Fault diagnosis positioning method and system for oil tank assembly of converter transformer

Through the combination of multimodal data, digital twins and artificial intelligence technology, the internal defect diagnosis problem of converter transformer fuel tank components is solved, and a fault inference flow chart is formed, which can accurately locate and improve the efficiency of the fault.

CN118913372APending Publication Date: 2024-11-08GLOBAL ENERGY INTERCONNECTION RES INST CO LTD +6
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411292665.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-14
Publication Date
2024-11-08

AI Technical Summary

Technical Problem

The prior art is difficult to effectively diagnose and locate internal defects of converter transformer oil tank components, such as metal suspension discharge, abnormal temperature of oil tanks and metal structural parts, and oil leakage, resulting in difficult evaluation of operating status and fault handling.

Method used

Multimodal data is used to combine digital twins and artificial intelligence technology, and through online monitoring and power outage tests, gas content, fuel tank appearance images and hot spot temperature data are collected, fault diagnosis and maintenance suggestions are carried out, and fault inference flow chart is formed.

Benefits of technology

It realizes accurate diagnosis and positioning of the oil tank assembly fault of the converter transformer, improves the fault positioning efficiency, provides reasonable maintenance suggestions, and enhances the operating status evaluation and fault resolution capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118913372A_ABST
    Figure CN118913372A_ABST
Patent Text Reader

Abstract

The invention belongs to the technical field of power equipment operation and maintenance, and provides a converter transformer oil tank assembly fault diagnosis positioning method and system, and the method comprises the steps: 1, collecting the gas content in the insulating oil of a converter transformer, the appearance image information of an oil tank, and the hot-spot temperature of the oil tank; 2, diagnosing whether metal suspension discharge in the oil tank has defects or not based on the gas content; 3, diagnosing the temperatures of the oil tank and the metal structure based on the hot-spot temperature of the oil tank; and 4, performing oil leakage trace image recognition on the appearance image information of the oil tank through artificial intelligence, and diagnosing oil leakage of the oil tank. According to multi-modal data such as online monitoring and live monitoring data of the converter transformer, fault diagnosis and fault positioning of the oil tank assembly are derived step by step in combination with digital twinning and artificial intelligence technologies, maintenance suggestions are proposed, and a fault reasoning flow chart of the oil tank assembly of the converter transformer is formed; therefore, a user can directly observe the defects in the oil tank, and the fault positioning efficiency is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of power equipment operation and maintenance, and more specifically, relates to a method and system for diagnosing and locating faults of a converter transformer oil tank assembly. Background Art

[0002] The oil tank is the outer shell of the converter transformer. The iron core and coil are installed in the oil tank and filled with transformer oil. The transformer oil mainly plays the role of insulation and heat dissipation. The health status of the oil tank components will directly affect the safe and stable operation of the converter transformer. During the operation of the converter transformer, the oil tank components are prone to defects such as metal suspension discharge, abnormal temperature of the oil tank and metal structural parts, and oil leakage. However, the defects inside the oil tank cannot be directly observed and need to be inferred and diagnosed through online monitoring and live detection methods.

[0003] Therefore, how to effectively and correctly perform reasoning diagnosis on fuel tank components is an urgent problem that needs to be solved. Summary of the invention

[0004] In order to solve the deficiencies in the prior art, the present invention provides a method and system for diagnosing and locating the status fault of a converter transformer oil tank assembly, which combines multimodal data such as online monitoring and power outage tests to perform defect diagnosis, fault location and maintenance suggestions, and construct a fault reasoning flowchart.

[0005] The present invention adopts the following technical solution. A first aspect of the present invention provides a method for diagnosing and locating a fault of a converter transformer oil tank assembly, comprising:

[0006] Step 1, collecting the gas content in the insulating oil of the converter transformer, the oil tank appearance image information and the oil tank hot spot temperature, wherein the gas content includes C2H2 content, C2H4 content, CH4 content, H2 content and C2H6 content;

[0007] Step 2, based on the C2H2 content, C2H4 content, CH4 content, H2 content and C2H6 content in step 1, diagnose whether the metal suspension discharge inside the fuel tank is defective, when C2H2 / C2H4≥3 and the C2H2 content exceeds the attention value, determine that the fuel tank component fault is metal suspension discharge;

[0008] Step 3: Based on the hot spot temperature of the fuel tank obtained in step 1, diagnose the temperature of the fuel tank and the metal structure. When the hot spot temperature of the fuel tank is greater than the temperature threshold, determine that the fuel tank component failure is due to abnormal temperature of the fuel tank and the metal structure. The temperature threshold is obtained through digital twin calculation.

