Quality evaluation method and system for multivariate heterogeneous data of intelligent substation of power system

Abstract

The invention discloses a multivariate heterogeneous data quality evaluation method and system for an intelligent substation of a power system, and belongs to the field of intelligent substations, and the method comprises the steps: accessing bay level equipment, station level equipment and a data transmission module, obtaining a model file accessed by a data communication gateway machine and the bay level equipment, and cutting the model file; a station level heterogeneous data transmission module is accessed, heterogeneous data streams of a station level are received, and mirror image processing is carried out on all the heterogeneous data streams; performing data integration on the clipped model file and the mirrored heterogeneous data stream, and performing data concurrence after data integration; and evaluating the concurrent data through the data sections, wherein the evaluation comprises the steps of performing heterogeneous data accuracy evaluation, heterogeneous data concurrence robustness evaluation and heterogeneous data quality evaluation on the data sections of the unified time intercept points. The method adapts to the development demand of the intelligent power grid technology, and meets the test and evaluation demand of the intelligent substation of the power system.

Description

Technical Field

[0001] The present invention provides a method and system for evaluating the quality of multi-dimensional heterogeneous data of an intelligent substation in an electric power system, and belongs to the field of evaluation of digitalization and quality of electric power systems. Background Art

[0002] As a key computing node in the power system, smart substations must differ from traditional substations in design and operation. This shift in approach will lead to new challenges in the development and operation of smart substations.

[0003] With the continuous development of big models and AI technologies, the data processing, data processing, and data analysis capabilities of the control layer systems of smart substations in power systems are continuously improving. Smart stations can now simultaneously process various types of protocol data, dynamic image data, audio data, and graphic and image data. Smart substations in power systems have evolved from traditionally uploading and issuing commands and data information to assisting in power grid system analysis and decision-making. Smart substations in power systems have become the distributed intelligent brains of smart grid operations. This also leads to higher technical requirements for data throughput, concurrent multi-data processing, and corresponding capabilities of smart substations in power systems, which are also important indicators for measuring the operational effectiveness of smart substations in power systems.

[0004] Smart substations in power systems represent significant differences from traditional substations in terms of functionality, performance, and event handling. They are capable of sophisticated data processing and handling massive amounts of data in a short period of time. Furthermore, incorporating artificial intelligence technology, they possess powerful learning capabilities and the ability to process high-dimensional, heterogeneous data. Existing testing and evaluation methods are no longer sufficient to handle the maintenance, operation, and troubleshooting of smart substations, creating significant challenges for their associated testing and evaluation.

[0005] Smart substations in power systems process large amounts of heterogeneous data, a feature not available in traditional smart substations. This large amount of differentially structured data is composed of various types of heterogeneous data. The ability to process heterogeneous data is a key factor in evaluating the data processing capabilities of smart substations, a crucial factor in assessing the operational health of smart grids, and a crucial indicator of the accuracy and speed of data processing in smart substations.

[0006] The power system is constantly evolving and iterating in this strategic goal. The composition structure, design method, and operation mode of intelligent substations in the power system are very different from those of traditional intelligent substations. The intelligent substation of the power system has many control links for data streams, rich processing content, complex coupling of various application functions within the station, and its operating characteristics are restricted by hardware resource conditions and superimposed by various advanced application module controls, resulting in a huge information processing capacity of the intelligent substation of the power system. The dispatching end has more stringent requirements for data processing, information classification, and fault handling at the station end, and the data processing granularity is finer.

[0007] With the continuous improvement of the requirements for lean management of secondary systems, in order to meet the business needs of remote centralized monitoring of intelligent substations in the power system, higher requirements are put forward for the breadth of substation equipment information collection, the depth of equipment perception ability, and the fineness of equipment operation and maintenance management. The huge changes in the internal output and external input environment of intelligent substations in the power system have also led to insufficient breadth and depth of equipment monitoring, insufficient support for remote monitoring, a closed software architecture of the station control system, and insufficient automation operation and maintenance technology. Summary of the Invention

[0008] To solve the problems existing in the prior art, the present invention proposes a method and system for evaluating the quality of multi-source heterogeneous data in an intelligent substation of a power system. This method adapts to the development needs of smart grid technology and meets the testing and evaluation needs of intelligent substations in the power system.

