A power transformation equipment test data dynamic mapping method and system based on digital twinning

Abstract

This invention relates to the field of digital mapping technology for power equipment, specifically a method and system for dynamic mapping of power equipment test data based on digital twins. The method corrects physical wiring and digital twin endpoint identifiers, constructs a connection network, searches for conduction paths based on switch states to match data, compares three-dimensional coordinates to extract spatial offset positions, verifies the spacing between adjacent points and corrects the mapping along the ideal direction, analyzes connectivity changes in conjunction with equipment time-series status, dynamically replaces virtual positions and reconfigures test records, and generates dynamic mapping results for test data. This invention corrects the correspondence between endpoints and connection identifiers through cross-checking, compresses the range of virtual-physical misalignment, identifies the real path based on conduction status to prevent data from spreading along incorrect branches, and superimposes three-dimensional offset recognition, adjacency constraints, and continuous time slice connectivity correction to converge spatial deviations and jump risks. This makes the test record connection positions more stable, the evolution path clearer, and anomalies easier to trace in dynamic scenarios.

Description

Technical Field

[0001] This invention relates to the field of digital mapping technology for power equipment, and in particular to a method and system for dynamic mapping of power equipment test data based on digital twins. Background Technology

[0002] The field of digital mapping technology for power equipment encompasses aspects such as power system equipment status modeling, data acquisition, data association, and virtual-real synchronization. This technical field revolves around the correspondence between the operating status of power equipment and its digital model. By collecting, recording, and organizing the operating parameters of primary and secondary equipment, it unifies and structures the multi-source data, including current, voltage, temperature, partial discharge spectrum response, and insulation parameters. Based on equipment ledger information, it establishes a correspondence between equipment entities and data. This field also involves data time series alignment, spatial location calibration, equipment structural feature description, and the integrated management of experimental and operational data, thereby forming a data expression system for the entire equipment lifecycle, providing fundamental support for subsequent status analysis and data mapping.

[0003] Among them, the dynamic mapping method for test data of substation equipment based on digital twin refers to a processing method that uses the digital twin model of substation equipment as a carrier to map various test data acquired during the test according to the time tag, equipment number, measurement point location, and test item. The technical matters involved include the test data acquisition unit recording AC withstand voltage test data, DC resistance test data, dielectric loss data, and partial discharge test data; associating the data with the corresponding equipment entity according to the equipment coding rules; realizing the positioning of data in the three-dimensional model by establishing the correspondence between the measurement point coordinates and the equipment structural model; serializing and organizing multiple batches of test data according to the timestamp order; mapping data of different stages to the corresponding model state nodes in combination with the equipment operation stage division rules; and using data field matching rules and structure mapping rules to complete the continuous updating and association processing of test data in the digital model.

[0004] Existing technologies focus on encoding association, field matching, and coordinate positioning. In practice, the mapping often relies on static correspondence to maintain data connection. When there is ambiguity in the wiring identification, the path changes with the working conditions, or there is cumulative deviation in the measurement point position, the mapping path is prone to become disconnected from the actual conduction relationship. Although the data can enter the model, it may fall into a nearby but inconsistent position. There may also be intermittent connection and result drift between consecutive time periods, which will affect the reliability of anomaly comparison, trend judgment, and test conclusion verification. Summary of the Invention

[0005] To address the technical problems existing in the prior art, embodiments of the present invention provide a method and system for dynamic mapping of substation test data based on digital twins. The technical solution is as follows: A method for dynamic mapping of substation test data based on digital twins includes the following steps: S1: Obtain the wiring diagrams and connection relationships of circuit breakers, disconnect switches, busbars, transformers, and cables in the substation, as well as the test injection terminal and measurement terminal numbers and connection records. Form connection endpoint identifiers and verify the consistency of start and end point identifiers. Filter equipment information to correct the identifier correspondence and generate topology coordinate consistency error information. S2: Based on the topological coordinate consistency error information, construct a connection network, determine the virtual positions corresponding to the test injection terminal and the measurement terminal, combine the path connection information and the on / off indicators of the switch and disconnect switch to search for a fully conductive effective path, match the test voltage and test current records, and generate digital twin path mapping content; S3: Based on the digital twin path mapping content, extract the three-dimensional coordinates of the circuit breaker contacts, busbar connection points and transformer connection points corresponding to each virtual position of the effective path, compare them point by point with the three-dimensional coordinates of the measurement terminals calibrated on site, determine the spatial offset position exceeding the set threshold, and generate the three-dimensional spatial offset distribution result. S4: Based on the three-dimensional spatial offset distribution results, retrieve the direct connection virtual position and three-dimensional coordinates of the virtual position corresponding to each measurement terminal in the connection network, compare the current mapping point spacing with the ideal point spacing and correct it along the ideal connection direction to generate the adjacency constraint correction mapping result. S5: Based on the adjacency constraint correction mapping result, call the time series of the operating mode status of the circuit breaker, disconnector and bus, check the connection status of adjacent time slices and replace the virtual position, rematch the test voltage record and test current record, and generate the dynamic mapping result of the substation test data.

[0006] As a further aspect of the present invention, the step of obtaining S1 is as follows: S11: Obtain the wiring diagrams and connection relationships of circuit breakers, disconnectors, busbars, transformers and cables in the substation, test injection terminal numbers, measurement terminal numbers and connection records in the digital twin environment. Based on the wiring diagram connection relationships, test injection terminal numbers and measurement terminal numbers, the numbers are spliced ​​together to form a wiring diagram connection endpoint sequence. The connection endpoint sequence is numbered and aligned, and the corresponding position sequence number of the number combination at both ends of each connection is recorded to obtain the endpoint sequence number set. S12: Based on the set of endpoint numbers, call the connection endpoint sequence of the wiring diagram and the connection record in the digital twin environment, check the combination of the start point identifier, end point identifier and the corresponding number of each position number in the connection record, determine the consistent state and the difference state, for the connection with the difference state, retrieve the equipment name, phase and voltage level and merge the same items, record the difference items corresponding to the position number of each connection, and obtain the identification deviation item; S13: Based on the aforementioned identification deviation item, call the difference status connection position sequence number, device name, phase, voltage level and connection record in the digital twin environment, filter connection records with the same device name, phase, and voltage level, swap the position of the connection endpoint identifier, connection start identifier, and connection end identifier in the wiring diagram, record the difference in position sequence number before and after the swap and the corresponding status of the identifier, and generate topology coordinate consistency error information.

[0007] As a further aspect of the present invention, the step of obtaining S2 is as follows: S21: Based on the corrected connection endpoints in the topology coordinate consistency error information, construct a dynamic mapping connection network for power equipment test data, determine the virtual positions corresponding to the test injection terminals and the measurement terminals, extract the path connection information corresponding to the connection network, circuit breaker virtual positions, and bus virtual positions, record the starting virtual positions, ending virtual positions, and path continuation order of each connection segment, and obtain the path continuation table. S22: Based on the path connection table, collect the on / off identifiers of each switch and disconnector switch under the current substation operating status, call the virtual starting position and virtual ending position of each connection segment, and check the subsequent connection segment by segment along the path connection information starting from the virtual position corresponding to the test injection terminal. Retain the path that passes through the conducting switch and has a continuous connection relationship, and filter out the path that passes through the disconnector switch to obtain the conducting path sequence. S23: Based on the conduction path sequence, select paths where the starting point and the virtual position corresponding to the test injection terminal are consistent, the ending point and the virtual position corresponding to the measurement terminal are consistent, and the entire path is conductive. Assign sequential numbers to the virtual positions within the effective path according to their order. Sequentially associate the virtual positions corresponding to each sequential number, the test item number, the test voltage record and the test current record within the corresponding time range to generate digital twin path mapping content.

[0008] As a further aspect of the present invention, the step of recording the virtual starting position, virtual ending position, and path continuation order of each connecting line segment to obtain the path continuation table is specifically as follows: The segments of the connection in the content corresponding to the endpoints of the corrected connection are sorted according to the connection relationship. The starting virtual position, ending virtual position and the connection sequence between adjacent connections are extracted for each segment of the connection. The segments of the connection that are connected to the circuit breaker virtual position and the busbar virtual position are grouped into the same connection record to form the path connection table. The specific steps of checking subsequent connections segment by segment along the path information, starting from the virtual position corresponding to the test injection terminal, are as follows: Based on the correspondence between the virtual endpoint of the previous connection segment and the virtual starting point of the next connection segment in the path continuation table, each connection segment is continuously checked, and the on / off indicators of each switch and disconnector switch involved in the continuous check process are verified accordingly.

[0009] As a further aspect of the present invention, the screening starting point is consistent with the virtual position corresponding to the test injection terminal, the ending point is consistent with the virtual position corresponding to the measurement terminal, and the entire conductive path is specifically as follows: Extract paths from the conduction path sequence where the starting virtual position matches the virtual position corresponding to the test injection terminal, the ending virtual position matches the virtual position corresponding to the measurement terminal, and the on / off indicators of each switch and disconnector corresponding to each segment of the connection are all conducting. Assign sequential numbers according to the order of each virtual position within the effective path, and sequentially associate each sequential number with the corresponding virtual position, test item number, and test voltage and test current records within the corresponding time range.

