switchgear.

[0072] For example, Figure 2 is a flow chart of an optional air-insulated switchgear detection method provided in an embodiment of the present application, such as Figure 2 As shown, the optional air-insulated switchgear detection method includes the following steps:

[0073] Step 1: Obtain target characteristic gas based on molecular dynamics simulation.

[0074] The number of atoms in the system is controlled to about 3000, which improves the calculation efficiency while ensuring the calculation accuracy. First, build a three-dimensional structure model of the four gas molecules, and optimize the structure of the established three-dimensional model so that each molecule is in its lowest energy state. Then build a periodic amorphous model. The gas molecules contained in the model are set according to the proportions under normal air. The constructed simulation model is in a higher energy state, and relaxation and dynamic optimization are required before simulation. First, the system is energy minimized, and then the NPT ensemble (isothermal and isobaric system) is used for dynamic optimization to make the constructed system reach equilibrium. With a time step of 0.1fs, after a 5ps equilibrium phase process, the system density is 0.0012g / cm 3 , which is close to the air density under the corresponding conditions in reality. After 5 ps of equilibration, molecular dynamics simulations were performed for 200 ps in the NPT ensemble (isothermal and isobaric system) at different target temperatures.

[0075] After the simulation is completed, the products are counted. In the simulation calculation results of this embodiment, the gases that are relatively stable and have a relatively significant content are NO2, CO, O3 and NO. Therefore, when testing the air-insulated switchgear, the content of the four characteristic gases, NO2, CO, O3 and NO, should be focused on.

[0076] In addition, this embodiment also provides another example of molecular dynamics simulation of reactions involving solid insulating materials. In the first example, only characteristic components are generated by air medium under discharge or overheating defects, while actual insulation defects often involve solid medium. In the second example, when constructing the reaction system, epoxy resin and silicone rubber molecules are added to the reaction system. Similarly, simulation calculations are performed according to the steps in the first example. The products obtained in this example include a large number of hydrocarbons, among which the main small molecule products are CH4, CH2O, CO and O3, etc. Therefore, when the insulation defects involve solid insulating materials, the above reaction products should be focused on.

[0077] Step 2: Obtain the content and variation law of characteristic components under different defect types and degrees. The common defects in air-insulated switchgear are counted, mainly metal protrusion defects, insulator metal contamination defects, insulator air gap defects and overheating defects. A switchgear simulation test platform is built, and the characteristic components and contents under different defects are counted in the test air chamber. The above four types of defects are simulated respectively. The ambient temperature and humidity conditions should be kept unchanged during the test. By adjusting the discharge applied voltage and overheating temperature to simulate defects of different degrees, the content and variation law of each characteristic component are continuously detected. It should be noted that the characteristic components including NO2, CO, O3, NO, CH4 and CH2O are determined in step 1. In order to control costs and reduce the volume of the detection device, the characteristic components detected in step 2 are mainly NO2, CO, O3 and NO. According to different defect degrees and different defect types, the content and variation law of characteristic components under different conditions are obtained. At the same time, for defects of the same type and degree, the ambient temperature and humidity conditions are changed to obtain the types, contents and variation laws of characteristic components under different ambient temperature and humidity conditions. It should be noted that this step is a simulation test, and its volume is basically consistent with the actual switch cabinet compartment, but there are still differences in structure and component distribution from the physical switch cabinet. Therefore, in step 3, the content differences of characteristic components in different areas of the switch cabinet are explored.

