A non-intrusive master station monitoring information acceptance method and system based on real-time data slice analysis

Abstract

This invention discloses a non-intrusive master station monitoring information acceptance method and system based on real-time data slice analysis, belonging to the field of power system monitoring technology. This invention securely obtains equipment and point number information sequences by parsing master station slice files; establishes a standardized point number mapping table through cluster analysis; uses this table to associate master and substation data, forming aligned numerical description pairs; then performs a two-dimensional intelligent comparison of numerical differences and text similarity to identify configuration misalignments and inconsistencies in hidden descriptions, generating enhanced verification records; finally, it constructs a tamper-proof full-link traceability log and closed-loop verification result report, and adaptively optimizes mapping rules and data paths based on report feedback. This achieves a fundamental transformation in monitoring information acceptance from manual verification to automatic diagnosis, and from single operations to continuous optimization, significantly improving the security, accuracy, and intelligence level of the acceptance process.

Description

Technical Field

[0001] This invention relates to the field of power system monitoring technology, specifically a non-intrusive master station monitoring information acceptance method and system based on real-time data slicing analysis. Background Technology

[0002] Power dispatch automation systems are core infrastructure for ensuring the safe and stable operation of the power grid. The master station, acting as the nerve center, is responsible for collecting and processing massive amounts of real-time monitoring information from substations across the country. The accuracy of this information directly determines the reliability of dispatch instructions and the ability to prevent power grid accidents. With the continuous expansion of the power grid and the rapid increase in the proportion of renewable energy integration, the acceptance of master station monitoring information has become increasingly frequent and complex, becoming an unavoidable and crucial part of daily operation and maintenance. Currently, the acceptance of master station monitoring information generally relies on manual methods, comparing substation measured values ​​with master station displayed values ​​through telephone coordination and point-to-point verification. This method is not only time-consuming and inefficient, but also prone to human error or omissions. More importantly, to obtain real-time data from the master station, many acceptance methods have to directly connect to or access the master station's production database. This operation poses significant security risks, as any improper access could interfere with running business systems and even lead to data tampering or system paralysis. Therefore, the industry urgently needs a means to securely and reliably obtain real-time master station data without accessing the production database.

[0003] However, to protect core business operations, master station systems typically do not expose standard interfaces for directly reading real-time data. The only usable data is often binary data slice files generated periodically within the system. These files are originally designed for internal snapshot backups or fault analysis, and their structure is highly complex and heavily reliant on the master station vendor's proprietary encoding rules. To extract real-time values ​​for a specific monitoring point, it is necessary to first crack the complex encoding logic of the data identifiers and correctly restore the seemingly random string of numbers to the actual device identifier and point number information. Because the encoding rules of master station systems from different vendors vary greatly, and different versions of the same vendor may also adjust their encoding methods, it is difficult to establish a stable and reliable parsing mapping relationship even if the slice files are obtained. This has become the biggest obstacle to non-intrusive data acquisition. Meanwhile, there are often discrepancies between the monitoring information descriptions from the main station and the signal descriptions reported by the substations. Simple numerical comparisons cannot uncover hidden configuration misalignments. To verify the consistency of descriptions, it is necessary to simultaneously obtain the theoretical values ​​from the substations, the measured values ​​from the main station, and the description texts from both sides, while ensuring that these information are strictly aligned in time. How to obtain real-time slice data, parse out accurate point number correspondences, achieve automatic dual comparison of numerical values ​​and descriptions, and form traceable closed-loop verification results without interfering with the main station's production environment has become a key issue that urgently needs to be addressed in the intelligent acceptance of main station monitoring information. Summary of the Invention

[0004] The purpose of this invention is to provide a non-intrusive main station monitoring information acceptance method and system based on real-time data slicing analysis, which significantly improves the security, accuracy and intelligence of the acceptance process.

[0005] The objective of this invention can be achieved through the following technical solutions:

[0006] This application provides a non-intrusive method for accepting master station monitoring information based on real-time data slicing analysis, including the following steps:

[0007] S1. Obtain the binary slice file generated by the main station as the initial input source, determine the pattern distribution of the encoding rules by parsing the composite encoding structure inside the slice file, and obtain the device identifier and point number information sequence;

[0008] S2. Based on the device identifier and point number information sequence, a clustering algorithm is used to group similar coding patterns in the sequence, and a standardized point number correspondence mapping table is obtained by comparing it with a preset manufacturer difference threshold.

[0009] S3. Based on the point number correspondence mapping table, associate the signal description text and theoretical value reported by the sub-station with the real-time value in the main station slice file to obtain the aligned numerical description pair dataset.

[0010] S4. For the numerical description pair dataset, a matching algorithm is used to compare the degree of difference between the real-time values ​​of the main station and the theoretical values ​​of the sub-station. If the difference exceeds the preset threshold, it is marked as a configuration misalignment, and a preliminary comparison result set is obtained.

[0011] S5. Based on the preliminary comparison result set, calculate the string similarity between the description text of the main station side and the sub-station side, determine whether the similarity is lower than the preset threshold to identify the inconsistency of hidden descriptions, and obtain the enhanced comparison verification record.

[0012] S6. By verifying the enhanced comparison records, generate a timestamp-aligned traceability log file, determine the integrity of all verification records in the log file, and obtain a closed-loop verification result report.

[0013] It also includes: S7, updating the dynamic adjustment parameters of the point number correspondence mapping table according to the closed-loop verification result report; if the adjustment parameters indicate a change in the encoding rules, the slice file is re-parsed to obtain an optimized real-time data acquisition path.

[0014] This application provides a non-intrusive master station monitoring information acceptance system based on real-time data slice analysis, used to implement a non-intrusive master station monitoring information acceptance method based on real-time data slice analysis, including:

[0015] The data monitoring and parsing module is used to monitor and parse the binary real-time data slice files generated by the main station. By recognizing the composite coding structure and the domain number zeroing algorithm, it converts the data ID into the device identifier, forms a preliminary mapping relationship with the monitoring information point table, and outputs the device identifier and point number information sequence after cluster analysis and multi-source verification.

[0016] The mapping relationship management module is used to extract coding features based on the sequence and perform unsupervised clustering and grouping, combine preset difference thresholds with sub-station configuration files for cross-validation and pattern merging, generate standard coding sequences and establish a global standardized mapping table, and form a standardized point number correspondence mapping table after verification with the main station authoritative point table.

[0017] The collaborative acceptance processing module is used to receive signal metadata reported by the substation through an independent closed-loop channel, map the substation number to the main station data ID using the standardized mapping table, extract the measured value and description in the corresponding time scale window of the slice file, complete the two-dimensional comparison and logical verification of the numerical value and description, integrate them into a structured numerical description pair dataset and feed it back to the substation in real time.

[0018] The difference analysis module is used to apply the corresponding difference calculation strategy to the dataset according to the signal type, compare it with the tolerance defined in the point table, mark the out-of-limit numerical pairs as configuration misalignment, perform pattern analysis and grouping on the misalignment sequence, combine multi-source configuration to preliminarily locate the root cause, and generate a preliminary comparison result set and pre-verification report.

