Converter transformer fault case processing method and system based on multi-dimensional information fusion
Through multi-dimensional information fusion and text optimization processing, the characteristics of the converter rheological failure cases are extracted, the correlation matrix is constructed and visualized, which solves the problem of low processing efficiency of converter rheological failure cases in the prior art, and achieves the effect of quickly and accurately obtaining the contact information of the fault case.
Patent Information
- Application Number
- CN202411498741.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-25
- Publication Date
- 2025-02-28
AI Technical Summary
The existing converter failure cases are not efficient in handling, and technicians need to spend a lot of time reading and understanding a large amount of text records to obtain the connection information between the failure cases.
Using a multi-dimensional information fusion method, the features of the fault case are extracted through text optimization processing, the case feature vector and multi-dimensional feature correlation matrix are constructed, and visualized, including drawing of the fault pie chart.
The efficiency of handling converter change failure cases is improved. Through multi-dimensional information fusion and visual analysis, technicians can quickly and accurately obtain the contact information between failure cases, thereby making targeted adjustments to avoid failures.
Smart Images
Figure CN119537908A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power systems, and in particular to a method and system for processing converter transformer fault cases based on multi-dimensional information fusion. Background Art
[0002] In the UHV power transmission system, the converter transformer equipment is at the core of power conversion. It is expensive and has a complex structure. Its safe operation directly affects whether the entire power transmission system can work continuously and reliably. Therefore, the converter transformer is one of the most important equipment in the UHV power transmission project, and its reliability is crucial to ensure the safety and stability of the entire system.
[0003] However, the endless quality problems of various manufacturers during the production and manufacturing stage of converter transformers have seriously affected the safe and stable operation of UHV projects. By analyzing the fault case characteristics of faulty converter transformers in combination with their production and manufacturing stages, it is possible to find deficiencies and defects in equipment design, and provide equipment manufacturers with a basis for improving the design, thereby improving the overall performance and reliability of the equipment. Therefore, visual analysis of converter transformer failure cases has extremely important practical significance.
[0004] The existing converter transformer fault case processing is to record all the data in text. The text content is relatively large, and technical personnel in this field need to read a large amount of text records to obtain the internal correlation of each fault case. Due to the large amount of text records, technical personnel need to spend a lot of time and energy to read and understand the details of each case, and some key information may be missed in the reading process.
[0005] It can be seen that how to quickly and accurately obtain the connection information between fault cases and make targeted adjustments to the converter transformers of the current power grid to avoid failures has become a technical problem that technical personnel in this field need to solve urgently. Summary of the invention
[0006] The present invention provides a converter transformer fault case processing method and system based on multi-dimensional information fusion, so as to solve the problem of low efficiency of existing converter transformer fault case processing and improve the efficiency of converter transformer fault case processing.
[0007] In order to solve the above technical problems, an embodiment of the present invention provides a method for processing converter transformer fault cases based on multi-dimensional information fusion.
[0008] During the case import process, the initial fault case text data of each imported converter transformer fault case is optimized to obtain the text data of each fault case;
[0009] Extracting each case feature of each converter transformer fault case based on each fault case text data, performing standardization processing on each case feature, and obtaining a feature label of each case feature; the case features include each factory feature and each fault feature;
[0010] Constructing a case feature vector for each case feature based on all the feature labels of the same case feature, and constructing a multi-dimensional feature correlation matrix based on the correlation features between all the case feature vectors;
[0011] Cross-classification is performed based on all the factory characteristics to obtain various factory categories, and the fault occurrence ratios of each fault characteristic corresponding to each factory category are counted. A fault pie chart of each fault characteristic is drawn based on all the fault occurrence ratios, and the fault pie chart and the multidimensional feature correlation matrix are visualized.
[0012] As an embodiment of the present application, the factory characteristics include the project name, manufacturer, product type and product grade, the fault characteristics include the problem type, problem location and problem cause, and the fault occurrence ratio includes the problem type ratio, problem location ratio and problem cause ratio.