[0010] As a further aspect of the present invention, the step of obtaining S3 is as follows: S31: Based on each virtual location within the effective path in the digital twin path mapping content, extract the three-dimensional coordinates of the circuit breaker contacts, busbar connection points, and transformer connection points corresponding to each virtual location in the digital twin three-dimensional environment, call the on-site calibration three-dimensional coordinates of the measurement terminals to perform position alignment, record the order of the coordinate groups corresponding to each virtual location and the coordinate groups of the measurement terminals, and obtain a coordinate correspondence list. S32: According to the coordinate correspondence list, call the three-dimensional coordinates corresponding to each virtual position and the three-dimensional coordinates calibrated on-site at the measurement terminal, check the coordinate difference in the three directions point by point, record the coordinate difference of each virtual position in the horizontal, vertical and longitudinal directions, determine the correspondence between the coordinate difference in each direction and the spatial offset threshold, extract the virtual positions that exceed the spatial offset threshold, and obtain the list of offset positions. S33: Based on the list of offset positions, call the order of virtual positions within the effective path, the coordinate differences of each spatial offset position in three directions, and the coordinate number of the corresponding measurement terminal. Rearrange the coordinate differences of each spatial offset position according to the order within the effective path, associate each spatial offset position with the path order number, and generate a three-dimensional spatial offset distribution result.

[0011] As a further aspect of the present invention, the step of obtaining S4 is as follows: S41: Based on the spatial offset position in the three-dimensional spatial offset distribution result, retrieve the virtual position directly connected to the virtual position corresponding to each measurement terminal in the dynamic mapping connection network of the test data of the power equipment, extract the three-dimensional coordinates corresponding to the virtual position directly connected, and arrange the coordinates of the current mapping point of the measurement terminal and the virtual position directly connected in the order of the effective path to obtain the point adjacency list. S42: Based on the point adjacency list, call the current mapped point coordinates of adjacent measurement terminals, the three-dimensional coordinates of the virtual position of the direct connection and the effective path order, check the spacing relationship of the current mapped points of adjacent measurement terminals on the same effective path, and compare it with the three-dimensional coordinates of the virtual position of the direct connection to form an ideal point spacing relationship to obtain the spacing deviation list. S43: Based on the spacing deviation list, call the current mapping point coordinates, ideal connection direction, effective path sequence and three-dimensional coordinates of the virtual position of the direct connection of the adjacent measurement terminals, correct the current mapping point along the ideal connection direction, and continuously check the connection relationship of the mapping points of the adjacent measurement terminals after correction, determine the sequence of virtual positions of the effective path after correction and the corresponding mapping points, and generate the adjacency constraint correction mapping result.

[0012] As a further aspect of the present invention, the step of obtaining S5 is as follows: S51: Based on the effective path virtual position sequence and corresponding mapping point in the adjacency constraint correction mapping result, call the time records of circuit breaker opening and closing status, disconnecting switch position status and bus operation mode status recorded in the digital twin environment in time sequence, arrange the on and off status and operation mode mark of each virtual position according to the time slice correspondence, check the changes in the arrangement of virtual positions of the same effective path in continuous time slices, and obtain the time sequence status list. S52: According to the time sequence status list, call the virtual position arrangement corresponding to the continuous time slice, the connection relationship of each segment in the connection network, the circuit breaker opening and closing status, the disconnecting switch position status and the bus operation mode status, check the connection status of the corresponding virtual position between adjacent time slices segment by segment, determine the location of the connection status change, select the virtual position in the corresponding time slice that is in the conducting state and maintains connection with the path of the previous time slice, and obtain the replacement candidate list. S53: Based on the candidate replacement list, call the connection state change position, candidate virtual position, original virtual position, corresponding time slice and effective path arrangement order, replace the original virtual position with the candidate virtual position, retrieve the test voltage record and test current record according to the arrangement of the replaced virtual position and the corresponding time slice, check the correspondence between virtual position, mapping point, voltage record and current record, and generate dynamic mapping result of power equipment test data.

[0013] A dynamic mapping system for substation test data based on digital twins, the system comprising: The connection verification module obtains the wiring diagrams and connection relationships of circuit breakers, disconnect switches, busbars, transformers, and cables in the substation, as well as the test injection terminal and measurement terminal numbers and connection records. It generates connection endpoint identifiers and verifies the consistency of the start and end point identifiers, filters equipment information to correct the identifier correspondence, and generates topology coordinate consistency error information. The path mapping module constructs a connection network based on the topological coordinate consistency error information, determines the virtual positions corresponding to the test injection terminal and the measurement terminal, searches for a fully conductive effective path by combining the path connection information and the on / off indicators of the switches and disconnect switches, matches the test voltage and test current records, and generates digital twin path mapping content. The coordinate comparison module, based on the digital twin path mapping content, extracts the three-dimensional coordinates of the circuit breaker contacts, busbar connection points and transformer connection points corresponding to each virtual position of the effective path, compares them point by point with the three-dimensional coordinates calibrated on-site at the measurement terminals, determines the spatial offset position exceeding the set threshold, and generates a three-dimensional spatial offset distribution result. The adjacency correction module, based on the three-dimensional spatial offset distribution results, retrieves the direct connection virtual positions and three-dimensional coordinates of the virtual positions corresponding to each measurement terminal in the connection network, compares the current mapping point spacing with the ideal point spacing and corrects it along the ideal connection direction, and generates adjacency constraint correction mapping results. The timing update module, based on the adjacency constraint correction mapping result, calls the time series of the operating mode status of circuit breakers, disconnectors and busbars, checks the connectivity status of adjacent time slices and replaces the virtual positions, rematches the test voltage records and test current records, and generates dynamic mapping results of substation test data.

[0014] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: In this invention, by cross-checking and correcting the correspondence between endpoint identifiers and virtual connection identifiers, the range of misaligned virtual and real connections can be compressed. By combining the connection status to identify the real path between the injection end and the measurement end segment by segment, the spread of test data along incorrect branches can be avoided. By superimposing three-dimensional offset recognition, adjacent spacing constraints and continuous time slice connectivity correction, the spatial landing point deviation and path jump risk can be converged simultaneously, making the connection position of the test record in the dynamic scene more stable, the evolution path clearer, and the source of anomalies easier to trace. Attached Figure Description

[0015] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a flowchart illustrating the acquisition process of S1 in this invention; Figure 3 This is a flowchart illustrating the acquisition process of S2 in this invention; Figure 4 This is a flowchart illustrating the acquisition process of S3 in this invention; Figure 5 This is a flowchart illustrating the acquisition process of S4 in this invention; Figure 6 This is a flowchart of the acquisition process for S5 of the present invention. Detailed Implementation

[0016] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0017] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0018] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the difference between them, their intended meanings are consistent. Similarly, the terms "of," "correlation (ponding)," and "correlation (ponding)" may sometimes be used interchangeably. It should be noted that, without emphasizing the difference between them, their intended meanings are consistent.