[0019] The text verification module is used to extract the description text of the main station and the sub-station, calculate the consistency using a hybrid model that integrates semantics and literal similarity, determine hidden inconsistencies using dynamic thresholds, and perform contextual business verification using device logic, classification rules and terminology dictionary. After excluding non-substantive differences, it generates enhanced comparison verification records.

[0020] The traceability and reporting module is used to generate time-stamped structured log units based on verification records, and to construct tamper-proof full-link traceability log files based on hash algorithms. It integrates and analyzes logs from the main station, sub-stations and closed-loop channels to verify whether a strict closed loop is formed from the triggering of the signal to the feedback, and generates a closed-loop verification result report.

[0021] The adaptive optimization module is used to quantify the abnormal conclusions into parameter adjustment instructions for the point number mapping table based on the closed-loop verification result report, update the mapping table and generate a new configuration version after self-learning, analyze whether the encoding rules have undergone structural shifts and trigger incremental re-parsing, evaluate the performance of each data channel and then use a dynamic weight algorithm to select the optimal real-time data acquisition path.

[0022] The beneficial effects of this invention are as follows:

[0023] By using non-intrusive real-time data slicing parsing and intelligent coding rule learning, the systemic security risks caused by the need to directly access the main station production database in traditional acceptance methods are completely resolved. By utilizing the inherent data backup mechanism within the main station system, by listening to and parsing the binary slice files it generates, and by using a unique coding pattern recognition and adaptive mapping technology, real-time monitoring data can be obtained safely and reliably without interfering with the operation of the business system. This enables real-time data acquisition without touching the operating system, providing a safe and reliable technical foundation for the automation of power main station systems.

[0024] By constructing a closed-loop verification channel for master-slave station collaboration and a dual-dimensional intelligent comparison engine, this technology overcomes the industry challenge of finding hidden configuration logic errors and inconsistencies in descriptions during manual acceptance. It not only compares the correctness of numerical values ​​but also introduces semantic and text similarity analysis to intelligently verify signal descriptions. This technology can accurately identify deep-seated problems caused by forwarding mapping errors, coefficient configuration deviations, and inconsistent terminology, enabling the acceptance process to leap from superficial numerical verification to in-depth configuration and logic diagnosis, significantly improving the intrinsic quality and reliability of power grid signals before commissioning.

[0025] By establishing a full-chain digital traceability and adaptive optimization mechanism, the backward state of traditional acceptance processes, which were unauditable and difficult to accumulate experience, has been fundamentally changed. Standardized logs with tamper-proof verification chains are automatically generated, fully recording every link from signal triggering to result feedback, realizing electronic archiving and full lifecycle management of the acceptance process. At the same time, based on historical acceptance results, self-learning and optimization analysis rules are formed to form a sustainable improvement closed loop of operation-feedback-optimization, transforming one-time debugging tasks into knowledge assets that support the continuous evolution of the system, greatly improving the level of intelligent and lean management of operation and maintenance. Attached Figure Description

[0026] To better understand and implement this application, the technical solution is described in detail below with reference to the accompanying drawings.

[0027] Figure 1 A flowchart illustrating a non-intrusive master station monitoring information acceptance method based on real-time data slice analysis provided in Embodiment 1 of this application;

[0028] Figure 2 This is a schematic diagram of the structure of a non-intrusive master station monitoring information acceptance system based on real-time data slicing analysis, provided in Embodiment 2 of this application. Detailed Implementation

[0029] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, exemplary embodiments will be described in detail below, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0030] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used herein are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0031] The following detailed description of the specific implementation methods, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided in detail.

[0032] Example 1

[0033] Please see Figure 1 This embodiment provides a non-intrusive main station monitoring information acceptance method based on real-time data slice analysis, including the following steps:

[0034] S1. Obtain the binary slice file generated by the main station as the initial input source. Determine the pattern distribution of the encoding rules by parsing the composite encoding structure inside the slice file, and obtain the device identifier and point number information sequence.

[0035] Further, step S1 specifically includes:

[0036] In response to the periodic data backup command triggered by the main station application system, it listens for the binary real-time data slice files generated by the main station, performs incremental capture based on the file creation timestamp and signature, and obtains the target slice file to be parsed;

[0037] The binary stream of the target slice file is read, the composite encoding structure is identified, and the table number, domain number, area number and record number contained in the data ID are parsed out according to the predefined field partitioning rules. The data ID is converted into the corresponding device identifier by setting the domain number to zero, and the record number is associated with the pre-stored monitoring information point table to form a preliminary device identifier and point number information mapping pair.

[0038] The coding sequence features in the preliminary mapping pair are extracted. Based on the preset manufacturer coding rule library, a clustering algorithm is used to group coding sequences with similar features and compare them with a preset difference threshold to distinguish the coding patterns of different manufacturers. The coding sequences belonging to the same pattern are standardized to generate a standardized correspondence between device identifiers and point number information.

[0039] The standardized correspondence is then used for multi-source association and intelligent matching with the database point table configuration obtained from the master station system and the substation configuration file (SCD / RCD) on the substation side. The consistency of the point number, equipment identifier and signal description in the correspondence is verified in each source data. Invalid or conflicting mapping entries are identified and removed. Based on the verification results, the sequence of equipment identifier and point number information that has passed multi-source verification is output.

[0040] Furthermore, based on predefined field partitioning rules, the table number, field number, region number, and record number contained in the data ID are parsed out. This includes: loading the acquired binary data ID as a fixed-length bit string (e.g., 64 bits); then, according to the starting bit and length of each field defined in the rules, the corresponding bit segments are extracted sequentially through bitmasking and shifting operations; for example, bits 1 to 16 are extracted as the table number, bits 17 to 32 as the field number, bits 33 to 40 as the region number, and the remaining bits as the record number; then, the extracted binary values ​​of each field are converted according to the system's preset byte order (big-endian or little-endian) to obtain the corresponding integer values; finally, a structured parsing result is output, generating a quadruple containing the table number, field number, region number, and record number for each data ID, thus completing the precise deconstruction from the original encoding to the explicit semantic fields.

[0041] The algorithm that sets the domain number to zero converts the data ID into a corresponding device identifier. This includes: generating a mask based on a predefined field bit order, where only the bit corresponding to the domain number is 0 and the rest are 1; performing a bitwise AND operation between this mask and the original data ID to accurately remove the domain number information while fully preserving the table number, area number, and record number fields. After this operation, the original data ID is converted into a pure device identifier that uniquely corresponds to the physical device and no longer contains signal type details. This identifier can be directly used to stably associate with devices in the monitoring information point table, laying the foundation for subsequent point number mapping. The core function of this process is to reverse-engineer the dynamic classification field in its private coding structure and extract a persistent and stable device identity identifier without relying on the main station's public interface.

[0042] Extracting the encoding sequence features from the preliminary mapping pairs, and grouping encoding sequences with similar features using a clustering algorithm based on a preset manufacturer encoding rule library, includes: extracting multi-dimensional features of the encoding sequence composed of dot information based on the preliminary mapping pairs. These features include, but are not limited to, total encoding length, field separator position and type, numerical or alphabetic distribution patterns of characters within each segment, presence of fixed prefix / suffix or check bit patterns, and presence of specific positional mapping relationships or numerical offset patterns with associated device identifiers. The preset manufacturer encoding rule library serves as prior knowledge, storing common encoding style templates and typical feature combinations from different manufacturers. When using clustering algorithms, each dot sequence is vectorized according to the extracted features, and then similarity is calculated based on the distance metric of the feature vectors (such as edit distance, Euclidean distance, or custom distance based on key position matching). During the unsupervised learning process, the algorithm combines the typical patterns provided by the rule base as references or constraints for the initial cluster centers, iteratively groups all sequences, and finally aggregates dot sequences with highly similar features that match the coding style of a certain type of manufacturer in the rule base into the same cluster, thereby realizing the automatic discovery and classification of unknown coding patterns.