[0078] Step 3: Obtain the diffusion characteristics of the characteristic components inside the switch cabinet, so as to obtain the optimal detection position during the actual detection of the switch cabinet (corresponding to the target detection position mentioned above). The most important parameters for state detection of the air switch cabinet based on characteristic components are the content and change law of each characteristic component, and the content distribution of gas in various areas inside the switch cabinet is different. Therefore, in order to improve the accuracy of the state diagnosis results, it is necessary to determine the detection position. First, construct a three-dimensional model of the physical switch cabinet, and set a gas source at a position prone to defects. The characteristic gas generation position set in this embodiment is the C-phase branch busbar, and the gas production rate is 1*10 -10 kg / s, this value is obtained based on the data in step 2. Based on the mass conservation equation, momentum conservation equation, energy conservation equation and component transport equation, the diffusion characteristics of NO2 and CO are simulated and calculated, and it is found that NO2 gas generally diffuses downward and gradually distributes to various areas in the compartment, while CO gas diffuses upward and gradually distributes evenly. Due to the differences in diffusion coefficient and gas production rate, the gas mole fractions of NO2 gas at the three measuring points vary greatly, and the numerical values ​​are The difference between the upper and middle measuring points of CO gas is small, but the difference between the upper and middle measuring points is large, and the numerical value is Simulation calculations cannot completely simulate the real situation. At the same time, in order to enrich and improve the changes in the content of characteristic components under different defects, different defects are set at different positions inside the switch cabinet, and different detection points are set at different positions on the surface of the switch cabinet. The content and type of characteristic components measured at each detection position within a certain period of time are statistically analyzed to obtain the detection area with the most types and the highest content of characteristic components. Moreover, this embodiment has statistically analyzed the content of characteristic components at different measurement points under various defect positions. The results show that the measurement points in the bottom area of ​​the switch cabinet have the highest content of characteristic components as a whole, which can be more effectively applied to actual field applications. For example, Figure 3 Schematic diagram of the best detection position for state detection of air-insulated switchgear in this embodiment. Figure 3 As shown, the shaded portion of the bottom area of ​​each compartment of the switch cabinet can be used as the optimal detection position (corresponding to the above-mentioned target detection position) during actual detection of the switch cabinet.

[0079] Step 4: Construct a characteristic gas atlas library under switch cabinet insulation defects, and construct a state evaluation model (corresponding to the above-mentioned target evaluation model). Collect the characteristic component content data in steps 2 and 3, and build an association relationship with the defect type and defect degree to form a switch cabinet insulation state evaluation model (corresponding to the above-mentioned target evaluation model). First, the detected characteristic component content is used as the original data, and the feature expansion is performed. The expanded characteristic parameters include the content value of each characteristic component, the ratio of the content between each characteristic component, and the gas production rate, etc. Then, the first five characteristic parameters that contribute the most to the recognition result are selected by the feature selection method as the input parameters of the evaluation model (corresponding to the above-mentioned target evaluation model). The input parameters selected in this embodiment are NO2 / O3, NO2, NO2 / SUM, CO / SUM and NO2 / CO. The BO-SVM algorithm (a support vector machine algorithm based on Bayesian optimization) is trained through a large number of data sets, and the recognition accuracy is improved by changing the hyperparameters and the number of iterations. When the recognition accuracy reaches more than 95%, it can be considered that this evaluation model (corresponding to the above-mentioned target evaluation model) can be applied to switch cabinet state detection. The trained model (corresponding to the target evaluation model) is placed on the detection device (corresponding to Figure 4 In the identification and diagnosis module of the air-insulated switchgear state detection device shown in FIG. Figure 4 is a schematic diagram of an optional air-insulated switch cabinet state detection device provided according to an embodiment of the present application, that is, the structure of the detection device is as follows Figure 4As shown, the detection device includes a gas detection module, an identification and diagnosis module, an operation and maintenance strategy recommendation module and a human-computer interaction interface. The gas detection module in this embodiment can be composed of a sampling probe, a built-in air pump, a gas sensor, a temperature and humidity sensor and a data processing unit, wherein the sampling probe can be composed of a high stainless steel sampling handle, a dust filter, and a long-distance telescopic hose, which is convenient for operators to operate flexibly according to the on-site equipment environment; the gas sensors recommended in this embodiment are electrochemical sensors for target gases NO2, CO and NO and ultraviolet sensors for target gas O3. Preferably, the sensor range is 20μL / L and the resolution is 0.001μL / L; the temperature and humidity sensor is used to detect the temperature and humidity in the air-insulated switch cabinet environment. The data processing unit converts the gas component signal of the gas sensor into an electrical signal, and then converts it into a concentration value and a concentration ratio between each component as an input parameter of the identification and diagnosis module. The identification and diagnosis module embeds the trained BO-SVM evaluation model (corresponding to the above-mentioned target evaluation model) into it, and intelligently outputs the defect type according to the component concentration information detected by the gas detection module, and the output defect type may include local discharge defects and local overheating defects. Furthermore, local discharge defects can be divided into surface discharge, internal discharge, corona discharge and other defect types, and according to the identification results (the defect type and defect degree of the identified insulation defects), the operator can specifically find the location where the defect may occur to improve the maintenance efficiency. The operation and maintenance strategy recommendation module intelligently recommends corresponding operation and maintenance strategies according to the degree of defects to guide operators and on-site staff to perform operations, and a recommended criterion is provided in this embodiment, as shown in Table 1, and Table 1 is a table of optional defect degree criteria provided according to this embodiment. In addition, as Figure 4 The air-insulated switchgear state detection device shown also has an alarm function. When the component concentration reaches the warning value, it will alarm through sound, light and other signals to prompt the operator. In addition, the human-computer interaction interface can change the component concentration display, such as units, signal change curves, etc., and can change the detection mode, such as long-term monitoring and timing detection modes. At the same time, it can also perform equipment self-test, zero point calibration, data storage, reading and other functions.

