Method and system for cross-professional report data sharing matching based on ontology mapping

Abstract

This invention discloses a cross-disciplinary report data sharing and matching method and system based on ontology mapping, belonging to the field of power system data processing technology. This invention deeply integrates data anti-tampering and loss recovery mechanisms with the real-time power flow calculation unique to the power grid. While completing the cross-disciplinary data record sequence completion, it automatically triggers underlying physical reversal tracking. Then, when tracing affected distribution paths, it accurately identifies power flow reversals and automatically generates reverse power receiving tags for relevant nodes. Business data containing directional dimension features is accurately synchronized to the marketing asset system through an ontology mapping dictionary. This not only solves the pain point of abnormal line loss calculations caused by business systems relying on fixed incorrect topologies, achieving adaptive consistency between the line loss model and the actual power flow of the physical power grid, but also greatly reduces the workload of manual verification caused by changes in power grid operation modes, improving the high-concurrency processing efficiency and intelligence level of data traceability.

Description

Technical Field

[0001] This invention belongs to the field of power system data processing technology, and specifically relates to a cross-disciplinary report data sharing and matching method and system based on ontology mapping. Background Technology

[0002] Distribution networks often employ a daisy-chain ring network power supply architecture to ensure the reliability of regional power supply. Under this architecture, fault tripping or load transfer operations can easily trigger dynamic changes in the local topology of the power grid, resulting in loop-closing or loop-breaking operation conditions. Accompanying these operations, the information system backend generates a large number of state change records and remote signaling evolution sequences. The IT infrastructure periodically scans data log files and performs multi-copy comparison and mapping recovery for missing data to maintain the integrity and consistency of power grid business data. In existing cross-disciplinary data sharing and recovery solutions for the power grid, on the one hand, simple disaster recovery technologies often only focus on the literal verification and compensation of underlying database fields. Their recovery results merely reflect the mechanical repair of logical states such as switch opening or closing, completely detached from the power flow transfer and physical attributes of the underlying power grid, making it difficult for the recovered data to support complex business applications. On the other hand, when the power grid experiences loop-breaking or loop-closing power transfer leading to a large directional shift in physical power flow, the existing marketing system, after receiving cross-disciplinary synchronized change data, cannot automatically detect the reversal of the underlying current direction. This not only makes it difficult for the marketing system to adaptively complete the dynamic topology reconstruction of the line loss calculation model, but also causes serious calculation deviations or even negative line losses in the metering data read by smart meters under the original rigid formula, which seriously restricts the efficiency of promoting lean line loss management in the same period. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides a cross-disciplinary report data sharing and matching method and system based on ontology mapping to solve the aforementioned technical problems.

[0004] The cross-disciplinary report data sharing and matching method based on ontology mapping includes the following steps:

[0005] S1: Scan the real-time remote signaling data files and topology node files in the power grid operation system, extract data change record points, and construct the first record sequence reflecting the status of primary power equipment and the connection status of topology nodes;

[0006] S2: Traverse the first record sequence and calculate the timestamp difference between adjacent data change record points. If the timestamp difference exceeds a preset time interval threshold, it is determined that data is missing. Extract the key identifier of the missing data. The key identifier includes at least the device ID and the missing time window.

[0007] S3: For the key identifier, retrieve the corresponding backup data from multiple heterogeneous backup data sources, extract candidate compensation data for the corresponding storage location from the multiple backup data, and analyze the field integrity rate of the candidate compensation data;

[0008] S4: If the field completeness rate reaches the preset completeness threshold, then based on the pre-built cross-professional association mapping dictionary, the ontology features of the candidate compensation data are aligned and mapped with the ontology features of the first record sequence, and the mapped candidate compensation data is spliced ​​into the missing time window to generate the completed second record sequence.

[0009] S5: Extract data features at the splicing boundary, check for smoothness and logical consistency, calculate the matching score, and when the matching score reaches the set standard, confirm that the second record sequence is valid and output it as the cross-professional data recovery result.

[0010] Preferably, before scanning the real-time telemetry data files and topology node files in the power grid operation system and extracting data change record points, a telemetry anti-jitter preprocessing step is also included, as shown below:

[0011] Obtain the mechanical anti-jitter time window of the target circuit breaker device;

[0012] If multiple remote signaling status reversal records of the same circuit breaker device are received within the mechanical anti-shake time window, the telemetry current change of the circuit breaker device in the corresponding time period is extracted and a blocking verification is performed.

[0013] False state reversal record points that are not accompanied by corresponding telemetry current abrupt changes are removed, and the remaining true state reversal record points are used as data change record points to construct the first record sequence.

[0014] Preferably, after outputting the cross-disciplinary data recovery results, a dynamic trend direction identification and matching step is also included, as shown below:

[0015] When the cross-disciplinary data recovery results reflect that the state change of the displacement device causes the power grid topology to close or open loop, the power grid topology reconstruction and real-time power flow transfer calculation are triggered.

[0016] Based on the real-time power flow transfer calculation results, the affected power distribution paths that experienced power flow reversal are traced in the completed second record sequence;

[0017] For primary power equipment that was originally in a passive or forward-receiving state on the affected power distribution path, if it is determined that its power flow direction has reversed, a reverse power receiving state label will be added to the primary power equipment in the cross-professional data recovery result.

