Signal identification matching method and device for anti-error logic verification of transformer substation

Abstract

The invention discloses a signal identification matching method and device for anti-error logic verification of a transformer substation. The method comprises the following steps: collecting power text data of the whole transformer substation; extracting combined word elements in the power text data of the whole substation, performing classification according to key feature information of the combined word elements, and establishing a power text comparison database; identifying electric power text content from the electric power text data of the whole substation, comparing a database based on the electric power text, and determining association degrees between the electric power text content and different key feature information; establishing a power text index system to preferentially index the combined character elements under the classification corresponding to the key feature information with the highest association degree with the input power text content; and obtaining and identifying the power text content contained in the to-be-processed signal, and indexing the power text content contained in the to-be-processed signal in the power text comparison database through the power text indexing system to obtain a corresponding matching result. According to the method, comprehensiveness, accuracy and efficiency of anti-error logic verification of the transformer substation can be ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of natural language processing, and particularly relates to a signal recognition and matching method and device for substation anti-error logic verification. Background Art

[0002] Substation anti-error logic verification is an important link to ensure the safe operation of substations. Its purpose is to prevent electrical misoperations and ensure the safe and stable operation of the power system. By verifying the correctness of the anti-error logic, it can be ensured that when operating personnel perform switching operations, they follow the correct operation sequence and rules, avoiding serious consequences such as equipment damage, personal injury or large-scale power outages caused by misoperations.

[0003] Substation anti-error logic verification mainly includes a preparation stage, a simulated operation stage, an analysis and verification stage, and a subsequent processing stage. In the preparation stage, relevant data needs to be collected and an anti-error logic verification model is established; this model should be able to simulate the actual operation of the substation, including the state changes of equipment, operation sequences, etc. Anti-error logic rules are set in the model, and these rules are based on the safety regulations and operating procedures of the power system to prevent misoperations. In the simulated operation stage, potential problems need to be discovered by simulating the operation process and recording the results. In the analysis and verification stage, the simulated operation results are deeply analyzed and the effectiveness of the anti-error logic rules is verified; in the subsequent processing stage, rectifications are made according to the problems found in the verification report and the operation of the substation is continuously monitored and evaluated.

[0004] Currently, in each stage of substation anti-error logic verification, manual operation is required. Due to the huge amount of data, the manual processing efficiency is low and the accuracy decreases. Therefore, how to achieve the accurate extraction and efficient matching of the entire substation's power text data (including, for example, the entire substation model file, design drawings, and power-specific naming data, etc.), so as to ensure the comprehensiveness, accuracy, and efficiency of substation anti-error logic verification, is a technical problem to be solved urgently. Summary of the Invention

[0005] The purpose of the present invention is to provide a signal recognition and matching method and device for substation anti-error logic verification, so as to achieve the accurate extraction and matching of the entire substation model file, design drawings, and power-specific naming data, thereby ensuring the comprehensiveness, accuracy, and efficiency of substation anti-error logic verification.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions:

[0007] According to the first aspect of the present invention, a signal recognition and matching method for substation anti-error logic verification is provided. The method includes the following steps:

[0008] Step 1: Collect the entire substation's power text data;

[0009] Step 2: Extract the combined character elements from the substation's full-station power text data, classify the combined character elements according to the key feature information of the combined character elements, and establish a power text comparison database based on the classified combined character elements;

[0010] Step 3: Identify the power text content from the substation's full-station power text data, and determine the degree of association between the power text content and different key feature information based on the power text comparison database;

[0011] Step 4: Establish a power text indexing system based on the degree of association, so that the power text indexing system preferentially indexes the combined character elements classified under the key feature information with the highest degree of association with the input power text content;

[0012] Step 5: Identify the power text content included in the signal to be processed, and match the power text content included in the signal to be processed in the power text comparison database through the power text indexing system.

[0013] Further, Step 1 includes:

[0014] Step 11: Collect various types of pictures related to the substation's full station, including: model file pictures, design pictures, and power-specific naming class file pictures;

[0015] Step 12: Based on the fully convolutional neural network, extract text information from the collected various types of pictures through symbol recognition, text recognition, connection recognition, and feature element classification and marking as the substation's full-station power text data;

[0016] Step 13: Clean the collected power text data, remove redundant and incorrect information, and format the cleaned data to meet the input requirements of natural language processing technology.