[0043] Specifically, by employing non-intrusive slice monitoring and intelligent encoding parsing, the system addresses the security risks associated with direct access to the main site's production database and the challenges of reliable data parsing due to complex and opaque encoding rules. Specifically, it monitors internal backup files, parses the data ID structure, and extracts device identifiers using a domain number zeroing algorithm. Combined with clustering analysis and multi-source configuration cross-validation, the system automatically and accurately converts raw binary slices into structured dot sequence sequences without interfering with the main site's operation. This achieves secure, reliable, and automated real-time data acquisition, laying a solid foundation for subsequent intelligent acceptance testing throughout the entire process.

[0044] S2. Based on the obtained equipment identifier and point number information sequence, a clustering algorithm is used to group similar coding patterns in the sequence, and it is determined whether the grouped patterns meet the preset manufacturer difference threshold to obtain a standardized point number correspondence mapping table.

[0045] Further, step S2 specifically includes:

[0046] Extract the encoding structure features of each point number in the device identifier and point number information sequence. The features include at least the encoding length, segmentation structure, character set at a specific position, and association rules with the device identifier. Based on the multi-dimensional features, an unsupervised clustering algorithm is used to automatically group all point number information sequences to form multiple initial pattern clusters.

[0047] For each initial pattern cluster, the edit distance or number of different characters between all pairs of dot sequence within the cluster is calculated. If the difference metric of all sequence pairs within a pattern cluster is less than the preset intra-manufacturer difference threshold, the cluster is determined to be a consistent encoding pattern. For multiple pattern clusters that are determined to be consistent, the dot sequence of the multiple pattern clusters is cross-mapped and verified with the SCD file and RCD file obtained from the substation side. If the verification is successful, these pattern clusters are merged to form a unified candidate standard pattern.

[0048] From each of the unified candidate standard patterns, a central sequence is selected or a standard coding sequence is generated as a benchmark through rule synthesis. A global point number standardization mapping table is established from each original point number sequence to its corresponding pattern standard coding sequence. The global point number standardization mapping table is then subjected to a secondary association verification with the point table configuration in the main station database. Based on the verification results, the mapping entries are dynamically adjusted, abnormal associations are eliminated, and missing mappings are supplemented, ultimately forming a standardized point number correspondence mapping table.

[0049] Furthermore, an unsupervised clustering algorithm is employed to automatically group all point number information sequences, forming multiple initial pattern clusters. This includes: quantizing each point number information sequence into a multi-dimensional feature vector based on its encoding structure features (such as total length, delimiter position, character type distribution of each field, and fixed offset relationship with associated device IDs); then selecting an unsupervised clustering algorithm suitable for processing sequence data (e.g., hierarchical clustering based on a custom edit distance metric or the DBSCAN algorithm) using these feature vectors as input; calculating the pairwise distance matrix between all point number sequences; and then automatically aggregating sequences with high feature similarity into the same set based on their distance, forming multiple initial pattern clusters. In this process, the algorithm does not rely on any pre-labeled manufacturer tags, but only performs discovery-based grouping based on the structural similarity of the data itself, thereby initially revealing potential different encoding styles or manufacturer patterns.

[0050] The point number sequences of multiple pattern clusters are cross-mapped and verified with SCD and RCD files obtained from the substation side. This includes: parsing the SCD file from the substation side to obtain the IED (Intelligent Electronic Device) logical nodes and data object models defined therein; simultaneously parsing the RCD file to clarify the mapping relationship between each forwarding point number configured by the remote control device and the specific signal path in the SCD; then, for each pattern cluster to be verified, all its point number sequences are searched in the point number-SCD signal path mapping library established in the RCD file. The core of the verification is to determine whether most of the point numbers within the same pattern cluster are mapped to the same manufacturer or the same type of IED device model. If the point numbers within a cluster show a high degree of device model consistency in the engineering configuration, the verification passes, and the cluster is considered to represent a real encoding pattern that matches the configuration; otherwise, it indicates that the clustering results may be biased.

[0051] A global dot number standardization mapping table is established, connecting each original dot number sequence to its corresponding pattern's standard encoding sequence. This includes: for each verified and merged candidate standard pattern, a standard encoding sequence for that pattern is generated; the generation method is either selecting the central dot number sequence of the pattern cluster (i.e., the sequence with the smallest average distance to all other sequences in the cluster), or synthesizing a standard template by analyzing all sequences in the cluster and summarizing their common feature rules (such as fixed prefixes, field lengths, and wildcard positions); the system creates a global mapping table, where each record uses the original dot number sequence as the key and its corresponding pattern's standard encoding sequence as the value; through this table, any dot number from the main site's slice file with different original encoding forms is standardized into a unified, standard identifier, thereby eliminating encoding inconsistencies caused by differences in manufacturers or versions, and establishing a unique index benchmark for subsequent accurate comparisons.

[0052] Among them, the master station database point table configuration refers to the authoritative configuration table containing the point numbers, descriptions and equipment association information of all monitoring signals, which is obtained from the production configuration library of the master station scheduling automation system through a read-only interface or export tool. It is used as the final benchmark to verify the accuracy and consistency of the point number mapping relationship intelligently derived in the preceding steps.

[0053] Specifically, an unsupervised clustering algorithm is used to automatically group point number sequences based on multiple features. This, combined with sub-site configuration files (SCD / RCD) for cross-validation and pattern merging, solves the industry problem of significant differences in encoding rules across different vendors and versions of the main site system, making it difficult to establish stable mapping relationships. This step automatically identifies and unifies various proprietary encoding patterns, generating a highly consistent standardized point number mapping table. It represents a fundamental shift from manual experience-based matching to intelligent algorithmic induction, establishing a reliable indexing foundation for subsequent accurate and efficient automatic comparison.

[0054] S3. Using the obtained point number correspondence mapping table, collect the reported signal description text and theoretical values ​​from the substation side, determine the correspondence between the real-time values ​​in the main station slice file and the substation text, and obtain the aligned numerical description pair dataset.

[0055] Furthermore, step S3 specifically includes:

[0056] S31. In response to the acceptance start command sent by the automatic point-to-point device at the substation through the independent private protocol closed-loop channel, the automatic acceptance system at the master station synchronously starts the real-time data monitoring task of the power station; it receives and parses the signal metadata reported by the substation with the accompanying signal triggering action in real time through the closed-loop channel. The metadata includes the signal point number, signal type, preset theoretical value, precise timestamp, and standard signal description text on the substation side, which is used to construct a signal triggering event queue.