[0013] For details, see Figure 3 , Figure 3 A fault pie chart showing the problem type in one embodiment of the present invention is shown. For details, see Figure 4 , Figure 4 The following is a pie chart showing the fault location of a problem in one embodiment of the present invention. Figure 5 , Figure 5 A pie chart showing the causes of problems in one embodiment of the present invention.
[0014] As one of the preferred solutions, the multi-dimensional feature correlation matrix is constructed based on the correlation features between all the case feature vectors, including:
[0015] All the case feature vectors are combined in pairs, and the Pearson correlation coefficient between the two case feature vectors in each combination is used as the feature strong correlation coefficient between the corresponding case features;
[0016] Each of the feature strong correlation coefficients is used as the element value at the corresponding position in the multidimensional feature correlation matrix.
[0017] As one of the preferred solutions, the text optimization processing is performed on the initial fault case text data of each converter transformer fault case imported to obtain the text data of each fault case, including:
[0018] Using regular expressions to remove spaces, punctuation marks and special symbols in the initial fault case text data to obtain standard text data of each fault case;
[0019] Each of the standard text data of the fault case is segmented to obtain various fault words, each of the fault words is matched with a pre-built fault case dictionary to obtain a standard word for each of the fault words, and each of the fault words in the standard text data of each of the fault cases is replaced with a corresponding standard word to obtain various fault case text data.
[0020] As one of the preferred solutions, the standardization process is performed on each case feature to obtain a feature label of each case feature, including:
[0021] Classify and summarize each of the case features, and count all the converter transformer failure cases according to the category corresponding to each of the case features, to obtain the number of failure cases in each category under each of the case features;
[0022] Each of the categories is graded based on the number of all the fault cases to obtain a feature label for each of the case features.
[0023] As one of the preferred solutions, constructing a case feature vector for each case feature based on all the feature labels of the same case feature includes:
[0024] Arrange all the converter transformer failure cases in ascending chronological order to obtain a number for each converter transformer failure case;
[0025] All the feature labels of the same case feature are arranged according to the serial numbers of the corresponding converter transformer fault cases as the case feature vector of each case feature.
[0026] Another embodiment of the present invention provides a converter transformer fault case processing system based on multi-dimensional information fusion, the system comprising a fault case editing module, a case feature summarization module, a case association analysis module and a fault case display module;
[0027] The fault case editing module is used to perform text optimization processing on the initial fault case text data of each converter transformer fault case imported during the case import process to obtain the text data of each fault case;
[0028] The case feature summarization module extracts each case feature of each converter transformer fault case based on each fault case text data, performs standardization processing on each case feature, and obtains a feature label of each case feature; the case features include each factory feature and each fault feature;
[0029] The case association analysis module constructs a case feature vector for each case feature based on all the feature labels of the same case feature, and constructs a multi-dimensional feature correlation matrix based on the correlation features between all the case feature vectors;
[0030] The fault case display module cross-classifies all the factory characteristics to obtain various factory categories, counts the fault occurrence ratios of each fault characteristic corresponding to each factory category, draws a fault pie chart of each fault characteristic based on all the fault occurrence ratios, and visualizes the fault pie chart and the multi-dimensional feature correlation matrix.
[0031] As one of the preferred solutions, the multi-dimensional feature correlation matrix is constructed based on the correlation features between all the case feature vectors, including:
[0032] All the case feature vectors are combined in pairs, and the Pearson correlation coefficient between the two case feature vectors in each combination is used as the feature strong correlation coefficient between the corresponding case features;
[0033] Each of the feature strong correlation coefficients is used as the element value at the corresponding position in the multidimensional feature correlation matrix.