[0080] Table 1

[0081]

[0082]

[0083] In summary, the detection method of air-insulated switchgear provided in the embodiment of the present application detects whether there is a characteristic gas in the air-insulated switchgear based on the type information of the gas in the air-insulated switchgear; if there is a characteristic gas in the air-insulated switchgear, it is determined that there is a target defect in the air-insulated switchgear, and the gas concentration information in the air-insulated switchgear is obtained, wherein the gas concentration information includes at least one of the following: the concentration value of the characteristic gas and the concentration ratio between each characteristic gas; the gas concentration information in the air-insulated switchgear is input into the target evaluation model for processing to obtain the defect type and defect degree of the target defect in the air-insulated switchgear, wherein the target evaluation model is a model constructed based on the Bayesian optimization algorithm and the support vector machine algorithm, which solves the problem of low accuracy in detecting whether there are defects such as local discharge and local overheating inside the air-insulated switchgear in the related art. By detecting the characteristic gas in the air-insulated switchgear and inputting the concentration information of the detected characteristic components in the air-insulated switchgear into the state assessment model for processing, the defect type and defect degree of the insulation defects inside the air-insulated switchgear are intelligently output. The state assessment model is a model constructed based on two intelligent algorithms, the Bayesian optimization algorithm and the support vector machine algorithm. This can avoid the situation in which the pulse current method used in the related technology to detect defects inside the air-insulated switchgear is easily affected by electromagnetic interference factors, resulting in low detection accuracy, thereby achieving the effect of improving the accuracy of detecting whether there are defects such as local discharge and local overheating inside the air-insulated switchgear.

[0084] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0085] The embodiment of the present application also provides a detection device for an air-insulated switch cabinet. It should be noted that the detection device for an air-insulated switch cabinet in the embodiment of the present application can be used to execute the detection method for an air-insulated switch cabinet provided in the embodiment of the present application. The detection device for an air-insulated switch cabinet provided in the embodiment of the present application is introduced below.

[0086] Figure 5 Schematic diagram of a detection device for an air-insulated switchgear provided according to an embodiment of the present application. Figure 5 As shown, the device includes: a first detection unit 501, a first processing unit 502 and a second processing unit 503.