[0018] After converting the data format of the reverse power receiving status label based on the cross-professional association mapping dictionary, the cross-professional data recovery result with the reverse power receiving status label is synchronously sent to the power grid dispatch system and the marketing asset system.

[0019] Preferably, the process of triggering grid topology reconfiguration and real-time power flow transfer calculation also includes a cross-voltage level pruning step, as shown below:

[0020] Extract the winding connection group characteristics of the transformer equipment on the affected power distribution path;

[0021] Based on the isolation / transmission characteristics of the zero-sequence current and phase according to the winding connection group characteristics, determine whether the power flow transfer penetrates to the low-voltage side topology across voltage levels;

[0022] If yes, the original tracing range is maintained; otherwise, the tracing range of the affected power distribution path is subject to boundary trimming across voltage levels.

[0023] Preferably, after synchronously sending the data to the power grid dispatching system and the marketing asset system, the following steps are also included:

[0024] A line loss reconstruction instruction is issued to the marketing asset system so that the marketing asset system reverses the calculation logic of the power outflow and inflow of primary power equipment with reverse power receiving status label, and regenerates and updates the line loss report of the primary power equipment.

[0025] Preferably, if the field integrity rate does not reach the integrity threshold, the backup data is determined to be invalid, and the missing data causal deduction recovery step is executed, as shown below:

[0026] Extract the primary power equipment corresponding to the missing data, obtain the relay protection action report associated with the primary power equipment in its upstream topology, and the zero-sequence current telemetry waveform report associated with its downstream topology.

[0027] Based on the preset relay protection logic and the short-circuit current characteristics of the zero-sequence current telemetry recording report, the actual switching state of the primary power equipment in the missing time window is calculated by reverse deduction.

[0028] After converting the actual separation and combination state into ontological features, they are then added to the second record sequence.

[0029] Preferably, after determining that data loss has occurred, if there are transient fault records in the current power grid operating system, the following alignment steps are performed:

[0030] Extract high-frequency sampling files of fault recordings from the power grid operation system, identify the arrival time of transient traveling waves, and use them as the reference physical zero point;

[0031] Based on the reference physical zero point, the start time of the missing time window recorded in the first recording sequence is timestamped and aligned, and the backup data is retrieved based on the aligned missing time window.

[0032] Preferably, when retrieving corresponding backup data from multiple heterogeneous backup data sources, if it is determined that there is a state conflict among the multiple heterogeneous backup data sources, the following steps are performed:

[0033] Based on the device ID, obtain the preset authority weight of each heterogeneous backup data source, and calculate the time decay factor of the generation time of the candidate compensation data provided by each data source relative to the current time.

[0034] Calculate the confidence score for each candidate compensation data. The confidence score is positively correlated with the authority weight and negatively correlated with the time decay factor.

[0035] The candidate compensation data with the highest confidence score is selected to generate the second record sequence.

[0036] Preferably, the cross-professional association mapping dictionary is pre-established based on the public information model standard and is used to represent the data dimensions of the power grid dispatching system and the marketing asset system. It includes a device resource naming specification mapping table, cross-system measurement attribute mapping rules, and topology connection relationship conversion templates.

[0037] A cross-disciplinary report data sharing and matching system based on ontology mapping includes the following:

[0038] The sequence construction module is used to scan the real-time remote signaling data files and topology node files in the power grid operation system, extract data change record points, and construct the first record sequence reflecting the status of primary power equipment and the connection status of topology nodes;

[0039] The missing data location module is used to traverse the first record sequence, calculate the timestamp difference between adjacent data change record points, and if the timestamp difference exceeds a preset time interval threshold, it is determined that data is missing and the key identifier of the missing data is extracted. The key identifier includes at least the device ID and the missing time window.

[0040] The backup retrieval and evaluation module is used to retrieve corresponding backup data from multiple heterogeneous backup data sources for the key identifier, extract candidate compensation data for the corresponding storage location in the multiple heterogeneous backup data sources, and analyze the field integrity rate of the candidate compensation data.

[0041] The mapping completion module is used to align and map the ontology features of the candidate compensation data with the ontology features of the first record sequence based on a pre-built cross-professional association mapping dictionary when the field completeness rate reaches a preset completeness threshold, and then splice the mapped candidate compensation data into the missing time window to generate a completed second record sequence.

[0042] The output verification module is used to extract data features at the splicing boundary, verify smoothness and logical consistency, calculate the verification matching score, and output the cross-professional data recovery result when the matching score reaches the set standard.

