Modular double-layer prefabricated substation system based on digital twin

Abstract

This invention discloses a modular, two-layer prefabricated substation system based on digital twins, relating to the field of digital twin power equipment technology. It aims to address the challenges of new energy integration and rapid urban load growth. Through semantic handshakes to eliminate data naming conflicts, time base solidification to unify the time reference, correlation modeling to reveal the inherent connections between data, drift detection to identify deviations between the model and the physical system, version patches to synchronize firmware and the model, and policy updates to optimize protection settings and lifespan assessment, it improves the deployment efficiency, operational reliability, and intelligence level of substations. This lays a solid foundation for the promotion and application of digital distribution networks and meets the demands of modern power systems for efficient and intelligent operation and maintenance.

Description

Technical Field

[0001] This invention relates to the field of digital twin power equipment technology, specifically to a modular, two-layer prefabricated substation system based on digital twins. Background Technology

[0002] Against the backdrop of rapid growth in new energy integration and urban load, modular double-layer prefabricated substations are being widely adopted. These substations consist of functional enclosures pre-commissioned in the factory, stacked layer by layer, and ready for energization upon installation. A digital twin platform maps the status of each primary, secondary, and auxiliary device in real time, enabling remote monitoring and predictive maintenance. However, because the enclosures are supplied by different manufacturers, the models of sensing devices, sampling rates, and communication semantics vary widely. Once on-site installation is complete, the edge device must immediately package all state variables into a unified model and push it to the cloud twin within milliseconds; otherwise, the thermo-mechanical-electrical coupling analysis, protection setting verification, and lifespan assessment relied upon by maintenance personnel will be distorted. Existing research indicates that prefabricated substations have short construction cycles and numerous but inconsistent interfaces. A digital twin framework can provide full lifecycle visualization and analysis support for the entire substation, and object-modeled data semantics are the core of achieving this goal.

[0003] However, the heterogeneous data streams generated at the moment of access across different enclosures remain a pain point in the industry: the large power data sent by the upper-layer primary equipment and the operating condition information uploaded by the lower-layer secondary and auxiliary equipment have naming conflicts and time reference drift. Traditional offline resynchronization methods can only align waveforms after the fact. Although the edge gateway can perform protocol conversion from Modebass to message queue telemetry transmission, it is difficult to complete semantic remapping on-site. When this desynchronization continues to accumulate, the twin model will gradually deviate from the real object, eventually leading to misjudgment of state assessment, delayed fault warning, and failure of self-healing control. Even if Kalman estimation or fault warning algorithms based on sensor drift detection are introduced, they can only alleviate rather than cure the problem, because version updates and manufacturer expansions will continuously add new differences.

[0004] In current practice, version control should be integrated into asset model management. By using a git-like maintenance approach to link each firmware upgrade with the evolution of the twin object, consistency between the virtual and physical systems can be fundamentally maintained. However, existing prefabricated substations lack a rapid calibration and continuous delivery mechanism for cross-layer modules, making the twin-physical coupling chain most vulnerable during the initial site commissioning phase. Therefore, bridging the cross-layer data gap immediately after installation to ensure steady-state alignment between the twin and the physical entity has become a core technical challenge that must be prioritized for digital twin-driven modular dual-layer prefabricated substation systems. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] To address the shortcomings of existing technologies, this invention provides a modular, two-layer prefabricated substation system based on digital twins. It eliminates data naming conflicts through semantic handshakes, unifies the time base through time-base solidification, reveals the intrinsic connections between data through correlation modeling, detects drift discrepancies between the model and the physical system, synchronizes firmware and model through version patches, and optimizes protection settings and lifespan assessment through policy updates. This improves the deployment efficiency, operational reliability, and intelligence level of substations, meeting the demands of modern power systems for efficient and intelligent operation and maintenance; and solves the technical problems described in the background section.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention is implemented through the following technical solution: a modular double-layer prefabricated substation system based on digital twins, including a semantic handshake unit, an edge controller that listens to the first report frame of each enclosure, calls the IEC61850 logic node template to complete the three-element registration of equipment, measurement points and functions, generates the first version of the semantic list and marks the hierarchical source;

[0009] The time-base solidification unit allows the edge gateway to insert highly stable timestamps based on the semantic list, complete the mapping from Modebass to the message queue telemetry protocol, and output a consistent time series stream across layers;

[0010] After the time series stream is written into the semantic buffer, the system constructs a multi-dimensional correlation matrix according to the hierarchy of the correlation modeling unit, and maps the primary power quantity and secondary operating condition quantity into a dynamic spectrum in real time.

[0011] The drift discrimination unit generates a residual vector and simultaneously calculates the residual spectral entropy index and the time delay jitter index. After the random forest model outputs the track reliability coefficient, the low threshold triggers the patch signal.

[0012] The version patch unit receives a trigger signal, submits the current firmware fingerprint and twin model version to the GitOps pipeline, generates a difference patch after approval, and pushes it back to the edge.

[0013] The strategy refresh unit refreshes the correlation matrix and broadcasts the state change to the cloud microservices as soon as the patch is applied. The engine then recalculates the protection settings and lifespan curves accordingly, keeping on-site control and twin analysis synchronized.

[0014] Furthermore, the edge controller listens to the first report frame of a newly accessed enclosure and receives data packets, verifies the integrity of the data packets and records the reception time, calls the IEC61850 logic node template to parse the first report frame, extracts device information, measurement point information and function information, and generates a structured dataset containing device identifiers, measurement points and functions.

[0015] Furthermore, the system performs equipment registration, measurement point registration, and function registration, registering equipment identifiers, measurement point identifiers, measurement types, hierarchical sources, and function descriptions to the database and generating a relational table.

[0016] Generate an initial semantic manifest based on the relation table, recording device identifiers, measurement point identifiers, measurement types, hierarchical sources, and functional descriptions.

[0017] Furthermore, the edge gateway extracts the measurement point identifier, measurement type, and hierarchical source from the initial semantic list, and ensures data semantic consistency based on the initial semantic list; the edge gateway inserts a highly stable timestamp into each received data packet to record the precise time of data packet reception.

[0018] Furthermore, the edge gateway converts data packets from the Modebest protocol to the Message Queuing Telemetry Transport Protocol, generating topics and payloads; the edge gateway sorts the data packets according to high-stability timestamps, generating a cross-layer consistent time-series stream, which is then pushed to the cloud-based digital twin platform in real time.

[0019] Furthermore, the edge gateway writes the cross-layer consistent time series stream into the semantic buffer in high-stability timestamp order, extracts the time series data from the semantic buffer, and distinguishes the primary power measurement data generated by the upper-layer primary equipment from the secondary operation data generated by the lower-layer secondary and auxiliary equipment based on the hierarchical source in the first version of the semantic list.

[0020] For each pair of primary power measurement data and secondary operation data, the Pearson correlation coefficient is calculated to generate a multidimensional correlation matrix. The correlation pairs in the multidimensional correlation matrix and the correlation strength represented by the Pearson correlation coefficient are mapped to a dynamic spectrum. Graph database technology is used to store the dynamic spectrum and update it in real time.