[0087] Specifically, the first detection unit 501 is used to detect whether there is a characteristic gas in the air-insulated switchgear according to the type information of the gas in the air-insulated switchgear;

[0088] The first processing unit 502 is configured to determine that a target defect exists in the air-insulated switchgear if a characteristic gas exists in the air-insulated switchgear, and obtain gas concentration information in the air-insulated switchgear, wherein the gas concentration information includes at least one of the following: a concentration value of the characteristic gas and a concentration ratio between each characteristic gas;

[0089] The second processing unit 503 is used to input the gas concentration information in the air-insulated switchgear into the target evaluation model for processing to obtain the defect type and defect degree of the target defect in the air-insulated switchgear, wherein the target evaluation model is a model constructed based on the Bayesian optimization algorithm and the support vector machine algorithm.

[0090] In summary, the detection device for air-insulated switchgear provided in the embodiment of the present application detects whether there is a characteristic gas in the air-insulated switchgear according to the type information of the gas in the air-insulated switchgear through the first detection unit 501; if the characteristic gas exists in the air-insulated switchgear, the first processing unit 502 determines that there is a target defect in the air-insulated switchgear, and obtains the gas concentration information in the air-insulated switchgear, wherein the gas concentration information includes at least one of the following: the concentration value of the characteristic gas and the concentration ratio between each characteristic gas; the second processing unit 503 inputs the gas concentration information in the air-insulated switchgear into the target evaluation model for processing, and obtains the defect type and defect degree of the target defect in the air-insulated switchgear, wherein the target evaluation model is a model constructed based on the Bayesian optimization algorithm and the support vector machine algorithm, which solves the problem of low accuracy in detecting whether there are defects such as local discharge and local overheating inside the air-insulated switchgear in the related art. By detecting the characteristic gas in the air-insulated switchgear and inputting the concentration information of the detected characteristic components in the air-insulated switchgear into the state assessment model for processing, the defect type and defect degree of the insulation defects inside the air-insulated switchgear are intelligently output. The state assessment model is a model constructed based on two intelligent algorithms, the Bayesian optimization algorithm and the support vector machine algorithm. This can avoid the situation in which the pulse current method used in the related technology to detect defects inside the air-insulated switchgear is easily affected by electromagnetic interference factors, resulting in low detection accuracy, thereby achieving the effect of improving the accuracy of detecting whether there are defects such as local discharge and local overheating inside the air-insulated switchgear.

[0091] Optionally, in the detection device of the air-insulated switchgear provided in the embodiment of the present application, the target evaluation model is obtained by the following units: a first determination unit, used to obtain a training data set based on multiple concentration values ​​of characteristic gases in the air-insulated switchgear, wherein the multiple concentration values ​​include at least: the concentration values ​​of characteristic gases in the air-insulated switchgear under different defect types and different defect degrees, and the concentration values ​​of characteristic gases at different detection positions in the air-insulated switchgear; a first construction unit, used to construct an original evaluation model for evaluating defects in the air-insulated switchgear based on a Bayesian optimization algorithm and a support vector machine algorithm; and a first training unit, used to use the training data set to learn and train the original evaluation model to obtain a target evaluation model.

[0092] Optionally, in the detection device of the air-insulated switchgear provided in the embodiment of the present application, the first determination unit includes: a first processing module, used to perform feature expansion processing on multiple concentration values ​​of characteristic gases in the air-insulated switchgear to obtain processed gas concentration information, wherein the processed gas concentration information at least includes: the concentration value of the characteristic gas and the concentration ratio between each characteristic gas; a second processing module, used to perform feature selection processing on the processed gas concentration information to obtain multiple concentration information, wherein the multiple concentration information is concentration information selected from the processed gas concentration information with an importance higher than a preset importance; the first determination module is used to obtain a training data set based on the multiple concentration information.

[0093] Optionally, in the detection device of the air-insulated switchgear provided in the embodiment of the present application, the first processing unit includes: a third processing module, used to determine the target detection position, and collect characteristic gases in the air-insulated switchgear based on the target detection position; a second determination module, used to determine the concentration signal of the characteristic gas in the air-insulated switchgear based on the characteristic gas collected from the air-insulated switchgear; and a fourth processing module, used to convert the concentration signal of the characteristic gas in the air-insulated switchgear into an electrical signal, and obtain the gas concentration information in the air-insulated switchgear based on the electrical signal.