[0043] The beneficial effects of this invention are as follows: By deeply integrating the data anti-tampering and loss recovery mechanism with the real-time power flow calculation unique to the power grid, it automatically triggers underlying physical reversal tracking while completing cross-professional data record sequence completion. Then, when tracing affected distribution paths, it accurately identifies power flow reversals and automatically generates reverse power receiving tags for relevant nodes. Business data containing directional dimension features is accurately synchronized to the marketing asset system through an ontology mapping dictionary. This not only solves the pain point of abnormal line loss calculations caused by business systems relying on fixed incorrect topologies, achieving adaptive consistency between the line loss model and the actual power flow of the physical power grid, but also greatly reduces the workload of manual verification caused by changes in power grid operation modes, improving the high-concurrency processing efficiency and intelligence level of data traceability. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0045] Figure 1 A flowchart illustrating the steps of the cross-disciplinary report data sharing and matching method based on ontology mapping provided by this invention;

[0046] Figure 2 This is a schematic diagram of the structure of the cross-disciplinary report data sharing and matching method based on ontology mapping provided by the present invention. Detailed Implementation

[0047] The following disclosure provides many different embodiments or examples for implementing various embodiments of the invention. To simplify the disclosure, specific embodiments are described below. Of course, these are merely examples and are not intended to limit the scope of the invention.

[0048] The embodiments of the invention will now be described in detail with reference to the accompanying drawings.

[0049] like Figure 1 As shown, the cross-disciplinary report data sharing and matching method based on ontology mapping includes the following steps:

[0050] S1: Scan the real-time remote signaling data files and topology node files in the power grid operation system, extract data change record points, and construct the first record sequence reflecting the status of primary power equipment and the connection status of topology nodes;

[0051] S2: Traverse the first record sequence and calculate the timestamp difference between adjacent data change record points. If the timestamp difference exceeds a preset time interval threshold, it is determined that data is missing. Extract the key identifier of the missing data. The key identifier includes at least the device ID and the missing time window.

[0052] S3: For the key identifier, retrieve the corresponding backup data from multiple heterogeneous backup data sources, extract candidate compensation data for the corresponding storage location from the multiple backup data, and analyze the field integrity rate of the candidate compensation data;

[0053] S4: If the field completeness rate reaches the preset completeness threshold, then based on the pre-built cross-professional association mapping dictionary, the ontology features of the candidate compensation data are aligned and mapped with the ontology features of the first record sequence, and the mapped candidate compensation data is spliced ​​into the missing time window to generate the completed second record sequence.

[0054] S5: Extract data features at the splicing boundary, check for smoothness and logical consistency, calculate the matching score, and when the matching score reaches the set standard, confirm that the second record sequence is valid and output it as the cross-professional data recovery result.

[0055] In practical implementation, the system scans the remote signaling and topology node files in the power grid SCADA system in real time to construct a first record sequence reflecting the primary power equipment and topology connections. When the timestamp difference between adjacent record points exceeds a preset time interval threshold, the system extracts the device ID and missing time window involved in the missing data and retrieves the corresponding backup data from heterogeneous backup data sources such as PMUs. The time interval threshold here is set based on a specific multiple of the power grid standard polling cycle, such as setting it to 3 times the normal 3-second communication cycle, i.e., 9 seconds, to eliminate misjudgments caused by normal network jitter.

[0056] After confirming that the data field completeness rate meets the standard, the system calls a cross-disciplinary association mapping dictionary to align the multi-source data. The ontology features here specifically include three main dimensions: equipment class attributes, measured physical quantity attributes, and spatiotemporal labels. For example, the ontology mapping implementation might be: an integer measurement point identified as "10kV_Breaker_Pos" with a data format of "0 / 1" in the scheduling system is semantically mapped to a character measurement point identified as "SW_10kV_State" with a data format of "Open / Close" in the marketing asset system, thereby eliminating cross-system semantic barriers. Subsequently, the system extracts telemetry values ​​and state quantities at the splicing boundary to verify and calculate the matching score. The calculation formula is as follows:

[0057]

[0058] Among them, the matching score is checked. Used to comprehensively evaluate the data quality and reasonableness after candidate compensation data is spliced ​​into the missing time window; smoothness The smoothness score represents the stability of telemetry values ​​before and after the splicing boundary. It is calculated by determining the rate of change of the derivative of the telemetry values ​​before and after the splicing boundary. The smaller the rate of change, the closer it is to physical inertia, and the higher the smoothness score. Logical consistency score... Whether the data at the splicing boundary conforms to the basic physical logic of the power system is determined by comparing the node state conflict situation. For example, when the boundary state quantity display switch is "separation", if the corresponding telemetry current is not zero, it is judged as a logical conflict and the score is 0. This is the smoothness weighting coefficient, used to adjust the proportion of the smoothness score in the final total score; This is the logical consistency weighting coefficient, used to adjust the proportion of logical consistency score in the final total score; in practical applications, it usually satisfies... When checking the matching score The system will only determine that the match meets the standard and output the recovery result when the value exceeds 0.85. The principle behind this design is based on the dual constraints of semantic web ontology alignment technology and physical laws. It solves the problem of data dialect incompatibility between heterogeneous systems through ontology dictionaries, and intercepts bad data pollution caused by forced splicing by using physical smoothness and logical consistency checks. Compared with traditional single linear interpolation or manual estimation and filling techniques, this solution integrates data evolution completion and real-time power flow inference to automatically generate reverse power receiving tags for power flow reversal nodes and perform cross-professional synchronization, which completely solves the problem of abnormal line loss calculation and greatly improves the accuracy of power grid data sharing and matching and the efficiency of business collaboration.

