Substation four-remote information point table generation method and related device

Abstract

The invention discloses a transformer substation four-remote information point table generation method and a related device, and relates to the technical field of power system automation, and the method comprises the steps: collecting multi-modal data used for generating a transformer substation four-remote information point table, the multi-modal data comprises drawing graphic data, text data, equipment identification, network communication acquisition data and standard basic data; carrying out data processing and analysis on the multi-modal data based on a self-adaptive AI algorithm; designing a template library based on a natural language processing technology, and generating a transformer substation four-remote information point table based on the template library and in combination with the multi-modal data analyzed by the data processor; the transformer substation four-remote information point table and the transformer substation SCD file are subjected to synchronous verification, the transformer substation four-remote information point table passing the synchronous verification is output, and the synchronous verification comprises version consistency verification. The method solves the problems that in the prior art, efficiency and accuracy are low, and the version consistency of the four-remote information point table of the transformer substation cannot be guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system automation, and in particular to a substation four-remote information point table generation method and related device. BACKGROUND

[0002] In the new construction and expansion project of the substation, the four-remote (telemetry, remote signaling, remote control, remote adjustment) information point table of the substation automation system is the key basis for realizing the remote monitoring and automatic operation of the substation. These information details record the information collection points and control command points of various devices in the substation, ensuring that the dispatch center can accurately obtain the real-time operation status of the substation equipment and correctly monitor and control the equipment remotely.

[0003] At present, the generation of the four-remote information point table of the substation automation is mainly completed by manual sorting. However, due to the large number of substation devices and complex types, the number of four-remote information point tables involved in the substation is huge; the version management of the information point table is chaotic, and when the point table and the SCD file version are not synchronized, repeated corrections are needed, the manual checking efficiency is low and easy to miss (such as CRC check missing); and the process of manually generating the four-remote information point table is complicated, and is prone to information omission, errors, and duplication, and has low editing and generating efficiency. Therefore, it is urgent to design a four-remote information point table generation method that can realize high-precision version management and is efficient and accurate. SUMMARY

[0004] The present application provides a substation four-remote information point table generation method and related device, which is used to solve the problems of low efficiency and accuracy of the prior art and inability to ensure the consistency of the version of the substation four-remote information point table.

[0005] Therefore, the first aspect of the present application provides a substation four-remote information point table generation method applied to a multi-modal robot, which comprises:

[0006] Collecting multi-modal data for generating a substation four-remote information point table, the multi-modal data comprising: drawing graphic data, text data, device identification, network communication collection data, and specification standard basic data;

[0007] Performing data processing and analysis on the multi-modal data based on an adaptive AI algorithm;

[0008] Designing a template library based on natural language processing technology, generating a substation four-remote information point table based on the template library and combining the multi-modal data analyzed by the data processing machine;

[0009] Synchronously checking the substation four-remote information point table and the substation SCD file, and outputting the substation four-remote information point table that passes the synchronous check, the synchronous check comprising: version consistency check.

[0010] Optionally, the network communication collection data includes: a total station SCD file, a running state of a device, measurement data, and event information.

[0011] Optionally, the data processing and analysis of the multi-modal data based on the adaptive AI algorithm includes:

[0012] information in the drawing graphic data is recognized by using a deep learning algorithm image recognition technology and is converted into structured data;

[0013] interval key information in the drawing graphic data is extracted by combining an image detection OCR technology and a regular expression matching method;

[0014] key information of the text data is extracted by using a natural language processing method.

[0015] Optionally, the information in the drawing graphic data is recognized by using a deep learning algorithm image recognition technology and is converted into structured data, and the method includes:

[0016] the image in the drawing graphic data is preprocessed, and the preprocessing includes: noise reduction, grayscale, and binarization;

[0017] the preprocessed image is subjected to feature extraction processing, and the feature extraction processing includes: edge detection, region segmentation, and texture feature extraction;

[0018] the extracted features are input into a pre-trained recognition model, an identification result of an element in the image is output, and the identification result is converted into structured data.

[0019] Optionally, the design of the template library based on the natural language processing technology, the generation of the substation four-remote information point table based on the template library and in combination with the multi-modal data analyzed by the data processing machine include:

[0020] the template library based on the natural language processing technology is designed, the multi-modal data is converted into structured information by taking the template in the template library as a framework, and the substation four-remote information point table is generated by using a customized rule of a natural language parsing engine.