[0017] Further, Step 2 includes:

[0018] Step 21: Use natural language processing tools to perform word segmentation and part-of-speech tagging on the text in the power text data;

[0019] Step 22: Based on the word segmentation results, professional knowledge in the power field, and actual requirements, extract combined character elements with practical meanings;

[0020] Step 23: Classify the combined character elements according to the key feature information of the extracted combined character elements; the key feature information includes: equipment name information, operation instruction information, and equipment status information.

[0021] Further, Step 3 includes:

[0022] Determine the degree of association between the current power text content A and the key feature information a based on the number of combined character elements in the current power text content A, the types of combined character elements related to the key feature information a contained in the current power text content A, and the number of occurrences of various combined character elements related to a in the current power text content A.

[0023] Furthermore, step 3 also includes:

[0024] Use the following formula to determine the degree of association between the current power text content A and the key feature information a:

[0025]

[0026] Where U represents the number of combined character elements in the power text content A; u(a) represents the number of combined character elements containing the key feature information a in the power text content A; x represents the x-th type of combined character element containing the key feature information a in the power text content A; s(x) represents the number of occurrences of the x-th type of combined character element in the power text content A; m represents the types of combined character elements appearing in the power text content A, and set a = {1, 2, 3}, where 1, 2, and 3 represent equipment name information, operation instruction information, and equipment status information respectively; n is the number of types of combined character elements containing the key feature information a in the power text content A.

[0027] Furthermore, in step 4, the establishment of the power text indexing system also includes:

[0028] By adopting methods such as regular expression matching, keyword matching, logical matching, and neural network recognition matching, make the power text indexing system perform further indexing under the classification corresponding to the key feature information with the highest degree of association in the power text comparison database, and obtain combined character elements that are the same as those contained in the input power text content.

[0029] Furthermore, in step 4, the establishment of the power text indexing system also includes:

[0030] Based on the interactive verification mechanism of neural network recognition, verify and screen the matching results;

[0031] Regularly update the substation's full-station power text data and the power text comparison database, and accordingly update the degree of association between the power text content and different key feature information, and verify and optimize the established power text indexing system based on the updated degree of association.

[0032] Furthermore, in step 5, identifying the power text content contained in the signal to be processed includes:

[0033] When the type of the signal to be processed is an image signal, computer vision technology is used to identify and process the image signal, extract the text information in the image signal and convert it into power text content;

[0034] When the type of the signal to be processed is a voice signal, speech recognition technology is used to identify and process the voice signal, extract the text information in the voice signal and convert it into power text content;

[0035] When the signal to be processed is a text signal, the power text content in the text signal is extracted through EP-NER technology.

[0036] According to the second aspect of the present invention, there is provided a signal recognition and matching device based on natural language processing using the method described in the first aspect of the present invention. The device includes:

[0037] A data collection module for collecting the power text data of the entire substation;

[0038] A database establishment module for extracting the combined character elements in the power text data of the entire substation, classifying the combined character elements according to the key feature information of the combined character elements, and establishing a power text comparison database based on the classified combined character elements;

[0039] An association degree determination module for identifying the power text content from the power text data of the entire substation and determining the association degree between the power text content and different key feature information based on the power text comparison database;

[0040] An index system establishment module for establishing a power text index system based on the association degree, so that the power text index system preferentially indexes the combined character elements classified under the key feature information with the highest association degree with the input power text content;

[0041] A matching module for identifying the power text content included in the signal to be processed and matching the power text content included in the signal to be processed in the power text comparison database through the power text index system.

[0042] According to the third aspect of the present invention, there is provided a terminal. The terminal includes a processor and a storage medium; the storage medium is used to store instructions; the processor is used to operate according to the instructions to execute the steps of the method described in the first aspect of the present invention.

[0043] According to the fourth aspect of the present invention, there is provided a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the steps of the method described in the first aspect of the present invention are implemented.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0045] 1. By extracting the combined character elements with actual meanings in the power text data of the entire substation, classifying the combined character elements according to the key feature information of the combined character elements to establish a power text comparison database. Then, based on the power text comparison database, determining the degree of association between the recognized power text content in the power text data of the entire substation and different key feature information. Furthermore, establishing a power text index system based on the degree of association, so that the power text index system preferentially indexes the combined character elements under the classification corresponding to the key feature information with the highest degree of association with the input power text content as the matching result. It is possible to determine the matching situation of each combined character element in the power text to be recognized and matched with each combined character element in the corresponding classification of the key feature information, achieving efficient, accurate, and comprehensive matching of various signals in the power system, thereby ensuring the comprehensiveness, accuracy, and efficiency of substation anti-misoperation logic verification.