[0057] S32. Based on the precise timestamps of each event in the signal-triggered event queue and combined with the preset transmission delay tolerance, determine the time window for matching and searching in the real-time data slice stream of the main station; using the standardized point number correspondence mapping table, map the point number information in the event to a unique main station data ID; within the corresponding slice file time window, locate and extract the measured value recorded by the data ID and the signal description stored in the main station system to form a preliminary association pair between the substation event and the main station record;

[0058] S33. The measured values ​​and descriptions of the master station in the preliminary association pair are compared with the theoretical values ​​and descriptions reported by the sub-stations in two dimensions: In terms of numerical values, compliance is judged based on the signal type using a difference threshold or relative error; in terms of description, a consistency score is calculated based on a text similarity algorithm; at the same time, the device context and forwarding path of the association pair are logically verified by calling the master station database point table and the sub-station SCD / RCD configuration pair; based on the above results, a comprehensive criterion is generated that includes numerical consistency status, description consistency status, and configuration logic status.

[0059] S34. Based on the comprehensive criteria, for the verified association pairs, integrate the theoretical value of the substation, the description of the substation, the measured value of the main station, the description of the main station, the time stamp, and various status markers into a standard record; aggregate all standard records according to the acceptance task to form a structured numerical description pair dataset; at the same time, feed back the statistical results and anomaly summaries of this association matching and verification to the substation end in real time through the closed-loop channel.

[0060] Furthermore, by combining the preset transmission delay tolerance, a time window for matching and searching in the real-time data slice stream of the main station is determined. This includes: based on the precise timestamp in the signal metadata reported by the substation, superimposed with the preset transmission delay tolerance (such as forward tolerance +ΔT1, reverse tolerance -ΔT2) according to historical data or network conditions, to construct a time window [T-ΔT2, T+ΔT1] that expands in both directions forward and backward with the substation trigger timestamp as the center; in the data slice stream continuously generated by the main station, all slice files whose timestamps fall within this window are selected as the target data pool for matching and searching, thereby ensuring that even if there is network jitter or a small delay in the main station's processing, the actual measured record of the main station corresponding to the signal can be captured.

[0061] Within the corresponding slice file time window, locate and extract the measured values ​​recorded by the data ID and the signal description stored by the main station system. This includes: in the determined target slice file, using the data ID as a key index, quickly locate the physical location of the record corresponding to the ID in the file; according to the known slice file storage format (such as fixed-length records, tagged binary blocks), parse and read the fields of the corresponding measured values ​​(such as floating-point numbers, integer status values) and the fields of the signal description (usually encoded text strings) in the record block; for the description text, it needs to be decoded according to the character encoding used by the main station system (such as GBK, UTF-8), and finally extract the measured values ​​and the main station side signal description text that can be directly used for comparison.

[0062] The consistency score is calculated based on a text similarity algorithm. This includes: preprocessing the extracted main station signal description text and the standard description text reported by the substation (such as removing extra spaces, unifying capitalization, and standardizing term abbreviations), and then using a text similarity algorithm (such as cosine similarity algorithm based on word vectors or edit distance algorithm oriented towards characters) to calculate the similarity score. The algorithm will convert the two texts into mathematical vectors or calculate the minimum cost of their mutual conversion, and finally output a similarity score between 0 and 1. This score quantifies the semantic or literal similarity between the two texts, providing an objective basis for judging the consistency of descriptions.

[0063] All standard records are aggregated according to the acceptance task to form a structured numerical description pair dataset. This includes: using a complete acceptance task (corresponding to a substation or a bay) as the organizational unit, all the standard records that have passed verification generated in step S34 are collected; each standard record is a row in the dataset, and its structured fields include at least: point number, substation theoretical value, substation description, master station measured value, master station description, signal time stamp, numerical verification status, description similarity score, and comprehensive status marker; finally, these records are integrated in the form of lists or tables, supplemented with task metadata (such as task ID, substation name, acceptance start and end time), to form a complete, ordered, structured dataset that can be used for subsequent statistical analysis, report generation, and process traceability.

[0064] Specifically, the system achieves real-time synchronization and precise correlation of signal data between master and slave stations through an independent closed-loop communication channel. By utilizing time-stamp alignment and a two-dimensional intelligent comparison mechanism, it solves the problems of coordination difficulties, low efficiency, and difficulty in discovering hidden configuration errors caused by the separate operation of master and slave stations and inconsistent signal descriptions in traditional manual acceptance. This enables automated and precise acceptance of master-slave station collaboration and forms a complete and traceable structured acceptance dataset.

[0065] S4. For the aligned numerical description pair dataset, a matching algorithm is used to compare the degree of difference between the real-time values ​​of the main station and the theoretical values ​​of the sub-station. If the degree of difference exceeds the preset threshold, it is marked as a configuration misalignment, and a preliminary comparison result set is obtained.

[0066] The degree of difference is determined based on the signal type: for telemetry signals, a state consistency judgment is used. If the real-time value of the master station is inconsistent with the theoretical value of the substation, such as split / combined or 0 / 1, then a difference is determined to exist. For telemetry signals, relative error or absolute error is used for calculation. The calculated error value is compared with the preset effective range and reasonable fluctuation threshold of the signal in the point table. If the error value exceeds the preset threshold or the measured value exceeds the effective range, then the degree of difference is determined to exceed the limit.

[0067] Furthermore, step S4 specifically includes:

[0068] For the aligned numerical description pair dataset, the corresponding difference calculation strategy is selected according to the signal type. For remote signaling signals, state consistency judgment is adopted, and for telemetry signals, relative error or absolute error calculation method is adopted, combined with the effective range and reasonable fluctuation threshold defined in the point table of the signal. If the difference between the measured value and the theoretical value exceeds the preset tolerance corresponding to its type, the current numerical pair is marked as misconfigured, and the difference details and judgment basis are recorded.

[0069] The numerical pairs marked as misaligned are sorted by timestamp, and the misalignment sequences appearing within a continuous time window are identified. For each continuous misalignment sequence, the consistency of the misalignment pattern is analyzed, including the direction of numerical deviation, the stability of the deviation magnitude, and whether there is accompanying inconsistency in description. Continuous misalignment sequences with similar patterns are grouped into the same misalignment group, forming a misalignment group set based on the dimensions of device, signal type, or forwarding link.

[0070] For each misaligned group, the potential configuration anomalies that cause the continuous signal misalignment are analyzed by combining the main station table configuration, sub-station SCD / RCD files and point number mapping relationship. These anomalies include, but are not limited to, coefficient configuration errors, inconsistent units, missing forwarding mapping, and data ID association deviations. A preliminary analysis summary containing the misalignment pattern, scope of impact, and suspected causes is generated.

[0071] All misalignment records, misalignment groupings, and preliminary root cause analysis results are integrated into a structured preliminary comparison result set, and a pre-verification report containing misalignment statistics, distribution of major problem types, and a list of key abnormal signals is generated for each substation acceptance task.

[0072] Furthermore, relative or absolute error calculation methods are employed, combined with the effective range and reasonable fluctuation thresholds defined in the signal point table. This includes: for telemetry signals, the error calculation strategy is automatically selected based on the signal's range, reference value, and engineering attributes defined in the point table: for signals near zero or measurement signals (such as voltage and current), absolute error calculation is prioritized; for signals with large variation ranges, such as power and load, relative error calculation is used. After calculation, the obtained error value is compared with the reasonable fluctuation threshold predefined for the signal in the point table (set based on normal operating conditions or accuracy requirements). Simultaneously, the measured value itself is compared with the effective range defined in the point table (such as upper and lower limits of the range, physical possible range). If the measured value exceeds the effective range, it is directly judged as abnormal. Ultimately, only when both conditions are met—error not exceeding the threshold and value within the effective range—is the value considered compliant; otherwise, it will be marked as misconfigured and the specific out-of-limit parameters will be recorded.