[0009] Step 4: Use artificial intelligence to perform oil leakage trace image recognition on the fuel tank appearance image information obtained in step 1. If the image recognition result is oil leakage, it is determined that the fuel tank component failure is fuel tank leakage.

[0010] Preferably, in the case where the oil tank component failure is oil tank leakage, if the total area of ​​oil stains on the outer wall of the transformer and the ground around the transformer is greater than a first threshold, the oil tank component failure is oil leakage in the secondary oil tank; if the total area of ​​oil stains on the outer wall of the transformer and the ground around the transformer is greater than a second threshold, the oil tank component failure is oil leakage in the primary oil tank; wherein the first threshold is less than the second threshold.

[0011] Preferably, in step 3, digital twin technology is used to simulate the complex situation inside the transformer; based on various parameters of the transformer and various sensor data, the temperature threshold of the hot spot temperature of the oil tank in different operating states is determined.

[0012] Preferably, step 4 includes: performing convolution, pooling and full connection operations on the oil tank appearance image to extract the feature vector of the abnormal area; if the matching degree between the feature vector of the abnormal area and the feature vector of the oil leakage defect exceeds a matching degree threshold, determining that the image recognition result is oil leakage, and determining that the oil tank component failure is oil tank leakage.

[0013] Preferably, step 1 further comprises encoding the gas content to obtain a gas content code, wherein the first digit of the gas content code is a C2H2 / C2H4 code, the second digit is a CH4 / H2 code, and the third digit is a C2H4 / C2H6 code.

[0014] Preferably, encoding the gas content comprises:

[0015] When C2H2 / C2H4<0.1, its coding value is 0; CH4 / H2<0.1, its coding value is 1; C2H4 / C2H6<0.1, its coding value is 0;

[0016] When 0.1≤C2H2 / C2H4<1, its code is assigned to 1; 0.1≤CH4 / H2<1, its code is assigned to 0; 0.1≤C2H4 / C2H6<1, its code is assigned to 0;

[0017] When 1≤C2H2 / C2H4<3, its code is assigned to 1; 1≤CH4 / H2<3, its code is assigned to 2; 1≤C2H4 / C2H6<3, its code is assigned to 1;

[0018] When C2H2 / C2H4≥3, its coding value is 2; when CH4 / H2≥3, its coding value is 2; when C2H4 / C2H6≥3, its coding value is 2.

[0019] Preferably, in step 2, whether the metal suspension discharge inside the fuel tank is defective is diagnosed based on the gas content code. When the first digit of the gas content code is 2 and the C2H2 content exceeds the warning value, it is determined that the fuel tank component failure is metal suspension discharge.

[0020] Preferably, the method further comprises:

[0021] Step 5: Determine the status of the fuel tank component and maintenance suggestions based on the fuel tank component fault, and form a displayable fuel tank component fault reasoning flowchart by combining the fault judgment process, fuel tank component status and maintenance suggestions of steps 2, 3 and 4.

[0022] Preferably, in step 5, when the fuel tank component failure is metal suspension discharge and / or abnormal temperature of the fuel tank and metal structural parts and / or oil leakage in the secondary fuel tank, the fuel tank component state is abnormal, and the fuel tank component state is displayed in the fuel tank component failure reasoning flow chart in yellow code form as a reminder;

[0023] When the oil tank component failure is oil leakage in the primary oil tank, the oil tank component status is abnormal, and the oil tank component status is displayed in the oil tank component failure reasoning flowchart in the form of a red code, and a power outage for maintenance is prompted.

[0024] The second aspect of the present invention provides a fault diagnosis and positioning system for a converter transformer oil tank assembly, comprising: an acquisition module, a first determination module, a second determination module and a third determination module. The acquisition module is used to collect the gas content in the insulating oil of the converter transformer, the oil tank appearance image information and the oil tank hot spot temperature, wherein the gas content includes C2H2 content, C2H4 content, CH4 content, H2 content and C2H6 content. The first determination module is used to diagnose whether the metal suspension discharge inside the oil tank is defective based on the gas content. When C2H2 / C2H4≥3 and the C2H2 content exceeds the attention value, the oil tank assembly fault is determined to be metal suspension discharge. The second determination module is used to diagnose the temperature of the oil tank and the metal structure based on the oil tank hot spot temperature. When the oil tank hot spot temperature is greater than the temperature threshold, the oil tank assembly fault is determined to be abnormal temperature of the oil tank and the metal structure; the temperature threshold is obtained by digital twin calculation. The third determination module is used to perform oil leakage trace image recognition on the oil tank appearance image information through artificial intelligence. If the image recognition result is oil leakage, the oil tank assembly fault is determined to be oil leakage.