[0019] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0020] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0021] Please see Figure 1 This invention provides a technical solution: a method for dynamic mapping of substation test data based on digital twins, comprising the following steps: S1: Obtain the wiring diagrams and connection relationships of circuit breakers, disconnectors, busbars, transformers, and cables in the substation, as well as the test injection terminal numbers, measurement terminal numbers, and connection records in the digital twin environment. Based on the wiring diagram connection relationships, test injection terminal numbers, and measurement terminal numbers, form wiring diagram connection endpoint identifiers. Using the wiring diagram connection endpoint identifiers as a benchmark, check the start and end identifiers in the connection records to determine whether the wiring diagram connection endpoint identifiers are consistent with the digital twin connection start and end identifiers. For connections with inconsistent identifiers, based on the equipment name, phase, and voltage level in the wiring diagram, filter connection records with consistent equipment names, phases, and voltage levels, correct the correspondence between the wiring diagram connection endpoint identifiers and the connection start and end identifiers, and generate topology coordinate consistency error information. S2: Based on the corrected connection endpoints in the topology coordinate consistency error information, construct a connection network for dynamic mapping of substation test data, determine the virtual positions corresponding to test injection terminals and measurement terminals, extract the path connection information corresponding to the connection network, circuit breaker virtual positions, and bus virtual positions, collect the on / off indicators of each switch and disconnector switch under the current substation operating status, determine the starting virtual position and ending virtual position corresponding to each connection segment from the connection network, take the virtual position corresponding to the test injection terminal as the starting point, search for subsequent connections segment by segment according to the path connection information and the connection relationship in the connection network, determine whether the switch corresponding to the virtual position is in the conducting state, terminate the current direction search when encountering an open switch, and continue the search when encountering a conducting switch. After completing the path search, filter the valid paths where the starting point is consistent with the virtual position corresponding to the test injection terminal, the ending point is consistent with the virtual position corresponding to the measurement terminal, and the entire path is conducting. Assign sequential numbers according to the order of each virtual position in the valid path, determine the virtual position and test item number corresponding to each sequential number in turn, match the test voltage record and test current record according to the corresponding time range, and generate digital twin path mapping content. S3: Based on each virtual location within the effective path in the digital twin path mapping content, extract the three-dimensional coordinates of the circuit breaker contacts, busbar connection points, and transformer connection points corresponding to each virtual location in the digital twin 3D environment. Compare the three-dimensional coordinates corresponding to each virtual location with the field calibration three-dimensional coordinates of the corresponding measurement terminals point by point to determine the offset of each virtual location in three directions. For virtual locations whose offset exceeds the preset spatial offset threshold, determine them as spatial offset locations. Arrange the offsets of each spatial offset location according to the order in the effective path to generate a three-dimensional spatial offset distribution result. S4: Based on the spatial offset position in the three-dimensional spatial offset distribution result, retrieve the virtual position directly connected to the virtual position corresponding to each measurement terminal in the connection network of the dynamic mapping of the test data of the power equipment, extract the corresponding three-dimensional coordinates, and check the spacing relationship between the current mapping points of adjacent measurement terminals on the same effective path in the order of the effective path. Compare it with the ideal point spacing relationship formed by the three-dimensional coordinates corresponding to the directly connected virtual positions. When the two types of spacing relationships are inconsistent, correct the current mapping point along the ideal connection direction, and continuously check the connection relationship between the mapping points of adjacent measurement terminals after correction. Determine the corrected effective path virtual position sequence and corresponding mapping point, and generate the adjacency constraint correction mapping result. S5: Based on the effective path virtual position sequence and corresponding mapping points in the adjacency constraint correction mapping results, call the time series of circuit breaker opening and closing status, disconnector switch position status and bus operation mode status recorded in the digital twin environment in chronological order, analyze the changes of the virtual position sequence of the same effective path in continuous time slices, check the connectivity status of the corresponding virtual positions between adjacent time slices in the connection network of dynamic mapping of substation test data segment by segment, when the connectivity status changes, determine the candidate virtual position that is in the conducting state in the corresponding time slice and is connected to the path of the previous time slice, and replace the original virtual position with the candidate virtual position. Rematch the test voltage record and test current record according to the replaced virtual position sequence and the corresponding time slice to generate the dynamic mapping result of substation test data.

[0022] Please see Figure 2 The steps to obtain S1 are as follows: S11: Obtain the wiring diagrams and connection relationships of circuit breakers, disconnectors, busbars, transformers and cables in the substation, test injection terminal numbers, measurement terminal numbers and connection records in the digital twin environment. Based on the wiring diagram connection relationships, test injection terminal numbers and measurement terminal numbers, the numbers are spliced ​​together to form a wiring diagram connection endpoint sequence. The connection endpoint sequence is numbered and aligned, and the corresponding position sequence number of the number combination at both ends of each connection is recorded to obtain the endpoint sequence number set. This process involves acquiring wiring diagrams and connection relationships for circuit breakers, disconnectors, busbars, transformers, and cables within the substation, as well as test injection terminal numbers, measurement terminal numbers, and connection records within the digital twin environment. During this data acquisition, OCR text recognition technology is used to extract wiring diagram connections from the substation engineering drawings. An RFID scanner is used to read the on-site test injection terminal number "T_IN_01" and the measurement terminal number. Connection records within the digital twin environment are also extracted from the substation's BIM building information model server. Finally, a character concatenation command is invoked to combine the acquired circuit breaker wiring diagram connection identifiers with the test injection terminal numbers. The input terminal number is concatenated using a string concatenation operation to extract the measurement terminal number "M_OUT_05" and associate it with the cable physical attribute label on the current transformer side. The above strings are then concatenated bit-by-bit according to the equipment type identifier, physical location code, and terminal sequence to form a one-dimensional character vector. The number string is converted into a decimal numerical sequence based on ASCII code. A one-hot encoding rule is used to perform a high-dimensional sparse matrix mapping on the equipment type identifier to form a discrete feature vector. A bit-by-bit comparison is performed on the concatenated feature vector, and a discretized index mapping is established on each feature dimension to extract the corresponding terminal from the wiring diagram. The connection from circuit breaker Q1 to bus B1 is numbered at both ends. The ASCII values ​​of the starting end number "Q1_L1_01" and the ending end number "B1_L1_05" are stored in a one-dimensional array buffer in memory. This is compared with the corresponding connection record fields in the digital twin environment database. The cosine similarity between the input feature vector and the feature vector of the connection record in the database is calculated. The iteration step size of the position number is set to 1. Starting from the first data entry, each line is traversed. When a record with a cosine similarity of 1 and complete overlap of the starting and ending fields is found, a match is found. If no completely overlapping record is found, it indicates the existence of an unrecorded orphan in the network. Upon establishing a node, the system will automatically trigger a manual review alarm. If the retrieval is successful, the memory address offset of the data entry will be extracted and assigned as the position number. For example, when traversing to the 12th connection record corresponding to the A phase outlet of the circuit breaker, its position number will be assigned as 12. Using loop control instructions, all connection end numbers involving disconnect switches and cables will be mapped to their corresponding position numbers in a bidirectional key-value pair manner. A dictionary table based on hash addressing will be constructed. The endpoint number sequence in the storage medium will be called, and each position number will be mapped to the corresponding physical link feature vector space. Array concatenation and feature dimension expansion operations will be performed to obtain the endpoint number set.

[0023] S12: Based on the endpoint sequence set, call the wiring diagram connection endpoint sequence and the connection record in the digital twin environment, check the start point identifier, end point identifier and the corresponding number combination of each position sequence number in the connection record, determine the consistent state and the difference state, for the difference state connection, retrieve the equipment name, phase and voltage level and merge the same items, record the difference items corresponding to each connection position sequence number, and obtain the identification deviation item; Based on the endpoint sequence set, the connection endpoint sequence from the wiring diagram and the connection records within the digital twin environment are retrieved. The standard number combination "CB_01_BUS_02" is extracted from the connection record with position sequence number 15. The starting point identifier field "CB_01" and the ending point identifier field "BUS_03" with index value 15 are retrieved from the environment database. These fields are converted into one-dimensional character arrays, and an element-by-element equality judgment operation is performed. The code values ​​at the same index positions of the two arrays are compared. If the difference in code values ​​at all index positions is 0, the state is considered consistent. If an element with a non-zero difference is detected in the ending point identifier array, the state is considered different. For connections determined to be different, the corresponding database table's device name, phase, and voltage level fields are retrieved, and a category label merging operation is performed. Records with the same label are stored in an exception queue. Text features are extracted by comparing the character overlap of the difference items, and a dimensionless threshold for label deviation judgment is set. The threshold value is 0.85. This threshold parameter is extracted from a pre-constructed supervised learning sample dataset. Its value is calculated by taking the character edit distance from the feature vectors of 1000 manually annotated anomaly records. and the corresponding maximum string length The ratio is used to extract the probability density distribution interval that appears most frequently in the sample space. The arithmetic mean of the input samples is assigned to the threshold parameter to calculate the bias coefficient of the current input sample. Assuming the edit distance between a measurement terminal identifier "M_01" and the comparison identifier "Meas_01" is 3, and the corresponding maximum string length is 7, the result is obtained through division and subtraction operations. , will calculate accurately and Perform a numerical comparison operation. If 0.57 is less than 0.85, it is determined that the attribute has a calibration deviation. If the result is greater than or equal to 0.85, it means that there is only a slight spelling difference or inconsistent format. The system performs automatic regular expression replacement and alignment without recording it as a serious deviation. Call the write protection command to enumerate the categories of the difference items corresponding to the serial number 15 at position 15 and enter them into the log file to obtain the identification deviation item.

[0024] S13: Based on the identification deviation item, call the connection position sequence number, equipment name, phase, voltage level and the connection record in the digital twin environment of the difference status connection, filter the connection record with the same equipment name, phase, voltage level, and perform position swapping on the connection end point mark, connection start mark, and end point mark of the wiring diagram, record the difference of the position sequence number before and after the swapping and the corresponding status of the mark, and generate topology coordinate consistency error information. Based on the identification deviation item, the system retrieves the connection location sequence number, equipment name, phase, voltage level, and connection records within the digital twin environment, along with the aforementioned equipment name, phase, and voltage level attribute data. This data is directly exported from the corresponding equipment ledger database through the station's SCADA data acquisition and monitoring control system. During the dataset preprocessing for training the machine learning classification model, all connection record entries in the environment database are traversed. The equipment name "Isolating Switch QS1" from the record to be corrected is used as the search keyword to construct a Boolean query condition. The phase "C phase" and voltage level "220kV" are extracted to construct a conditional logical expression, which is then executed via a bitwise AND operation. When the target record is retrieved... When the number of intersection elements between the feature attribute set and the item to be corrected equals the total feature dimension, it is considered a complete overlap. If the number of intersection elements does not equal the total feature dimension, it indicates that the device attribute is missing a mapping in three-dimensional space, and it is recorded as a black-box node and highlighted in the twin model. In the case of complete overlap, the memory pointer addresses corresponding to the start identifier "QS1_C_01" and the end identifier "BUS_C_10" of the target record are extracted. The pointer addresses of the original identifier array are swapped with the pointer addresses of the newly filtered identifiers. The column index of the two-dimensional data table is reorganized, and the position index integer variables before and after the column swap are extracted. The absolute difference of the position index values ​​before and after the pointer swap is calculated. Assuming the original position number variable is 45 and the new position number variable is 102, subtraction is performed and the absolute value is checked to obtain the difference variable 57. The relationship matrix table corresponding to the state is updated, the corrected topological relationship feature vector and the initial topological relationship feature vector are extracted, and a bitwise XOR operation is performed on the two vectors. The XOR operation result vector is traversed to extract the index positions of elements with non-zero values. The device coordinate component vectors associated with the positions, the topological difference scalar, and the logical link change are extracted. A matrix splicing operation is performed to synthesize a multi-dimensional data feature tensor and generate topological coordinate consistency error information.