[0021] Furthermore, by extracting the actual correlation strength between the upper-level primary equipment measurement points and the lower-level secondary and auxiliary equipment measurement points from the dynamic spectrum, and obtaining the predicted correlation strength from the digital twin platform, the difference between the two is calculated to generate a residual vector.

[0022] Apply the Fast Fourier Transform to the residual vector to generate a frequency domain representation. After calculating the normalized power spectral density, use the Shannon entropy formula to calculate the residual spectral entropy exponent.

[0023] Furthermore, high-stability timestamps are extracted from the measurement points of the primary equipment in the upper layer and the measurement points of the secondary and auxiliary equipment in the lower layer from the time series stream. Cross-layer timestamp differences are calculated, and the improved Allan variance is applied to calculate the delay jitter index.

[0024] The residual spectral entropy index and the time delay jitter index are input into the random forest model, and the track alignment confidence coefficient is output. When the track alignment confidence coefficient is lower than the preset threshold, a patch signal is triggered.

[0025] Furthermore, the edge controller continuously monitors the patch signal. When the patch signal is 1, the edge controller initiates the version patching process, collects the device's firmware fingerprint and the model version identifier obtained from the cloud digital twin platform, packages the firmware fingerprint and model version identifier into a version information package, and submits it to the GitOps pipeline.

[0026] Furthermore, after receiving the version information packet, the GitOps pipeline compares the firmware fingerprint with the latest firmware fingerprint in the Git repository and the model version identifier with the latest model version identifier to generate firmware difference patches and model difference patches.

[0027] The GitOps pipeline submits firmware and model difference patches for approval, and after approval, integrates them into the final difference patch.

[0028] Furthermore, the GitOps pipeline pushes the final difference patch back to the edge controller; the edge controller applies the firmware difference patch to update the device firmware and applies the model difference patch to update the digital twin model, records the new firmware fingerprint and model version identifier, and notifies the cloud digital twin platform to update synchronously.

[0029] Furthermore, the edge controller receives differential patches pushed back from the GitOps pipeline, updates the device firmware and digital twin model, generates a new firmware fingerprint and model version identifier, and records them.

[0030] The edge controller recalculates the correlation strength in the multidimensional correlation matrix based on the updated digital twin model, generating a matrix that reflects the relationships between devices.

[0031] Furthermore, the edge controller packages the new firmware fingerprint, the new model version identifier, and the refreshed multidimensional association matrix into a state change package, which is then broadcast to the cloud microservices via a secure communication channel.

[0032] The cloud-based microservice receives the status change packet, updates the firmware fingerprint and model version identifier, adjusts the protection algorithm parameters and lifetime prediction model parameters according to the refreshed multidimensional correlation matrix, recalculates the protection setpoint and lifetime curve, and sends them to the edge controller.

[0033] (III) Beneficial Effects

[0034] This invention provides a modular, two-layer prefabricated substation system based on digital twins, which has the following advantages:

[0035] The semantic handshake listens to the first frame reported by the enclosure through the edge controller and calls the IEC61850 logical node template to complete the three-element registration of the device, measurement point and function, generate the first version of the semantic list and mark the hierarchical source, effectively eliminating naming conflicts of cross-layer data streams. The time base solidification inserts a high-stability timestamp on the basis of the first version of the semantic list and completes the mapping from Modebs to the message queue telemetry protocol, outputting a consistent time series stream across layers, eliminating time base drift, and ensuring the data semantic consistency and time accuracy of the system under complex multi-layer architecture.

[0036] Association modeling writes consistent time-series streams across layers into a semantic buffer, constructs a multi-dimensional association matrix, and maps the primary power quantities of the upper layer and the secondary operating conditions of the lower layer into a dynamic spectrum in real time. Through dynamic topology support, it provides an efficient means for real-time monitoring and anomaly analysis of the system status.

[0037] Drift detection is achieved by generating residual vectors and calculating residual spectral entropy and latency jitter exponents. A random forest model is used to output a track alignment reliability coefficient, enabling accurate detection and quantification of drift between the digital twin model and the actual device state. After receiving the drive signal from the track alignment reliability coefficient, the version patch submits the current firmware fingerprint and the twin model version to the GitOps pipeline. Upon approval, a difference patch is generated and pushed back to the edge, achieving synchronous updates between the device firmware and the digital twin model. This maintains consistency between the virtual and real systems, ensures continuous alignment between the digital twin and the actual device state, and improves the system's adaptability.

[0038] Version patches, through efficient GitOps pipeline management, differentiate firmware fingerprints and twin model versions and push them back to edge devices, ensuring the synchronous evolution of assets and models. The linkage between version patches and drift detection forms a closed-loop mechanism from drift detection to calibration, providing technical support for the continuous optimization of modular dual-layer prefabricated substation systems.

[0039] After the version patch is implemented, the policy update refreshes the multidimensional correlation matrix and broadcasts the state change to the cloud microservices. It also recalculates protection settings and lifetime curves, ensuring continuous synchronization between on-site control and twin analysis, thereby improving security margins and cost-effectiveness. The collaboration between policy updates, correlation modeling, and version patches forms a complete chain from data mapping to policy optimization, guaranteeing the system's efficiency and reliability during dynamic operation. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the modular double-layer prefabricated substation system structure based on digital twins according to the present invention. Detailed Implementation

[0041] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] Please see Figure 1 This invention provides a modular, two-layer prefabricated substation system based on digital twins, comprising:

[0043] Step 1: In the modular double-layer prefabricated substation system, the edge controller listens to the first report frame of the newly connected box and receives the data packet, verifies the integrity of the data packet and records the reception time, calls the IEC61850 logic node template to parse the first report frame, extracts equipment information, measurement point information and function information, generates a structured dataset containing equipment identifiers, measurement points and functions, performs equipment registration, measurement point registration and function registration, registers the equipment identifier, measurement point identifier, measurement type, hierarchical source and function description to the database, and generates a relational table;

[0044] The initial semantic list is generated based on the relation table, recording the device identifier, measurement point identifier, measurement type, hierarchical source and function description to ensure data semantic consistency. The initial semantic list is stored in the version control database, marked with a timestamp and version number. The version record is updated when the device configuration changes, and historical data is retained.

[0045] Step one includes the following:

[0046] Step 101: Device Access and First Report Frame Monitoring

[0047] In a modular double-layer prefabricated substation system, once a new enclosure is hoisted into place and its physical interface is connected to the edge controller, the edge controller activates a preset listening mechanism to capture the first report frame sent by the enclosure.

[0048] The first report frame is an initialization data packet actively sent by the enclosure when it first connects to the system. It contains the device's unique identifier, type, manufacturer information, and a description of the data point. The edge controller continuously monitors data traffic through a pre-configured communication interface, such as Ethernet or a serial bus, and immediately receives and stores the data packet when the first report frame is detected. During reception, the edge controller verifies the integrity of the data packet, ensures no transmission errors, and records the reception time to guarantee data real-time performance.

[0049] In use, an automated monitoring mechanism promptly captures the initialization data of newly connected enclosures, enabling automatic device identification and access management. The edge controller continuously scans the status of the communication interface, and when it identifies a data packet matching the characteristics of the first reporting frame, it triggers the receiving program, stores the data packet in a temporary buffer, and performs integrity verification. First reporting frame monitoring enables automatic discovery of new devices, avoiding manual intervention and ensuring a rapid system response to modular enclosure access.