[0046] 2. Based on the number of combined character elements in the current power text content, the types of combined character elements related to key feature information a included in the current power text content, and the number of occurrences of various combined character elements related to a in the current power text content, determining the degree of association between the current power text content and key feature information a can better quantify the degree of association between the power text content and each key feature information, contribute to further improving the efficiency and accuracy of information retrieval, quickly locate and extract key information in the text, and further improve the information processing efficiency.

[0047] 3. By introducing an interactive verification mechanism based on neural network recognition, verifying and screening the matching results, and regularly verifying and optimizing the calculation results of the degree of association and the power text index system, it is possible to further ensure the accuracy and reliability of the matching results, so as to improve the accuracy and reliability of signal recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0049] Figure 1 FIG. shows the flowchart of a signal recognition and matching method for substation anti-misoperation logic verification in an embodiment of the present invention;

[0050] Figure 2 FIG. shows the schematic diagram of a signal recognition and matching device for substation anti-misoperation logic verification in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0051] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0052] Embodiment 1

[0053] This embodiment provides a signal recognition and matching method based on natural language processing, as Figure 1 shown, the method specifically includes the following steps:

[0054] Step 1: Collect the power text data of the entire substation.

[0055] Specifically, Step 1 includes:

[0056] Step 11: Collect various types of pictures related to the entire substation, including: model file pictures, design pictures, and power-specific naming file pictures.

[0057] Step 12: Based on the fully convolutional neural network, extract text information from the collected various pictures as the power text data of the entire substation through the comprehensive application of symbol recognition, text recognition, connection recognition, and feature element classification and marking.

[0058] Step 13: Clean the collected power text data, remove redundant and incorrect information, and format the cleaned data to meet the input requirements of natural language processing technology.

[0059] Step 2: Extract the combined character elements in the power text data of the entire substation, classify the combined character elements according to the key feature information of the combined character elements, and establish a power text comparison database based on the classified combined character elements.

[0060] Specifically, Step 2 includes:

[0061] Step 21: Use natural language processing tools to perform word segmentation and part-of-speech tagging on the text in the power text data.

[0062] Among them, natural language processing tools can be, for example, Jieba, HanLP, LAC, etc.

[0063] Step 22: Based on the word segmentation results and the professional knowledge and actual needs in the power field, extract the combined character elements with actual meanings.

[0064] Among them, when extracting power text data based on the word segmentation results, professional knowledge and actual needs in the power field, words and phrases in the power text data can usually be obtained, denoted as word elements. Word elements are divided into single word elements and combined word elements. Single word elements usually refer to single characters, while combined word elements usually refer to words or phrases composed of multiple single word elements and having actual meanings. Among the obtained word elements, word elements without actual meanings need to be deleted, usually single word elements.

[0065] Step 23: Classify the combined word elements according to the key feature information of the extracted combined word elements. These key feature information include: equipment name information, operation instruction information, and equipment status information.

[0066] Based on the classified combined word elements, a power text comparison database can be established.

[0067] Step 3: Identify the power text content from the substation's full-station power text data, and determine the degree of association between the power text content and different key feature information based on the power text comparison database.

[0068] Among them, based on the EP-NER technology, that is, the power named entity recognition technology, specific content related to the power system, equipment, and industry can be accurately identified from the text, denoted as power text content.

[0069] Using the power text comparison database to retrieve and compare the power text content based on natural language processing technology, decompose the power text content into text content composed of multiple combined word elements, retain the content of each combined word element, and record the number of occurrences of each combined word element.

[0070] Based on the number of combined word elements in the current power text content A, the number of types of combined word elements related to the key feature information a included in the current power text content A, and the number of occurrences of various combined word elements related to a in the current power text content A, the following formula can be used to determine the degree of association between the current power text content A and the key feature information a:

[0071]

[0072] Among them, F(A, a) represents the degree of association between the power text content A and the key feature information a, U represents the number of combined character elements of the power text content A, u(a) represents the number of combined character elements containing the key feature information a in the power text content A, x represents the x-th combined character element containing the key feature information a in the power text content A, s(x) represents the number of times the x-th combined character element appears in the power text content A, m represents the number of types of combined character elements that appear in the power text content A. Set a = {1, 2, 3}, where 1, 2, and 3 represent equipment name information, operation instruction information, and equipment status information respectively; n is the number of types of combined character elements containing the key feature information a in the power text content A.