[0025] Please see Figure 3 The steps to obtain S2 are as follows: S21: Based on the corrected connection endpoints in the topology coordinate consistency error information, construct a dynamic mapping connection network for substation test data, determine the virtual positions corresponding to the test injection terminals and the measurement terminals, extract the path connection information corresponding to the connection network, circuit breaker virtual positions, and bus virtual positions, record the starting virtual positions, ending virtual positions, and path continuation order of each connection segment, and obtain the path continuation table. Record the virtual starting and ending positions of each connecting line segment, as well as the path continuation order, to obtain the path continuation table as follows: The segments of the connection in the content corresponding to the endpoints of the corrected connection are sorted according to the connection relationship. The starting virtual position, ending virtual position and the connection sequence between adjacent connections are extracted for each segment of the connection. The segments of the connection that are connected to the circuit breaker virtual position and the busbar virtual position are grouped into the same connection record to form a path connection table. Based on the corrected connection endpoints in the topology coordinate consistency error information, the mapping data records of the corrected circuit breaker terminal Q1_1 and bus-side terminal B1_5 are extracted. The three-dimensional coordinate data is collected and extracted using UAV oblique photography and LiDAR point cloud fusion modeling technology. The coordinate components of the two ends of the connection are mapped to the preset three-dimensional graph grid index space. The virtual three-dimensional coordinate vector of the test injection terminal T_IN is determined to be (120.5, 45.0, 10.2) and the virtual three-dimensional coordinate vector of the measurement terminal M_OUT is determined to be (135.2, 50.8, 10.2) in the topology node dictionary table. The multi-dimensional node feature set composed of circuit breakers, buses and transformers is extracted from the substation topology graph data structure. For each corrected connection vector, the topology adjacency relationship sorting operation based on the adjacency matrix is ​​performed. The end coordinate components of the first connection "T_IN to circuit breaker Q1" and the starting coordinate components of the second connection "Q1 to bus B1" are substituted into the Euclidean distance calculation formula with spatial geometric distance physical meaning. When the distance modulus is in meters When the distance modulus is 0, it indicates that the two nodes have a physical connection. If the distance modulus is not 0, it means that there is a geometric discontinuity in the physical space of the current topology. At this time, the system marks the connection in red and freezes subsequent connection searches. For the nodes that are determined to be connected, the circuit breaker internal contact position identifier field and the bus branch connection point position identifier field are combined into the same logical linked list through a union operation. If the starting point of a connection vector is (125.0, 46.5, 10.2) and the ending point is (128.0, 47.0, 10.2), a new data block space is allocated in memory, and so on. The topology weight floating-point number, device type code character, and connection port number of the connection segment are pushed into the sequence. A breadth-first traversal is performed on all virtual location nodes of power equipment involved in the calculation. All connection vectors with the same bus segment number "BUS_SEC_01" are grouped into the same two-dimensional connection matrix. The start pointer and end pointer of each connection object are traversed by recursive function calls. Intermediate node variables are temporarily stored through a stack structure. The arithmetic mean of the three-axis coordinate components corresponding to each conductive medium segment is recorded as the scalar feature of the path node to obtain the path continuation table.

[0026] S22: Based on the path connection table, collect the on / off identifiers of each switch and disconnector switch in the current substation operating status, call the virtual starting position and virtual ending position of each connection segment, and check the subsequent connection segment by segment along the path connection information starting from the virtual position corresponding to the test injection terminal. Retain the path that passes through the conducting switch and has a continuous connection relationship, and filter out the path that passes through the disconnector switch to obtain the conducting path sequence. Starting from the virtual position corresponding to the test injection terminal, the subsequent connections are checked segment by segment along the path connection information. Specifically: Based on the correspondence between the virtual endpoint of the previous connection segment and the virtual starting point of the next connection segment in the path connection table, each connection segment is continuously checked, and the on / off markings of each switch and disconnector switch involved in the continuous check process are verified accordingly. Based on the path connection table, the on / off status register values ​​of each switch and disconnector switch are read from the underlying monitoring database. These register values ​​are acquired in real-time at high frequency through hard-wiring of the substation remote terminal unit (RTU) and monitoring and control devices, or through IEC61850MMS protocol message parsing technology. Status word binarization is performed, mapping the hexadecimal register values ​​to a Boolean vector of 0s and 1s, where 1 represents closed conduction and 0 represents disconnection. The endpoint variable "Node_n_End" and the starting variable "Node_{n+1}_Start" of the nth connection segment recorded in the path connection table are extracted. Memory address consistency verification is performed, and the test injection terminal pair is then... The corresponding virtual location coordinate vector (120.5, 45.0, 10.2) is set as the root node of the depth-first search tree. The child nodes are traversed level by level according to the continuation index order specified in the path continuation table. At each level node, the on / off identification array of the associated switch object is extracted. If the status bit of circuit breaker Q1 is found to be equal to 1 and the status bit of disconnector QS1 is equal to 1, the connection object is stored in the dynamic array corresponding to the candidate path set. If the status bit of disconnector QS2 is found to be equal to 0, the search branch is determined to be in an interrupted state and the branch elimination instruction is executed to pop it from the stack. The Boolean values ​​of each node in the path are extracted, and the path connectivity score scalar with series loop logic and engineering significance is calculated. , in the formula For the first on the path The dimensionless on / off Boolean value of a switch. Given the total number of switches corresponding to the current search depth, compare the calculated result with the value 1. When strictly equal to 1, the node index of the complete path is pushed onto the valid result stack. When the value is strictly equal to 0, garbage collection is performed to clean up the memory of the path. This means that the path contains breakpoints and the current topology cannot form a valid conductive loop. The system records the breakpoint device and terminates the analog mapping of the path. It iterates through all possible topological adjacency combinations in the graph structure, extracts the path node vector sequence with complete connection relationship and a connectivity score product of 1 between the injection terminal T_IN and the measurement terminal M_OUT, flattens and expands it for storage, and obtains the conductive path sequence.

[0027] S23: Based on the conduction path sequence, select paths where the starting point and the virtual position corresponding to the test injection terminal are consistent, the ending point and the virtual position corresponding to the measurement terminal are consistent, and the entire path is conductive. Assign sequential numbers to the virtual positions within the effective path according to their order, and sequentially associate the virtual positions corresponding to each sequential number, the test item number, the test voltage record and the test current record within the corresponding time range to generate digital twin path mapping content. The screening start point and the virtual position corresponding to the test injection terminal are consistent, the end point and the virtual position corresponding to the measurement terminal are consistent, and the entire conductive path is as follows: Extract paths from the conduction path sequence where the starting virtual position matches the virtual position corresponding to the test injection terminal, the ending virtual position matches the virtual position corresponding to the measurement terminal, and the on / off indicators of each switch and disconnector corresponding to each segment of the connection are all conducting. Assign sequential numbers according to the order of each virtual position within the effective path, and sequentially associate each sequential number with the corresponding virtual position, test item number, and test voltage and test current records within the corresponding time range. Based on the conduction path sequence, extract the first and last node elements from each path vector list. Perform vector subtraction on the virtual position coordinate vector of the first element and the standard reference coordinate vector (120.5, 45.0, 10.2) of the test injection terminal. Extract the Euclidean modulus of the difference vector in meters and subtract it from the minimum deviation threshold constant. Perform numerical comparison, set The value is 0.001. This comparison represents the verification of the spatial coincidence of physical endpoints within the engineering tolerance range. If the modulus value is less than... Then determine if the starting attributes of the paths are consistent, and if the magnitude is greater than or equal to... This indicates a misalignment between the test path start point and the physical access point. In this case, a path rejection operation is performed, and a misalignment deviation log is output. The difference attribute is determined by comparing the coordinate vector of the path end with the standard reference coordinate vector (135.2, 50.8, 10.2) of the measurement terminal using the same subtraction and comparison operations. Paths with a modulus greater than or equal to the specified value are rejected. For non-matching path records, for valid path vectors stored in memory, according to the unidirectional topological flow from the injection end to the measurement end, the local variable counter is initialized to 1, and an integer sequence number field is assigned to each virtual location node as the node traversal increments. The project code string "TEST_SEQ_001" is retrieved from the external test database, and the dataset that matches the sampling round of the path in the full-pass state is extracted from the corresponding data sampling sequence table. Read the test voltage floating-point sequence transmitted in the data bus and test current floating-point sequence The voltage and current data are acquired by high-frequency sampling at 10kHz using a microcomputer protection tester and a data acquisition card DAQ installed in the secondary circuit. The voltage amplitude variable, current amplitude variable, and three-axis coordinate variable are combined with the feature vectors of the virtual position nodes with corresponding sequential numbers to generate a multi-dimensional hybrid tensor set containing three-dimensional coordinate features, sampling round index, and electrical parameter features. The binary byte stream corresponding to the tensor set is extracted and stored to generate digital twin path mapping content.