[0050] Step 102: Parsing the first report frame based on the IEC61850 standard

[0051] After receiving the first report frame, the edge controller calls the logical node template based on the IEC61850 standard to perform structured parsing of the data packet content.

[0052] The IEC 61850 standard is a common communication specification in the field of substation automation, defining the logical nodes and data object models of equipment. The parsing process extracts three types of information from the first frame: the first type is equipment information, including the equipment's unique identifier, model, and manufacturer name; the second type is measurement point information, covering the measurement points supported by the equipment, such as voltage, current, and temperature, with each measurement point accompanied by its corresponding data type and unit; the third type is functional information, describing the types of functions supported by the equipment, such as protection, control, or monitoring. After parsing, the edge controller stores the extracted information according to a predefined structure, generating a structured dataset containing equipment identifiers, measurement points, and functions.

[0053] In use, the edge controller reads the contents of the first report frame field by field according to the data model of the IEC61850 standard, maps each field to the corresponding logical node attribute, and generates a structured list containing equipment information, measurement point information and functional information. This ensures the interoperability of equipment data from different manufacturers and guarantees the universality and consistency of the parsing results. Through a standardized parsing process, the system can accurately extract key equipment information, providing a reliable basis for subsequent data management and application.

[0054] Step 103: Based on the parsing results of the first reported frame, the edge controller executes the three-element registration process to register the extracted information into the system database.

[0055] The three-dimensional registration process comprises three sub-processes: First, device registration, which inputs the device's unique identifier, model, and manufacturer information into the edge-side device management database, forming a device list within the system; second, measurement point registration, which assigns a unique measurement point identifier to each measurement point and records its physical quantity type, such as voltage or current, as well as the data source level, categorized as upper-level primary equipment or lower-level secondary equipment; and finally, function registration, which associates the functions supported by the device with the corresponding device identifier and measurement point identifier, and records the applicable scope of the function, such as the triggering conditions for protection functions. After registration, the edge controller generates a relationship table containing device identifiers, measurement point identifiers, and function associations for subsequent data reference.

[0056] During use, a structured registration process establishes the relationships between devices, measurement points, and functions, enabling systematic data management. The edge controller assigns a unique identifier to each device, measurement point, and function, records attribute information and relationships in the database, and generates a relationship table containing all registered items. This generated relationship table supports rapid querying and retrieval of device data, providing an efficient data organization method for system operation.

[0057] Step 103: Semantic Inventory Generation

[0058] The edge controller generates an initial semantic list based on the relationship table completed by the ternary registration, which is used to record the semantic description of the data.

[0059] The semantic manifest is organized in tabular form, with each record containing the following fields: Device Identifier (a unique identifier for the device); Measurement Point Identifier (a unique identifier for the measurement point); Measurement Type (describing the measured physical quantity, such as voltage or current); Hierarchical Source (indicating whether the data originates from the primary or secondary layer); and Functional Description (explaining the actual function's purpose, such as overcurrent protection or status monitoring). During generation, the edge controller uses standardized naming rules to ensure conflict-free naming of different devices and measurement points, thereby guaranteeing semantic consistency across layers. Upon completion, the initial version of the semantic manifest is stored as the system's data description baseline.

[0060] When in use, standardized semantic descriptions are generated based on registration information to eliminate data semantic differences between different devices; the edge controller integrates the registration information in the relationship table into a table, recording the device identifier, measurement point identifier, measurement type, hierarchical source and functional description one by one, ensuring the integrity and consistency of each record; the generated semantic list provides a standardized reference for the system's data exchange and analysis, improving the understandability and compatibility of the data.

[0061] After the initial semantic manifest is generated, the edge controller stores it in the version control database and marks it as the initial version. For version management, each semantic manifest record includes a timestamp and version number, identified by the semantic manifest version number plus a timestamp, such as "semantic manifest_v1_timestamp_v1". When device firmware is updated or configuration changes occur, the edge controller re-executes the parsing and registration process based on the new initial report frame, generating an updated semantic manifest and recording it as the new version, while retaining records of previous versions. The version control database maintains the change history of all versions, ensuring the system can track the evolution of the semantic manifest.

[0062] When in use, a version control mechanism manages the change history of the semantic manifest, ensuring data traceability and consistency. The edge controller assigns a timestamp and version number to each semantic manifest, stores them in the version control database, and updates the version record with each change, preserving historical data; this enhances the maintainability of the system and ensures the accuracy and reliability of the semantic manifest in long-term operation.

[0063] In a modular, double-layer prefabricated substation system, the specific processing logic of semantic handshake is achieved through five steps: device access and first report frame monitoring, first report frame parsing based on the IEC61850 standard, ternary registration, semantic list generation, and version control. This enables automatic identification, data parsing, and unified semantic management of newly accessed enclosures, thereby solving the problem of inconsistent data semantics across enclosures after hoisting. It ensures the standardization and traceability of system data and provides efficient and reliable technical support for the operation of the modular substation system.

[0064] Step 2: The edge gateway extracts the measurement point identifier, measurement type, and hierarchical source from the initial semantic list generated in Step 1, and ensures data semantic consistency based on the initial semantic list; the edge gateway inserts a high-stability timestamp into each received data packet to record the precise time of data packet reception;

[0065] The edge gateway converts data packets from the Modbus protocol to the Message Queuing Telemetry Transport Protocol, generates topics and payloads, and achieves data standardization and efficient transmission. The edge gateway sorts data packets according to high-stability timestamps, generates a consistent time-series stream across layers, and pushes it to the cloud digital twin platform in real time. This solves the problem of data stream drift on the time reference across layers, improves the real-time performance and accuracy of the system, and provides a reliable data foundation for subsequent steps.

[0066] Step two includes the following:

[0067] Step 201, Semantic List Reference

[0068] The edge gateway first extracts key information from the initial semantic manifest generated in step one.

[0069] The initial semantic manifest contains measurement point identifiers, measurement types, and hierarchical sources. This information is used to identify and process heterogeneous data from upper-layer primary devices and lower-layer secondary and auxiliary devices. The edge gateway queries the initial semantic manifest to identify measurement point identifiers in data packets and associates them with their corresponding measurement types and hierarchical sources. The edge gateway loads the initial semantic manifest, extracts the measurement point identifiers, measurement types, and hierarchical sources, and stores them as local data structures for easy and rapid querying.

[0070] The first version of the semantic manifest provides standardized data descriptions, ensuring that edge gateways accurately understand the semantics of cross-layer data. As a unified data reference, the first version of the semantic manifest can eliminate semantic differences in data from different levels of devices, avoiding data misunderstandings or processing errors. By providing clear semantic context, it ensures that edge gateways can correctly parse and utilize data in subsequent processing, enhancing system compatibility and stability.

[0071] Step 202: Inserting a Highly Stable Timestamp

[0072] The edge gateway inserts a highly stable timestamp for each received data packet to record the precise moment the data packet was received.