[0073] Step 4: Establish a power text indexing system so that the power text indexing system preferentially indexes the combined character elements under the classification corresponding to the key feature information with the highest degree of association with the input power text content.

[0074] Among them, as an example, when the key feature information of the combined character elements includes equipment name information, operation instruction information, and equipment status information, the classification corresponding to the key feature information with the highest degree of association with the input power text content is selected from one of the equipment name information, operation instruction information, and equipment status information.

[0075] Furthermore, establishing the power text indexing system also includes: by adopting various technical means such as regular expression matching, keyword matching, logical matching, and neural network recognition matching, enabling the power text indexing system to further index under the classification corresponding to the key feature information with the highest degree of association in the power text comparison database, and obtaining the combined character elements that are the same as the combined character elements contained in the input power text content as the matching result.

[0076] To implement the above functions of the power text indexing system, including preferentially indexing the combined character elements under the classification corresponding to the key feature information with the highest degree of association with the input power text content, and further indexing under the classification corresponding to the key feature information with the highest degree of association, obtaining the combined character elements that are the same as the combined character elements contained in the input power text content as the matching result. Among them, the power text indexing system can be implemented by constructing a neural network and training the constructed neural network based on the calculated degree of association result and the model matching result (i.e., the matching results obtained by means such as regular expression matching, keyword matching, logical matching, and neural network recognition matching). The specific training method will not be elaborated here.

[0077] Furthermore, the establishment of the power text indexing system also includes: an interactive verification mechanism based on neural network recognition to verify and screen the matching results; and regularly updating the power text data of the entire substation and the power text comparison database, and accordingly updating the association degree between the power text content and different key feature information, and verifying and optimizing the established power text indexing system based on the updated association degree.

[0078] Among them, the steps of interactive verification are as follows: randomly divide the input-output data set of the model into a training set and a test set; use the training set to train the machine learning model; use the test set to evaluate the performance of the model, such as indicators like accuracy, recall rate, F1 score, etc. Reflect the model matching performance through performance indicators.

[0079] Through verification and optimization and regular updates, the accuracy and reliability of the matching results can be further ensured to improve the accuracy and reliability of signal recognition.

[0080] Step 5: Obtain and identify the power text content included in the signal to be processed, and index the power text content included in the signal to be processed in the power text comparison database based on the power text indexing system to obtain corresponding matching results.

[0081] Among them, the signal to be processed refers to various signals related to power system operations, instructions, and states that may be involved in the anti-misoperation logic verification process. These signals may contain power text content and need to be identified, indexed, and matched through a specific processing process. For example, dispatching instruction signals, operation record signals, equipment status signals, fault alarm signals, system log signals, etc. The signal to be processed can be obtained from various sensors, controllers, and communication interfaces in the power system and includes various types. For example, it can be an image signal, a voice signal, or a text signal.

[0082] By combining computer vision technology, speech recognition technology, and natural language processing technology in AI technology, image signals, voice signals, and text signals can be uniformly converted into data containing only power text content.

[0083] Correspondingly, in step 5, identifying the power text content included in the signal to be processed may include:

[0084] When the type of the signal to be processed is an image signal, use computer vision technology to identify and process the image signal, extract the text information in the image signal and convert it into power text content;

[0085] When the type of the signal to be processed is a voice signal, use speech recognition technology to identify and process the voice signal, extract the text information in the voice signal and convert it into power text content;

[0086] When the signal to be processed is a text signal, the power text content in the text signal is extracted by the EP-NER technology.

[0087] In addition, after extracting the power text content from various types of signals to be processed, preprocessing and format conversion can be performed to ensure that it meets the input requirements of the power text comparison database.

[0088] Finally, the recognized and converted power text content is matched with the text content in the power text comparison database, and the matching degree and relevance can be determined.

[0089] It should be noted that although the steps of the method in the present invention are numbered, the purpose is to better explain each step, rather than to limit the order of each step. For example, the establishment of the power text indexing system in step 4 of the present invention is actually a process of continuous update and optimization. As described above, it is usually necessary to regularly update the power text data of the entire substation, update the power text comparison database, and optimize and update the power text indexing system on this basis, and then identify and match the signal to be processed based on the updated power text indexing system.