[0034] As one of the preferred solutions, the text optimization processing is performed on the initial fault case text data of each converter transformer fault case imported to obtain the text data of each fault case, including:
[0035] Using regular expressions to remove spaces, punctuation marks and special symbols in the initial fault case text data to obtain standard text data of each fault case;
[0036] Each of the standard text data of the fault case is segmented to obtain various fault words, each of the fault words is matched with a pre-built fault case dictionary to obtain a standard word for each of the fault words, and each of the fault words in the standard text data of each of the fault cases is replaced with a corresponding standard word to obtain various fault case text data.
[0037] As one of the preferred solutions, the standardization process is performed on each case feature to obtain a feature label of each case feature, including:
[0038] Classify and summarize each of the case features, and count all the converter transformer failure cases according to the category corresponding to each of the case features, to obtain the number of failure cases in each category under each of the case features;
[0039] Each of the categories is graded based on the number of all the fault cases to obtain a feature label for each of the case features.
[0040] As one of the preferred solutions, constructing a case feature vector for each case feature based on all the feature labels of the same case feature includes:
[0041] Arrange all the converter transformer failure cases in ascending chronological order to obtain a number for each converter transformer failure case;
[0042] All the feature labels of the same case feature are arranged according to the serial numbers of the corresponding converter transformer fault cases as the case feature vector of each case feature.
[0043] Compared with the prior art, the embodiments of the present invention have the following advantages:
[0044] (1) The fault case editing module is used to optimize the text of the initial fault case text data of each converter transformer fault case imported during the case import process to obtain the text data of each fault case. By removing spaces, punctuation marks and special symbols in the initial fault case text data, each fault word in the standard text data of each fault case is replaced with a corresponding standard word, so that the standard text data of the fault case is more in line with the processing requirements of the computer and the consistency of the fault words is ensured, thereby improving the reliability of the subsequent processing of the converter transformer fault case;
[0045] (2) Considering that the converter transformer fault cases involve a large amount of complex text descriptions, in order to facilitate the subsequent converter transformer fault case analysis, the case feature summary module extracts each case feature of each converter transformer fault case based on each fault case text data. The case features include each factory feature and each fault feature. First, the important features in the fault case text data are extracted, and then each case feature is standardized to obtain the feature label of each case feature, so as to adaptively classify the extracted same features and convert them into corresponding digital formats, so as to facilitate the subsequent processing of the converter transformer fault cases;
[0046] (3) The case correlation analysis module constructs a case feature vector for each case feature based on all feature labels of the same case feature, and constructs a multidimensional feature correlation matrix based on the correlation characteristics between all case feature vectors. The multidimensional feature correlation matrix reflects the correlation characteristics between different case features.
[0047] (4) The fault case display module cross-classifies all factory features to obtain various factory categories, counts the fault occurrence ratios of various fault features corresponding to each factory category, draws a fault pie chart of each fault feature based on all fault occurrence ratios, and visualizes the fault pie chart and the multi-dimensional feature correlation matrix. This not only allows intuitive observation of the association between different case features, but also allows timely acquisition of the occurrence ratios of various faults under different factory categories, thereby improving the efficiency of processing converter transformer fault cases. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] Figure 1 It is a flow chart of a method for processing a converter transformer fault case based on multi-dimensional information fusion in one embodiment of the present invention;
[0049] Figure 2 This is a structural block diagram of a converter transformer fault case processing system based on multi-dimensional information fusion in one embodiment of the present invention;
[0050] Figure 3 is a fault pie chart of problem types in one embodiment of the present invention;
[0051] Figure 4 This is a fault pie chart of a problem location in one embodiment of the present invention;
[0052] Figure 5 It is a fault pie chart showing the causes of problems in one embodiment of the present invention. DETAILED DESCRIPTION
[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0054] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.
[0055] In the description of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two elements. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are only for illustrative purposes, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.
[0056] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.