[0073] The high-stability timestamp originates from the clock source built into the edge gateway, which maintains microsecond-level time accuracy by synchronizing with an external high-precision clock.

[0074] The processing steps include: the edge gateway parsing the data packet and extracting the measurement point identifier; querying the initial semantic list based on the measurement point identifier to obtain the corresponding measurement type and hierarchical source; and then adding the high-stability timestamp as an additional field to the data packet, with the timestamp format being seconds plus microsecond precision.

[0075] After receiving data packets, the edge gateway parses their measurement point identifiers, queries the initial semantic manifest, generates a highly stable timestamp, and updates the data packet structure. By inserting the highly stable timestamp, it achieves precise marking and alignment of data packets in the time dimension. The highly stable timestamp solves the inconsistency problem caused by time base drift in cross-layer data streams, ensuring the time traceability of all data.

[0076] Step 203, Protocol Mapping

[0077] The edge gateway converts data packets from the Modebest protocol to the Message Queuing Telemetry Transport protocol to achieve data standardization and efficient transmission. The conversion process includes: mapping register addresses in the Modebest protocol to topics in the Message Queuing Telemetry Transport protocol, with the topics consisting of a combination of device identifiers and measurement point identifiers; and converting data values ​​in the Modebest protocol into a JSON-formatted payload, which contains the measurement values ​​and a highly stable timestamp.

[0078] The edge gateway parses the Modebs data packets, generates corresponding topics and payloads, encapsulates them into message queue telemetry transmission messages, and publishes them to a pre-configured message queue. Through protocol mapping, heterogeneous data is converted into a unified format, which facilitates cross-system transmission and processing. The message queue telemetry transmission protocol is lightweight and has high real-time performance, which can meet the data transmission needs in IoT scenarios.

[0079] Step 204: Time series stream output

[0080] The edge gateway sorts data packets from all measurement points based on high-stability timestamps, generating a consistent time-series stream across layers. Indexed by high-stability timestamps, the time-series stream includes measurement data from upper-layer primary devices and lower-layer secondary and auxiliary devices, ensuring data alignment in the time dimension. The edge gateway employs streaming processing technology to push the time-series stream to the cloud-based digital twin platform in real time. After receiving data packets, the edge gateway sorts them by high-stability timestamps, constructs the time-series stream, and continuously outputs it to the cloud. This time-series stream output ensures the temporal continuity and cross-layer consistency of the data.

[0081] When used, time-series streams provide a complete view of data organized in chronological order, facilitating state analysis and modeling on cloud platforms. The generated time-series data streams improve data quality and support accurate mapping and real-time monitoring of device status on cloud-based digital twin platforms.

[0082] Step 3: The edge gateway writes the cross-layer consistent time series stream into the semantic buffer according to the high-stability timestamp order, extracts the time series data from the semantic buffer, and distinguishes the primary power measurement data generated by the upper-layer primary equipment from the secondary operation data generated by the lower-layer secondary and auxiliary equipment based on the hierarchical source in the first version of the semantic list. For each pair of primary power measurement data and secondary operation data, the Pearson correlation coefficient is calculated to generate a multidimensional correlation matrix. The correlation pairs in the multidimensional correlation matrix and the correlation strength represented by the Pearson correlation coefficient are mapped to a dynamic graph. The dynamic graph is stored and updated in real time using graph database technology.

[0083] Step three includes the following:

[0084] Step 301: Write the time series stream to the semantic buffer.

[0085] The edge gateway writes cross-layer consistent time series streams into the semantic buffer in the order of highly stable timestamps.

[0086] The semantic buffer is a pre-configured high-speed read / write storage area specifically designed for temporarily storing time-series streaming data. The writing process requires data packets to be arranged in ascending order of timestamps to ensure that the data maintains temporal continuity and order during storage, so that subsequent steps can extract and process the data as needed.

[0087] By writing time-series stream data into the semantic buffer in an orderly manner, the integrity of the time order of the data is ensured during storage and subsequent processing. After receiving the time-series stream, the edge gateway first reads the high-stability timestamp of each data packet, then sorts the data packets according to the size of the timestamp, and finally stores the sorted data into the predetermined position of the semantic buffer.

[0088] The semantic buffer, as an intermediate storage area for data processing, provides efficient data access and retrieval capabilities. At the same time, the timestamp sorting ensures the temporal consistency of the data, enabling subsequent correlation matrix construction and dynamic graph mapping to be performed based on accurate time order, thereby improving the reliability and accuracy of the entire modeling process.

[0089] Step 302: Construct a multidimensional correlation matrix according to hierarchy.

[0090] After extracting time-series stream data from the semantic buffer, the data is divided into upper-layer primary equipment data and lower-layer secondary and auxiliary equipment data according to the hierarchical sources defined in the first version of the semantic manifest.

[0091] For each measurement point of the upper-level primary equipment and the lower-level secondary and auxiliary equipment, correlation pairs are established, and the correlation strength between them is calculated. The correlation strength is calculated using the Pearson correlation coefficient method. The specific process is as follows: extract the measurement value sequence of each correlation pair, calculate the covariance of the two sequences, then calculate the standard deviation of each sequence, and finally divide the covariance by the product of the two standard deviations to obtain the Pearson correlation coefficient, which serves as the quantitative value of the correlation strength. All correlation pairs and their correlation strengths form a multidimensional correlation matrix.

[0092] By quantifying the correlation strength between measurement points of primary equipment at the upper layer and measurement points of secondary and auxiliary equipment at the lower layer, the intrinsic connections between cross-layer data are revealed, providing a structured data foundation for subsequent dynamic graph mapping. Measurement points of primary equipment at the upper layer and secondary and auxiliary equipment at the lower layer are distinguished based on the initial semantic list. Corresponding measurement value sequences are extracted from the semantic buffer. For each correlation pair, the mean of the measurement value sequence is calculated. Based on the mean, the covariance and standard deviation are calculated. Finally, the covariance is divided by the product of the two standard deviations to obtain the Pearson correlation coefficient, which is then filled into the corresponding position in the multidimensional correlation matrix.

[0093] Step 303: Real-time mapping to dynamic graph

[0094] The association pairs and association strengths in the multidimensional association matrix are converted into a structured representation of the dynamic graph.

[0095] Specifically, the measurement points of the upper-level primary equipment and the measurement points of the lower-level secondary and auxiliary equipment are defined as nodes in the dynamic graph. Each association pair is defined as an edge connecting the two nodes, and the weight of the edge is determined by the corresponding association strength (i.e., the Pearson correlation coefficient). The dynamic graph is stored using graph database technology, which supports real-time updates and efficient queries of nodes, edges, and their weights. The transformation process is as follows: the two measurement points and their association strength of each association pair are read from the multidimensional association matrix, the measurement points are mapped to nodes, the association strength is mapped to the edge weights, and stored in the graph database. The state of the nodes and edges is adjusted in real time according to the updates of the time series stream.