[0028] Please see Figure 4 The steps to obtain S3 are as follows: S31: Based on each virtual location within the effective path in the digital twin path mapping content, extract the three-dimensional coordinates of the circuit breaker contacts, busbar connection points, and transformer connection points corresponding to each virtual location in the digital twin three-dimensional environment, call the on-site calibration three-dimensional coordinates of the measurement terminals to perform position alignment, record the order of the coordinate groups corresponding to each virtual location and the coordinate groups of the measurement terminals, and obtain the coordinate correspondence list. Based on the virtual locations within the effective path in the digital twin path mapping content, the corresponding circuit breaker contact center point, busbar connection point bolt axis, and transformer secondary side connection point are extracted in the Cartesian coordinate system within the digital twin 3D environment. , , The three-axis floating-point numerical values ​​were used to extract the three-dimensional components of the virtual position of the circuit breaker contacts into coordinate tuples (125.40, 46.20, 10.50), and the three-dimensional components of the virtual position of the bus connection point into coordinate tuples (128.60, 47.10, 10.50). The dataset of three-dimensional floating-point coordinates for on-site calibration of the measurement terminals, generated based on laser point cloud modeling, was read from the external file system. These on-site calibration coordinates were acquired through on-site measurements using a 3D panoramic laser scanner installed at the station in conjunction with an RTK high-precision total station. Multiple table operations were then performed within the relational database. The primary key index join operation retrieves the measured coordinate vector (125.45, 46.15, 10.42) of the field measurement terminal with the number M1. It then performs a one-to-one pairing combination with the circuit breaker contact coordinates in the virtual location tuple. For each pairing, a hash function mapping operation is performed between the virtual and real coordinates. If the corresponding measured coordinates of the measurement point cannot be found in the relational database, it means that the measurement point is physically obstructed or has not been scanned on-site. In this case, the system automatically performs spatial neighborhood interpolation estimation to fill the gap and adds an estimation label, recording the virtual location coordinate vector as... ; The corresponding on-site calibration coordinate vector is denoted as... For example, the virtual three-dimensional coordinates of the circuit breaker contact with sequence index 1 are combined with the field-calibrated three-dimensional coordinates to form a one-dimensional vector and stored in the first row of the result matrix. The virtual three-dimensional coordinates (128.60, 47.10, 10.50) of the bus connection point with sequence index 2 are combined with the field measured values ​​(128.75, 47.05, 10.45) and stored in the second row of the result matrix. The above coordinate alignment and splicing operation is performed by traversing all connected nodes in the effective path through a loop iteration statement. The feature number of each node, the three components of the virtual coordinates, and the three components of the measured coordinates are pushed into a linear structured array according to the order of the topological flow direction to obtain the coordinate correspondence list.

[0029] S32: Based on the coordinate correspondence list, call the corresponding three-dimensional coordinates of each virtual position and the on-site calibration three-dimensional coordinates of the measurement terminal, check the coordinate difference in the three directions point by point, record the coordinate difference of each virtual position in the horizontal, vertical and longitudinal directions, determine the correspondence between the coordinate difference in each direction and the spatial offset threshold, extract the virtual positions that exceed the spatial offset threshold, and obtain the list of offset positions. Based on the coordinate correspondence list, retrieve the corresponding 3D coordinate tuples for each virtual location. Field calibration of three-dimensional coordinate units with measurement terminals Perform vector subtraction operations on each coordinate component dimension to extract the difference variable. , , The subtraction operation results extracted from the circuit breaker contact nodes yielded a horizontal coordinate difference variable of -0.05, a vertical coordinate difference variable of 0.05, and a vertical coordinate difference variable of 0.08. The numerical deviation results of each virtual position in the three directions were stored in a three-dimensional deviation matrix cache. The spatial offset absolute threshold parameter with units of meters, preset in the configuration file, was then read. The threshold parameter is set to 0.10, and this parameter setting refers to the tolerance range for support point installation in the construction specifications. The mean of the Gaussian distribution is obtained by taking the absolute value of the calculated coordinate differences in each direction. , , The three absolute floating-point numbers mentioned above are respectively compared with... The parameters are compared using logical logic. If the absolute value of any variable in either direction is greater than 0.10, a logical OR operation is performed and returns true, indicating that the node is in a deviation state exceeding the threshold. For example, if the calculated absolute value of the vertical coordinate difference of a connection point is 0.12, and the comparison shows 0.12 is greater than 0.10, the unique identifier of the connection point and its corresponding coordinate difference vector sequence are appended and stored in the abnormal deviation memory queue. A hard-boundary-based quantization classification operation is then performed on the deviation vectors in the memory queue, and a conditional statement is set to determine if the offset falls within the specified range. Assign a category label of 0 to determine if it falls within the range. Assign a category label of 1 if the value is greater than 0.10, then assign a category label of 2 if the difference variable falls within the range. When a category label value of 0 is assigned, it means that the on-site installation accuracy fully meets the high fidelity requirements of digital twins, and there is no need to trigger displacement correction operations. The significant offset node features with category label value of 2 are extracted through logical filtering to obtain the list of offset positions.

[0030] S33: Based on the list of offset positions, call the order of virtual positions within the effective path, the coordinate differences of each spatial offset position in three directions, and the coordinate number of the corresponding measurement terminal. Rearrange the coordinate differences of each spatial offset position according to the order within the effective path, associate each spatial offset position with the path order number, and generate a three-dimensional spatial offset distribution result. Based on the list of offset positions, the system retrieves the integer index of the virtual position features within the valid path, the floating-point components of the three-axis coordinate deviations corresponding to each spatial offset position, and the corresponding measurement terminal coordinate string identification code. Following the flow sequence of the valid path from start to finish, the system executes a position reordering command on all entries in the list, extracting the offset position coordinate difference vector corresponding to path number 5. Move its position in the memory linked list to the offset position with the corresponding path index attribute of 8, using the coordinate difference vector. In the pre-storage address space, the corresponding path sequence number variable is filled into the attribute prefix field of each recombined offset position record. A database primary-foreign key association matching operation is performed to extract the three-axis deviation components of each offset point. These components are then substituted into the modulus formula to calculate the scalar of the comprehensive Euclidean distance deviation relative to the ideal path node. The calculation is verified to yield: The comprehensive offset floating-point number is approximately 0.116, thus correcting the previously inherited numerical error. The bubble sort algorithm is applied to the set of scalars to achieve descending order. If there are cases where the comprehensive offset is equal, it means that multiple nodes are deviating from the geometric center to the same degree. In this case, a second ascending order is performed according to the path order index to ensure topological continuity. The sequence number parameter after the arrangement is extracted, and the coordinate difference corresponding to the path order variable 5 and the comprehensive offset floating-point number 0.116 obtained by the operation are assigned to the first row of the data structure of the result feature matrix. For all the offset node elements corresponding to the list, the spatial coordinate feature dimension and the path sequence identifier dimension are merged and assigned. The recombined high-dimensional matrix is ​​exported to persistent storage to generate the three-dimensional spatial offset distribution result.

[0031] Please see Figure 5 The steps to obtain S4 are as follows: S41: Based on the spatial offset position in the three-dimensional spatial offset distribution results, retrieve the virtual positions directly connected to the virtual positions corresponding to each measurement terminal in the dynamic mapping connection network of the test data of the power equipment, extract the three-dimensional coordinates corresponding to the virtual positions directly connected, and arrange the coordinates of the current mapping point of the measurement terminal and the virtual positions directly connected in the order of the effective path to obtain the point adjacency list. Based on the spatial offset position in the 3D spatial offset distribution results, extract the node feature tuple with the identifier field "Circuit Breaker Contact Q1_C" from the offset position feature list. Perform a reverse graph traversal search in the adjacency matrix of the connected network loaded into memory, and read the column index integer of the non-zero element in the corresponding row vector of the adjacency matrix. If no node with a non-zero weight is found in the reverse graph traversal, it means that there is an unclosed floating terminal in the topology network. At this time, a connection loss alarm will be automatically sent to the handheld terminal of the maintenance personnel. Extract the virtual position coordinate vector (125.40,46) corresponding to the measurement terminal M1. The target coordinates of the direct connection virtual position with non-zero weights (.20, 10.50) are used. Assuming the direct model coordinate tuple extracted to the bus connection point B1 is (128.60, 47.10, 10.50), the three-dimensional coordinate floating-point components corresponding to the extracted direct connection virtual position are used as the reference vector parameters. The current real-time dynamic mapping point coordinate tuple (125.45, 46.15, 10.42) of the measurement terminal is read and paired with the reference vector of the upstream node obtained by reverse retrieval. The effective path sequence field is extracted, and the process is performed from the index variable. arrive The ascending serialization operation merges the current mapped coordinates (125.45, 46.15, 10.42) of the path node with index 5 with the ideal connection point coordinates (128.60, 47.10, 10.50) in the digital twin environment feature library into a multidimensional array. The loop body statement performs the above reverse feature retrieval and merging operation for each measurement terminal in the offset position list. In the computer heap space, a nested dictionary structure object is instantiated with the path index number as the hash key and the merged array of mapped point position and ideal coordinates as the hash value. According to the predefined directed edge pointing attributes in the topology graph, the key-value pairs in the dictionary are traversed and the three-dimensional space combination vector of all adjacent nodes are linearly serialized and assigned values ​​to obtain the point adjacency list.