[0073] Specifically, by employing a signal-type-based differential calculation strategy and misalignment pattern analysis, the problem of systematically identifying and locating telemetry and teleindication signal configuration errors during manual acceptance testing has been solved. This enables automatic identification, classification, and preliminary root cause analysis of deep configuration anomalies such as coefficient errors and dimensional inconsistencies, significantly improving the diagnostic depth and efficiency of signal configuration quality.

[0074] S5. Based on the preliminary comparison result set, obtain the descriptive text fragments from the main site and perform string similarity calculation with the text from the sub-site. Determine whether the similarity is lower than the preset threshold to identify hidden description inconsistencies and obtain the enhanced comparison verification record.

[0075] Furthermore, step S5 specifically includes:

[0076] S51. From the preliminary comparison result set, extract the signal description text of the main station side and the corresponding description text of the sub-station side respectively. Perform preprocessing on each pair of texts, including word segmentation, removal of stop words, standardization of units and units, and extract multi-dimensional feature vectors including keyword sequence, text length, signal type identifier, and device context to form a standardized set of text description pairs.

[0077] S52. For each text description pair, a hybrid calculation model that integrates word vector semantic similarity and edit distance is adopted to comprehensively evaluate the consistency between the two in semantic expression and literal composition. A dynamic threshold is set in combination with signal type and device type. If the similarity score is lower than the corresponding threshold, it is determined that there is a hidden inconsistency in the description pair, and the inconsistent text fragments, difference types and confidence scores are recorded.

[0078] S53. For description pairs determined to be hidden inconsistencies, contextual verification is performed by combining the device logic association, signal classification rules, and terminology standardization dictionary extracted from SCD, RCD, and master station configuration to exclude non-substantive inconsistencies such as naming habits and differences in full abbreviations; for entries confirmed to be true description mismatches, their corresponding numerical misalignment information is associated to form a unified anomaly record containing both numerical and descriptive anomalies.

[0079] S54. Integrate all confirmed non-conformities with the preliminary comparison result set to generate enhanced comparison verification records; each record shall include at least the point number, numerical comparison result, descriptive similarity score, inconsistency type, associated configuration context, and correction suggestions.

[0080] Furthermore, a hybrid computational model integrating word vector semantic similarity and edit distance is adopted to comprehensively evaluate the consistency between the two in terms of semantic expression and literal composition. This includes: using word vector models (such as Word2Vec or BERT) to map the two preprocessed descriptive texts to a high-dimensional semantic space, calculating the average vector of their word vector sequences or obtaining the cosine similarity of sentence vectors through an attention mechanism to assess the closeness between the two at the semantic level; simultaneously, calculating the standardized edit distance (such as Lewinstein distance) between the two texts and converting it into a similarity score to quantify the differences between the two in terms of literal composition and character order, and using a preset fusion weight (such as semantic weight α, literal weight β, where α+β=1), weighted summing of the semantic similarity score and the literal similarity score to generate a comprehensive similarity score between 0 and 1; by taking into account both semantics and literal meaning, this model can effectively identify complex inconsistencies such as circuit breaker and switch (semantically close, literally distant) or ABC switch and ABD switch (semantically distant, literally close), thereby making a comprehensive evaluation of descriptive consistency.

[0081] For description pairs identified as having hidden inconsistencies, contextual verification is performed by combining the device logical associations, signal classification rules, and terminology standardization dictionaries extracted from the SCD, RCD, and master station configurations. This includes: locating the IED device and logical node type to which the description belongs from the SCD file based on the point number; confirming its forwarding path from the RCD file; constructing a standard device context framework for the signal based on the device naming conventions in the master station configuration; normalizing the terms in the description text using a pre-built power system terminology standardization dictionary (e.g., mapping the main transformer to the main transformer and the PT to the voltage transformer); and determining whether the current description pair conforms to the conventional expression logic of its signal type based on signal classification rules (e.g., protection action signals, status indication signals, and measurement signals should have specific description patterns). Finally, a comprehensive logical verification is performed: if the description difference only exists between synonym substitution, standard terminology and colloquialism, or conforms to the alternating naming convention under the same device context, it is determined to be a non-substantial inconsistency and is excluded; if the difference leads to conflicting device attribution, confusion of signal nature, or violation of classification rules, it is confirmed as a genuine description mismatch. This process elevates plain text comparison to intelligent diagnosis driven by business knowledge, significantly reducing the false alarm rate.

[0082] Specifically, by integrating a hybrid model that combines semantic and literal similarity with intelligent verification of power business context, the problem of hidden inconsistencies in the signal description texts of the main station and substations, which are difficult to detect through manual verification, has been solved. This has enabled the automatic elimination of non-substantive differences such as terminology differences and naming conventions, as well as the accurate identification of mismatches in actual descriptions, significantly improving the accuracy and reliability of signal description consistency verification.

[0083] S6. By verifying the enhanced comparison records, generate a timestamp-aligned traceability log file, determine the integrity of all verification records in the log file, and obtain a closed-loop verification result report.

[0084] Furthermore, step S6 specifically includes:

[0085] In response to the enhanced comparison and verification record, for each signal acceptance event, a structured log unit is generated according to its process nodes, including substation triggering, remote telemetry forwarding, master station receiving, data parsing, intelligent comparison and result feedback status. Each unit carries a precise timestamp and a globally unique event identifier.

[0086] All log units under the same acceptance task are linked together with the event logic in chronological order, and an associated integrity verification chain is generated based on a hash algorithm to form a full-link traceability log file with anti-tampering capabilities.

[0087] Based on the traceability log file, the continuity of the hash chain is automatically verified, and the internal logs of the main station system, the trigger logs of the substation device, and the closed-loop channel communication logs are integrated and analyzed to verify whether the entire process from signal triggering to result feedback forms a closed loop consistency in terms of time sequence and business logic.

[0088] Specifically, when the hash chain of the trace log file is verified to be continuous and complete, the time stamps of each link meet the strict monotonically increasing condition and do not exceed the preset delay threshold, and the main station internal logs, sub-station trigger logs and closed-loop channel communication logs corroborate each other and the business link is completely closed, it is automatically determined that the entire process from signal triggering to result feedback has formed a closed-loop consistency.

[0089] Based on the results of integrity verification and closed-loop consistency verification, a structured closed-loop verification result report is generated. The report includes the overall acceptance conclusion, quantitative performance indicators of each stage, and detailed tracing paths and related context information for all abnormal events.