[0059] More specifically, before scanning the real-time remote signaling data files and topology node files in the power grid operation system to extract data change record points, a remote signaling anti-jitter preprocessing step is also included, as detailed below:

[0060] Obtain the mechanical anti-jitter time window of the target circuit breaker device;

[0061] If multiple remote signaling status reversal records of the same circuit breaker device are received within the mechanical anti-shake time window, the telemetry current change of the circuit breaker device in the corresponding time period is extracted and a blocking verification is performed.

[0062] False state reversal record points that are not accompanied by corresponding telemetry current abrupt changes are removed, and the remaining true state reversal record points are used as data change record points to construct the first record sequence.

[0063] In practical implementation, the system first retrieves the inherent mechanical anti-jitter time window of the target circuit breaker from the equipment ledger. For example, a certain model of 10kV vacuum circuit breaker is calibrated to 40 milliseconds. When the system receives multiple remote signaling state reversal records of "open-close-open-close" reported by the circuit breaker within this 40-millisecond time window, the anti-jitter module immediately extracts the high-frequency telemetry current mutation amount of the circuit breaker in the corresponding time period for mutation amount blocking verification. Specifically, if a state reversal is found not to be accompanied by a significant telemetry current mutation amount, it is determined that the state reversal is purely a false signal generated by contact mechanical bounce and is discarded. If a tripping record is accompanied by a physical mutation of current dropping sharply to zero, it is determined that... This design identifies the actual action as a valid change record point in the first record sequence. The underlying physical coupling logic of "mechanical displacement of power equipment is always accompanied by sudden changes in electrical quantities" is based on this principle. It utilizes continuous analog quantities as the absolute physical criterion for discrete switching quantities, breaking the limitations of traditional single-signal data's isolated self-verification. Compared to existing pure software delay filtering algorithms that blindly shield signals by setting fixed delays, this solution can completely filter out over 95% of hardware jitter-related dirty data from the source. This avoids the waste of system computing power caused by frequent topology reconstruction triggered by false signals and ensures the absolute physical authenticity of subsequent cross-disciplinary data mapping and sequence entry.

[0064] More specifically, after outputting the cross-disciplinary data recovery results, the process also includes a dynamic identification and matching step for the trend direction, as shown below:

[0065] When the cross-disciplinary data recovery results reflect that the state change of the displacement device causes the power grid topology to close or open loop, the power grid topology reconstruction and real-time power flow transfer calculation are triggered.

[0066] Based on the real-time power flow transfer calculation results, the affected power distribution paths that experienced power flow reversal are traced in the completed second record sequence;

[0067] For primary power equipment that was originally in a passive or forward-receiving state on the affected power distribution path, if it is determined that its power flow direction has reversed, a reverse power receiving state label will be added to the primary power equipment in the cross-professional data recovery result.

[0068] After converting the data format of the reverse power receiving status label based on the cross-professional association mapping dictionary, the cross-professional data recovery result with the reverse power receiving status label is synchronously sent to the power grid dispatch system and the marketing asset system.

[0069] In practical implementation, the system continuously monitors the status of restored changeover devices. When it detects that a bus tie circuit breaker in a substation has switched from open to closed, the topology analysis engine determines the 10kV distribution network loop-closing operation triggered by this. It then calls the grid state estimation module to perform real-time power flow transfer calculations, accurately quantifying the affected distribution paths due to the current path reconfiguration caused by the loop-closing operation. Based on this, the system traces upstream along the electrical path in the completed second record sequence, identifying the #3 ring network cabinet incoming line cabinet, originally supplied by a single power source from substation A. Due to the reverse power inflow after loop closure, power flow calculations confirm that its active power flow direction has changed from forward power reception to reverse power feedback. The anti-misjudgment module further verifies the electrical quantity characteristics of the device. After confirming that it meets the physical conditions for reverse power reception, it automatically adds a standardized reverse power reception status label to the incoming line cabinet. Subsequently, the system... The label is converted into a character format recognizable by the marketing asset system by calling a cross-disciplinary association mapping dictionary. The fully labeled cross-disciplinary data recovery results are then synchronized in real time to the dispatch SCADA system and the marketing line loss calculation system. The principle behind this design is based on Kirchhoff's law that changes in power grid topology inevitably lead to the redistribution of energy flow. Through a triple mechanism of proactive sensing, precise tracing, and cross-system semantic synchronization, the real power flow direction changes in the physical world are seamlessly mapped to the digital system, eliminating blind spots in direction recognition caused by professional boundaries. Compared to the industry problem of existing technologies where the dispatch and marketing systems independently determine the power flow direction, resulting in abnormal fluctuations in line loss calculation values ​​or even negative losses during the loop closure period, this solution can achieve millisecond-level automatic labeling and precise cross-system synchronization of power flow direction status in complex topology change scenarios, effectively eliminating metering deviations caused by misjudgment of direction.