[0021] Optionally, the synchronous verification of the substation four-remote information point table and the substation SCD file includes:

[0022] the key information and the check code in the substation four-remote information point table and the substation SCD file are compared by using a cyclic redundancy check method, and it is judged whether the versions of the two are consistent according to a comparison result.

[0023] Optionally, the multi-modal robot adopts a pluggable interface design, carries several data reading modules and is equipped with several communication interfaces.

[0024] The second aspect of the present application provides a substation four-remote information point table generation system applied to a multi-modal robot, the system comprising:

[0025] A collection unit is configured to collect multi-modal data for generating a substation four-remote information point table, the multi-modal data comprising: drawing graphic data, text data, equipment identification, network communication collection data and specification standard basic data.

[0026] A processing unit is configured to perform data processing and analysis on the multi-modal data based on an adaptive AI algorithm.

[0027] A generation unit is configured to design a template library based on natural language processing technology, and generate a substation four-remote information point table based on the template library and the multi-modal data analyzed by the data processing unit.

[0028] An output unit is configured to perform synchronous verification on the substation four-remote information point table and a substation SCD file, and output the substation four-remote information point table that passes the synchronous verification, the synchronous verification comprising: version consistency verification.

[0029] The third aspect of the present application provides a substation four-remote information point table generation device, the device comprising a processor and a memory:

[0030] The memory is configured to store program code and transmit the program code to the processor.

[0031] The processor is configured to execute the steps of the substation four-remote information point table generation method according to the instructions in the program code.

[0032] The fourth aspect of the present application provides a computer readable storage medium for storing program code, the program code being used to execute the substation four-remote information point table generation method of the first aspect.

[0033] From the above technical solutions, the present application has the following advantages:

[0034] The substation four-remote information point table generation method provided by the embodiment of the present application first collects various data of the substation site through a multi-modal robot, which covers the running state, parameters and other aspects of the equipment. During the collection process, the robot uses various sensors equipped to obtain data in a high-precision and high-efficiency manner, ensuring the integrity and accuracy of the data. Then, deep learning AI algorithms are used to deeply mine and analyze the data, so that the key information of the substation can be quickly and accurately obtained, and structured data can be generated for the automatic generation of the information point table, greatly improving the adaptability. Then, through a natural language parsing engine, the key parameters of the four-remote information point table are accurately and quickly generated according to customized expression rules. Finally, the preliminary table generated is audited and proofread again, and compared with historical data and industry standards to ensure the accuracy and consistency of the information. Further, the four-remote information point table that passes the audit can be presented in a clear and standardized format, which is convenient for operators to view and use, and can be stored in a database according to actual needs for subsequent query and management. Thus, the problems of low efficiency and accuracy and inability to guarantee the version consistency of the substation four-remote information point table in the prior art are solved. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or the prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0036] Figure 1 The flowchart of the substation four-remote information point table generation method provided by the embodiment of the present application is shown in the figure.

[0037] Figure 2 The structural diagram of the substation four-remote information point table generation system provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0038] In order to make the purposes, features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the following described embodiments are only some of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0039] Please refer to Figure 1The method for generating a four-remote information point table of a transformer substation provided in the embodiment of the application is applied to a multi-modal robot, and the method comprises the following steps:

[0040] In one embodiment, the multi-modal robot of the application adopts a pluggable interface design, carries several data reading modules, and is equipped with several communication interfaces.

[0041] It should be noted that, in order to facilitate on-site operation of staff and obtain data of various devices on site, the multi-modal robot of the application carries multiple data reading modules, has the characteristics of convenient carrying and real-time operation. Moreover, the multi-modal robot adopts a pluggable interface design, can carry multiple data reading modules such as an RFID tag scanner and a multifunctional barcode scanner, and is equipped with multiple communication interfaces including but not limited to an Ethernet interface, an RS485 / RS232 interface, a USB interface, a Bluetooth WIFI interface and the like, so as to facilitate real-time collection of data of various devices on site. The pluggable interface design enables the multi-modal robot to be conveniently matched with different devices and sensors, so as to realize rapid collection and transmission of data; the multiple communication interfaces make the multi-modal robot be able to adapt to various application scenarios, whether it is in communication with an automation system in a transformer substation or in data interaction with external devices, it can be easily realized.

[0042] In step 101, multi-modal data for generating a four-remote information point table of a transformer substation is collected, and the multi-modal data comprises drawing graphic data, text data, device identification, network communication collection data and specification standard basic data.