[0009] To solve the above technical problems, the technical solution adopted by the present invention is as follows: In a first aspect, the present invention provides a method for evaluating the quality of multi-source heterogeneous data in an intelligent substation of a power system, including: Connect to interval layer devices, station control layer devices, and data transmission modules, obtain the model files accessed by the data communication gateway and interval layer devices, and trim the model files; Connect to the heterogeneous data transmission module of the station control layer, receive the heterogeneous data streams of the station control layer, and perform mirroring processing on all the heterogeneous data streams; Integrate the trimmed model files and the mirrored heterogeneous data streams, and perform data concurrency after data integration; Evaluate the concurrent data through data sections, including: evaluating the accuracy of heterogeneous data, the robustness of heterogeneous data concurrency, and the data quality of heterogeneous data for the data sections at the same time point.

[0010] As a further improvement of the present invention, the device models accessed by the data communication gateway and interval layer devices are designed to be maximized according to the same level of the substation, and the data nodes of the device models trim the model files according to the maximum load-bearing capacity of the device performance.

[0011] As a further improvement of the present invention, the heterogeneous data stream includes image files, audio files, video files, and miscellaneous irregular files.

[0012] As a further improvement of the present invention, the data quality evaluation of the heterogeneous data includes: Compare the data measurement accuracy, control accuracy, telemetry accuracy, and data source, and evaluate the accuracy and precision after a large amount of data throughput concurrency; for the transmitted pictures, graphics, audio, video, and multimedia data, compare them with the data source to evaluate the reduction degree.

[0013] As a further improvement of the present invention, the data quality evaluation of the heterogeneous data further includes: Based on the measurement accuracy, control accuracy, telemetry accuracy, and multimedia data in the concurrent heterogeneous data results, restore and analyze the health status, system trends, familial defects, and operation factors of individual devices of the power system intelligent substation system.

[0014] As a further improvement of the present invention, the data quality evaluation of the heterogeneous data further includes: Based on the analysis results, conduct early warning evaluation of the system status, perform abnormal diagnosis analysis and display of the power automation system, count the defects of system modules in the integrated monitoring system, and sense and predict the risks that the power system intelligent substation system may pose to the power grid.

[0015] As a further improvement of the present invention, the data quality evaluation of the heterogeneous data further includes: Evaluate the equipment quality of in-station station control equipment, secondary equipment, network equipment, and data networks.

[0016] In a second aspect, the present invention provides a power system intelligent substation multi-source heterogeneous data quality evaluation system, including: A clipping module, used to access interval layer devices, station control layer devices, and data transmission modules, obtain the model files accessed by the data communication gateway and interval layer devices, and clip the model files; A mirroring module, used to access the station control layer heterogeneous data transmission module, receive the heterogeneous data stream of the station control layer, and perform mirroring processing on all the heterogeneous data streams; A concurrency module, used to perform data integration on the clipped model files and the mirrored heterogeneous data stream, and perform data concurrency after data integration; An evaluation module, used to evaluate the concurrent data through data sections, including: evaluating the accuracy of heterogeneous data, the robustness of heterogeneous data concurrency, and the data quality evaluation of heterogeneous data for data sections at the same time point.

[0017] In a third aspect, the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for evaluating the quality of multi-source heterogeneous data of the power system intelligent substation is implemented.

[0018] In a fourth aspect, the present invention provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the method for evaluating the quality of multi-source heterogeneous data of the power system intelligent substation is implemented.

[0019] In a fifth aspect, the present invention provides a computer program product including computer instructions, and the computer instructions direct a computer to execute the method for evaluating the quality of multi-source heterogeneous data of the power system intelligent substation.