[0070] More specifically, the process of triggering grid topology reconfiguration and real-time power flow transfer calculation also includes a cross-voltage level pruning step, as detailed below:

[0071] Extract the winding connection group characteristics of the transformer equipment on the affected power distribution path;

[0072] Based on the isolation / transmission characteristics of the zero-sequence current and phase according to the winding connection group characteristics, determine whether the power flow transfer penetrates to the low-voltage side topology across voltage levels;

[0073] If yes, the original tracing range is maintained; otherwise, the tracing range of the affected power distribution path is subject to boundary trimming across voltage levels.

[0074] In practical implementation, when the system detects a power flow transfer caused by the closing of the bus tie switch at a 220kV substation, it automatically extracts the winding connection group characteristics of all transformer equipment along the affected path. For example, if the 110 / 10kV main transformer is identified as having a YNd11 connection, based on the characteristic in transformer electromagnetic theory that "the delta winding of the YNd11 connection forms a closed loop for the zero-sequence current," it determines that the zero-sequence component cannot penetrate to the 10kV low-voltage side, and only the positive-sequence component can be transmitted. Based on this, the system performs cross-voltage level boundary trimming on the tracing range, limiting the tracing range that originally required a full network scan. The network is designed for voltage levels of 110kV and above, completely excluding 10kV distribution network equipment. The principle behind this design is to strictly follow the physical boundary conditions of transformer electromagnetic coupling, that is, different winding connection groups form a natural zero-sequence network isolation barrier. The scope of power flow transfer is limited by the physical feasibility of electromagnetic energy transfer rather than the simple topological connection relationship. Compared with the extensive calculation mode of "full topology indiscriminate tracing" in the existing technology, this solution reduces the computational complexity of power flow tracing by 2-3 orders of magnitude while ensuring the completeness of the calculation through precise tailoring driven by physical rules.

[0075] More specifically, after the data is simultaneously sent to the power grid dispatch system and the marketing asset system, the following steps are also included:

[0076] A line loss reconstruction instruction is issued to the marketing asset system so that the marketing asset system reverses the calculation logic of the power outflow and inflow of primary power equipment with reverse power receiving status label, and regenerates and updates the line loss report of the primary power equipment.

[0077] In practical implementation, the system automatically sends a standardized line loss reconstruction instruction message to the marketing asset system, including a list of device IDs to be reconstructed and a direction change timestamp. After receiving the instruction, the marketing asset system identifies the 10kV ring main unit #3 incoming line cabinet with the tag "Reverse_Flow_Flag=1", reverses the power metering logic of this device from the default "power supply - power sales" to "power sales - power supply", swaps the numerator and denominator terms in the original line loss calculation formula, and rereads the real-time current and voltage curves pushed by the SCADA system. It then recalculates the power flow direction of this device during the power flow reversal period with a 15-minute granularity, generating a corrected time-segmented line loss report, such as 08:30-10:45. In a case of loop-closing switching operation in a coastal city distribution network, the traditional method failed to identify... The reverse power supply state of the #3 ring main unit incorrectly included its 2180kWh output in the loss calculation, causing the line loss rate in this area to surge abnormally to 22.3%. After reconstruction using this solution, the line loss rate accurately returned to the theoretical value of 3.8%. The principle behind this design is based on the strong coupling relationship between the law of conservation of energy and the power flow direction. Through dynamic inverted metering logic, it ensures that regardless of changes in power flow direction, the line loss calculation always conforms to the physical principle of "input energy - output energy = loss energy". Compared to the lagging processing mode of existing technologies that rely on manual detection of anomalies and then make monthly manual adjustments, this solution achieves millisecond-level automatic adaptation between power flow direction changes and the line loss metering model, completely eliminating the phenomenon of false high loss or negative loss caused by misjudgment of direction, and upgrading cross-professional data sharing from simple state synchronization to the collaborative evolution of calculation logic.

[0078] More specifically, if the field integrity rate does not reach the integrity threshold, the backup data is determined to be invalid, and the missing data causal deduction recovery steps are executed, as shown below:

[0079] Extract the primary power equipment corresponding to the missing data, obtain the relay protection action report associated with the primary power equipment in its upstream topology, and the zero-sequence current telemetry waveform report associated with its downstream topology.

[0080] Based on the preset relay protection logic and the short-circuit current characteristics of the zero-sequence current telemetry recording report, the actual switching state of the primary power equipment in the missing time window is calculated by reverse deduction.

[0081] After converting the actual separation and combination state into ontological features, they are then added to the second record sequence.