[0025] A third aspect of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when loaded into the processor, implements the above-mentioned converter transformer oil tank assembly fault diagnosis and positioning method.

[0026] A fourth aspect of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned converter transformer oil tank assembly fault diagnosis and positioning method.

[0027] Compared with the prior art, the beneficial effects of the present invention include at least:

[0028] (1) Based on multimodal data such as online monitoring and power outage test data of converter transformers, the present invention combines digital twin and artificial intelligence technology to gradually derive the status evaluation, fault diagnosis, and fault location of the oil tank assembly, and puts forward further maintenance suggestions, forming a converter transformer oil tank assembly fault reasoning flow chart.

[0029] (2) The present invention combines online oil chromatography data and gas content coding to gradually infer the occurrence of metal suspension discharge defects inside the oil tank, evaluate the operating status of the converter transformer when the defect occurs, and form a metal suspension discharge defect fault inference flow chart inside the oil tank.

[0030] (3) The present invention combines live detection data and digital twin technology to gradually infer the occurrence of abnormal temperature defects in the converter transformer oil tank and metal structural parts, evaluate the operating status of the converter transformer when the defects occur, and form a flow chart for inferring abnormal temperature faults in the oil tank and metal structural parts.

[0031] (4) The present invention utilizes online detection of multimodal data and artificial intelligence technology to deduce the oil leakage defect of the converter transformer tank, evaluate and encode the operating status of the converter transformer, and then provide reasonable maintenance suggestions to form an oil tank leakage defect fault reasoning flow chart.

[0032] (5) The fuel tank oil leakage defect fault reasoning flowchart formed by the present invention can enable users to more intuitively observe the defects occurring inside the fuel tank assembly, thereby improving the efficiency of fault location and fault resolution. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a flowchart schematic diagram of a method for diagnosing and locating a fault of a converter transformer oil tank assembly provided by an embodiment of the present invention;

[0034] Figure 2 It is a schematic diagram of a converter transformer oil tank component defect fault reasoning flow chart provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The described embodiments are only embodiments of a part of the present invention, not all embodiments. Based on the spirit of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the protection scope of the present invention.

[0036] The present invention proposes a method and system for diagnosing and locating the fault of the core clamp lead assembly for the converter transformer. Based on the state quantity data such as the online monitoring and live detection data of the converter transformer, the state evaluation, fault diagnosis and fault location of the core clamp lead assembly are realized by combining digital twin and artificial intelligence means, and further maintenance suggestions are put forward.

[0037] Combination Figure 1 and Figure 2 As shown, embodiment 1 of the present invention provides a method for diagnosing and locating a fault of a converter transformer oil tank assembly, comprising the following steps:

[0038] Step 1: Collect the gas content in the insulating oil of the converter transformer, the oil tank appearance image information and the hot spot temperature of the oil tank, wherein the gas content includes C2H2 content, C2H4 content, CH4 content, H2 content and C2H6 content.

[0039] Step 1 specifically includes:

[0040] Step 1.1, collecting the gas content in the converter transformer insulating oil, including: C2H2 (acetylene), C2H4 (ethylene), CH4 (methane), H2 (hydrogen) and C2H6 (ethane), preferably but not limited to, obtained through online oil chromatography monitoring of the converter transformer.

[0041] Preferably, encoding is performed based on the gas content in the insulating oil of the converter transformer to obtain the gas content encoding. The gas content encoding rules are shown in Table 1:

[0042] Table 1

[0043]

[0044] Among them, the first digit of the gas content code is C2H2 / C2H4 code, the second digit is CH4 / H2 code, and the third digit is C2H4 / C2H6 code.