[0032] S42: Based on the point adjacency list, call the current mapped point coordinates of adjacent measurement terminals, the three-dimensional coordinates of the virtual position of the direct connection, and the effective path order, check the spacing relationship of the current mapped points of adjacent measurement terminals on the same effective path, and compare it with the three-dimensional coordinates of the virtual position of the direct connection to form an ideal point spacing relationship, and obtain the spacing deviation list; Based on the point adjacency list, extract the floating-point components (125.45, 46.15, 10.42) of node A and (128.75, 47.05, 10.45) of node B from the adjacent measurement terminal mapping point array, and substitute them into the three-dimensional Euclidean distance formula to obtain the current mapping point spacing scalar in meters: The floating-point result variable obtained by the verification operation is 3.421. The ideal coordinate combinations of the corresponding direct connection virtual positions (125.40, 46.20, 10.50) and (128.60, 47.10, 10.50) in the data structure are read and substituted into the same formula to obtain the scalar value of the ideal point spacing. Calling the subtraction and absolute value operators will and Combined with the calculation of the spacing deviation scalar The calculation yields an assignment parameter of 0.097, which has engineering significance in determining the length deformation of rigid and flexible conductors. The absolute threshold for adjacency constraint determination is then retrieved from the parameter configuration file. The threshold constant is set to 0.05, and this value is called from the parameter set of allowable range of standard deviation for elongation compensation of connecting hardware. The median characteristic value is used to call the greater than sign comparison instruction to... Variable values ​​and The constant is used to compare the numerical values. If the value is greater than 0.05, it is determined that the adjacent points in this group violate the spacing consistency constraint. The deviation level determiner is initialized, and the deviation value is matched to the left-closed and right-closed intervals. The first type of deviation flag is assigned. If the deviation value falls within this interval, it means that the current spacing difference is caused by reasonable deformation due to thermal expansion and contraction of the equipment. In this case, no translation correction is performed and the current mapping state is maintained. Matching is performed in the left-open-right-closed interval. The system assigns a second type of deviation flag to the data, and assigns a third type of deviation flag to values ​​exceeding 0.05. It then iterates through the list data to extract the index numbers of all measurement terminal objects with the third type of deviation flag. The system performs subtraction on each three-dimensional coordinate component of the offset object to obtain the positive and negative polarity signs. It uses the combination of sign bits to record the tensor of the stretching or compression state between nodes in space. Finally, it performs batch packaging and serialization operations on the array elements of the collected violation points to obtain the spacing deviation list.

[0033] S43: Based on the spacing deviation list, call the coordinates of the current mapped point of the adjacent measurement terminal, the ideal connection direction, the effective path sequence and the three-dimensional coordinates of the virtual position of the direct connection, correct the current mapped point along the ideal connection direction, and continuously check the connection relationship of the mapped points of the adjacent measurement terminals after correction, determine the sequence of virtual positions of the effective path after correction and the corresponding mapped points, and generate the adjacency constraint correction mapping result. Based on the spacing deviation list, extract the combined coordinate vector of the current mapped points of adjacent measurement terminals, extract and substitute it into the three-dimensional coordinate components of the virtual position of the direct connection to calculate the direction vector of the ideal connection direction: Extract the valid path order attribute field to locate the starting node index variable where the offset attribute occurs. Calculate the direction vector The Euclidean modulus is calculated and a division operation is performed to obtain the unit direction vector. The coordinate translation transformation operator is invoked to perform linear displacement correction on the coordinates of the currently offset mapping point along the unit direction vector; Application of translation correction calculation formula The new coordinate variables are calculated. This corrected formula has clear physical engineering significance, namely, it restores the ideal topological coordinates of a rigid connection by deriving the position of the subsequent point based on the previous anchor point and the theoretical direction. It is assumed that the unit vector is extracted and calculated. Substituting the pre-anchor point, the ideal point spacing scalar parameter 3.324, and the extracted direction components, matrix multiplication and addition are performed. The resulting new coordinate array is reassigned to replace the original array, and the corrected 3D feature tuple (125.40, 46.20, 10.50) is saved. For the corrected new coordinate point data, the coordinate arrays of its mapping points with the adjacent measurement terminals are extracted, and a second round of distance difference threshold verification is performed. If the difference variable still exceeds the preset tolerance, the recursive call instruction is extracted and the cascade displacement correction operation is performed along the directed edge topology downstream nodes. The loop condition is set until the verification difference is less than the tolerance and the loop is exited. If the difference in the cyclic verification cannot be less than the tolerance and exceeds the maximum recursion depth, it means that there is non-rigid mechanical torsion damage on site that cannot be solved by a simple translation algorithm. At this time, the system executes local physical disconnection marking and outputs maintenance work orders that need to be manually checked on site. It extracts and assembles the coordinate attribute floating-point number sequence of the corrected node in each dimension, performs hash association mapping processing on the sequence and the unique identification code of the equipment entity object, extracts the corrected path list to check the connectivity weight scalar, filters and removes abnormal breakpoint elements with a weight of 0, extracts the final valid path virtual position feature object sequence combination that meets the feature consistency verification, and generates the adjacency constraint correction mapping result.

[0034] Please see Figure 6 The steps to obtain S5 are as follows: S51: Based on the effective path virtual position sequence and corresponding mapping point in the adjacency constraint correction mapping result, call the time records of circuit breaker opening and closing status, disconnecting switch position status and bus operation mode status recorded in the digital twin environment in time sequence, arrange the on and off status and operation mode mark of each virtual position according to the time slice correspondence, check the changes in the arrangement of virtual positions of the same effective path in continuous time slices, and obtain the time sequence status list. Based on the effective path virtual location sequence and corresponding mapping points in the adjacency constraint correction mapping results, the spatial position parameter combinations such as the coordinate vectors of the 5th path node (125.40, 46.20, 10.50) and the 8th path node (128.60, 47.10, 10.50) after correction are extracted. The operating mode characteristic byte sequences of specific circuit breakers Q1, specific disconnectors QS1, and corresponding busbars are retrieved from the equipment operation history status word record database. This status word data is acquired using the high-speed timestamp message extraction technology of the substation microcomputer anti-misoperation interlocking system. The sampling frequency of the extraction time axis is set with a step size parameter, and the continuously collected status word strings are divided into arrays according to a fixed-length sequence period to extract the sequence period. The corresponding circuit breaker open / close status bit element value is 1, the disconnector switch position status bit element value is 1, and the bus single operating status code identifier integer is used. The extracted binary value array is then sorted and aligned with the feature vectors of each virtual location object within the valid path using index correspondence. A one-dimensional array sliding window is initialized, and for continuous sequence periodic intervals... arrive Extracting consecutive elements from a continuous loop involves performing a translation comparison function call to extract... Virtual position on / off state array variable under period ,extract The state array variable under the period: Perform a bitwise XOR operation between two arrays and sum the non-zero elements in the result array to calculate the total difference. Call the conditional statement for comparison Check if the variable value is greater than 0. A variable value of 0 indicates that the path structure is stable and there are no changes in electrical connectivity within the sequence period. The state transition recording step is skipped, and the original configuration is maintained. If the value is greater than 0, it indicates that a state transition has occurred in the virtual position logical arrangement attribute within the current feature sequence period. This assumes that during the traversal to the sequence period... Timely verification and calculation The value is equal to 1. The extracted features indicate that the state flag bit of the isolating switch element of the 3rd node in the path is reversed from 1 to 0. The sequence period label value and its corresponding state transition bit device object encoding string are recorded. The transition records calculated within all step intervals are extracted and written into the multi-dimensional timing topology matrix table to obtain the timing state list.