[0020] The beneficial effects of the present invention compared with the prior art are as follows: Through the full access of the interval layer device, the station control layer device and the data transmission module, the mirror processing and the model pruning technology work together, which not only retains the original data characteristics, but also eliminates redundant interference, making the data quality evaluation result closer to the actual operation state. It supports the parallel processing of multi-protocol heterogeneous data, and realizes cross-platform evaluation through a standardized data section, which is applicable to hybrid architecture substations. The data concurrency mechanism adopts a distributed caching technology, supports millisecond-level time point synchronization, and can still maintain more than 90% data integrity under high-load scenarios. The evaluation results are directly connected to the operation and maintenance traceability system, providing data support for equipment status early warning and maintenance decision-making, and forming an iterative closed loop of "evaluation - optimization - verification". Through automatic data quality scoring and anomaly location, the manual verification workload is reduced, and the fault troubleshooting efficiency is improved by about 40%. This method solves the problems of difficult fusion of multi-source heterogeneous data and single evaluation dimension in traditional substations through hierarchical data integration and dynamic quality evaluation, and provides an extensible technical framework for the digital operation and maintenance of intelligent substations. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following introduces the relevant technical solution drawings in the embodiments of the present invention or the prior art. It should be understood that the drawings introduced below are only for conveniently and clearly expressing some embodiments of the technical solutions in the present invention, and those skilled in the art can also obtain other drawings based on these drawings without creative efforts.

[0022] Figure 1 It is a flowchart of a method for evaluating the quality of multi-source heterogeneous data of a power system intelligent substation provided by the present invention; Figure 2Schematic diagram of the principle for evaluating the quality of multi - heterogeneous data in an intelligent substation of a power system provided by an embodiment of the present invention; Figure 3 Flowchart for evaluating the quality of data provided by the present invention; Figure 4 A system for evaluating the quality of multi - heterogeneous data in an intelligent substation of a power system provided by the present invention; Figure 5 Schematic diagram of an electronic device provided by the present invention. Detailed implementation manners

[0023] The embodiments of the present invention will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention. For the step numbers in the following embodiments, they are only set for the convenience of description and illustration, and no limitation is imposed on the order between the steps. The execution order of each step in the embodiments can be adjusted adaptively according to the understanding of those skilled in the art.

[0024] In the description of the present invention, unless otherwise clearly defined, terms such as "set", "installed", "connected", etc. should be understood in a broad sense. Those skilled in the art can reasonably determine the specific meanings of the above terms in the present invention in combination with the specific content of the technical solution.

[0025] For the operation of an intelligent substation in a power system, higher levels of robustness, accuracy, real - time performance, and stability are required to meet the operation requirements of the smart grid. One of the essential requirements for all of this boils down to the data - processing ability of the intelligent substation in the power system. The data stream in the intelligent substation of the power system includes data streams of traditional protocols, as well as heterogeneous data streams such as videos, photos, and audios. The data stream has a large flow, rich types, and different processing requirements, and the data - processing ability of the data stream is also a key indicator for testing the intelligent substation of the power system.

[0026] As Figure 1 shown, the first object of the present invention is to provide a method for evaluating the quality of multi - heterogeneous data in an intelligent substation of a power system, including: S1, accessing interval - layer devices, sub - station control - layer devices, and data - transmission modules, obtaining the model files accessed by the data communication gateway and interval - layer devices, and trimming the model files; S2, accessing the heterogeneous data - transmission module of the sub - station control layer, receiving the heterogeneous data stream of the sub - station control layer, and performing mirroring processing on all the heterogeneous data streams; S3, performing data integration on the trimmed model files and the mirrored heterogeneous data streams, and performing data concurrency after data integration; In the above solution, through the full access of the interval layer device, substation control layer device and data transmission module, combined with the model pruning technology, the standardized mapping of device data is realized. Model pruning adapts to the communication protocol differences of devices from different manufacturers by removing redundant parameters and retaining key feature quantities. The mirror processing technology completely replicates the original data stream to avoid data loss or tampering during transmission.

[0027] Furthermore, in the data concurrency stage, the time point synchronization technology is adopted to achieve the alignment of cross-system data sections. The accuracy evaluation is completed by comparing the semantic consistency between the mirror data and the original data; the robustness evaluation is quantitatively analyzed based on indicators such as the data loss rate and latency rate under high-concurrency scenarios.