[0045] More specifically, it can be seen from Table 1 that when C2H2 / C2H4<0.1, its coding value is 0; CH4 / H2<0.1, its coding value is 1; C2H4 / C2H6<0.1, its coding value is 0; when 0.1≤C2H2 / C2H4<1, its coding value is 1; 0.1≤CH4 / H2<1, its coding value is 0; 0.1≤C2H4 / C2H6<1, its coding value is 0; when 1≤C2H2 / C2H4<3, its coding value is 1; 1≤CH4 / H2<3, its coding value is 2; 1≤C2H4 / C2H6<3, its coding value is 1; when C2H2 / C2H4≥3, its coding value is 2; CH4 / H2≥3, its coding value is 2; C2H4 / C2H6≥3, its coding value is 2.

[0046] Step 1.2, collecting the hot spot temperature of the fuel tank, preferably but not limited to, using infrared charging detection to obtain the hot spot temperature of the fuel tank. Step 1.2, collecting the appearance image information of the fuel tank, preferably but not limited to, obtaining it through online monitoring of the appearance of the fuel tank.

[0047] Step 2: Based on the C2H2 content, C2H4 content, CH4 content, H2 content and C2H6 content in step 1, diagnose whether the metal suspension discharge inside the fuel tank is defective. When C2H2 / C2H4≥3 and the C2H2 content exceeds the attention value, it is determined that the fuel tank component failure is metal suspension discharge.

[0048] Specifically, based on the obtained gas content code, a diagnosis is made as to whether there is a defect in the metal suspension discharge inside the fuel tank. Specifically, it includes:

[0049] Based on the gas content code, the nature of the transformer fault is judged, as shown in Table 2:

[0050] Table 2

[0051]

[0052] When the first bit of the gas content code is set to 2, the nature of the transformer fault is judged based on the gas content code, and the fault type is judged as low-energy discharge.

[0053] Preferably, but not restrictively, if the online monitored acetylene content exceeds the warning value and the first bit of the gas content code is set to 2, it can be inferred that a metal suspension discharge defect occurs inside the oil tank. At this time, the operating status of the converter transformer is evaluated as "yellow code" and is in an abnormal operating state.

[0054] Preferably, but not limited to, the noted value is 0.5 μL / L.

[0055] Step 3: Based on the hot spot temperature of the fuel tank obtained in step 1, diagnose the temperature of the fuel tank and the metal structure. When the hot spot temperature of the fuel tank is greater than the temperature threshold, it is determined that the failure of the fuel tank component is abnormal temperature of the fuel tank and the metal structure. The temperature threshold is obtained through digital twin calculation.

[0056] Specifically, based on the obtained hot spot temperature of the oil tank, the temperature of the oil tank and the metal structure is diagnosed. Specifically including:

[0057] Digital twin technology is used to simulate the complex situation inside the transformer, and the most appropriate threshold of the hot spot temperature of the oil tank under different operating conditions is analyzed in real time based on various transformer parameters and various sensor data.

[0058] Specifically, consider the effects of electromagnetic field, temperature field, stress field and other multi-physical fields on the transformer. Transformer parameters, such as oil tank size, oil number, specific heat capacity, thermal conductivity, density, viscosity and thermal diffusion coefficient of the oil tank material, and monitor the temperature of the transformer oil tank wall and the external environment temperature.

[0059] For example, the finite element method is used to calculate the transformer temperature and oil flow distribution, and it is possible to calculate the heat exchange between the oil tank wall and the air, the radiation heat dissipation on the transformer oil tank wall and the heat sink surface, and the influence of gravity acceleration and oil properties on the transformer oil circulation at different ambient temperatures.

[0060] A large oil-immersed transformer is simulated and calculated by setting the number of iterations. The distribution of the internal temperature field and flow field of the oil-immersed transformer and the temperature distribution on the surface of the oil tank and heat sink of the oil-immersed transformer are obtained under different ambient temperatures and different loads, and the most appropriate threshold value of the hot spot temperature of the oil tank under different operating conditions is obtained.

[0061] Furthermore, if the digital twin calculates that the hot spot temperature of the infrared temperature measurement oil tank during power outage monitoring is higher than the temperature threshold D1, it can be inferred that abnormal temperature defects have occurred in the oil tank and metal structural parts, and the operating status of the converter transformer is assessed as "yellow code", which means it is in an abnormal operating state.

[0062] Step 4: Use artificial intelligence to perform oil leakage trace image recognition on the fuel tank appearance image information obtained in step 1. If the image recognition result is oil leakage, it is determined that the fuel tank component failure is fuel tank leakage.