[0094] Optionally, in the detection device of the air-insulated switchgear provided in the embodiment of the present application, the third processing module includes: a first acquisition submodule, used to acquire the concentration value of the characteristic gas at different detection positions in the air-insulated switchgear; a first determination submodule, used to determine the detection area with the most types of characteristic gases and the highest concentration of characteristic gases based on the concentration values ​​of the characteristic gases at different detection positions in the air-insulated switchgear; and a second determination submodule, used to determine the target detection position based on the detection area with the most types of characteristic gases and the highest concentration of characteristic gases.

[0095] Optionally, in the detection device for air-insulated switchgear provided in an embodiment of the present application, the device also includes: a second determination unit, for inputting the gas concentration information in the air-insulated switchgear into a target evaluation model for processing, and obtaining the defect type and defect degree of the target defect in the air-insulated switchgear, and then determining a target strategy for performing maintenance processing on the air-insulated switchgear according to the defect type and defect degree of the target defect in the air-insulated switchgear; and a first sending unit, for sending the target strategy to a target object, and prompting the target object to perform maintenance processing on the air-insulated switchgear according to the target strategy.

[0096] Optionally, in the detection device for air-insulated switchgear provided in the embodiment of the present application, the device also includes: a first judgment unit, used to input the gas concentration information in the air-insulated switchgear into the target evaluation model for processing, and after obtaining the defect type and defect degree of the target defect in the air-insulated switchgear, determine whether the defect degree is higher than a preset defect degree; a second sending unit, used to send an alarm prompt information to the target object if the defect degree is higher than the preset defect degree, wherein the alarm prompt information is used to prompt the target object to perform maintenance on the air-insulated switchgear; and a second detection unit, used to continue to detect whether there is a target defect in the air-insulated switchgear if the defect degree is not higher than the preset defect degree.

[0097] The detection device of the air-insulated switchgear includes a processor and a memory. The above-mentioned first detection unit 501, first processing unit 502 and second processing unit 503 are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions.

[0098] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the accuracy of detecting whether there are defects such as partial discharge and local overheating inside the air-insulated switchgear can be improved by adjusting the kernel parameters.

[0099] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0100] An embodiment of the present invention provides a computer-readable storage medium on which a program is stored. When the program is executed by a processor, the detection method of the air-insulated switchgear is implemented.

[0101] An embodiment of the present invention provides a processor, which is used to run a program, wherein the detection method of the air-insulated switchgear is executed when the program is run.

[0102] like Figure 6 As shown, an embodiment of the present invention provides an electronic device, the device includes a processor, a memory and a program stored in the memory and executable on the processor, and the processor implements the following steps when executing the program: based on the type information of the gas in the air-insulated switch cabinet, detecting whether there is a characteristic gas in the air-insulated switch cabinet; if there is a characteristic gas in the air-insulated switch cabinet, determining that there is a target defect in the air-insulated switch cabinet, and obtaining the gas concentration information in the air-insulated switch cabinet, wherein the gas concentration information includes at least one of the following: the concentration value of the characteristic gas and the concentration ratio between each characteristic gas; inputting the gas concentration information in the air-insulated switch cabinet into the target evaluation model for processing, and obtaining the defect type and defect degree of the target defect in the air-insulated switch cabinet, wherein the target evaluation model is a model constructed based on the Bayesian optimization algorithm and the support vector machine algorithm.

[0103] When the processor executes the program, the following steps are also implemented: the target evaluation model is obtained in the following manner: a training data set is obtained based on multiple concentration values ​​of characteristic gases in the air-insulated switchgear, wherein the multiple concentration values ​​at least include: concentration values ​​of characteristic gases in the air-insulated switchgear under different defect types and different defect degrees, and concentration values ​​of characteristic gases at different detection positions in the air-insulated switchgear; based on the Bayesian optimization algorithm and the support vector machine algorithm, an original evaluation model for evaluating defects in the air-insulated switchgear is constructed; and the training data set is used to learn and train the original evaluation model to obtain a target evaluation model.