[0028] S4, evaluate the concurrent data through data sections, including: evaluating the accuracy of heterogeneous data, the robustness of heterogeneous data concurrency, and the data quality of heterogeneous data for the data sections at the same time point.

[0029] In the above solution, the data quality evaluation integrates multi-dimensional indicators and uses a weighted scoring model to generate a comprehensive quality index. Combining with the historical data pattern library, the system can dynamically adjust the evaluation threshold to adapt to the quality requirements under different operating conditions.

[0030] The present invention is to meet the development needs of smart grid technology, study the key technologies for evaluating the quality of multi-source heterogeneous data in intelligent substations of power systems, evaluate the system defense capabilities of intelligent substations in power systems, the safety of equipment bodies, whether there are blind spots in safety monitoring, and the credible immunity capabilities of equipment connections, so as to meet the testing and evaluation requirements for intelligent substations in power systems.

[0031] The present invention will deeply study the technology for evaluating the quality of multi-source heterogeneous data in intelligent substations of power systems. Based on power big data and grid dispatching control rules, by using technologies such as data classification and heterogeneous data stripping, it provides means and methods suitable for testing and evaluating heterogeneous data in intelligent substations of power systems.

[0032] Next, the method for evaluating the quality of multi-source heterogeneous data in the intelligent substation of the power system of the present invention will be further described in detail with reference to the accompanying drawings and specific embodiments.

[0033] As Figure 2As shown, this invention, based on a smart power substation, uses the smart substation data transmission system as the entry point. By fully integrating bay-layer equipment, station-control-layer equipment, and data transmission modules, it comprehensively evaluates the multi-faceted, heterogeneous concurrent processing capabilities of the smart substation. With full integration of bay-layer, station-control-layer equipment, and data transmission modules, the smart substation forms a highly integrated, standardized information exchange system that supports real-time monitoring, rapid protection, and intelligent operation and maintenance. Furthermore, through modular design and low-carbon technologies, it adapts to the needs of emerging power systems.

[0034] In the above scheme, the bay layer equipment includes measurement and control devices, protection devices, relay protection devices, fault recorders, PMUs (synchronized phasor measurement units), etc., which are responsible for real-time data acquisition, protection logic execution, and local control functions.

[0035] The measurement and control device monitors primary equipment operating parameters (such as voltage, current, and power) in real time and executes remote control commands (such as circuit breaker opening and closing operations). It supports communication between the station control layer and the process layer, transmitting data via optical fiber to achieve digital information exchange across the entire station.

[0036] Protection devices quickly identify power system faults (such as short circuits and overloads) and trigger circuit breakers to isolate the faulted area, ensuring safe grid operation. They adhere to the DL / T860 standard and support the GOOSE (Generic Object Oriented Substation Event) communication protocol, enabling rapid transmission of protection signals and multi-device collaboration.

[0037] PMUs (synchronized phasor measurement units) use a high-precision synchronized clock (such as GPS) to measure voltage, current, and frequency at grid nodes for dynamic monitoring of grid stability. This data can be connected to a wide-area measurement system (WAMS) to support real-time grid analysis and control decisions.

[0038] Bay-layer equipment must maintain normal communication with the process layer (such as merging units and intelligent terminals) and the station control layer (such as monitoring systems) to ensure reliable transmission of SV (sampled value) and GOOSE signals. Bay-layer equipment, through its highly integrated and standardized design, supports intelligent substations in power systems, achieving advanced functions such as real-time control, condition monitoring, and collaborative interaction.

[0039] Station control layer equipment: covers monitoring hosts, data servers, integrated application servers, operator stations, etc., to achieve full-station data integration, visual monitoring and advanced applications (such as intelligent alarms and energy efficiency analysis).

[0040] Data transmission module: realizes cross-layer data interaction through the station control layer network (Ethernet) and the process layer network (fiber optic communication), and supports GOOSE (Generic Object Oriented Substation Event) and SV (Sampled Value) protocol transmission.