[0063] Based on the image information of the oil tank appearance, the oil leakage defect of the iron oil tank is diagnosed. Specifically including:

[0064] Image recognition and calculation are performed through artificial intelligence, and artificial intelligence models include but are not limited to LeNet-5, AlexNet, VGGNet, SeNet, Transformer, ConViT and other image recognition algorithm models. If image recognition finds traces of oil leakage in the oil tank, it can be determined that the oil tank has an oil leakage defect. At this time, the converter transformer is in an abnormal operating state, and the operating state is evaluated as "yellow code".

[0065] Taking the convolutional neural network (CNN) as an example, the oil tank appearance image is convolved, pooled, fully connected, and other operations are performed to extract the feature vector of the abnormal part. If the feature vector of the abnormal area has a high degree of matching with the feature vector of the known oil leakage defect, it is preliminarily determined that the oil tank has an oil leakage defect. In addition, target detection algorithms such as Faster R-CNN and YOLO (You Only Look Once) series can be further introduced to accurately locate the specific location of the oil leakage, and quickly lock possible leakage points by scanning and evaluating different areas in the image.

[0066] Furthermore, if artificial intelligence image recognition and calculation reveal serious oil leakage in the oil tank, the operating status of the converter transformer will be rated as "red code" and is in a serious state, and the operation and maintenance personnel should promptly carry out power outages.

[0067] Specifically, oil stains are found in the oil tank appearance image. If the total area of ​​the oil stains on the outer wall of the transformer and the ground around the transformer is greater than a first threshold, it is determined that the oil tank has an oil leakage defect, and the oil tank leakage is a second-level oil tank leakage; if the total area of ​​the oil stains on the outer wall of the transformer and the ground around the transformer is greater than a second threshold, it is determined that the oil tank has a serious oil leakage, and the oil tank leakage is a first-level oil tank leakage.

[0068] In a preferred but not limited embodiment, the first threshold is 2 cm 2 , the second threshold is 10cm 2 .

[0069] Preferably, the method further comprises:

[0070] Step 5: Determine the status of the fuel tank component and maintenance suggestions based on the fuel tank component fault, and form a displayable fuel tank component fault reasoning flowchart by combining the fault judgment process, fuel tank component status and maintenance suggestions of steps 2, 3 and 4.

[0071] like Figure 2 As shown, the oil leakage of the fuel tank includes the primary fuel tank leakage and the secondary fuel tank leakage. In step 5, when the fuel tank component failure is metal suspension discharge and / or abnormal temperature of the fuel tank and metal structural parts and / or secondary fuel tank leakage, the fuel tank component state is abnormal, and the fuel tank component state is displayed in the fuel tank component failure reasoning flow chart in yellow code form as a reminder;

[0072] When the oil tank component failure is oil leakage in the primary oil tank, the oil tank component status is abnormal, and the oil tank component status is displayed in the oil tank component failure reasoning flowchart in the form of a red code, and a power outage for maintenance is prompted.

[0073] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0074] Embodiment 2 of the present invention provides a fault diagnosis and positioning system for a converter transformer oil tank assembly, comprising: an acquisition module, a first determination module, a second determination module and a third determination module. The acquisition module is used to collect the gas content in the insulating oil of the converter transformer, the oil tank appearance image information and the oil tank hot spot temperature, wherein the gas content includes C2H2 content, C2H4 content, CH4 content, H2 content and C2H6 content. The first determination module is used to diagnose whether the metal suspension discharge inside the oil tank is defective based on the gas content. When C2H2 / C2H4≥3 and the C2H2 content exceeds the attention value, the oil tank assembly fault is determined to be metal suspension discharge. The second determination module is used to diagnose the temperature of the oil tank and the metal structure based on the oil tank hot spot temperature. When the oil tank hot spot temperature is greater than the temperature threshold, the oil tank assembly fault is determined to be abnormal temperature of the oil tank and the metal structure; the temperature threshold is obtained by digital twin calculation. The third determination module is used to perform oil leakage trace image recognition on the oil tank appearance image information through artificial intelligence. If the image recognition result is oil leakage, the oil tank assembly fault is determined to be oil leakage.

[0075] Embodiment 3 of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the computer program is loaded into the processor, the method for diagnosing and locating faults of a converter transformer oil tank assembly described in Embodiment 1 is implemented.

[0076] Embodiment 4 of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for diagnosing and locating faults of a converter transformer oil tank assembly according to Embodiment 1 is implemented.

[0077] Embodiment 5 of the present invention provides a practical verification example of a method for diagnosing and locating faults of a converter transformer oil tank assembly.