[0096] Graph database technology is used to transform the relationships in a multidimensional correlation matrix into an intuitive topological structure, facilitating the visualization and analysis of complex connections between cross-layer data. Each correlation pair is extracted from the multidimensional correlation matrix, identifying the upper-layer primary equipment measurement points and the lower-layer secondary and auxiliary equipment measurement points within each pair. These measurement points are recorded as nodes in the graph database. Correlation pairs are then defined as edges connecting the two nodes, and the corresponding Pearson correlation coefficients are assigned as edge weights. Finally, the data structure of these nodes and edges is stored in the graph database, completing the construction of a dynamic graph.

[0097] Writing time-series data to a semantic buffer ensures temporal consistency across layers. A multi-dimensional correlation matrix is ​​constructed hierarchically, and data is extracted from the semantic buffer to quantify the correlation strength between measurement points of primary equipment in the upper layer and secondary and auxiliary equipment in the lower layer, forming a structured correlation representation. Finally, real-time mapping to a dynamic graph transforms the quantization results of the multi-dimensional correlation matrix into a graph structure, enabling dynamic visualization and real-time management of correlation relationships.

[0098] Step 4: Extract the actual correlation strength between the upper-level primary equipment measurement points and the lower-level secondary and auxiliary equipment measurement points from the dynamic spectrum, obtain the predicted correlation strength from the digital twin platform, and calculate the difference between the two to generate a residual vector;

[0099] A fast Fourier transform is applied to the residual vector to generate a frequency domain representation. After calculating the normalized power spectral density, the residual spectral entropy exponent is calculated using the Shannon entropy formula. High-stability timestamps of the upper-layer primary equipment measurement points and the lower-layer secondary and auxiliary equipment measurement points are extracted from the time series stream. Cross-layer timestamp differences are calculated, and the delay jitter exponent is calculated using the improved Allan variance. The residual spectral entropy exponent and the delay jitter exponent are input into the random forest model, and the track alignment reliability coefficient is output. When the track alignment reliability coefficient is lower than a preset threshold, a patch signal is triggered.

[0100] Step four includes the following:

[0101] Step 401: Residual Vector Generation

[0102] The actual correlation strength between the measuring points of the upper-level primary equipment and the measuring points of the lower-level secondary and auxiliary equipment is extracted from the dynamic spectrum generated in step three. This actual correlation strength is calculated using the Pearson correlation coefficient. The calculation process for the Pearson correlation coefficient is as follows:

[0103] The covariance of the time series data for each of the two measurement points is calculated, and then divided by the product of their respective standard deviations to obtain a value between -1 and 1. Simultaneously, the predicted correlation strength of the same measurement point pair is obtained from the cloud-based digital twin platform. This predicted correlation strength is generated by the digital twin model based on historical data and physical rules. For each pair of measurement points, the difference between the actual correlation strength and the predicted correlation strength is calculated, and this difference is defined as the residual. The residuals of all measurement point pairs are arranged in order of the measurement point pairs to form a residual vector.

[0104] By calculating the difference between the actual and predicted correlation strengths, the degree of deviation between the digital twin model and the actual equipment state in terms of correlation is quantified, providing data for subsequent drift detection. For each pair of measurement points, the actual and predicted correlation strengths are obtained, and the difference between them is calculated as the residual. Then, the residuals of all measurement point pairs are sequentially combined into a residual vector. The residual vector can intuitively reflect the deviation between the digital twin model and the actual equipment state in terms of correlation. The generation of the residual vector provides a quantitative description of the degree of drift, facilitating automated detection and processing by the system.

[0105] Step 402: Calculation of Residual Spectral Entropy Index

[0106] Perform a Fast Fourier Transform on the residual vector to convert it from the time domain to the frequency domain, thus obtaining its frequency domain representation.

[0107] The Fast Fourier Transform (FFT) process involves decomposing the residual vector into a superposition of sine and cosine waves, calculating the amplitude and phase of each frequency component, and generating a frequency domain representation. The power spectral density is then calculated based on this frequency domain representation. The power spectral density is calculated by squaring the amplitude of each frequency component to obtain the power value.

[0108] The power spectral density is normalized by dividing each power value by the sum of the power values ​​to obtain the normalized power spectral density components. The entropy value of each normalized power spectral density component is calculated using the Shannon entropy formula, defined as the residual spectral entropy exponent. The Shannon entropy calculation process is as follows: take the logarithm to base 2 for each component in the normalized power spectral density, multiply it by the component itself, sum all the products, and take the negative value to obtain the residual spectral entropy exponent.

[0109] The residual spectral entropy index is used to evaluate the complexity of the residual vector in the frequency domain, reflecting the dynamic characteristics of the state drift between the digital twin model and the actual device. First, a fast Fourier transform is applied to the residual vector to generate a frequency domain representation. Then, the power spectral density of the frequency domain representation is calculated and normalized. Finally, the entropy value is calculated according to the Shannon entropy formula to obtain the residual spectral entropy index.

[0110] When used, the residual spectral entropy index, as a dimensionless value, is easy to compare and integrate with other indicators, thereby improving the comprehensiveness and accuracy of drift detection.

[0111] Step 402: Calculation of latency jitter index

[0112] Extract the high-stability timestamps from the primary equipment measurement points and the secondary and auxiliary equipment measurement points in the time series stream generated in step two. These high-stability timestamps are time stamps calibrated using time synchronization technology. Calculate the timestamp difference between the primary equipment measurement points and the secondary and auxiliary equipment measurement points in the lower layer to generate a cross-layer timestamp difference sequence. Apply an improved Allan variance calculation method to the cross-layer timestamp difference sequence to obtain the delay jitter index.

[0113] The improved calculation process of Allan variance is as follows: Select a fixed average time window length, slide the window on the cross-layer timestamp difference sequence, calculate the integral of the difference value in each window; then calculate the difference between the integral values ​​of adjacent windows, sum the squares of these differences and take the average value to obtain the time delay jitter index.

[0114] The latency jitter index is used to evaluate the stability of cross-layer data streams on a time base, detect latency fluctuations in time series streams, extract high-stability timestamps from measurement points of primary equipment in the upper layer and measurement points of secondary and auxiliary equipment in the lower layer, calculate the timestamp difference between the two, and generate a cross-layer timestamp difference sequence. Then, the improved Allan variance formula is applied to calculate the variance of the difference sequence under different time windows to obtain the latency jitter index.

[0115] When used, the time delay jitter index, as a dimensionless value, quantifies the time base drift of the time series stream, providing a time-dimensional evaluation basis for drift detection.

[0116] Step 403: Calculation of track reliability coefficient

[0117] The residual spectral entropy index and the time delay jitter index are used as input features and fed into a pre-trained random forest model.

[0118] The random forest model consists of multiple decision trees, each of which makes independent predictions based on the input features. The model calculates a dimensionless value between 0 and 1 by majority voting or averaging the predictions of all decision trees. This value is defined as the alignment confidence coefficient, which indicates the degree of alignment between the digital twin model and the actual device state. The larger the value, the higher the degree of alignment.

[0119] A random forest model is used to comprehensively evaluate the residual spectral entropy index and the time delay jitter index, generating an alignment confidence coefficient to automatically determine the degree of drift. The residual spectral entropy index and the time delay jitter index are input into the random forest model, which calculates and outputs the alignment confidence coefficient by integrating the prediction results of multiple decision trees.