[0090] Furthermore, an integrity verification chain is generated based on a hash algorithm, forming a fully traceable log file with tamper-proof capabilities. This includes: after generating the first log unit, calculating the hash value (such as SHA-256) of all its contents (including timestamps, event identifiers, status data, etc.), which serves as the digital fingerprint of the unit and is stored in the unit's metadata; when generating the second log unit, in addition to calculating its own content hash, the hash value of the previous unit is also used as part of the input for joint calculation, and the resulting new hash value is also stored in the current unit; this process is iterative, so that the hash value of each new log unit inherits and contains the hash fingerprints of all preceding units, thus forming an interlocking hash chain; any tampering with the historical log content will cause unpredictable chain changes in the hash values ​​of that unit and all subsequent units; during verification, it is only necessary to recalculate and compare the hash values ​​chain by chain to quickly and reliably locate any chain break or data tampering point, thus forming a fully traceable log file with cryptographic tamper-proof protection.

[0091] By integrating and analyzing the internal logs of the main station system, the trigger logs of the substation devices, and the communication logs of the closed-loop channel, the system verifies whether the entire process from signal triggering to result feedback forms a strict closed loop in terms of time sequence and business logic. This includes: aligning the timestamps of each key node (substation triggering, main station receiving, and result feedback) in the traceability logs with the precise records of MMS signal transmission in the substation device triggering logs, the corresponding network message reception and processing records in the internal logs of the main station system, and the transmission and reception records of private protocol instructions in the closed-loop channel communication logs. Furthermore, cross-validation ensures that substation triggering events are corroborated by source device logs, main station receiving and processing events are corroborated by internal system logs, and result feedback events are corroborated by channel communication logs. Finally, based on the correlated multi-source evidence chain, the system verifies that the complete event sequence from triggering to feedback not only satisfies the logic of strictly monotonically increasing timestamps, but also that the intervals between each stage do not exceed the preset physical transmission and processing delay thresholds. Thus, under the premise of ensuring the authenticity of evidence, compliance with timing, and logical coherence, the system determines whether a single signal acceptance has formed an irrefutable and strict business closed loop.

[0092] Specifically, by constructing a hash-based anti-tampering traceability log and a multi-source log fusion verification mechanism, the problems of unauditable traditional acceptance processes and difficulty in backtracking results are solved. This achieves a full-link, verifiable closed-loop record from signal triggering to result feedback, ensuring the integrity, traceability, and immutability of the acceptance process, and providing a reliable data evidence chain and structured report for project acceptance.

[0093] S7. Based on the closed-loop verification result report, update the dynamic adjustment parameters of the point number correspondence mapping table. If the adjustment parameters indicate a change in the encoding rules, re-parse the slice file to obtain the optimized real-time data acquisition path.

[0094] Furthermore, step S7 specifically includes:

[0095] In response to the generation of the closed-loop verification result report, statistical conclusions and specific entries regarding configuration misalignment, inconsistent descriptions, and link anomalies are parsed from it. Based on preset rules and historical learning models, the conclusions and entries are quantified into dynamic adjustment weights and correction vectors for specific parameters in the point number correspondence mapping table. The adjustment weights and correction vectors are applied to update the corresponding entries in the mapping table, generating an adjusted mapping configuration version with self-learning capabilities.

[0096] Based on the adjusted mapping configuration version, analyze whether the dot encoding pattern of the same device or manufacturer has undergone structural shift. The judgment criteria include, but are not limited to, changes in the length of the encoding field, the addition of fixed character segments, changes in field semantics, or a decrease in the mapping success rate with the latest configuration file on the sub-site side. If it is determined that the encoding rules have changed, the incremental re-parsing process of the main site slice file is triggered. Only the corresponding data ID segments involved in the rule change are located and re-decoded, and the encoding pattern benchmark library in the mapping table is updated synchronously.

[0097] Based on the updated file structure data obtained after incremental re-parsing, and combined with the timeliness, stability, and load indicators of each data channel fed back in the closed-loop verification result report, the efficiency of the complete data acquisition path from file listening, capture to parsing is re-evaluated; a dynamic weighting algorithm is used to calculate and update the priority of each potential path, and the path with the best overall performance is selected as the main real-time data acquisition path for the next round of acceptance tasks.

[0098] The adjusted mapping configuration version, the updated encoding rule benchmark library, and the optimized real-time data acquisition main path parameters are packaged into a unified system configuration file version. This system configuration file version is then released and integrated into the main station's automatic acceptance system. It serves as the initial parameters and execution basis for the next round of monitoring information acceptance tasks, thereby enabling the acceptance system to achieve a closed loop of self-iteration and performance optimization based on feedback during continuous operation.

[0099] Furthermore, based on preset rules and a historical learning model, the conclusions and entries are quantified into dynamic adjustment weights and correction vectors for specific parameters in the point number correspondence mapping table. This includes: first, attributing and classifying each abnormal entry according to preset rules, for example, classifying continuously misaligned values ​​with constant deviations as coefficient configuration errors, and classifying the absence of point numbers on the main station side as mapping omissions; the historical learning model is activated, which calculates two core quantitative outputs by analyzing the frequency of similar anomalies in history, self-repair status, and correlation strength with specific equipment manufacturers: one is the adjustment weight, a value between 0 and 1, representing the urgency and confidence of correcting the entry or rule (for example, the weight of frequently occurring anomalies approaches 1); the other is the correction vector, a mathematical vector that specifically indicates which parameter(s) in the mapping table (such as dimension conversion coefficients, field offsets, description matching rules) needs adjustment, as well as the theoretical direction and magnitude of the adjustment; this process transforms qualitative acceptance issues into precise mathematical instructions that drive the evolution of the system's core parameters.

[0100] The adjustment weights and correction vectors are applied to update the corresponding entries in the mapping table, generating an adjusted mapping configuration version with self-learning capabilities. This includes: performing update operations on the target parameters based on the adjustment weights as the learning rate or update intensity coefficient. For example, if an entry in the mapping table is marked due to inconsistent descriptions, its description-related confidence parameter will be decayed proportionally according to the weights; if it is determined to be an error in the boundary of the encoded parsing field, its start and end parameters will be directly corrected according to the value specified by the correction vector; all these update operations are performed in a memory copy of the current mapping configuration, generating an adjusted mapping configuration version containing all new parameter values; this version not only records the new state of the parameters, but its generation process itself constitutes a complete self-learning cycle of the system's internal knowledge base based on feedback evidence.

[0101] Based on the adjusted mapping configuration version, analyze whether the dot encoding pattern of the same device or manufacturer has undergone structural shift, including: 1) Horizontal consistency scan: compare the new and old dot numbers in the cluster to see whether there are collective variations in core structural features (such as total length, number and position of field separators, fixed prefix / suffix); 2) Semantic field stability test: analyze whether the encoding value domain or encoding dictionary of key information fields (such as device serial number, type code) has undergone discontinuous and systematic changes; 3) External mapping success rate verification: re-evaluate the success rate of automatic mapping of all dot numbers in the cluster with the latest version of the substation project configuration file (SCD / RCD). If the success rate drops significantly and continuously compared with the historical baseline, it constitutes strong external counter-evidence; when the above multiple analyses all point to the same conclusion, that is, the basic generation logic of encoding has changed, it is determined that a structural shift requiring rule base upgrade has occurred.