[0090] Embodiment 2

[0091] This embodiment provides a signal recognition and matching device for substation anti-misoperation logic verification, which utilizes the method described in Embodiment 1 of the present invention. As Figure 2 shown, the device includes:

[0092] [[ID=2,0]]A data collection module for collecting the power text data of the entire substation;

[0093] A database establishment module for extracting the combined character elements from the power text data of the entire substation, classifying the combined character elements according to the key feature information of the combined character elements, and establishing a power text comparison database based on the classified combined character elements;

[0094] An association degree determination module for identifying the power text content from the power text data of the entire substation and determining the association degree between the power text content and different key feature information based on the power text comparison database;

[0095] An indexing system establishment module for establishing a power text indexing system based on the association degree, so that the power text indexing system preferentially indexes the combined character elements corresponding to the classification of the key feature information with the highest association degree with the input power text content;

[0096] A matching module for identifying the power text content included in the signal to be processed and matching the power text content included in the signal to be processed in the power text comparison database through the power text indexing system.

[0097] Example 3

[0098] Example 3 of the present invention provides an electronic device.

[0099] An electronic device includes a memory, a processor, and a program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps in the signal recognition and matching method for substation anti-error logic verification as described in Example 1 of the present invention.

[0100] The detailed steps are the same as those of the signal recognition and matching method for substation anti-error logic verification provided in Example 1, and will not be repeated here.

[0101] Example 4

[0102] Example 4 of the present invention provides a computer-readable storage medium.

[0103] A computer-readable storage medium stores a program, which when executed by a processor implements the steps in the signal recognition and matching method for substation anti-error logic verification as described in Example 1 of the present invention.

[0104] The detailed steps are the same as those of the signal recognition and matching method for substation anti-error logic verification provided in Example 1, and will not be repeated here.

[0105] In summary, compared with the prior art, the beneficial effects of the present invention are as follows:

[0106] 1. By extracting the combined character elements with actual meanings in the substation-wide power text data, classifying the combined character elements according to the key feature information of the combined character elements to establish a power text comparison database. Then, based on the power text comparison database, determining the degree of association between the recognized power text content in the substation-wide power text data and different key feature information, and further establishing a power text indexing system based on the degree of association, so that the power text indexing system preferentially indexes the combined character elements under the classification corresponding to the key feature information with the highest degree of association with the input power text content as the matching result. It is possible to determine the matching situation of each combined character element in the power text to be recognized and matched with each combined character element in the corresponding classification of the key feature information, realizing efficient, accurate, and comprehensive matching of various signals in the power system, thus ensuring the comprehensiveness, accuracy, and efficiency of substation anti-error logic verification.

[0107] 2. Based on the number of combined character elements in the current power text content, the types of combined character elements related to the key feature information a included in the current power text content, and the number of occurrences of various combined character elements related to a in the current power text content, determine the degree of association between the current power text content and the key feature information a, which can better quantify the degree of association between the power text content and each key feature information, contribute to further improving the efficiency and accuracy of information retrieval, quickly locate and extract the key information in the text, and further improve the information processing efficiency.

[0108] 3. By introducing an interactive verification mechanism based on neural network recognition, verify and screen the matching results, and regularly verify and optimize the calculation results of the degree of association and the power text indexing system, which can further ensure the accuracy and reliability of the matching results to improve the accuracy and reliability of signal recognition.

[0109] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to the specific embodiments described. Obviously, many modifications and variations can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the present invention, so that those skilled in the relevant technical fields can understand and utilize the present invention well. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A signal recognition and matching method for substation anti-error logic verification, characterized in that Including: Step 1: Collect the power text data of the entire substation; Step 2: Extract the combined character elements from the power text data of the entire substation, classify the combined character elements according to the key feature information of the combined character elements, and establish a power text comparison database based on the classified combined character elements; Step 3: Identify the power text content from the power text data of the entire substation, and determine the degree of association between the power text content and different key feature information based on the power text comparison database; Step 4: Establish a power text indexing system so that the power text indexing system preferentially indexes the combined character elements under the classification corresponding to the key feature information with the highest degree of association with the input power text content; Step 5: Obtain and identify the power text content included in the signal to be processed, and index the power text content included in the signal to be processed in the power text comparison database through the power text indexing system to obtain the corresponding matching result.