[0043] It should be noted that four-remote refers to telemetry, remote signaling, remote control and remote adjustment. The drawing graphic data is obtained by a high-definition camera carried by the robot, and the high-definition camera obtains graphic information such as a primary wiring diagram of a transformer substation, a display screen screenshot of a system or device or an electronic graphic file made by a manufacturer; the text data is derived from paper or electronic documents such as design drawing instructions of a transformer substation, device parameter manuals, operation records and initial point table materials; the device identification is obtained by reading device RFID tag data by an RFID tag scanner carried by the robot and reading device barcode or two-dimensional code data by a multifunctional barcode scanner; the specification standard basic data (including standards and specifications related to the power industry, technical regulations and the like) of a power grid company obtained by reading from a built-in storage module or network reading provides a basis for monitoring information point table generation; the network communication collection data is obtained by connecting the multi-modal robot with a communication network in a transformer substation, and real-time acquisition of data such as a full-station SCD file, a running state of a device, measurement data and event information. The multi-modal data obtained by the application in various ways is integrated and used for subsequent generation of a four-remote information point table of a transformer substation, so as to comprehensively and accurately reflect the actual situation of a transformer substation and improve scene adaptability and correctness of point table generation.

[0044] Step 102, data processing and analysis of multi-modal data based on adaptive AI algorithm.

[0045] In one embodiment, step 102 includes:

[0046] Step 1021, after recognizing the information in the drawing graphic data using deep learning algorithm image recognition technology, the information is converted into structured data.

[0047] Specifically, step 1021 includes:

[0048] The image in the drawing graphic data is preprocessed, which includes denoising, graying, and binarization.

[0049] It should be noted that the specific method of preprocessing is as follows: (1) denoising: first, use a filtering algorithm to remove noise in the image, such as Gaussian filtering to smooth the image and reduce random noise interference. (2) Grayscale: If there are color pictures, the color image needs to be converted to a grayscale image to simplify subsequent processing, and the color of each pixel point is represented by a gray value. (3) Binarization: By setting a threshold, the grayscale image is converted to a binary image, separating the objects and background in the image, making it easier to extract features later.

[0050] The preprocessed image is subjected to feature extraction processing, which includes edge detection, region segmentation, and texture feature extraction.

[0051] It should be noted that the specific method of feature extraction processing is as follows: (1) Edge detection: Canny algorithm (Canny algorithm is a computer vision algorithm for edge detection) is used to detect the edges in the image, obtaining the edge information of lines, device contours, etc. in the wiring diagram. (2) Region segmentation: Threshold segmentation, clustering, and other methods are used to segment the image into different regions, such as dividing different devices and lines into their respective regions. (3) Texture feature extraction: Analyze the texture information of the image, such as the texture of the device surface and the thickness of the line, to assist in recognition.

[0052] The extracted features are input into a pre-trained recognition model, which outputs the recognition results of the elements in the image, and the recognition results are converted into structured data.

[0053] It should be noted that the model training and identification of the recognition software specifically includes: (1) selecting a model: using a model suitable for image recognition such as a convolutional neural network (CNN) which has the ability to automatically extract image features. (2) Data labeling: label a large number of substation primary wiring diagrams, mark different equipment, lines and other elements as training data. (3) Model training: input the labeled data into the model for training, adjust the parameters of the model so that it can accurately identify various elements. (4) Identification and classification: for example, input the wiring diagram to be identified into the trained model, and the model outputs the identification result of the elements in the image, such as determining which equipment it is and which line it belongs to. Further, the image recognition result can be optimized and post-processed, including: (1) Result optimization: using adaptive algorithms to automatically adjust the parameters or feature extraction method of the model according to the difference between the recognition result and the actual situation, to improve the recognition accuracy. (2) Post-processing: analyze and process the recognition result, such as removing misidentified elements, merging adjacent elements of the same type, etc., to obtain the final accurate recognition result.