[0035] S52: Based on the time sequence status list, call the virtual position arrangement corresponding to the continuous time slice, the connection relationship of each segment in the connection network, the circuit breaker opening and closing status, the disconnecting switch position status and the bus operation mode status, check the connection status of the corresponding virtual position between adjacent time slices segment by segment, determine the location of the connection status change, select the virtual position in the corresponding time slice that is in the conducting state and maintains connection with the path of the previous time slice, and obtain the replacement candidate list. Based on the time-series state list, extract the periods that belong to continuous sequences from the list. and The Boolean vector corresponding to the virtual position of the record row is arranged. The topology weight two-dimensional connection matrix corresponding to the bus B1 pointed to by circuit breaker Q1 is retrieved from the memory heap of the graph connection network. This topology weight data is directly extracted and called through the site's GIS geographic information system and the topology error prevention logic library. The sequence period in the memory is extracted. The system retrieves the real-time parameter set where the circuit breaker flag variable is 1 and the disconnector QS2 flag variable is 0. It then reads the node connection subarray from segment 3 to segment 4 in the path object list, calls the adjacency matrix search function to perform a depth-first graph traversal to find spare edges, queries the non-zero connected element objects in the connection matrix, extracts the coordinate array of candidate compensation nodes that have Boolean mutual exclusion logic with the original detected disconnected node and a topological distance of 1, and reads the set connectivity evaluation benchmark coefficient. The coefficient is set to 1.0. Its calculation relies on the binary logic network evaluation model. The model extracts fully conductive samples along the path, calculates the closure coefficient of a single element as 1.0 and the discontinuity coefficient as 0, traverses the candidate path, and performs a product operation on the coefficients of all associated switch nodes. , judge if If the result of the calculation is exactly equal to 1.0, the overall connectivity feature of the path is determined to be valid, within the given sequence period. Within the retrieval space, the spare component disconnect switch QS3 flag bit is set to 1, and its predecessor node and sequence period are extracted and calculated. If the index of the second node in the record path is consistent, calculate the Euclidean distance offset variable between the three-dimensional coordinates of the candidate compensation node and the original path node. The value is assigned to 0.15, and it is determined that it falls within the preset numerical adjacency tolerance range. If no compensation node with a flag of 1 is found within the specified adjacent tolerance range, it means that the current path has been hard interrupted and there is no backup ring network branch. The circuit breaker mechanism is directly triggered to interrupt the twin current mapping of the path. The judgment condition combination is executed. If the tolerance range is met and the connectivity product feature result is equal to 1.0, a replaceable label is attached to the QS3 node feature object of the candidate virtual location. All node index features, equipment classification identifiers and corresponding sequence period parameters containing the label are packaged into a matrix object to obtain the replacement candidate list.

[0036] S53: Based on the candidate replacement list, call the connection state change position, candidate virtual position, original virtual position, corresponding time slice and effective path arrangement order, replace the original virtual position with the candidate virtual position, retrieve the test voltage record and test current record according to the arrangement of the replaced virtual position and the corresponding time slice, check the correspondence between virtual position, mapping point, voltage record and current record, and generate dynamic mapping result of power equipment test data; Based on the candidate substitution list, extract the integer variable corresponding to the path position index where a connectivity state bit change occurs. The corresponding candidate virtual location coordinate vector group (126.10, 46.50, 10.50) and the memory block of the original fault point virtual location object where the logical break originally occurred at this index are extracted and corresponding to the sequence period. For the specified complete slice range of the sampling set, execute the data overwrite update instruction operation for the corresponding memory object, extract the target effective path feature sequence, and directly overwrite and assign the original virtual position and its associated measurement point mapping coordinate parameters using the position parameters of the candidate object. Then, update the new... An index is built based on the feature groups of the three-axis coordinate sequence. A join query is performed in the database based on the period number of the currently established joint sequence and the object's primary key to retrieve the retrieved data. Floating-point variables of test voltage sampling parameters recorded within the period interval And retrieve the floating-point variables of the test current sampling parameters. The high-order tensor assembly module function interface is called to combine the three-dimensional features of spatial points, the one-dimensional features of time-series periods, and the two-dimensional features of voltage and current electrical parameters to construct a four-dimensional data tensor object. The updated node codes and electrical parameters are extracted and matched with the updated and covered values ​​for consistency verification. The values ​​are then substituted into the feature calculation formula to generate a dimensionless scalar index for data distribution consistency evaluation. This indicator is obtained by calculating the reciprocal of the variance between different sampled floating-point values ​​within a unified period and the standard deviation of the model's preset ideal voltage. The benchmark scalar constant 0.95 is read and compared to obtain the calculated result. Furthermore, if the logical expression with a value greater than 0.95 is true, it indicates that the newly established multidimensional mapping relationship features are a perfect match. A baseline scalar value less than or equal to 0.95 indicates that the mapped electrical parameter data is distorted or subject to severe electromagnetic interference from the field. The current mapping relationship is determined to be unreliable, and the data frame is marked as dirty data. At the same time, the Kalman filter algorithm is started to perform secondary data smoothing. All evolution node sequence objects that have undergone sequence transfer mutations are processed through the outer traversal loop statement to perform the attribute alignment operation. The spatial point coordinate tensor, equipment status identifier vector and analog quantity record matrix after the coverage verification are extracted and persisted to generate the dynamic mapping result of the substation test data.

[0037] A dynamic mapping system for substation test data based on digital twins, the system comprising: The connection verification module obtains the wiring diagrams and connection relationships of circuit breakers, disconnect switches, busbars, transformers, and cables in the substation, as well as the test injection terminal and measurement terminal numbers and connection records. It generates connection endpoint identifiers and verifies the consistency of the start and end point identifiers, filters equipment information to correct the identifier correspondence, and generates topology coordinate consistency error information. The path mapping module constructs a connection network based on the topological coordinate consistency error information, determines the corresponding virtual positions of the test injection terminal and the measurement terminal, searches for a fully conductive effective path by combining the path connection information and the on / off status of the switch and disconnector switch, matches the test voltage and test current records, and generates digital twin path mapping content. The coordinate comparison module, based on the digital twin path mapping content, extracts the three-dimensional coordinates of circuit breaker contacts, busbar connection points and transformer connection points corresponding to each virtual position of the effective path, compares them point by point with the three-dimensional coordinates calibrated on-site at the measurement terminals, determines the spatial offset position exceeding the set threshold, and generates a three-dimensional spatial offset distribution result. The adjacency correction module, based on the three-dimensional spatial offset distribution results, retrieves the direct connection virtual position and three-dimensional coordinates of the virtual position corresponding to each measurement terminal in the connection network, compares the current mapping point spacing with the ideal point spacing and corrects it along the ideal connection direction, generating the adjacency constraint correction mapping result. The timing update module, based on the adjacency constraint correction mapping results, calls the time series of circuit breaker, disconnector and bus operation mode status, checks the connectivity status of adjacent time slices and replaces the virtual position, rematches the test voltage record and test current record, and generates dynamic mapping results of substation test data.

[0038] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for dynamic mapping of substation test data based on digital twins, characterized in that, Includes the following steps: S1: Obtain the wiring diagrams and connection relationships of circuit breakers, disconnect switches, busbars, transformers, and cables in the substation, as well as the test injection terminal and measurement terminal numbers and connection records. Form connection endpoint identifiers and verify the consistency of start and end point identifiers. Filter equipment information to correct the identifier correspondence and generate topology coordinate consistency error information. S2: Based on the topological coordinate consistency error information, construct a connection network, determine the virtual positions corresponding to the test injection terminal and the measurement terminal, combine the path connection information and the on / off indicators of the switch and disconnect switch to search for a fully conductive effective path, match the test voltage and test current records, and generate digital twin path mapping content; S3: Based on the digital twin path mapping content, extract the three-dimensional coordinates of the circuit breaker contacts, busbar connection points and transformer connection points corresponding to each virtual position of the effective path, compare them point by point with the three-dimensional coordinates of the measurement terminals calibrated on site, determine the spatial offset position exceeding the set threshold, and generate the three-dimensional spatial offset distribution result. S4: Based on the three-dimensional spatial offset distribution results, retrieve the direct connection virtual position and three-dimensional coordinates of the virtual position corresponding to each measurement terminal in the connection network, compare the current mapping point spacing with the ideal point spacing and correct it along the ideal connection direction to generate the adjacency constraint correction mapping result. S5: Based on the adjacency constraint correction mapping result, call the time series of the operating mode status of the circuit breaker, disconnector and bus, check the connection status of adjacent time slices and replace the virtual position, rematch the test voltage record and test current record, and generate the dynamic mapping result of the substation test data.

2. The method for dynamic mapping of substation test data based on digital twins according to claim 1, characterized in that: The steps for obtaining S1 are as follows: S11: Obtain the wiring diagrams and connection relationships of circuit breakers, disconnectors, busbars, transformers and cables in the substation, test injection terminal numbers, measurement terminal numbers and connection records in the digital twin environment. Based on the wiring diagram connection relationships, test injection terminal numbers and measurement terminal numbers, the numbers are spliced ​​together to form a wiring diagram connection endpoint sequence. The connection endpoint sequence is numbered and aligned, and the corresponding position sequence number of the number combination at both ends of each connection is recorded to obtain the endpoint sequence number set. S12: Based on the set of endpoint numbers, call the connection endpoint sequence of the wiring diagram and the connection record in the digital twin environment, check the combination of the start point identifier, end point identifier and the corresponding number of each position number in the connection record, determine the consistent state and the difference state, for the connection with the difference state, retrieve the equipment name, phase and voltage level and merge the same items, record the difference items corresponding to the position number of each connection, and obtain the identification deviation item; S13: Based on the aforementioned identification deviation item, call the difference status connection position sequence number, device name, phase, voltage level and connection record in the digital twin environment, filter connection records with the same device name, phase, and voltage level, swap the position of the connection endpoint identifier, connection start identifier, and connection end identifier in the wiring diagram, record the difference in position sequence number before and after the swap and the corresponding status of the identifier, and generate topology coordinate consistency error information.