[0057] An embodiment of the present invention provides a method for processing a converter transformer fault case based on multi-dimensional information fusion. For details, see Figure 1 , Figure 1 The flowchart of a converter transformer fault case processing method based on multi-dimensional information fusion in one embodiment of the present invention is shown. The converter transformer fault case processing method based on multi-dimensional information fusion includes:
[0058] During the case import process, the initial fault case text data of each imported converter transformer fault case is optimized to obtain the text data of each fault case;
[0059] Extracting each case feature of each converter transformer fault case based on each fault case text data, performing standardization processing on each case feature, and obtaining a feature label of each case feature; the case features include each factory feature and each fault feature;
[0060] Constructing a case feature vector for each case feature based on all the feature labels of the same case feature, and constructing a multi-dimensional feature correlation matrix based on the correlation features between all the case feature vectors;
[0061] Cross-classification is performed based on all the factory characteristics to obtain various factory categories, and the fault occurrence ratios of each fault characteristic corresponding to each factory category are counted. A fault pie chart of each fault characteristic is drawn based on all the fault occurrence ratios, and the fault pie chart and the multidimensional feature correlation matrix are visualized.
[0062] Furthermore, in the above embodiment, the multi-dimensional feature correlation matrix is constructed based on the correlation features between all the case feature vectors, including:
[0063] All case feature vectors are combined in pairs, and the Pearson correlation coefficient between the two case feature vectors in each combination is used as the feature strong correlation coefficient between the corresponding case features, and each feature strong correlation coefficient is used as the element value at the corresponding position in the multidimensional feature correlation matrix.
[0064] Furthermore, in the above embodiment, the text optimization processing is performed on the initial fault case text data of each converter transformer fault case imported to obtain the text data of each fault case, including:
[0065] Regular expressions are used to remove spaces, punctuation marks and special symbols in the initial fault case text data to obtain standard text data for each fault case.
[0066] Perform word segmentation on the standard text data of each fault case to obtain each fault vocabulary, perform vocabulary matching on each fault vocabulary and a pre-built fault case dictionary to obtain the standard vocabulary of each fault vocabulary, replace each fault vocabulary in the standard text data of each fault case with the corresponding standard vocabulary to obtain each fault case text data.
[0067] Furthermore, in the above embodiment, the standardization process is performed on each of the case features to obtain a feature label for each of the case features, including:
[0068] Each case feature is classified and summarized, and all converter transformer fault cases are counted according to the category corresponding to each case feature to obtain the number of fault cases in each category under each case feature. Each category is graded based on the number of all fault cases to obtain the feature label of each case feature.
[0069] Furthermore, in the above embodiment, constructing a case feature vector for each case feature based on all the feature tags of the same case feature includes:
[0070] All converter transformer fault cases are arranged in ascending chronological order to obtain the number of each converter transformer fault case, and all feature labels of the same case feature are arranged according to the number of the corresponding converter transformer fault case as the case feature vector of each case feature.
[0071] Another embodiment of the present invention provides a converter transformer fault case processing system based on multi-dimensional information fusion. For details, see Figure 2 , Figure 2 The structure block diagram of a converter transformer fault case processing system based on multi-dimensional information fusion in one embodiment of the present invention is shown. The converter transformer fault case processing system based on multi-dimensional information fusion includes:
[0072] The fault case editing module 11 is used for performing text optimization processing on the initial fault case text data of each converter transformer fault case imported during the case import process to obtain the text data of each fault case;
[0073] Case feature summarization module 12: extracting each case feature of each converter transformer fault case based on each fault case text data, performing standardization processing on each case feature, and obtaining a feature label of each case feature; the case features include each factory feature and each fault feature;
[0074] Case correlation analysis module 13: constructing a case feature vector of each case feature based on all the feature labels of the same case feature, and constructing a multi-dimensional feature correlation matrix based on the correlation features between all the case feature vectors;
[0075] Fault case display module 14: cross-classify all the factory characteristics to obtain various factory categories, count the fault occurrence ratios of each fault characteristic corresponding to each factory category, draw a fault pie chart of each fault characteristic based on all the fault occurrence ratios, and visualize the fault pie chart and the multi-dimensional feature correlation matrix.