[0054] Further, the recognition result is converted into structured data, thereby facilitating subsequent data storage, query and analysis. The structured data can be presented in the form of a table, with each row representing an identified element and each column corresponding to different attributes of the element, such as device name, device type, line affiliation, coordinate position, etc. Such clear structure helps to efficiently manage and utilize these data. In practical applications, the generated structured data can be used to create a substation four-remote information point table. Through further processing and integration of the structured data, the identified device and line information are associated with the four-remote information. For example, according to the device type and location information, determine the corresponding telemetry, remote signaling, remote control, and remote adjustment information points. At the same time, structured data can also provide strong support for the operation and maintenance management of substations. Operation and maintenance personnel can query structured data to quickly understand the distribution and status of devices within the substation. When a fault occurs, the fault device and related lines can be quickly located based on these data, improving the efficiency of fault diagnosis and repair. In addition, it is also feasible to interface the structured data with other systems. For example, integrate with the monitoring system of the substation to realize real-time data sharing and interaction, so that the monitoring system can more accurately reflect the actual operation of the substation. It can also be connected with the asset management system to provide data basis for the whole life cycle management of equipment.

[0055] Step 1022, combine image detection OCR technology and regular expression matching method to extract interval key information in drawing graphic data.

[0056] It should be noted that the image detection OCR technology and the regular expression matching method are combined to realize the intelligent extraction of interval key information. After accurately recognizing the text information in the image by the OCR technology, the pre-defined regular expression rules are used to accurately match and extract the interval data. This method not only can effectively process interval information of different formats, but also can ensure the accuracy and consistency of key data extraction, significantly improving the automation level and efficiency of interval information processing.

[0057] The image detection OCR technology and the regular expression matching method are combined to extract the interval key information in the drawing graphic data. The specific method is as follows: first, the image detection technology is used to pre-process the drawing graphic, enhance the image definition and contrast, and remove noise interference, so that the OCR technology can more accurately recognize the text. After pre-processing, the OCR technology is used to recognize the text in the drawing graphic and convert it into editable text information. Then, according to the pre-designed regular expression rules, the converted text information is filtered and matched. These regular expression rules are customized according to the characteristics and format of the interval key information, and can accurately locate and extract the required key information. For some complex drawing graphics, there may be multiple different regions containing interval key information, at which time image detection and OCR recognition need to be performed on different regions, and then regular expression matching is performed. In the matching process, context information can also be combined for further verification and screening to ensure the accuracy and integrity of the extracted interval key information. For errors or ambiguous information that occur during recognition and matching, a combination of manual intervention and automatic error correction is used to improve the reliability and efficiency of the entire extraction process. At the same time, in order to facilitate subsequent data processing and analysis, the extracted interval key information is stored and organized according to a certain format, and a corresponding database or data table is established to facilitate subsequent query and use.

[0058] Note: OCR (Optical Character Recognition) technology is a technology that converts the text of various documents, newspapers, books, manuscripts and other printed matter into image information through scanning and other optical input methods, and then converts the image information into editable text using text recognition technology.

[0059] The following is the recognition result obtained by recognizing the case picture:

[0060] Substation name: 110kV Haibin Substation Voltage (110kV), interval unit classification lists each interval information, including:

[0061] 110kV main transformer interval:

[0062]

[0063] Feeder bay:

[0064]

[0065] 110kV bus bay:

[0066]

[0067] 110kV section bay:

[0068]

[0069] Feeder bay (outgoing bay):

[0070]

[0071] Step 1023, extracting key information of the text data by using natural language processing method.

[0072] It should be noted that for text data, key information such as device name, function description, parameter value, etc. is extracted by using natural language processing method.

[0073] Note: Natural Language Processing (NLP) is a field at the intersection of computer science, artificial intelligence, and linguistics, aiming to enable computers to understand, process, and generate human language. It includes a series of techniques and tasks, such as: Natural language understanding: enabling computers to understand the meaning of text, such as text classification (classifying text into different categories), sentiment analysis (judging the positive, negative or neutral sentiment expressed in text), information extraction (extracting specific information from text) and other tasks. Natural language generation: enabling computers to generate natural and fluent text, such as machine translation (translating one language into another), text summarization (automatically generating summaries of text), dialogue systems (enabling computers to have natural conversations with humans) and other tasks.

[0074] Step 103, designing a template library based on natural language processing technology, generating a substation four-remote information point table based on the template library and combining multi-modal data analyzed by the data processing machine.

[0075] In one embodiment, step 103 includes:

[0076] Designing a template library based on natural language processing technology, using templates in the template library as a framework, converting multi-modal data into structured information, and generating a substation four-remote information point table using customized rules of a natural language parsing engine.