[0082] In practical implementation, the system extracts the target primary power equipment corresponding to the missing data, such as the No. 1 circuit breaker in the middle section of a 10kV feeder. It then automatically traverses upstream in the power grid topology model to obtain relay protection action reports from the associated substations, and simultaneously traverses downstream to obtain zero-sequence current telemetry waveform reports recorded by distribution smart terminals. The specific reverse deduction logic takes a single-phase ground fault scenario on a 10kV feeder as an example. If the system lacks the state change record of the No. 1 circuit breaker within the fault occurrence time window, the deduction engine will extract the downstream waveform report and find that the peak zero-sequence short-circuit current reaches 600A at the time of the fault. It will also read the upstream relay protection report to confirm a clear protection action event. Subsequently, the extracted short-circuit current characteristic value is compared with the preset "500A" relay protection action setting value of the No. 1 circuit breaker. When the condition of "downstream short-circuit current greater than the circuit breaker's setting value and upstream protection action coordination" is met, the system can eliminate the fault based on the inevitability of the fault. The logic, through reverse precise deduction, determines that the true state of circuit breaker No. 1 within the missing time window is "open". Finally, this "open" state is converted into the corresponding ontological characteristics of the target system and filled into the second record sequence, such as the numerical label "0" or the character label "Open". The principle of this design is based on the strict physical causality law and protection coordination timing constraints of the power network. It uses the electrical quantities and actions of the upstream and downstream related nodes that have not been lost as evidence chains, and realizes the rigorous inference of the state of the central node through reverse logic closed loop, avoiding absolute dependence on the communication state of a single node. Compared with the existing technology, which can only perform inefficient manual inspection or simply mark as bad data and discard it when faced with large-scale data loss, this solution can still achieve high-fidelity and fully automatic logical reconstruction of equipment status even under extreme communication interruption conditions. It eliminates the blind spots in the power grid topology during the fault period and ensures the data continuity and accuracy of cross-professional system line loss calculation and power outage range analysis.

[0083] More specifically, after determining that data loss has occurred, if there are transient fault records in the current power grid operating system, the following alignment steps are performed:

[0084] Extract high-frequency sampling files of fault recordings from the power grid operation system, identify the arrival time of transient traveling waves, and use them as the reference physical zero point;

[0085] Based on the reference physical zero point, the start time of the missing time window recorded in the first recording sequence is timestamped and aligned, and the backup data is retrieved based on the aligned missing time window.

[0086] In practical implementation, when a phase-to-phase short-circuit fault is detected on a 220kV line and the SCADA system is found to have a 5-8 second data gap, the system automatically extracts the 1MHz high-frequency fault waveform file recorded by the PMU device. Through wavelet transform, it accurately identifies the physical moment when the transient traveling wave first arrives at the substation bus as a reference physical zero point unaffected by clock drift. Subsequently, it calculates the clock offset between this reference physical zero point and the fault start time recorded by the SCADA system, and performs sub-millisecond correction and alignment on the start time of the missing time window. Based on the corrected and accurate time window, the system accurately retrieves perfectly matching backup data from the WAMS system and the fault information system, successfully restoring the entire process of circuit breaker tripping and voltage drop. The principle behind this design is to utilize the physical characteristics of fault traveling waves—propagating at the speed of light and with accurately measurable arrival times—to anchor the clock alignment of the distributed system to objective physical events rather than relying on network clock protocols, fundamentally avoiding IEEE 80 ... The 1588 protocol accumulates synchronization errors during transient processes. Compared to the defects of cross-system data "misalignment splicing" caused by relying on NTP / SNTP protocols for second-level time synchronization in existing technologies, this solution can achieve microsecond-level spatiotemporal alignment of multi-source data under extreme fault conditions. In a 220kV line fault case in a certain regional power grid, the reconstruction error of the circuit breaker change time was reduced from ±120ms in the traditional method to ±0.5ms, which significantly improved the accuracy of fault recording and protection action sequence restoration.

[0087] More specifically, when retrieving corresponding backup data from multiple heterogeneous backup data sources, if a state conflict is determined to exist between the multiple heterogeneous backup data sources, the following steps are performed:

[0088] Based on the device ID, obtain the preset authority weight of each heterogeneous backup data source, and calculate the time decay factor of the generation time of the candidate compensation data provided by each data source relative to the current time.

[0089] Calculate the confidence score for each candidate compensation data. The confidence score is positively correlated with the authority weight and negatively correlated with the time decay factor.

[0090] The candidate compensation data with the highest confidence score is selected to generate the second record sequence.

[0091] To address the pain point of data state conflicts arising between heterogeneous backup data sources during parallel operation of multiple power grid systems due to asynchronous acquisition frequencies or network latency, the system triggers an optimization mechanism based on dynamic confidence quantification when retrieving backup data and determining a conflict. The specific implementation is as follows: The system first retrieves the preset authority weight W for each heterogeneous backup data source based on the device ID. In one implementation, for the same substation circuit breaker, the authority weight of the PMU phasor measurement device is set to 0.95, while the weight of the conventional SCADA system is set to 0.85. Subsequently, the generation time of each source candidate compensation data is extracted, and the time difference between the generated data and the current time is calculated. The time difference is then directly substituted into the following mathematical expression to calculate the final confidence score for each candidate data. :

[0092]

[0093] in This is the time decay factor. This is the preset time decay constant.