3. The method for dynamic mapping of substation test data based on digital twins according to claim 1, characterized in that: The steps for obtaining S2 are as follows: S21: Based on the corrected connection endpoints in the topology coordinate consistency error information, construct a dynamic mapping connection network for power equipment test data, determine the virtual positions corresponding to the test injection terminals and the measurement terminals, extract the path connection information corresponding to the connection network, circuit breaker virtual positions, and bus virtual positions, record the starting virtual positions, ending virtual positions, and path continuation order of each connection segment, and obtain the path continuation table. S22: Based on the path connection table, collect the on / off identifiers of each switch and disconnector switch under the current substation operating status, call the virtual starting position and virtual ending position of each connection segment, and check the subsequent connection segment by segment along the path connection information starting from the virtual position corresponding to the test injection terminal. Retain the path that passes through the conducting switch and has a continuous connection relationship, and filter out the path that passes through the disconnector switch to obtain the conducting path sequence. S23: Based on the conduction path sequence, select paths where the starting point and the virtual position corresponding to the test injection terminal are consistent, the ending point and the virtual position corresponding to the measurement terminal are consistent, and the entire path is conductive. Assign sequential numbers to the virtual positions within the effective path according to their order. Sequentially associate the virtual positions corresponding to each sequential number, the test item number, the test voltage record and the test current record within the corresponding time range to generate digital twin path mapping content.

4. The method for dynamic mapping of substation test data based on digital twins according to claim 4, characterized in that: The path continuation table is obtained by recording the virtual starting position, virtual ending position, and path continuation order of each connecting line segment as follows: The segments of the connection in the content corresponding to the endpoints of the corrected connection are sorted according to the connection relationship. The starting virtual position, ending virtual position and the connection sequence between adjacent connections are extracted for each segment of the connection. The segments of the connection that are connected to the circuit breaker virtual position and the busbar virtual position are grouped into the same connection record to form the path connection table. The specific steps of checking subsequent connections segment by segment along the path information, starting from the virtual position corresponding to the test injection terminal, are as follows: Based on the correspondence between the virtual endpoint of the previous connection segment and the virtual starting point of the next connection segment in the path continuation table, each connection segment is continuously checked, and the on / off indicators of each switch and disconnector switch involved in the continuous check process are verified accordingly.

5. The method for dynamic mapping of substation test data based on digital twins according to claim 4, characterized in that: The screening start point corresponds to the virtual position of the test injection terminal, the end point corresponds to the virtual position of the measurement terminal, and the entire conductive path is as follows: Extract paths from the conduction path sequence where the starting virtual position matches the virtual position corresponding to the test injection terminal, the ending virtual position matches the virtual position corresponding to the measurement terminal, and the on / off indicators of each switch and disconnector corresponding to each segment of the connection are all conducting. Assign sequential numbers according to the order of each virtual position within the effective path, and sequentially associate each sequential number with the corresponding virtual position, test item number, and test voltage and test current records within the corresponding time range.

6. The method for dynamic mapping of substation test data based on digital twins according to claim 1, characterized in that: The steps for obtaining S3 are as follows: S31: Based on each virtual location within the effective path in the digital twin path mapping content, extract the three-dimensional coordinates of the circuit breaker contacts, busbar connection points, and transformer connection points corresponding to each virtual location in the digital twin three-dimensional environment, call the on-site calibration three-dimensional coordinates of the measurement terminals to perform position alignment, record the order of the coordinate groups corresponding to each virtual location and the coordinate groups of the measurement terminals, and obtain a coordinate correspondence list. S32: According to the coordinate correspondence list, call the three-dimensional coordinates corresponding to each virtual position and the three-dimensional coordinates calibrated on-site at the measurement terminal, check the coordinate difference in the three directions point by point, record the coordinate difference of each virtual position in the horizontal, vertical and longitudinal directions, determine the correspondence between the coordinate difference in each direction and the spatial offset threshold, extract the virtual positions that exceed the spatial offset threshold, and obtain the list of offset positions. S33: Based on the list of offset positions, call the order of virtual positions within the effective path, the coordinate differences of each spatial offset position in three directions, and the coordinate number of the corresponding measurement terminal. Rearrange the coordinate differences of each spatial offset position according to the order within the effective path, associate each spatial offset position with the path order number, and generate a three-dimensional spatial offset distribution result.

7. The method for dynamic mapping of substation test data based on digital twins according to claim 1, characterized in that: The steps for obtaining S4 are as follows: S41: Based on the spatial offset position in the three-dimensional spatial offset distribution result, retrieve the virtual position directly connected to the virtual position corresponding to each measurement terminal in the dynamic mapping connection network of the test data of the power equipment, extract the three-dimensional coordinates corresponding to the virtual position directly connected, and arrange the coordinates of the current mapping point of the measurement terminal and the virtual position directly connected in the order of the effective path to obtain the point adjacency list. S42: Based on the point adjacency list, call the current mapped point coordinates of adjacent measurement terminals, the three-dimensional coordinates of the virtual position of the direct connection and the effective path order, check the spacing relationship of the current mapped points of adjacent measurement terminals on the same effective path, and compare it with the three-dimensional coordinates of the virtual position of the direct connection to form an ideal point spacing relationship to obtain the spacing deviation list. S43: Based on the spacing deviation list, call the current mapping point coordinates, ideal connection direction, effective path sequence and three-dimensional coordinates of the virtual position of the direct connection of the adjacent measurement terminals, correct the current mapping point along the ideal connection direction, and continuously check the connection relationship of the mapping points of the adjacent measurement terminals after correction, determine the sequence of virtual positions of the effective path after correction and the corresponding mapping points, and generate the adjacency constraint correction mapping result.

8. The method for dynamic mapping of substation test data based on digital twins according to claim 1, characterized in that: The steps for obtaining S5 are as follows: S51: Based on the effective path virtual position sequence and corresponding mapping point in the adjacency constraint correction mapping result, call the time records of circuit breaker opening and closing status, disconnecting switch position status and bus operation mode status recorded in the digital twin environment in time sequence, arrange the on and off status and operation mode mark of each virtual position according to the time slice correspondence, check the changes in the arrangement of virtual positions of the same effective path in continuous time slices, and obtain the time sequence status list. S52: According to the time sequence status list, call the virtual position arrangement corresponding to the continuous time slice, the connection relationship of each segment in the connection network, the circuit breaker opening and closing status, the disconnecting switch position status and the bus operation mode status, check the connection status of the corresponding virtual position between adjacent time slices segment by segment, determine the location of the connection status change, select the virtual position in the corresponding time slice that is in the conducting state and maintains connection with the path of the previous time slice, and obtain the replacement candidate list. S53: Based on the candidate replacement list, call the connection state change position, candidate virtual position, original virtual position, corresponding time slice and effective path arrangement order, replace the original virtual position with the candidate virtual position, retrieve the test voltage record and test current record according to the arrangement of the replaced virtual position and the corresponding time slice, check the correspondence between virtual position, mapping point, voltage record and current record, and generate dynamic mapping result of power equipment test data.

9. A dynamic mapping system for substation test data based on digital twins, characterized in that, The system is used in the dynamic mapping method for substation test data based on digital twins as described in any one of claims 1-8, the system comprising: The connection verification module obtains the wiring diagrams and connection relationships of circuit breakers, disconnect switches, busbars, transformers, and cables in the substation, as well as the test injection terminal and measurement terminal numbers and connection records. It generates connection endpoint identifiers and verifies the consistency of the start and end point identifiers, filters equipment information to correct the identifier correspondence, and generates topology coordinate consistency error information. The path mapping module constructs a connection network based on the topological coordinate consistency error information, determines the virtual positions corresponding to the test injection terminal and the measurement terminal, searches for a fully conductive effective path by combining the path connection information and the on / off indicators of the switches and disconnect switches, matches the test voltage and test current records, and generates digital twin path mapping content. The coordinate comparison module, based on the digital twin path mapping content, extracts the three-dimensional coordinates of the circuit breaker contacts, busbar connection points and transformer connection points corresponding to each virtual position of the effective path, compares them point by point with the three-dimensional coordinates calibrated on-site at the measurement terminals, determines the spatial offset position exceeding the set threshold, and generates a three-dimensional spatial offset distribution result. The adjacency correction module, based on the three-dimensional spatial offset distribution results, retrieves the direct connection virtual positions and three-dimensional coordinates of the virtual positions corresponding to each measurement terminal in the connection network, compares the current mapping point spacing with the ideal point spacing and corrects it along the ideal connection direction, and generates adjacency constraint correction mapping results. The timing update module, based on the adjacency constraint correction mapping result, calls the time series of the operating mode status of circuit breakers, disconnectors and busbars, checks the connectivity status of adjacent time slices and replaces the virtual positions, rematches the test voltage records and test current records, and generates dynamic mapping results of substation test data.

Images