[0077] It should be noted that the template library based on natural language processing technology is designed to realize the bidirectional intelligent derivation of key parameters in the four-remote information generation process. Through the natural language analysis engine, the key parameters of the four-remote information point table are accurately generated according to the customized expression rules. Relying on the structured support of the template library, the system establishes a bidirectional interaction mechanism between the four-remote information template and the instance data, which runs through the whole process of information point generation. This mechanism not only supports the forward output of data, but also realizes the real-time reverse derivation of parameters.

[0078] It can be understood that the multi-modal data analyzed by the data processing machine is input into the preset natural language analysis engine, so that the natural language analysis engine generates and outputs the transformer substation four-remote information point table according to the customized expression rules. The specific method is as follows: first, the customized expression rules are deeply analyzed. The rules are formulated in combination with the characteristics and actual needs of the transformer substation four-remote information, covering data format, data correlation, data filtering conditions and other aspects. The purpose of analyzing the rules is to build an accurate processing framework for the natural language analysis engine, so that it can accurately understand the conversion logic between the input data and the output information point table. Then, the natural language analysis engine performs preliminary preprocessing on the input multi-modal data analyzed by the data processing machine. Multi-modal data may include images, texts, numerical values and other forms. The preprocessing process includes data cleaning, removing noise, duplicate data and invalid information; normalizing different modal data to unify the format and measurement standard of the data for more efficient processing in the future. Then, the engine extracts features from the preprocessed data according to the analyzed customized expression rules. The key features related to the transformer substation four-remote information point table are identified from the multi-modal data, such as device state features, power parameter features, etc. These features will serve as the basis data for generating the information point table. Then, the extracted features are integrated and converted using the correlation and logical operations in the rules. Different sources and types of features are combined into data entries that meet the format of the four-remote information point table according to the rules. In this process, the engine will perform real-time verification on the data to ensure that the generated data entries meet the constraints of the rules, ensuring the accuracy and consistency of the data. Finally, the integrated and converted data are organized and arranged according to the structure of the transformer substation four-remote information point table to generate a complete information point table. The generated information point table is finally audited and optimized to check whether there are any missing or incorrect information, and the data that do not conform to the actual situation are adjusted and corrected, and finally a high-quality transformer substation four-remote information point table is output.

[0079] Step 104, synchronously check the transformer substation four-remote information point table and the transformer substation SCD file, and output the transformer substation four-remote information point table that passes the synchronous check, and the synchronous check includes version consistency check.

[0080] In one embodiment, the step 104 of synchronously checking the substation four-remote information point table and the substation SCD file includes:

[0081] The cyclic redundancy check method is used to compare the key information and the check code in the substation four-remote information point table and the substation SCD file, and it is determined whether the versions of the two are consistent according to the comparison result.

[0082] It should be noted that the generated four-remote information point table is synchronously checked with the substation SCD (Substation Configuration Description) file. Specifically, after the point table is generated, the CRC (Cyclic Redundancy Check) and other checking algorithms are automatically used to perform version consistency checking on the SCD file and the point table. By comparing the key information and the check code in the SCD file and the point table, the consistency of the two versions is ensured, and errors and conflicts caused by inconsistent versions are avoided. This automatic checking method can effectively reduce manual review errors and improve the quality and reliability of the generated information point table.

[0083] The specific implementation steps of the synchronous checking are as follows:

[0084] 1) Parse the file:

[0085] (1) Four-remote information point table parsing: using the parsing program, according to the format specification of the point table, the telemetry, remote signaling, remote control, remote adjustment and other information in the four-remote information point table are extracted to form an operable data structure, for example, the name, number, function description, data type and other attributes of each information point are stored in the data objects or database tables in the memory.

[0086] (2) SCD file parsing: according to the IEC 61850 standard and the SCL configuration language specification, a special SCD parser is used to parse the SCD file into a model object in the memory. During the parsing process, the information related to four-remote is extracted, including the configuration information of IED device, signal connection relationship, data object definition, etc.

[0087] 2) Synchronous checking:

[0088] (1) Information comparison: the software compares the parsed four-remote information point table data with the corresponding four-remote information in the SCD file one by one. For telemetry information, compare whether the name, number, data type, range and other attributes of the measurement point are consistent; for remote signaling information, check whether the signal name, state value definition, associated device and other information are matched; for remote control and remote adjustment information, check whether the control object, operation command, parameter setting and other information are consistent.

[0089] (2) Connection relationship verification: Check whether the communication connection relationships between IED devices described in the SCD file are consistent with the devices involved in the four-remote information point table.