[0094] In one implementation, taking a node state conflict as an example, if the PMU provides a status of "closed" with a delay of 2 seconds, and the SCADA provides a status of "open" with a delay of 0.1 seconds, the preset... The score is 0.5. Calculated using the formula, the SCADA data score of 0.808 is higher than the PMU data score of 0.349. Therefore, the system automatically selects the SCADA "shutdown" data with the highest score to complete the second record sequence. The principle behind this design is that the data quality of heterogeneous systems is distributed in two dimensions. Simply relying on static hardware levels can easily lead to the mis-collection of highly authoritative but severely outdated ghost data. By introducing an exponentially decaying time penalty mechanism, mathematical decoupling and precise balance between static data reliability and dynamic real-time performance are achieved. Compared to the crude "majority rule" voting mechanism or single static priority arbitration rule in existing technologies, this solution can accurately intercept expired high-risk data in high-concurrency conflict conditions using purely mathematical quantification. It effectively solves the problem of timing disorder and topology misjudgment caused by clock asynchrony between cross-professional systems, ensuring the uniqueness and absolute correctness of data recovery under extreme conditions.

[0095] More specifically, the cross-professional association mapping dictionary is pre-established based on the public information model standard and is used to represent the data dimensions of the power grid dispatching system and the marketing asset system. It includes equipment resource naming specification mapping table, cross-system measurement attribute mapping rules, and topology connection relationship conversion template.

[0096] like Figure 2 As shown, the cross-disciplinary report data sharing and matching system based on ontology mapping includes the following:

[0097] The sequence construction module is used to scan the real-time remote signaling data files and topology node files in the power grid operation system, extract data change record points, and construct the first record sequence reflecting the status of primary power equipment and the connection status of topology nodes;

[0098] The missing data location module is used to traverse the first record sequence, calculate the timestamp difference between adjacent data change record points, and if the timestamp difference exceeds a preset time interval threshold, it is determined that data is missing and the key identifier of the missing data is extracted. The key identifier includes at least the device ID and the missing time window.

[0099] The backup retrieval and evaluation module is used to retrieve corresponding backup data from multiple heterogeneous backup data sources for the key identifier, extract candidate compensation data for the corresponding storage location in the multiple heterogeneous backup data sources, and analyze the field integrity rate of the candidate compensation data.

[0100] The mapping completion module is used to align and map the ontology features of the candidate compensation data with the ontology features of the first record sequence based on a pre-built cross-professional association mapping dictionary when the field completeness rate reaches a preset completeness threshold, and then splice the mapped candidate compensation data into the missing time window to generate a completed second record sequence.

[0101] The output verification module is used to extract data features at the splicing boundary, verify smoothness and logical consistency, calculate the verification matching score, and output the cross-professional data recovery result when the matching score reaches the set standard.

[0102] The operation and effect of the cross-professional report data sharing and matching system based on ontology mapping of the present invention are consistent with the above-mentioned cross-professional report data sharing and matching method based on ontology mapping. Therefore, the cross-professional report data sharing and matching system based on ontology mapping will not be described again here.

[0103] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented using necessary general-purpose hardware platforms, or it can be implemented through a combination of hardware and software. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a computer product. The present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and specification of the present invention.

Claims

1. A cross-disciplinary report data sharing and matching method based on ontology mapping, characterized in that, Includes the following steps: S1: Scan the real-time remote signaling data files and topology node files in the power grid operation system, extract data change record points, and construct the first record sequence reflecting the status of primary power equipment and the connection status of topology nodes; S2 S1: Traverse the first record sequence and calculate the timestamp difference between adjacent data change record points. If the timestamp difference exceeds a preset time interval threshold, it is determined that data is missing. Extract the key identifier of the missing data. The key identifier includes at least the device ID and the missing time window. S2: For the key identifier, retrieve the corresponding backup data from multiple heterogeneous backup data sources, extract candidate compensation data for the corresponding storage location in the multiple backup data, and analyze the field integrity rate of the candidate compensation data. S4: If the field completeness rate reaches the preset completeness threshold, then based on the pre-constructed cross-professional association mapping dictionary, the ontology features of the candidate compensation data are aligned and mapped with the ontology features of the first record sequence, and the mapped candidate compensation data is spliced ​​to the missing time window to generate the completed second record sequence; S5: Extract the data features at the splicing boundary for smoothness and logical consistency verification, calculate the verification matching score, and when the matching score reaches the set standard, confirm that the second record sequence is valid and output it as the cross-professional data recovery result.

2. The cross-disciplinary report data sharing and matching method based on ontology mapping according to claim 1, characterized in that, Before scanning the real-time telemetry data files and topology node files in the power grid operation system and extracting data change record points, a telemetry anti-jitter preprocessing step is also included, as follows: Obtain the mechanical anti-jitter time window of the target circuit breaker device; if multiple telemetry state flip records of the same circuit breaker device are received within the mechanical anti-jitter time window, the telemetry current mutation amount of the circuit breaker device in the corresponding time period is extracted and a blocking verification is performed. False state reversal record points that are not accompanied by corresponding telemetry current abrupt changes are removed, and the remaining true state reversal record points are used as data change record points to construct the first record sequence.