[0104] When the processor executes the program, the following steps are also implemented: based on multiple concentration values ​​of characteristic gases in the air-insulated switchgear, a training data set is obtained, including: performing feature expansion processing on the multiple concentration values ​​of characteristic gases in the air-insulated switchgear to obtain processed gas concentration information, wherein the processed gas concentration information at least includes: the concentration value of the characteristic gas and the concentration ratio between each characteristic gas; performing feature selection processing on the processed gas concentration information to obtain multiple concentration information, wherein the multiple concentration information is concentration information selected from the processed gas concentration information with an importance higher than a preset importance; based on the multiple concentration information, a training data set is obtained.

[0105] When the processor executes the program, the following steps are also implemented: Obtaining the gas concentration information in the air-insulated switchgear includes: determining the target detection position, and collecting characteristic gases in the air-insulated switchgear based on the target detection position; determining the concentration signal of the characteristic gas in the air-insulated switchgear based on the characteristic gas collected from the air-insulated switchgear; converting the concentration signal of the characteristic gas in the air-insulated switchgear into an electrical signal, and obtaining the gas concentration information in the air-insulated switchgear based on the electrical signal.

[0106] When the processor executes the program, the following steps are also implemented: determining the target detection position includes: obtaining the concentration values ​​of the characteristic gases at different detection positions in the air-insulated switchgear; determining the detection area with the most types of characteristic gases and the largest concentration of characteristic gases based on the concentration values ​​of the characteristic gases at different detection positions in the air-insulated switchgear; determining the target detection position based on the detection area with the most types of characteristic gases and the largest concentration of characteristic gases.

[0107] When the processor executes the program, the following steps are also implemented: after the gas concentration information in the air-insulated switchgear is input into the target evaluation model for processing and the defect type and defect degree of the target defect in the air-insulated switchgear are obtained, it also includes: determining the target strategy for repairing the air-insulated switchgear based on the defect type and defect degree of the target defect in the air-insulated switchgear; sending the target strategy to the target object, and prompting the target object to repair the air-insulated switchgear based on the target strategy.

[0108] When the processor executes the program, the following steps are also implemented: after the gas concentration information in the air-insulated switchgear is input into the target evaluation model for processing and the defect type and defect degree of the target defect in the air-insulated switchgear are obtained, it also includes: judging whether the defect degree is higher than the preset defect degree; if the defect degree is higher than the preset defect degree, sending an alarm prompt information to the target object, wherein the alarm prompt information is used to prompt the target object to perform maintenance on the air-insulated switchgear; if the defect degree is not higher than the preset defect degree, continuing to detect whether there is a target defect in the air-insulated switchgear.

[0109] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0110] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program that is initialized with the following method steps: based on the type information of the gas in the air-insulated switchgear, detecting whether there is a characteristic gas in the air-insulated switchgear; if there is a characteristic gas in the air-insulated switchgear, determining that there is a target defect in the air-insulated switchgear, and obtaining the gas concentration information in the air-insulated switchgear, wherein the gas concentration information includes at least one of the following: the concentration value of the characteristic gas and the concentration ratio between each characteristic gas; inputting the gas concentration information in the air-insulated switchgear into a target evaluation model for processing, and obtaining the defect type and defect degree of the target defect in the air-insulated switchgear, wherein the target evaluation model is a model constructed based on the Bayesian optimization algorithm and the support vector machine algorithm.

[0111] When executed on a data processing device, it is also suitable for executing a program initialized with the following method steps: a target evaluation model is obtained in the following manner: a training data set is obtained based on multiple concentration values ​​of characteristic gases in an air-insulated switchgear, wherein the multiple concentration values ​​at least include: concentration values ​​of characteristic gases in the air-insulated switchgear under different defect types and different defect degrees, and concentration values ​​of characteristic gases at different detection positions in the air-insulated switchgear; based on a Bayesian optimization algorithm and a support vector machine algorithm, an original evaluation model for evaluating defects in an air-insulated switchgear is constructed; and the training data set is used to learn and train the original evaluation model to obtain a target evaluation model.