[0120] When in use, the calculation of the track reliability coefficient provides a comprehensive quantitative indicator of the degree of drift, which facilitates the system to automatically trigger subsequent calibration operations based on the coefficient, ensuring the continuous consistency between the digital twin model and the actual equipment status.

[0121] Step 404, Patch Signal Trigger

[0122] A threshold for track reliability coefficient is pre-set, which is determined based on system requirements and historical data analysis.

[0123] When the track alignment confidence coefficient output by the random forest model is lower than a preset threshold, a patch signal is generated and sent to step five to notify the execution of the difference patch operation. The difference patch operation aims to calibrate the deviation between the digital twin model and the actual device state.

[0124] The alignment reliability coefficient is determined by a preset threshold, and the patching process is automatically triggered to ensure timely calibration between the digital twin model and the actual equipment status. The alignment reliability coefficient output by the random forest model is compared with the preset threshold. If the alignment reliability coefficient is less than the threshold, a patch signal is generated and sent to step five.

[0125] When in use, the patch signal triggering mechanism ensures that the digital twin model and the actual device state are continuously aligned, improving the system's adaptability and long-term reliability.

[0126] Step 5: The edge controller continuously monitors the patch signal generated in Step 4. When the patch signal is 1, the edge controller initiates the version patching process, collects the device's firmware fingerprint and the model version identifier obtained from the cloud digital twin platform, packages the firmware fingerprint and model version identifier into a version information package, and submits it to the GitOps pipeline. After receiving the version information package, the GitOps pipeline compares the firmware fingerprint with the latest firmware fingerprint in the Git repository and the model version identifier with the latest model version identifier, generating a firmware difference patch and a model difference patch. The GitOps pipeline submits the firmware difference patch and model difference patch for approval. After approval, they are integrated into the final difference patch. The GitOps pipeline pushes the final difference patch back to the edge controller. The edge controller applies the firmware difference patch to update the device firmware and applies the model difference patch to update the digital twin model, records the new firmware fingerprint and model version identifier, and simultaneously notifies the cloud digital twin platform to update synchronously.

[0127] Step five includes the following:

[0128] Step 501: Trigger signal reception

[0129] The edge controller continuously monitors the patch signal generated in step four. The patch signal is a binary value, where 1 indicates that the version patching process needs to be initiated, and 0 indicates that the version patching process does not need to be initiated. The edge controller makes a judgment by periodically reading the value of the patch signal: when the patch signal is detected as 1, the subsequent version patching operation is executed immediately; when the patch signal is 0, the monitoring state is maintained without executing any operation.

[0130] The edge controller automates the version patching process by monitoring patch signals in real time, ensuring timely calibration when the digital twin model and the actual device state drift apart. The edge controller reads the patch signal value at fixed time intervals and determines whether to start the version patching process based on the reading result: if the value is 1, the process is started; if the value is 0, monitoring continues.

[0131] In practice, the patch signal, as the output of the drift detection in step four, accurately reflects the synchronization status between the digital twin model and the actual device state. By using it as a trigger condition, the version patching process can be ensured to start only when necessary, thus avoiding unnecessary resource consumption. The automated triggering mechanism improves the system's response speed and adaptability, ensuring that the digital twin model and the actual device state can be aligned in a timely manner.

[0132] Step 502: Firmware fingerprint and model version submission

[0133] The edge controller first collects the current firmware fingerprint of the device. The firmware fingerprint is a fixed-length string generated by hashing the firmware file content and is used to uniquely identify the firmware status of the device.

[0134] The hashing process is as follows: the complete data of the firmware file is input into a hash function, and a fixed-length unique string is output as the firmware fingerprint. Simultaneously, the edge controller obtains the version identifier of the current digital twin model from the cloud-based digital twin platform; this version identifier serves as a unique number for the digital twin model. The edge controller packages the firmware fingerprint and model version identifier into a version information packet. The version information packet is structured as a set of key-value pairs containing the firmware fingerprint and model version identifier. This version information packet is submitted to the GitOps pipeline via a secure communication channel.

[0135] The edge controller provides the GitOps pipeline with the accurate information needed for version comparison by submitting firmware fingerprints and model version identifiers, ensuring the correctness of subsequent patch generation.

[0136] In use, the edge controller performs a hash operation on the firmware file to generate a firmware fingerprint, reads the model version identifier from the cloud-based digital twin platform, combines the two into a version information packet, and sends it to the GitOps pipeline via an encrypted communication protocol. The submission of the version information packet provides the GitOps pipeline with complete version data, ensuring that patch generation is based on the latest and most accurate information, thereby improving the reliability and accuracy of version synchronization.

[0137] Step 503, GitOps Pipeline Processing

[0138] After receiving the version information packet, the GitOps pipeline performs a version comparison operation.

[0139] The specific process is as follows: The GitOps pipeline extracts the firmware fingerprint and model version identifier from the version information packet, and then compares them with the latest firmware fingerprint and the latest model version identifier stored in the Git repository, respectively. The comparison method is string matching: the current firmware fingerprint is compared character by character with the latest firmware fingerprint; if an inconsistency is found, a firmware difference patch is generated; the current model version identifier is compared character by character with the latest model version identifier; if an inconsistency is found, a model difference patch is generated. The firmware difference patch contains the incremental update content of the firmware from the current version to the latest version, and the model difference patch contains the incremental update content of the digital twin model from the current version to the latest version.

[0140] GitOps pipelines identify differences between firmware and digital twin models through version comparison and generate corresponding patches to ensure that the patch content accurately reflects version changes. The GitOps pipeline extracts the firmware fingerprint and model version identifier from the version information package, compares them with the latest version information in the Git repository, and if differences are found, calls version control tools to generate corresponding firmware and model difference patches.

[0141] When used, GitOps pipelines improve the efficiency and accuracy of version management, ensuring that patch content is completely consistent with actual version changes and avoiding potential errors caused by human intervention.

[0142] Step 504: Approval and Patch Generation

[0143] The GitOps pipeline submits the generated firmware and model difference patches to the approval process. The approval process verifies the security and compliance of the patches based on pre-configured automated rules or manual review. Automated rules include steps such as checking the completeness of the patch content and verifying version compatibility; manual review involves operations personnel reviewing the patch content to ensure it meets system requirements. Once approved, the GitOps pipeline integrates the firmware and model difference patches into a final difference patch, which contains complete update data for both the firmware and the digital twin model.

[0144] The approval process verifies the security and compliance of patches, preventing unauthorized or insecure version changes from entering the system. The GitOps pipeline submits firmware and model difference patches to the approval system, which executes predefined automated rules or accepts human review results. Once approved, the GitOps pipeline integrates the two into the final difference patch.

[0145] When in use, the approval and patch generation mechanism enhances the controllability of version updates, ensuring that patch content is rigorously verified, thereby guaranteeing the stability and security of the system.