[0090] (3) Data consistency check: The device verifies whether the data values in the four-remote information point table are consistent with the default values or current values of the corresponding data objects in the SCD file. If there is a difference, it is recorded and further analyzed whether it is a configuration error or a difference caused by the actual running state.

[0091] 3) Version consistency verification:

[0092] (1) Version number extraction: The software extracts version number information from the four-remote information point table and the SCD file respectively. The version number is usually defined in the header or specific metadata area of the file, which can be obtained by parsing the relevant part of the file.

[0093] (2) Compare version numbers: The software compares the two version numbers extracted to determine whether they are the same. If the version numbers are consistent, it means that the two files are compatible in version; if the version numbers are different, further analysis of the difference is needed.

[0094] (3) Version compatibility check: Even if the version numbers are different, it is necessary to check whether there is compatibility between the two versions. In some cases, although the version number has been updated, it may only have made some small modifications or functional enhancements that do not affect the consistency of the four-remote information, at which point it can be determined whether such version differences can be accepted according to the specific modification content and system requirements.

[0095] 4) Result output and report generation:

[0096] (1) Record differences: During the synchronization verification and version consistency verification process, all differences and problems found are recorded in detail. This includes the specific content of the difference, the location (such as the specific information point, device name, etc.), the possible impact, etc.

[0097] (2) Generate report: According to the verification results, generate a detailed verification report. The report should include basic information about the verification (such as verification time, file versions involved in the verification, etc.), a summary of the verification results (whether the verification is passed, the number of problems, etc.), a specific list of differences and analysis, suggested solutions, etc.

[0098] Further, the multi-modal robotic device for field operation is described as follows:

[0099] The operator can view the generated four-remote information point table through the human-machine interface of the robot device and make necessary editing and adjustment. Since the device adopts a pluggable interface design, the operator can conveniently connect different equipment and sensors to obtain on-site equipment data according to actual needs; and the device is connected with a substation automation system or other external equipment or system by using various communication interfaces, so that data can be quickly transmitted and shared.

[0100] The substation four-remote information point table generation method provided in the embodiment of the application first collects various types of data on the site of the substation through a multi-modal robot, which covers the running state, parameters and other information of the equipment. During the collection process, the robot uses various sensors equipped by itself to obtain data in a high-precision and high-efficiency manner, so as to ensure the integrity and accuracy of the data. Then, a deep learning AI algorithm is used to deeply mine and analyze the data, so that the key information of the substation can be quickly and accurately obtained, and structured data can be generated for the automatic generation of the information point table, which greatly improves the adaptability. Then, the natural language parsing engine is used to accurately and quickly generate the key parameters of the four-remote information point table according to the customized expression rules. Finally, the preliminary table generated is rechecked and proofread, and is compared with historical data and industry standards, so as to ensure the accuracy and consistency of the information. Further, the four-remote information point table that passes the review can be presented in a clear and standardized format, which is convenient for the operator to view and use, and can be stored in a database according to actual needs for subsequent query and management. Thus, the problems of low efficiency and accuracy and the inability to guarantee the version consistency of the substation four-remote information point table in the prior art are solved.

[0101] The above is a substation four-remote information point table generation method provided in the embodiment of the application, and the following is a substation four-remote information point table generation system provided in the embodiment of the application.

[0102] Please refer to Figure 2 The substation four-remote information point table generation system provided in the embodiment of the application is applied to a multi-modal robot, and the system comprises:

[0103] The acquisition unit 201 is configured to acquire multi-modal data for generating a substation four-remote information point table, and the multi-modal data comprises drawing graphic data, text data, equipment identification, network communication acquisition data and specification standard basic data.

[0104] The processing unit 202 is configured to perform data processing and analysis on the multi-modal data based on an adaptive AI algorithm.

[0105] The generation unit 203 is configured to design a template library based on natural language processing technology, and generate a substation four-remote information point table based on the template library and the multi-modal data analyzed by the data processing machine.

[0106] The output unit 204 is configured to perform synchronization checking on the transformer substation four-remote information point table and the transformer substation SCD file, and output the transformer substation four-remote information point table that passes the synchronization checking, and the synchronization checking includes version consistency checking.

[0107] Further, the embodiment of the present application also provides a transformer substation four-remote information point table generation device, the device comprising a processor and a memory:

[0108] The memory is configured to store program code and transmit the program code to the processor.

[0109] The processor is configured to execute the steps of the transformer substation four-remote information point table generation method according to the instructions in the program code.