[0112] When executed on a data processing device, it is also suitable for executing an initialized program having the following method steps: obtaining a training data set based on multiple concentration values ​​of characteristic gases in an air-insulated switchgear, including: performing feature expansion processing on the multiple concentration values ​​of characteristic gases in the air-insulated switchgear to obtain processed gas concentration information, wherein the processed gas concentration information at least includes: the concentration value of the characteristic gas and the concentration ratio between each characteristic gas; performing feature selection processing on the processed gas concentration information to obtain multiple concentration information, wherein the multiple concentration information is concentration information selected from the processed gas concentration information with an importance higher than a preset importance; obtaining a training data set based on the multiple concentration information.

[0113] When executed on a data processing device, it is also suitable for executing a program initialized with the following method steps: obtaining gas concentration information in an air-insulated switch cabinet includes: determining a target detection position, and collecting characteristic gases in the air-insulated switch cabinet based on the target detection position; determining a concentration signal of the characteristic gases in the air-insulated switch cabinet based on the characteristic gases collected from the air-insulated switch cabinet; converting the concentration signal of the characteristic gases in the air-insulated switch cabinet into an electrical signal, and obtaining the gas concentration information in the air-insulated switch cabinet based on the electrical signal.

[0114] When executed on a data processing device, it is also suitable for executing a program initialized with the following method steps: determining the target detection position includes: obtaining the concentration value of the characteristic gas at different detection positions in the air-insulated switchgear; determining the detection area with the most types of characteristic gases and the largest concentration of characteristic gases based on the concentration values ​​of the characteristic gases at different detection positions in the air-insulated switchgear; determining the target detection position based on the detection area with the most types of characteristic gases and the largest concentration of characteristic gases.

[0115] When executed on a data processing device, it is also suitable for executing an initialization program having the following method steps: after inputting the gas concentration information in the air-insulated switchgear into the target assessment model for processing and obtaining the defect type and defect degree of the target defect in the air-insulated switchgear, it also includes: determining a target strategy for repairing the air-insulated switchgear based on the defect type and defect degree of the target defect in the air-insulated switchgear; sending the target strategy to the target object, and prompting the target object to repair the air-insulated switchgear based on the target strategy.

[0116] When executed on a data processing device, it is also suitable for executing an initialization program having the following method steps: after inputting the gas concentration information in the air-insulated switchgear into the target evaluation model for processing and obtaining the defect type and defect degree of the target defect in the air-insulated switchgear, it also includes: judging whether the defect degree is higher than the preset defect degree; if the defect degree is higher than the preset defect degree, sending an alarm prompt information to the target object, wherein the alarm prompt information is used to prompt the target object to perform maintenance on the air-insulated switchgear; if the defect degree is not higher than the preset defect degree, continuing to detect whether there is a target defect in the air-insulated switchgear.

[0117] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0118] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0119] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0120] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0121] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0122] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0123] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0124] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0125] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0126] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A method for detecting an air-insulated switchgear, characterized in that: include: Detecting whether characteristic gas exists in the air-insulated switchgear according to the type information of the gas in the air-insulated switchgear; If there is a characteristic gas in the air-insulated switchgear, it is determined that there is a target defect in the air-insulated switchgear, and gas concentration information in the air-insulated switchgear is obtained, wherein the gas concentration information includes at least one of the following: a concentration value of the characteristic gas and a concentration ratio between each characteristic gas; The gas concentration information in the air-insulated switchgear is input into a target evaluation model for processing to obtain the defect type and defect degree of the target defect in the air-insulated switchgear, wherein the target evaluation model is a model constructed based on a Bayesian optimization algorithm and a support vector machine algorithm.