[0146] Step 505: Patch Pushback and Application

[0147] The GitOps pipeline pushes the final difference patch back to the edge controller via a secure communication channel. Upon receiving the final difference patch, the edge controller first applies the firmware difference patch, upgrading the device firmware to the latest version by updating the firmware file content; then it applies the model difference patch, upgrading the digital twin model to the latest version by updating the model data. After the update is complete, the edge controller re-hashes the updated firmware file to generate a new firmware fingerprint, and simultaneously records the updated model version identifier, storing both locally. The edge controller then notifies the cloud-based digital twin platform via a communication interface to synchronously update the new firmware fingerprint and model version identifier, ensuring consistency between cloud data and the edge device.

[0148] By using patch pushback and application, the device firmware and digital twin model are updated synchronously, ensuring consistency between the virtual model and the actual device state. The edge controller receives the final difference patch, sequentially executes firmware update and model update operations, generates a new firmware fingerprint for the updated firmware, records the new model version identifier, and notifies the cloud-based digital twin platform of the results for synchronization. The patch pushback and application mechanism enables automatic synchronization updates of the device and model, improving the system's adaptability and long-term operational reliability.

[0149] Step 6: The edge controller receives the difference patch pushed back from the GitOps pipeline, updates the device firmware and digital twin model, generates a new firmware fingerprint and model version identifier, and records them.

[0150] The edge controller recalculates the correlation strength in the multidimensional correlation matrix based on the updated digital twin model, generating a matrix reflecting the relationships between devices. The edge controller packages the new firmware fingerprint, the new model version identifier, and the refreshed multidimensional correlation matrix into a state change package, which is then broadcast to the cloud microservice through a secure communication channel. The cloud microservice receives the state change package, updates the firmware fingerprint and model version identifier, adjusts the protection algorithm parameters and lifetime prediction model parameters according to the refreshed multidimensional correlation matrix, recalculates the protection settings and lifetime curves, and sends them back to the edge controller.

[0151] Step six includes the following:

[0152] Step 601: Patch deployment processing

[0153] After receiving the difference patch pushed back through the GitOps pipeline, the edge controller begins to execute the patch application operation.

[0154] First, the edge controller decompresses the difference patch, separating it into a firmware difference patch and a model difference patch. Next, the edge controller applies the firmware difference patch to the device firmware, updating it to the latest state by comparing the current firmware version with the patch content item by item. Then, the edge controller applies the model difference patch to the digital twin model, similarly updating the digital twin model to the latest state by comparing the current model version with the patch content. After the update is complete, the edge controller generates a new firmware fingerprint based on the updated firmware and a new model version identifier based on the updated model, recording both locally for subsequent version management and verification.

[0155] The edge controller achieves synchronous updates of the device firmware and digital twin model by decompressing the difference patch and applying the firmware difference patch and model difference patch respectively, while generating and recording the new firmware fingerprint and model version identifier.

[0156] The difference patch contains updates to both the firmware and the digital twin model. By applying these two patches separately, it's possible to ensure that the device's operational status and the digital twin model's status remain consistent after the update. Generating and recording new firmware fingerprints and model version identifiers provides a basis for subsequent version tracking and consistency verification.

[0157] When in use, it enables precise updates to firmware and models, avoiding state deviations caused by incomplete updates. At the same time, through fingerprint and version identification records, it can provide a traceable update history, improving the reliability and efficiency of maintenance.

[0158] Step 602: Refreshing the correlation matrix

[0159] The edge controller recalculates the multidimensional correlation matrix based on the updated digital twin model.

[0160] First, the edge controller extracts the correlation strength between the measurement points of the upper-layer primary equipment and the measurement points of the lower-layer secondary and auxiliary equipment from the updated digital twin model. The extraction of correlation strength is based on the physical relationships and data dependencies between measurement points defined in the model, and the specific value is determined by analyzing the degree of mutual influence between measurement point data.

[0161] Subsequently, the edge controller writes the extracted association strengths one by one into the multidimensional association matrix, updating the value of each element in the matrix accordingly. After the update is complete, the multidimensional association matrix reflects the latest topology and data association status of the devices after the update, which is used for subsequent status change broadcasting and cloud analysis.

[0162] The edge controller extracts the correlation strength between the measurement points of the upper-layer primary devices and the measurement points of the lower-layer secondary and auxiliary devices based on the updated digital twin model, and updates the multidimensional correlation matrix accordingly, generating a matrix reflecting the latest device relationships. The updated multidimensional correlation matrix accurately describes the correlation relationships between the updated devices, providing reliable data support for state change broadcasting and cloud computing, and improving the accuracy of the digital twin model and the credibility of the analysis results.

[0163] Step 603: Broadcasting the state change

[0164] The edge controller integrates the new firmware fingerprint, the new model version identifier, and the refreshed multidimensional association matrix into a state change package, which is used to deliver update information to cloud microservices.

[0165] Specifically, the edge controller packages the new firmware fingerprint, the new model version identifier, and the multidimensional association matrix into a complete data structure according to a predefined format. After packaging, the edge controller sends the state change package to the cloud microservice through a secure communication channel, ensuring the integrity and security of the data during transmission. Upon receiving the state change package, the cloud microservice can update its own data and model based on its contents.

[0166] The edge controller packages the new firmware fingerprint, the new model version identifier, and the updated multidimensional association matrix into a state change package, which is then broadcast to the cloud microservices via a secure communication channel. This state change package integrates the latest state information of the device and model, and its broadcast transmission to the cloud ensures data synchronization between the cloud microservices and the edge devices. The use of a secure communication channel guarantees the reliability and confidentiality of data transmission. This enables efficient information sharing between the edge and the cloud, ensuring that the cloud microservices can promptly obtain updated state data, thereby supporting subsequent calculations and analysis, and improving the overall coordination and responsiveness of the system.

[0167] Step 604: Recalculation of protection settings and life curves

[0168] After receiving the state change packet, the cloud-based microservice performs update and recalculation operations.

[0169] First, the cloud microservice replaces the current firmware fingerprint and model version identifier with the new firmware fingerprint and model version identifier from the state change package, completing version synchronization. Next, the cloud microservice recalculates the protection settings and lifetime curves based on the updated multidimensional correlation matrix in the state change package. For the recalculation of protection settings, the cloud microservice analyzes the correlation strength in the multidimensional correlation matrix and adjusts the parameters of the protection algorithm to adapt to the updated device relationships. For the recalculation of lifetime curves, the cloud microservice similarly updates the parameters of the device lifetime prediction model based on the correlation strength in the multidimensional correlation matrix, generating new lifetime prediction results. After recalculation, the cloud microservice distributes the new protection settings and lifetime curves to the edge controller for field control and status monitoring.

[0170] The cloud-based microservice updates the firmware fingerprint and model version identifier based on the state change package, adjusts the protection algorithm parameters and lifetime prediction model parameters based on the refreshed multidimensional correlation matrix, recalculates the protection settings and lifetime curves, and then distributes the results to the edge controller. Device updates may affect the accuracy of protection logic and lifetime prediction. By recalculating the protection settings and lifetime curves based on the multidimensional correlation matrix, it is ensured that the calculation results reflect the latest device status, improving the accuracy of control and prediction.

[0171] When in use, the recalculated protection settings and life curves are consistent with the updated equipment status, which improves the reliability of the system protection function and the accuracy of life assessment, providing precise technical support for on-site control and long-term operation and maintenance.