[0102] A dynamic weighting algorithm is employed to calculate and update the priority of each potential path, selecting the path with the best overall performance as the main real-time data acquisition path for the next round of acceptance tasks. This includes: firstly, normalizing the historical and real-time performance indicators of each candidate data acquisition path and constructing a multi-period performance dataset based on a sliding time window; secondly, using the entropy weighting method to objectively calculate the weights of indicators such as timeliness, stability, and load, the core of which lies in dynamically allocating weights according to the dispersion of each indicator value in the current dataset, the greater the dispersion, the higher the discrimination of the indicator, and the corresponding increase in weight; thirdly, the algorithm introduces a feedback reinforcement mechanism based on the success rate of historical tasks to fine-tune the objective weights, for example, the weights of key defect indicators of paths that have recently failed will be penalized and reinforced; finally, by combining the dynamic weights, the TOPSIS multi-attribute decision method is used to calculate and rank the relative closeness of each path to the ideal solution, and the path closest to the ideal solution is established as the main acquisition path for the next round of tasks, while the second-best path is set as a hot backup, thus forming an adaptive path optimization closed loop with objective evaluation, feedback learning, and flexible decision-making capabilities.

[0103] Specifically, based on the feedback of closed-loop verification results, by quantitatively analyzing abnormal items and dynamically adjusting mapping parameters, the problems of high maintenance costs and long-term reliability degradation caused by the inability of traditional systems to adapt to changes in coding rules and the solidification of data acquisition paths are solved. By self-learning to update mapping configuration, intelligently detecting rule offsets and triggering incremental re-parsing, and dynamically optimizing data acquisition paths, the acceptance system achieves continuous self-iteration and performance optimization during operation, significantly improving the adaptability and stability of the system in long-term operation.

[0104] Example 2

[0105] Please see Figure 2 This embodiment provides a non-intrusive master station monitoring information acceptance system based on real-time data slice analysis, used to implement a non-intrusive master station monitoring information acceptance method based on real-time data slice analysis, including:

[0106] The data monitoring and parsing module is used to monitor and parse the binary real-time data slice files generated by the main station. By recognizing the composite coding structure and the domain number zeroing algorithm, it converts the data ID into the device identifier, forms a preliminary mapping relationship with the monitoring information point table, and outputs the device identifier and point number information sequence after cluster analysis and multi-source verification.

[0107] The mapping relationship management module is used to extract coding features based on the sequence and perform unsupervised clustering and grouping, combine preset difference thresholds with sub-station configuration files for cross-validation and pattern merging, generate standard coding sequences and establish a global standardized mapping table, and form a standardized point number correspondence mapping table after verification with the main station authoritative point table.

[0108] The collaborative acceptance processing module is used to receive signal metadata reported by the substation through an independent closed-loop channel, map the substation number to the main station data ID using the standardized mapping table, extract the measured value and description in the corresponding time scale window of the slice file, complete the two-dimensional comparison and logical verification of the numerical value and description, integrate them into a structured numerical description pair dataset and feed it back to the substation in real time.

[0109] The difference analysis module is used to apply the corresponding difference calculation strategy to the dataset according to the signal type, compare it with the tolerance defined in the point table, mark the out-of-limit numerical pairs as configuration misalignment, perform pattern analysis and grouping on the misalignment sequence, combine multi-source configuration to preliminarily locate the root cause, and generate a preliminary comparison result set and pre-verification report.

[0110] The text verification module is used to extract the description text of the main station and the sub-station, calculate the consistency using a hybrid model that integrates semantics and literal similarity, determine hidden inconsistencies using dynamic thresholds, and perform contextual business verification using device logic, classification rules and terminology dictionary. After excluding non-substantive differences, it generates enhanced comparison verification records.

[0111] The traceability and reporting module is used to generate time-stamped structured log units based on verification records, and to build tamper-proof full-link traceability log files based on hash algorithms. It integrates and analyzes logs from the main station, sub-stations and closed-loop channels to verify whether the entire process from triggering to feedback of the signal forms closed-loop consistency, and generates a closed-loop verification result report.

[0112] The adaptive optimization module is used to quantify abnormal conclusions into parameter adjustment instructions for the point number mapping table based on the closed-loop verification result report, update the mapping table and generate a new configuration version after self-learning, analyze whether the encoding rules have undergone structural shifts and trigger incremental re-parsing, evaluate the performance of each data channel and then use a dynamic weight algorithm to select the optimal real-time data acquisition path to achieve continuous system optimization and closed-loop.

[0113] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention are within the scope of the present invention.

Claims

1. A non-intrusive main station monitoring information acceptance method based on real-time data slicing analysis, characterized in that: Includes the following steps: S1. Obtain the binary slice file generated by the main station as the initial input source, determine the pattern distribution of the encoding rules by parsing the composite encoding structure inside the slice file, and obtain the device identifier and point number information sequence; Step S1 specifically includes: Listen for periodic data backup commands triggered by the main station system, and capture real-time binary data slices to be parsed based on file timestamps and signature increments; Read the binary stream of the target slice file, identify its composite encoding structure, parse the table number, domain number, area number and record number in the data ID according to the predefined field rules, and convert the data ID into a device identifier through the domain number zeroing algorithm, and associate it with the pre-stored point table to form a preliminary device identifier and point number information mapping pair; Extract the coding sequence features from the mapping pairs, group similar sequences using a clustering algorithm based on a preset manufacturer coding rule library, distinguish different manufacturer coding patterns by combining the difference threshold, and standardize the same pattern sequences to generate a standardized correspondence between device identifiers and point number information. The standardized correspondence is associated with the main station database point table configuration and the substation configuration file on the substation side through multi-source association and intelligent matching. The consistency of point number, equipment identifier and signal description in each source data is verified. Invalid or conflicting mapping entries are identified and removed. The sequence of equipment identifier and point number information verified by multiple sources is output. S2. Based on the device identifier and point number information sequence, a clustering algorithm is used to group similar coding patterns in the sequence, and a standardized point number correspondence mapping table is obtained by comparing it with a preset manufacturer difference threshold. S3. Based on the point number correspondence mapping table, associate the signal description text and theoretical value reported by the sub-station with the real-time value in the main station slice file to obtain the aligned numerical description pair dataset. S4. For the numerical description pair dataset, a matching algorithm is used to compare the degree of difference between the real-time values ​​of the main station and the theoretical values ​​of the sub-station. If the difference exceeds the preset threshold, it is marked as a configuration misalignment, and a preliminary comparison result set is obtained. S5. Based on the preliminary comparison result set, calculate the string similarity between the description text of the main station side and the sub-station side, determine whether the similarity is lower than the preset threshold to identify the inconsistency of hidden descriptions, and obtain the enhanced comparison verification record. S6. By verifying the enhanced comparison records, generate a timestamp-aligned traceability log file, determine the integrity of all verification records in the log file, and obtain a closed-loop verification result report.

2. The non-intrusive master station monitoring information acceptance method based on real-time data slice analysis according to claim 1, characterized in that: Step S2 specifically includes: Extract the coding structure features of each point number in the device identifier and point number information sequence, and use an unsupervised clustering algorithm to group all point number sequences to form multiple initial pattern clusters; For each initial pattern cluster, calculate the edit distance between the point number sequences within the cluster. If the difference value is less than a preset threshold, it is determined to be a consistent encoding pattern, and cross-mapping verification is performed with the substation SCD and RCD files. From the verified patterns, a central sequence or a synthetic standard coding sequence is selected as the benchmark to establish a global point number standardized mapping table. This table is then used for secondary association verification with the point table in the main station database. After dynamically adjusting the entries, a standardized point number correspondence mapping table is formed.