2. The signal recognition and matching method for substation anti-misoperation logic verification according to claim 1, characterized in that Step 1 includes: Step 11: Collect various types of pictures related to the entire substation, including: model file pictures, design pictures, and power special naming class file pictures; Step 12: Based on the fully convolutional neural network, extract text information from the collected various types of pictures through symbol recognition, text recognition, connection recognition, and feature element classification and marking as the power text data of the entire substation; Step 13: Clean the collected power text data, remove redundant and incorrect information, and format the cleaned data to meet the input requirements of natural language processing technology.

3. The signal recognition and matching method for substation anti-misoperation logic verification according to claim 1, characterized in that Step 2 includes: Step 21: Use natural language processing tools to perform word segmentation and part-of-speech tagging on the text in the power text data; Step 23: Classify the combined character elements according to the key feature information of the extracted combined character elements; the key feature information includes: equipment name information, operation instruction information, and equipment status information.

4. The signal recognition and matching method for substation anti-misoperation logic verification according to claim 1, characterized in that Step 3 includes: Based on the number of combined character elements in the current power text content A, the types of combined character elements related to the key feature information a included in the current power text content A, and the number of occurrences of various combined character elements related to a in the current power text content A, determine the degree of association between the current power text content A and the key feature information a.

5. The signal recognition and matching method for substation anti-misoperation logic verification according to claim 4, characterized in that Step 3 further includes: Use the following formula to determine the degree of association between the current power text content A and the key feature information a: Wherein, U represents the number of combined character elements of the power text content A; u(a) represents the number of combined character elements containing the key feature information a in the power text content A; x represents the x-th combined character element containing the key feature information a in the power text content A; s(x) represents the number of times the x-th combined character element appears in the power text content A; m represents the number of types of combined character elements that appear in the power text content A. Set a = {1, 2, 3}, where 1, 2, and 3 represent equipment name information, operation instruction information, and equipment status information respectively; n is the number of types of combined character elements containing the key feature information a in the power text content A.

6. The signal recognition and matching method for substation anti-error logic verification according to claim 1, characterized in that In step 4, the establishment of the power text indexing system further includes: By adopting methods such as regular expression matching, keyword matching, logical matching, and neural network recognition matching, the power text indexing system performs further indexing under the classification corresponding to the key feature information with the highest degree of association in the power text comparison database, and obtains the combined character elements that are the same as the combined character elements included in the input power text content as the matching result.

7. The signal recognition and matching method for substation anti-error logic verification according to claim 6, characterized in that: In step 4, the establishment of the power text indexing system further includes: Based on the interactive verification mechanism of neural network recognition, verify and screen the matching results; Regularly update the power text data of the entire substation and the power text comparison database, and correspondingly update the degree of association between the power text content and different key feature information. Based on the updated degree of association, verify and optimize the established power text indexing system.

8. The signal recognition and matching method for substation anti-error logic verification according to claim 1, characterized in that In step 5, identifying the power text content included in the signal to be processed includes: When the type of the signal to be processed is an image signal, use computer vision technology to identify and process the image signal, extract the text information in the image signal and convert it into power text content; When the type of the signal to be processed is a voice signal, use speech recognition technology to identify and process the voice signal, extract the text information in the voice signal and convert it into power text content; When the signal to be processed is a text signal, extract the power text content in the text signal through EP-NER technology.

9. A signal recognition and matching device for substation anti-error logic verification using the method according to any one of claims 1-8, characterized in that, It includes: A data collection module for collecting the power text data of the entire substation; A database establishment module for extracting the combined character elements from the power text data of the entire substation, classifying the combined character elements according to the key feature information of the combined character elements, and establishing a power text comparison database based on the classified combined character elements; An association degree determination module for identifying the power text content from the power text data of the entire substation, and determining the association degree between the power text content and different key feature information based on the power text comparison database. An indexing system establishment module, configured to establish a power text indexing system based on the degree of association, so that the power text indexing system preferentially indexes combined character elements under the classification corresponding to the key feature information with the highest degree of association with the input power text content; A matching module, configured to identify the power text content included in the signal to be processed, and match the power text content included in the signal to be processed in the power text comparison database through the power text indexing system.

10. An electronic device, comprising a processor and a storage medium; characterized in that: The storage medium is used for storing instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-8.

Images