[0110] Further, the embodiment of the present application also provides a computer readable storage medium, the computer readable storage medium is configured to store program code, and the program code is configured to execute the transformer substation four-remote information point table generation method according to the above method embodiment.

[0111] Those skilled in the art can clearly understand that, for the convenience and brevity of the description, the specific working process of the above-described system and unit can refer to the corresponding process in the foregoing method embodiment, and will not be repeated here.

[0112] In several embodiments provided by the present application, it should be understood that the disclosed system, device and method can be implemented by other manners. For example, the above-described device embodiment is only schematic, for example, the division of the unit is only a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the displayed or discussed units can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.

[0113] The units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. According to actual needs, part or all of the units can be selected to achieve the purpose of the embodiment scheme.

[0114] In addition, each function unit in each embodiment of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software function unit.

[0115] If the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in the form of a contribution to the prior art, or all or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in each embodiment of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0116] The above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of each embodiment of the present application.

Claims

1. A method for generating a substation remote information point table, characterized in that, Applied to multimodal robots, the methods include: Multimodal data is collected to generate a table of remote sensing information points for substations. The multimodal data includes: drawing graphic data, text data, equipment identification, network communication data, and basic data of specifications and standards. The multimodal data is processed and analyzed based on an adaptive AI algorithm; Design a template library based on natural language processing technology, and generate a substation remote information point table based on the template library and multimodal data analyzed by a data processor. The substation remote information point table and the substation SCD file are synchronized and verified, and the substation remote information point table that passes the synchronization verification is output. The synchronization verification includes: version consistency verification.

2. The method for generating a substation remote information point table according to claim 1, characterized in that, The network communication data collected includes: the full-site SCD file, the device's operating status, measurement data, and event information.

3. The method for generating a substation remote information point table according to claim 1, characterized in that, The data processing and analysis of the multimodal data based on the adaptive AI algorithm includes: The information in the drawing graphic data is identified using deep learning algorithm image recognition technology and then converted into structured data; The key information of intervals in the drawing graphic data is extracted by combining image detection OCR technology with regular expression matching methods. Key information from the text data is extracted using natural language processing methods.

4. The method for generating a substation remote information point table according to claim 3, characterized in that, The process of using deep learning algorithm image recognition technology to identify information in the drawing graphic data and converting it into structured data includes: The images in the drawing graphic data are preprocessed, including noise reduction, grayscale conversion, and binarization. The preprocessed image is subjected to feature extraction processing, which includes edge detection, region segmentation, and texture feature extraction. The extracted features are input into a pre-trained recognition model, which outputs the recognition results of elements in the image and converts the recognition results into structured data.

5. The method for generating a substation remote information point table according to claim 1, characterized in that, The design is based on a template library using natural language processing technology. Based on this template library and combined with multimodal data analyzed by a data processing machine, a substation remote sensing information point table is generated, including: Design a template library based on natural language processing technology. Using the templates in the template library as a framework, convert multimodal data into structured information, and use customized rules of the natural language parsing engine to generate a substation remote information point table.

6. The method for generating a substation remote information point table according to claim 1, characterized in that, The synchronization verification of the substation remote information point table and the substation SCD file includes: The key information and check codes in the substation remote information point table and the substation SCD file are compared using the cyclic redundancy check method. The comparison results are used to determine whether the two versions are consistent.

7. The method for generating a substation remote information point table according to claim 1, characterized in that, The multimodal robot adopts a pluggable interface design and is equipped with several data reading modules and several communication interfaces.

8. A substation remote information point table generation system, characterized in that, The system is applied to multimodal robots and includes: The acquisition unit is used to acquire multimodal data for generating the substation remote information point table. The multimodal data includes: drawing graphic data, text data, equipment identification, network communication acquisition data, and standard data. The processing unit is used to perform data processing and analysis on the multimodal data based on an adaptive AI algorithm; The generation unit is used to design a template library based on natural language processing technology, and to generate a substation remote information point table based on the template library and combined with multimodal data analyzed by the data processor. The output unit is used to perform synchronous verification between the substation remote information point table and the substation SCD file, and output the substation remote information point table that has passed the synchronous verification. The synchronous verification includes: version consistency verification.

9. A device for generating a substation remote information point table, characterized in that, The device includes a processor and a memory: The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the substation remote information point table generation method according to any one of claims 1-7 according to the instructions in the program code.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store program code for executing the substation remote information point table generation method according to any one of claims 1-7.

Images