[0076] Furthermore, in the above embodiment, the multi-dimensional feature correlation matrix is constructed based on the correlation features between all the case feature vectors, including:
[0077] All case feature vectors are combined in pairs, and the Pearson correlation coefficient between the two case feature vectors in each combination is used as the feature strong correlation coefficient between the corresponding case features, and each feature strong correlation coefficient is used as the element value at the corresponding position in the multidimensional feature correlation matrix.
[0078] Furthermore, in the above embodiment, the text optimization processing is performed on the initial fault case text data of each converter transformer fault case imported to obtain the text data of each fault case, including:
[0079] Regular expressions are used to remove spaces, punctuation marks and special symbols in the initial fault case text data to obtain standard text data for each fault case.
[0080] Perform word segmentation on the standard text data of each fault case to obtain each fault vocabulary, perform vocabulary matching on each fault vocabulary and a pre-built fault case dictionary to obtain the standard vocabulary of each fault vocabulary, replace each fault vocabulary in the standard text data of each fault case with the corresponding standard vocabulary to obtain each fault case text data.
[0081] Furthermore, in the above embodiment, the standardization process is performed on each of the case features to obtain a feature label for each of the case features, including:
[0082] Each case feature is classified and summarized, and all converter transformer fault cases are counted according to the category corresponding to each case feature to obtain the number of fault cases in each category under each case feature. Each category is graded based on the number of all fault cases to obtain the feature label of each case feature.
[0083] Furthermore, in the above embodiment, constructing a case feature vector for each case feature based on all the feature tags of the same case feature includes:
[0084] All converter transformer fault cases are arranged in ascending chronological order to obtain the number of each converter transformer fault case, and all feature labels of the same case feature are arranged according to the number of the corresponding converter transformer fault case as the case feature vector of each case feature.
[0085] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.
Claims
1. A method for processing converter transformer fault cases based on multi-dimensional information fusion, characterized in that: The method comprises: During the case import process, the initial fault case text data of each imported converter transformer fault case is optimized to obtain the text data of each fault case; Extracting each case feature of each converter transformer fault case based on each fault case text data, performing standardization processing on each case feature, and obtaining a feature label of each case feature; the case features include each factory feature and each fault feature; Constructing a case feature vector for each case feature based on all the feature labels of the same case feature, and constructing a multi-dimensional feature correlation matrix based on the correlation features between all the case feature vectors; Cross-classification is performed based on all the factory characteristics to obtain various factory categories, and the fault occurrence ratios of each fault characteristic corresponding to each factory category are counted. A fault pie chart of each fault characteristic is drawn based on all the fault occurrence ratios, and the fault pie chart and the multidimensional feature correlation matrix are visualized.
2. A method for processing converter transformer fault cases based on multi-dimensional information fusion according to claim 1, characterized in that: The constructing of a multidimensional feature correlation matrix based on the correlation features between all the case feature vectors includes: All the case feature vectors are combined in pairs, and the Pearson correlation coefficient between the two case feature vectors in each combination is used as the feature strong correlation coefficient between the corresponding case features; Each of the feature strong correlation coefficients is used as the element value at the corresponding position in the multidimensional feature correlation matrix.
3. The method for processing converter transformer fault cases based on multi-dimensional information fusion according to claim 1 is characterized in that: The text optimization process is performed on the initial fault case text data of each converter transformer fault case imported to obtain the text data of each fault case, including: Using regular expressions to remove spaces, punctuation marks and special symbols in the initial fault case text data to obtain standard text data of each fault case; Each of the standard text data of the fault case is segmented to obtain various fault words, each of the fault words is matched with a pre-built fault case dictionary to obtain a standard word for each of the fault words, and each of the fault words in the standard text data of each of the fault cases is replaced with a corresponding standard word to obtain various fault case text data.