3. The cross-disciplinary report data sharing and matching method based on ontology mapping according to claim 1, characterized in that, After outputting the cross-disciplinary data recovery results, the dynamic identification and matching step of power flow direction is also included, as follows: When the cross-disciplinary data recovery results reflect that the state change of the displacement device causes the power grid topology to close or open loop operation, the power grid topology reconstruction and real-time power flow transfer calculation are triggered; combined with the real-time power flow transfer calculation results, the affected distribution path that has experienced power flow reversal is traced in the completed second record sequence. For primary power equipment on the affected power distribution path that was originally in a passive or forward power receiving state, if it is determined that its power flow direction has reversed, a reverse power receiving state label is added to the primary power equipment in the cross-disciplinary data recovery result; after converting the data format of the reverse power receiving state label based on the cross-disciplinary association mapping dictionary, the cross-disciplinary data recovery result with the reverse power receiving state label is synchronously sent to the power grid dispatch system and the marketing asset system.

4. The cross-disciplinary report data sharing and matching method based on ontology mapping according to claim 3, characterized in that, The process of triggering grid topology reconfiguration and real-time power flow transfer calculation also includes a cross-voltage level pruning step, as follows: extract the winding connection group characteristics of the transformer equipment on the affected distribution path; based on the isolation / transmission characteristics of the winding connection group characteristics for zero-sequence current and phase, determine whether the power flow transfer penetrates to the low-voltage side topology across voltage levels; if yes, maintain the original tracing range; if no, perform cross-voltage level boundary pruning on the tracing range of the affected distribution path.

5. The cross-disciplinary report data sharing and matching method based on ontology mapping according to claim 3, characterized in that, After being simultaneously sent to the power grid dispatch system and the marketing asset system, the following steps are also included: issuing a line loss reconstruction instruction to the marketing asset system so that the marketing asset system reverses the calculation logic of the power outflow and inflow for primary power equipment with reverse power receiving status labels, and regenerates and updates the line loss report of the primary power equipment.

6. The cross-disciplinary report data sharing and matching method based on ontology mapping according to claim 1, characterized in that, If the field integrity rate does not reach the integrity threshold, the backup data is determined to be invalid, and the missing data causal inference recovery step is executed, as follows: Extract the primary power equipment corresponding to the missing data, obtain the relay protection action report associated with the primary power equipment in its upstream topology, and the zero-sequence current telemetry waveform report associated with its downstream topology; Based on the preset relay protection logic and the short-circuit current characteristics of the zero-sequence current telemetry waveform report, reverse inference is performed to calculate the actual switching state of the primary power equipment in the missing time window; After converting the actual switching state into the ontological features, it is completed in the second record sequence.

7. The cross-disciplinary report data sharing and matching method based on ontology mapping according to claim 1, characterized in that, After determining that data loss has occurred, if there are transient fault records in the current power grid operating system, the following alignment steps are performed: extract the high-frequency sampling file of the fault waveform of the power grid operating system, identify the arrival time of the transient traveling wave, and use it as the reference physical zero point; based on the reference physical zero point, perform timestamp correction alignment on the start time of the missing time window recorded in the first recording sequence, and retrieve the backup data based on the corrected and aligned missing time window.

8. The cross-disciplinary report data sharing and matching method based on ontology mapping according to claim 1, characterized in that, When retrieving corresponding backup data from multiple heterogeneous backup data sources, if it is determined that there is a state conflict between the multiple heterogeneous backup data sources, the following steps are performed: obtain the preset authority weight of each heterogeneous backup data source based on the device ID, and calculate the time decay factor of the generation time of the candidate compensation data provided by each data source relative to the current time. Calculate the confidence score for each candidate compensation data, wherein the confidence score is positively correlated with the authority weight and negatively correlated with the time decay factor; select the candidate compensation data with the highest confidence score to generate the second record sequence.

9. The cross-disciplinary report data sharing and matching method based on ontology mapping according to claim 1, characterized in that, The cross-disciplinary association mapping dictionary is pre-established based on the public information model standard and is used to represent the data dimensions of the power grid dispatching system and the marketing asset system. It includes a mapping table of equipment resource naming conventions, cross-system measurement attribute mapping rules, and topology connection relationship conversion templates.

10. A cross-disciplinary report data sharing and matching system based on ontology mapping, characterized in that, The system includes the following components: a sequence construction module, used to scan real-time remote signaling data files and topology node files in the power grid operation system, extract data change record points, and construct a first record sequence reflecting the status of primary power equipment and the connection status of topology nodes; and a missing data location module, used to traverse the first record sequence, calculate the timestamp difference between adjacent data change record points, and if the timestamp difference exceeds a preset time interval threshold, determine that data is missing, extract the key identifier of the missing data, and the key identifier includes at least the device ID and the missing time window. The backup retrieval and evaluation module is used to retrieve corresponding backup data from multiple heterogeneous backup data sources for the key identifier, extract candidate compensation data for the corresponding storage location in the multiple heterogeneous backup data sources, and analyze the field integrity rate of the candidate compensation data. The mapping completion module is used to align and map the ontology features of the candidate compensation data with the ontology features of the first record sequence based on a pre-built cross-professional association mapping dictionary when the field completeness rate reaches a preset completeness threshold. The mapped candidate compensation data is then spliced ​​into the missing time window to generate a completed second record sequence. The verification output module is used to extract data features at the splicing boundary for smoothness and logical consistency verification, calculate the verification matching score, and output the cross-professional data recovery result when the matching score reaches a set standard.

Images