[0078] In this embodiment, the temperature threshold D1 is 85° C. The hot spot temperature of the oil tank of a converter transformer detected by infrared temperature measurement exceeds the threshold D1. According to the fault reasoning flow chart of the converter transformer oil tank component defect, it is inferred that the converter transformer oil tank and metal structural parts have abnormal temperature defects.

[0079] After further inspection on site, the operation and maintenance personnel determined that the cause of the fault was poor contact caused by oxidation of the U-shaped copper busbar surface and rust at the connection of the box cover. Under the superposition of the interphase leakage magnetic flux of the two-column coil, a circulating current was generated between the box cover and the oil tank, causing local overheating. This is consistent with the inference result of the converter transformer oil tank component defect fault reasoning flow chart.

[0080] It is worth noting that the outstanding essential features of the present invention and the significant progress brought to the prior art include at least:

[0081] (1) Based on multimodal data such as online monitoring and live detection data of converter transformers, the present invention combines digital twin and artificial intelligence technology to gradually derive the status evaluation, fault diagnosis, and fault location of the oil tank assembly, and puts forward further maintenance suggestions, forming a converter transformer oil tank assembly fault reasoning flow chart.

[0082] (2) The present invention combines online oil chromatography data and gas content coding to gradually infer the occurrence of metal suspension discharge defects inside the oil tank, evaluate the operating status of the converter transformer when the defect occurs, and form a metal suspension discharge defect fault inference flow chart inside the oil tank.

[0083] (3) The present invention combines live detection data and digital twin technology to gradually infer the occurrence of abnormal temperature defects in the converter transformer oil tank and metal structural parts, evaluate the operating status of the converter transformer when the defects occur, and form a flow chart for inferring abnormal temperature faults in the oil tank and metal structural parts.

[0084] (4) The present invention utilizes online detection of multimodal data and artificial intelligence technology to deduce the oil leakage defect of the converter transformer tank, evaluate and encode the operating status of the converter transformer, and then provide reasonable maintenance suggestions to form an oil tank leakage defect fault reasoning flow chart.

[0085] (5) The fuel tank oil leakage defect fault reasoning flowchart formed by the present invention can enable users to more intuitively observe the defects occurring inside the fuel tank assembly, thereby improving the efficiency of fault location and fault resolution.

[0086] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0087] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for diagnosing and locating faults of a converter transformer oil tank assembly, characterized in that: include: Step 1, collecting the gas content in the insulating oil of the converter transformer, the oil tank appearance image information and the oil tank hot spot temperature, wherein the gas content includes C2H2 content, C2H4 content, CH4 content, H2 content and C2H6 content; Step 2: Based on the C2H2 content, C2H4 content, CH4 content, H2 content and C2H6 content in step 1, diagnose whether the metal suspension discharge inside the fuel tank is defective; when C2H2 / C2H4≥3 and the C2H2 content exceeds the attention value, determine that the fuel tank component fault is metal suspension discharge; Step 3: Based on the hot spot temperature of the fuel tank obtained in step 1, diagnose the temperature of the fuel tank and the metal structure. When the hot spot temperature of the fuel tank is greater than the temperature threshold, determine that the fuel tank component failure is due to abnormal temperature of the fuel tank and the metal structure. The temperature threshold is obtained through digital twin calculation. Step 4: Use artificial intelligence to perform oil leakage trace image recognition on the fuel tank appearance image information obtained in step 1. If the image recognition result is oil leakage, it is determined that the fuel tank component failure is fuel tank leakage.

2. The method for diagnosing and locating a converter transformer oil tank assembly fault according to claim 1, characterized in that: In step 3, the transformer is simulated using digital twin technology; based on the transformer parameters and various sensor data, the temperature threshold of the hot spot temperature of the oil tank under different operating conditions is determined.

3. The method for diagnosing and locating a converter transformer oil tank assembly fault according to claim 1, characterized in that: Step 4 includes: Perform convolution, pooling, and full connection operations on the appearance image of the fuel tank to extract the feature vector of the abnormal area; If the matching degree between the feature vector of the abnormal area and the feature vector of the oil leakage defect exceeds the matching degree threshold, the image recognition result is determined to be an oil leakage, and the oil tank component failure is determined to be an oil tank leakage.