2. The method according to claim 1, characterized in that The target evaluation model is obtained by: A training data set is obtained according to a plurality of concentration values ​​of characteristic gases in the air-insulated switchgear, wherein the plurality of concentration values ​​at least include: concentration values ​​of characteristic gases in the air-insulated switchgear under different defect types and different defect degrees, and concentration values ​​of characteristic gases at different detection positions in the air-insulated switchgear; Based on Bayesian optimization algorithm and support vector machine algorithm, an original evaluation model for evaluating defects in air-insulated switchgear is constructed; The original evaluation model is trained using the training data set to obtain the target evaluation model.

3. The method according to claim 2, characterized in that According to multiple concentration values ​​of characteristic gases in the air-insulated switchgear, a training data set is obtained including: Performing feature expansion processing on multiple concentration values ​​of characteristic gases in the air-insulated switchgear to obtain processed gas concentration information, wherein the processed gas concentration information at least includes: concentration values ​​of characteristic gases and concentration ratios between characteristic gases; Performing feature selection processing on the processed gas concentration information to obtain a plurality of concentration information, wherein the plurality of concentration information is concentration information selected from the processed gas concentration information and having an importance higher than a preset importance; The training data set is obtained according to the multiple concentration information.

4. The method according to claim 1, characterized in that Acquiring the gas concentration information in the air-insulated switchgear includes: Determine a target detection position, and collect characteristic gas in the air-insulated switchgear according to the target detection position; Determining a concentration signal of the characteristic gas in the air-insulated switchgear based on the characteristic gas collected from the air-insulated switchgear; The concentration signal of the characteristic gas in the air-insulated switch cabinet is converted into an electrical signal, and the gas concentration information in the air-insulated switch cabinet is obtained based on the electrical signal.

5. The method according to claim 4, characterized in that Determining the target detection location includes: Acquiring concentration values ​​of characteristic gases at different detection positions in the air-insulated switchgear; According to the concentration values ​​of the characteristic gases at different detection positions in the air-insulated switchgear, determining the detection area with the most types of characteristic gases and the highest concentration of characteristic gases; The target detection position is determined based on the detection area with the most types of characteristic gases and the highest concentration of characteristic gases.

6. The method according to claim 1, characterized in that After inputting the gas concentration information in the air-insulated switchgear into the target evaluation model for processing to obtain the defect type and defect degree of the target defect in the air-insulated switchgear, the method further includes: Determining a target strategy for performing maintenance processing on the air-insulated switchgear according to the defect type and defect degree of the target defect in the air-insulated switchgear; The target strategy is sent to a target object, and the target object is prompted to perform maintenance processing on the air-insulated switchgear according to the target strategy.

7. The method according to claim 1, characterized in that After inputting the gas concentration information in the air-insulated switchgear into the target evaluation model for processing to obtain the defect type and defect degree of the target defect in the air-insulated switchgear, the method further includes: Determining whether the defect degree is higher than a preset defect degree; If the defect degree is higher than the preset defect degree, an alarm prompt message is sent to the target object, wherein the alarm prompt message is used to prompt the target object to perform maintenance on the air insulated switchgear; If the defect level is not higher than the preset defect level, continue to detect whether the target defect exists in the air-insulated switchgear.

8. A detection device for an air-insulated switch cabinet, characterized in that: include: A first detection unit, configured to detect whether a characteristic gas exists in the air-insulated switchgear according to type information of the gas in the air-insulated switchgear; A first processing unit is configured to determine that a target defect exists in the air-insulated switchgear if a characteristic gas exists in the air-insulated switchgear, and obtain gas concentration information in the air-insulated switchgear, wherein the gas concentration information includes at least one of the following: a concentration value of the characteristic gas and a concentration ratio between each characteristic gas; The second processing unit is used to input the gas concentration information in the air-insulated switchgear into the target evaluation model for processing to obtain the defect type and defect degree of the target defect in the air-insulated switchgear, wherein the target evaluation model is a model constructed based on the Bayesian optimization algorithm and the support vector machine algorithm.

9. A computer-readable storage medium, characterized in that: The storage medium stores a program, wherein the program executes the air-insulated switchgear detection method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the detection method for air-insulated switchgear as described in any one of claims 1 to 7.

Images