[0172] This system updates devices and models through patch implementation, generates the latest data relationships by refreshing the correlation matrix, achieves synchronization between the edge and cloud through state change broadcasting, and optimizes control and prediction parameters by recalculating protection settings and life curves. It effectively solves the problem of discrepancies between the digital twin model and the actual equipment state caused by firmware updates and configuration changes, ensuring continuous consistency between field control and twin analysis. This provides high-precision data support and technical assurance for remote monitoring, condition assessment, and predictive maintenance of modular, double-layer prefabricated substation systems.

[0173] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0174] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0175] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0176] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0177] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A modular, double-layer prefabricated substation system based on digital twins, characterized in that: include, The semantic handshake unit, the edge controller listens to the first report frame of each enclosure, calls the IEC61850 logic node template to complete the three-element registration of equipment, measurement points and functions, generates the first version of the semantic list and marks the hierarchical source; The time-based fixed unit, the edge gateway inserts a highly stable timestamp according to the semantic list, and completes the telemetry from Modebas to the message queue. transmission Protocol mapping outputs a consistent time series stream across layers; After the time series stream is written into the semantic buffer, the system constructs a multi-dimensional correlation matrix according to the hierarchy of the correlation modeling unit, and maps the primary power quantity and secondary operating condition quantity into a dynamic spectrum in real time. The drift discrimination unit generates a residual vector and simultaneously calculates the residual spectral entropy index and the time delay jitter index. When the track reliability coefficient output by the random forest model is lower than the preset threshold, a patch signal is triggered. The version patch unit receives a trigger signal, submits the current firmware fingerprint and twin model version to the GitOps pipeline, generates a difference patch after approval, and pushes it back to the edge. The strategy refresh unit refreshes the correlation matrix and broadcasts the state change to the cloud microservices as soon as the patch is applied. The engine then recalculates the protection setpoints and life curves accordingly, keeping on-site control and twin analysis synchronized. The edge gateway writes the cross-layer consistent time series stream into the semantic buffer in high-stability timestamp order, extracts the time series data from the semantic buffer, and distinguishes the primary power measurement data generated by the upper-layer primary equipment from the secondary operation data generated by the lower-layer secondary and auxiliary equipment based on the hierarchical source in the first version of the semantic list. For each pair of primary power measurement data and secondary operation data, the Pearson correlation coefficient is calculated to generate a multidimensional correlation matrix. The correlation pairs in the multidimensional correlation matrix and the correlation strength represented by the Pearson correlation coefficient are mapped to a dynamic graph. The dynamic graph is stored in a graph database and updated in real time.

2. The modular double-layer prefabricated substation system based on digital twin as described in claim 1, characterized in that: The edge controller listens to the first report frame of a newly connected enclosure and receives data packets, verifies the integrity of the data packets and records the reception time, calls the IEC61850 logic node template to parse the first report frame, extracts device information, measurement point information and function information, and generates a structured dataset containing device identifiers, measurement points and functions.

3. The modular double-layer prefabricated substation system based on digital twin as described in claim 2, characterized in that: Perform equipment registration, measurement point registration, and function registration, registering equipment identifiers, measurement point identifiers, measurement types, hierarchical sources, and function descriptions to the database and generating a relational table; Generate an initial semantic manifest based on the relation table, recording device identifiers, measurement point identifiers, measurement types, hierarchical sources, and functional descriptions.

4. The modular double-layer prefabricated substation system based on digital twin as described in claim 3, characterized in that: The edge gateway extracts the measurement point identifier, measurement type, and hierarchical source from the initial semantic list, and ensures data semantic consistency based on the initial semantic list. The edge gateway inserts a highly stable timestamp for each received data packet to record the precise moment the data packet was received.

5. The modular double-layer prefabricated substation system based on digital twin as described in claim 4, characterized in that: The edge gateway converts data packets from the Modbus protocol to the Message Queuing Telemetry Transport protocol, generating topics and payloads; The edge gateway sorts data packets according to high-stability timestamps, generates a consistent time-series stream across layers, and pushes it to the cloud-based digital twin platform in real time.

6. The modular double-layer prefabricated substation system based on digital twin as described in claim 1, characterized in that: The actual correlation strength between the upper-level primary equipment measurement points and the lower-level secondary and auxiliary equipment measurement points is extracted from the dynamic spectrum, and the predicted correlation strength is obtained from the digital twin platform. The difference between the two is calculated to generate a residual vector. Apply the Fast Fourier Transform to the residual vector to generate a frequency domain representation. After calculating the normalized power spectral density, use the Shannon entropy formula to calculate the residual spectral entropy exponent.

7. The modular double-layer prefabricated substation system based on digital twin as described in claim 6, characterized in that: Extract high-stability timestamps from the measurement points of the primary equipment in the upper layer and the measurement points of the secondary and auxiliary equipment in the lower layer from the time series stream, calculate the cross-layer timestamp difference, and apply the improved Allan variance to calculate the delay jitter index; The residual spectral entropy index and the time delay jitter index are input into the random forest model, and the track alignment confidence coefficient is output. When the track alignment confidence coefficient is lower than the preset threshold, a patch signal is triggered.

8. The modular double-layer prefabricated substation system based on digital twin as described in claim 7, characterized in that: The edge controller continuously monitors the patch signal. When the patch signal is 1, the edge controller starts the version patching process, collects the device's firmware fingerprint and the model version identifier obtained from the cloud digital twin platform, packages the firmware fingerprint and model version identifier into a version information package and submits it to the GitOps pipeline.

9. The modular double-layer prefabricated substation system based on digital twin as described in claim 8, characterized in that: After receiving the version information packet, the GitOps pipeline compares the firmware fingerprint with the latest firmware fingerprint in the Git repository and the model version identifier with the latest model version identifier to generate firmware difference patches and model difference patches. The GitOps pipeline submits firmware and model difference patches for approval, and after approval, integrates them into the final difference patch.

10. The modular double-layer prefabricated substation system based on digital twin as described in claim 9, characterized in that: The GitOps pipeline pushes the final difference patch back to the edge controller; the edge controller applies the firmware difference patch to update the device firmware and applies the model difference patch to update the digital twin model, records the new firmware fingerprint and model version identifier, and notifies the cloud digital twin platform to update synchronously.

11. The modular double-layer prefabricated substation system based on digital twin as described in claim 10, characterized in that: The edge controller receives differential patches pushed back from the GitOps pipeline, updates the device firmware and digital twin model, generates new firmware fingerprints and model version identifiers, and records them. The edge controller recalculates the correlation strength in the multidimensional correlation matrix based on the updated digital twin model, generating a matrix that reflects the relationships between devices.

12. The modular double-layer prefabricated substation system based on digital twin as described in claim 11, characterized in that: The edge controller packages the new firmware fingerprint, the new model version identifier, and the refreshed multidimensional association matrix into a state change package, and broadcasts it to the cloud microservices through a secure communication channel. The cloud-based microservice receives the status change packet, updates the firmware fingerprint and model version identifier, adjusts the protection algorithm parameters and lifetime prediction model parameters according to the refreshed multidimensional correlation matrix, recalculates the protection setpoint and lifetime curve, and sends them to the edge controller.

Images