3. The non-intrusive master station monitoring information acceptance method based on real-time data slice analysis according to claim 1, characterized in that: Step S3 specifically includes: In response to the acceptance command sent by the substation through an independent private protocol closed-loop channel, receive and parse the signal metadata reported by the substation, and construct a signal trigger event queue; The time window is determined based on the event timestamp and transmission delay tolerance. The point number is mapped to the main station data ID using a mapping table. The measured value and description are located in the slice file to form a preliminary association pair. The system performs a two-dimensional comparison of numerical differences and descriptive similarity between the associated pairs, and combines the configurations of the main site and sub-sites for logical verification to generate a comprehensive criterion. The records that pass the comprehensive judgment criteria are integrated and verified into a numerical description pair dataset, and the statistical results are fed back to the sub-station in real time.

4. The non-intrusive master station monitoring information acceptance method based on real-time data slice analysis according to claim 1, characterized in that: Step S4 specifically includes: For the numerical description pair dataset, select the corresponding difference calculation strategy and tolerance threshold according to the signal type, and mark the numerical pairs that exceed the tolerance as configuration misalignment; The misalignments are sorted by timestamp, consecutive misalignment sequences are identified, and sequences with similar patterns are merged into misalignment groups. For each misaligned group, analyze potential configuration anomalies by combining the main site and sub-site configuration files, and generate a preliminary analysis summary; Integrate all misaligned records and analysis results to form a preliminary comparison result set and a pre-verification report.

5. The non-intrusive master station monitoring information acceptance method based on real-time data slice analysis according to claim 1, characterized in that: Step S5 specifically includes: The descriptive text of the main station and sub-station is extracted from the preliminary comparison result set, preprocessed and multidimensional features are extracted to form a standardized set of text description pairs; For each text description pair, a hybrid model that integrates word vector semantics and edit distance is used to calculate similarity, and hidden inconsistencies are determined based on dynamic thresholds; For description pairs determined to be hidden inconsistencies, verification is performed in conjunction with the configuration context to exclude non-substantive differences, confirm genuine mismatches, and associate numerical misalignment information. Integrate real anomaly records with preliminary comparison results to generate enhanced comparison and verification records.

6. The non-intrusive master station monitoring information acceptance method based on real-time data slice analysis according to claim 1, characterized in that: Step S6 specifically includes: For each signal acceptance event, a structured log unit with timestamps and event identifiers is generated based on the process nodes; All log units under the same acceptance task are sequentially linked together, and an integrity verification chain is generated based on a hash algorithm to form a full-link traceability log file; By verifying the continuity of the hash chain and integrating and analyzing multi-source logs, we can verify whether a closed-loop consistency is formed throughout the entire process. Among them, when the hash chain of the trace log file is verified to be continuous and complete, the time stamp of each link meets the strict monotonically increasing condition and does not exceed the preset time delay threshold, and the main station internal log, sub-station trigger log and closed-loop channel communication log corroborate each other and the business link is completely closed, it is automatically determined that the whole process from signal triggering to result feedback has formed a closed-loop consistency. A structured closed-loop verification result report is generated based on the verification results.

7. The non-intrusive master station monitoring information acceptance method based on real-time data slice analysis according to claim 1, characterized in that: Also includes: S7. Based on the closed-loop verification result report, update the dynamic adjustment parameters of the point number correspondence mapping table. If the adjustment parameters indicate a change in the encoding rules, re-parse the slice file to obtain the optimized real-time data acquisition path.

8. The non-intrusive master station monitoring information acceptance method based on real-time data slice analysis according to claim 7, characterized in that: Step S7 specifically includes: Based on the closed-loop verification result report, the abnormal statistical entries are analyzed and quantified into dynamic adjustment weights and correction vectors of the mapping table parameters according to preset rules and historical models. The mapping table entries are then updated to generate an adjusted configuration version with self-learning capabilities. Based on the adjusted configuration version, analyze whether the device or manufacturer's encoding mode has undergone structural shift. If the judgment rule has changed, trigger incremental re-parsing of the main site slice file and update the encoding mode benchmark library synchronously. Based on the data obtained from re-analysis and the performance indicators of each channel reported in the report, the efficiency of the complete data acquisition path is re-evaluated. A dynamic weighting algorithm is used to calculate and select the path with the best overall performance as the main path for the next round of data acquisition. The adjusted configuration version, updated encoding rule library, and optimized main path parameters are packaged into a unified system configuration file version and integrated into the main site acceptance system. This serves as the initial execution basis for subsequent acceptance tasks, realizing a feedback-based self-iterative optimization closed loop for the system.

9. A non-intrusive master station monitoring information acceptance system based on real-time data slice analysis, applied to the non-intrusive master station monitoring information acceptance method based on real-time data slice analysis as described in any one of claims 1-8, characterized in that: include: The data monitoring and parsing module is used to monitor and parse the binary real-time data slice files generated by the main station. By recognizing the composite coding structure and the domain number zeroing algorithm, it converts the data ID into the device identifier, forms a preliminary mapping relationship with the monitoring information point table, and outputs the device identifier and point number information sequence after cluster analysis and multi-source verification. The mapping relationship management module is used to extract coding features based on the sequence and perform unsupervised clustering and grouping, combine preset difference thresholds with sub-station configuration files for cross-validation and pattern merging, generate standard coding sequences and establish a global standardized mapping table, and form a standardized point number correspondence mapping table after verification with the main station authoritative point table. The collaborative acceptance processing module is used to receive signal metadata reported by the substation through an independent closed-loop channel, map the substation number to the main station data ID using the standardized mapping table, extract the measured value and description in the corresponding time scale window of the slice file, complete the two-dimensional comparison and logical verification of the numerical value and description, integrate them into a structured numerical description pair dataset and feed it back to the substation in real time. The difference analysis module is used to apply the corresponding difference calculation strategy to the dataset according to the signal type, compare it with the tolerance defined in the point table, mark the out-of-limit numerical pairs as configuration misalignment, perform pattern analysis and grouping on the misalignment sequence, combine multi-source configuration to preliminarily locate the root cause, and generate a preliminary comparison result set and pre-verification report. The text verification module is used to extract the description text of the main station and the sub-station, calculate the consistency using a hybrid model that integrates semantics and literal similarity, determine hidden inconsistencies using dynamic thresholds, and perform contextual business verification using device logic, classification rules and terminology dictionary. After excluding non-substantive differences, it generates enhanced comparison verification records. The traceability and reporting module is used to generate time-stamped structured log units based on verification records, and to build tamper-proof full-link traceability log files based on hash algorithms. It integrates and analyzes logs from the main station, sub-stations and closed-loop channels to verify whether the entire process from triggering to feedback of the signal forms closed-loop consistency, and generates a closed-loop verification result report. The adaptive optimization module is used to quantify the abnormal conclusions into parameter adjustment instructions for the point number mapping table based on the closed-loop verification result report, update the mapping table and generate a new configuration version after self-learning, analyze whether the encoding rules have undergone structural shifts and trigger incremental re-parsing, evaluate the performance of each data channel and then use a dynamic weight algorithm to select the optimal real-time data acquisition path.

Images