4. The method for diagnosing and locating a converter transformer oil tank assembly fault according to claim 3 is characterized in that: In the case where the oil tank component failure is oil tank leakage, if the total area of ​​oil stains on the outer wall of the transformer and the ground around the transformer is greater than the first threshold, the oil tank component failure is secondary oil tank leakage; if the total area of ​​oil stains on the outer wall of the transformer and the ground around the transformer is greater than the second threshold, the oil tank component failure is primary oil tank leakage; wherein the first threshold is less than the second threshold.

5. The method for diagnosing and locating a converter transformer oil tank assembly fault according to claim 1, characterized in that: Step 1 also includes encoding the gas content to obtain a gas content code, wherein the first digit of the gas content code is a C2H2 / C2H4 code, the second digit is a CH4 / H2 code, and the third digit is a C2H4 / C2H6 code.

6. The method for diagnosing and locating a converter transformer oil tank assembly fault according to claim 4, characterized in that: Encoding of gas content includes: When C2H2 / C2H4<0.1, its coding value is 0; CH4 / H2<0.1, its coding value is 1; C2H4 / C2H6<0.1, its coding value is 0; When 0.1≤C2H2 / C2H4<1, its code is assigned to 1; 0.1≤CH4 / H2<1, its code is assigned to 0; 0.1≤C2H4 / C2H6<1, its code is assigned to 0; When 1≤C2H2 / C2H4<3, its code is assigned to 1; 1≤CH4 / H2<3, its code is assigned to 2; 1≤C2H4 / C2H6<3, its code is assigned to 1; When C2H2 / C2H4≥3, its coding value is 2; when CH4 / H2≥3, its coding value is 2; when C2H4 / C2H6≥3, its coding value is 2.

7. The method for diagnosing and locating a converter transformer oil tank assembly fault according to claim 6, characterized in that: In step 2, whether the metal suspension discharge inside the fuel tank is defective is diagnosed based on the gas content code. When the first digit of the gas content code is 2 and the C2H2 content exceeds the warning value, it is determined that the fuel tank component failure is metal suspension discharge.

8. The method for diagnosing and locating a converter transformer oil tank assembly fault according to any one of claims 4 to 7, characterized in that: The method further comprises: Step 5: Determine the status of the fuel tank component and maintenance suggestions based on the fuel tank component fault, and form a displayable fuel tank component fault reasoning flowchart by combining the fault judgment process, fuel tank component status and maintenance suggestions of steps 2, 3 and 4.

9. The method for diagnosing and locating a converter transformer oil tank assembly fault according to claim 8, characterized in that: In step 5, when the fuel tank component fault is metal suspension discharge and / or abnormal temperature of the fuel tank and metal structural parts and / or oil leakage in the secondary fuel tank, the fuel tank component state is abnormal, and the fuel tank component state is displayed in the fuel tank component fault reasoning flow chart in yellow code form as a reminder; When the oil tank component failure is oil leakage in the primary oil tank, the oil tank component status is abnormal, and the oil tank component status is displayed in the oil tank component failure reasoning flowchart in the form of a red code, and a power outage for maintenance is prompted.

10. A converter transformer oil tank assembly fault diagnosis and positioning system, characterized in that: include: The acquisition module is used to collect the gas content in the insulating oil of the converter transformer, the image information of the oil tank appearance and the hot spot temperature of the oil tank, wherein the gas content includes C2H2 content, C2H4 content, CH4 content, H2 content and C2H6 content; The first determination module is used to diagnose whether the metal suspension discharge inside the fuel tank is defective based on the C2H2 content, the C2H4 content, the CH4 content, the H2 content and the C2H6 content, and when C2H2 / C2H4≥3 and the C2H2 content exceeds the attention value, determine that the fuel tank component fault is metal suspension discharge; The second determination module is used to diagnose the temperature of the fuel tank and the metal structure based on the hot spot temperature of the fuel tank. When the hot spot temperature of the fuel tank is greater than the temperature threshold, the fuel tank component failure is determined to be abnormal temperature of the fuel tank and the metal structure. The temperature threshold is obtained through digital twin calculation; The third determination module is used to perform oil leakage trace image recognition on the appearance image information of the fuel tank through artificial intelligence. If the image recognition result is oil leakage, it is determined that the fuel tank component failure is oil tank leakage.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is loaded into a processor, the method for diagnosing and locating faults of a converter transformer oil tank assembly according to any one of claims 1 to 9 is implemented.

12. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for diagnosing and locating faults of a converter transformer oil tank assembly according to any one of claims 1 to 9 is implemented.