4. The method for processing converter transformer fault cases based on multi-dimensional information fusion according to claim 1 is characterized in that: The step of performing standardization on each of the case features to obtain a feature label for each of the case features includes: Classify and summarize each of the case features, and count all the converter transformer failure cases according to the category corresponding to each of the case features, to obtain the number of failure cases in each category under each of the case features; Each of the categories is graded based on the number of all the fault cases to obtain a feature label for each of the case features.
5. The method for processing converter transformer fault cases based on multi-dimensional information fusion according to claim 1 is characterized in that: The step of constructing a case feature vector for each case feature based on all the feature labels of the same case feature includes: Arrange all the converter transformer failure cases in ascending chronological order to obtain a number for each converter transformer failure case; All the feature labels of the same case feature are arranged according to the serial numbers of the corresponding converter transformer fault cases as the case feature vector of each case feature.
6. A converter transformer fault case processing system based on multi-dimensional information fusion, characterized in that: The system includes a fault case editing module, a case feature summarization module, a case association analysis module and a fault case display module; The fault case editing module is used to perform text optimization processing on the initial fault case text data of each converter transformer fault case imported during the case import process to obtain the text data of each fault case; The case feature summarization module extracts each case feature of each converter transformer fault case based on each fault case text data, performs standardization processing on each case feature, and obtains a feature label of each case feature; the case features include each factory feature and each fault feature; The case association analysis module constructs a case feature vector for each case feature based on all the feature labels of the same case feature, and constructs a multi-dimensional feature correlation matrix based on the correlation features between all the case feature vectors; The fault case display module cross-classifies all the factory characteristics to obtain various factory categories, counts the fault occurrence ratios of each fault characteristic corresponding to each factory category, draws a fault pie chart of each fault characteristic based on all the fault occurrence ratios, and visualizes the fault pie chart and the multi-dimensional feature correlation matrix.
7. A converter transformer fault case processing system based on multi-dimensional information fusion according to claim 6, characterized in that: The constructing of a multidimensional feature correlation matrix based on the correlation features between all the case feature vectors includes: All the case feature vectors are combined in pairs, and the Pearson correlation coefficient between the two case feature vectors in each combination is used as the feature strong correlation coefficient between the corresponding case features; Each of the feature strong correlation coefficients is used as the element value at the corresponding position in the multidimensional feature correlation matrix.
8. The converter transformer fault case processing system based on multi-dimensional information fusion according to claim 6 is characterized in that: The text optimization process is performed on the initial fault case text data of each converter transformer fault case imported to obtain the text data of each fault case, including: Using regular expressions to remove spaces, punctuation marks and special symbols in the initial fault case text data to obtain standard text data of each fault case; Each of the standard text data of the fault case is segmented to obtain various fault words, each of the fault words is matched with a pre-built fault case dictionary to obtain a standard word for each of the fault words, and each of the fault words in the standard text data of each of the fault cases is replaced with a corresponding standard word to obtain various fault case text data.
9. The converter transformer fault case processing system based on multi-dimensional information fusion according to claim 6 is characterized in that: The step of performing standardization on each of the case features to obtain a feature label for each of the case features includes: Classify and summarize each of the case features, and count all the converter transformer failure cases according to the category corresponding to each of the case features, to obtain the number of failure cases in each category under each of the case features; Each of the categories is graded based on the number of all the fault cases to obtain a feature label for each of the case features.
10. The converter transformer fault case processing system based on multi-dimensional information fusion according to claim 6, characterized in that: The step of constructing a case feature vector for each case feature based on all the feature labels of the same case feature includes: Arrange all the converter transformer failure cases in ascending chronological order to obtain a number for each converter transformer failure case; All the feature labels of the same case feature are arranged according to the serial numbers of the corresponding converter transformer fault cases as the case feature vector of each case feature.