[0041] As Figure 2 shown, the intelligent substation of the power system is connected to the data communication network shutdown device and the bay-level devices. The device models accessed by the data communication network shutdown device and the bay-level devices are designed to be maximized according to the same level of the substation. The data nodes of the device model file are trimmed according to the maximum bearing capacity of the device performance.

[0042] Specifically, the access models of the data communication network shutdown device and the bay-level devices (such as measurement and control, protection devices) need to follow the unified specifications of the substation voltage level to ensure that the model structure and function definition match the substation level, and support cross-device data interaction and standardized communication protocols (such as DL / T860). The device model should cover all data nodes, including basic functions such as telemetry, tele-signaling, and remote control, as well as advanced application interfaces, to meet the call requirements of the dispatching master station for the information of the entire station.

[0043] The model file trimming method is based on the device processing capabilities (such as CPU load, storage capacity, and communication bandwidth), trimming redundant data nodes and retaining the core function models. For example, the merging unit only retains the SV message configuration related to the transmission of sampled values, reducing unnecessary logical nodes. Dynamically adjust the model data scale according to the real-time operating status of the device. For example, in high-load scenarios, turn off non-critical monitoring functions to prioritize the transmission efficiency of protection and control data.

[0044] The model design of the data communication network shutdown device and the bay-level devices in the intelligent substation needs to take standardization and dynamic optimization as the core, and achieve efficient data interaction through unified modeling by DL / T860 and performance-driven trimming. During the design process, it is necessary to combine the device processing capabilities, communication requirements, and commissioning specifications to ensure that the model takes into account the operation reliability and resource utilization rate while maximizing compatibility.

[0045] As Figure 2 shown, the intelligent substation of the power system is connected to the heterogeneous data transmission module of the station control layer to receive image files, audio files, video files, and miscellaneous irregular files from the station control layer. All the above-mentioned heterogeneous data streams accessed by the intelligent substation of the power system are mirrored, and after data integration of the intelligent substation of the power system, data concurrency is performed.

[0046] Specifically, the image files come from the image file host module, the audio and video files come from the multimedia file host module, and the miscellaneous irregular files come from the irregular file host module. The corresponding host modules verify the data and connect it to the intelligent substation of the power system.

[0047] Among them, the mirroring technology implementation deploys mirror ports in the station control layer network (MMS protocol) and the process layer network (GOOSE / SV protocol) to losslessly copy the data streams generated by all station devices, covering diverse data sources such as protection action signals, sampled values, and status monitoring. The protocol parsing technology is used to classify and label the mirrored data according to the DL / T860 standard, distinguishing data streams of real-time control type (GOOSE), measurement type (SV), and management type (MMS).

[0048] Preferably, redundant data (such as repeated alarm signals) is filtered through traffic shaping technology to reduce the mirroring bandwidth occupancy rate and retain the integrity of key data (such as fault recording and PMU dynamic phasors). A priority queue is set up to ensure that the mirroring transmission delay of protection action instructions (such as trip GOOSE messages) is ≤2 ms, avoiding affecting the real-time control function due to mirroring processing.

[0049] For example, the unified data model construction is based on the DL / T860 standard to establish the information model of all station devices, map the mirrored multi-source data to unified logical devices (LD) and logical nodes (LN), and achieve cross-protocol data semantic alignment. The SCD (substation configuration description file) is used to define data association rules, such as associating PMU dynamic monitoring data with the protection device action records to support panoramic data analysis. A data integration middleware (such as a message bus) is deployed to support protocol conversions such as MQTT and IEC104, realizing seamless docking of the station control layer monitoring system, dispatching master station, and third-party platforms (such as photovoltaic monitoring). For the SV sampled value data of process layer devices (such as merging units), an interpolation algorithm is used to compensate for transmission jitter to ensure data time series consistency.

[0050] Among them, concurrent data processing deploys a distributed computing framework (such as Apache Kafka) in the station control layer server cluster. Through the partitioning mechanism, the data stream is divided into processing units according to intervals (such as line protection and transformer monitoring) to improve the concurrent throughput capacity. The CPU core and memory resource allocation are dynamically adjusted to prioritize the processing of protection-type data streams and limit the resource occupancy rate of non-critical services (such as historical data archiving) to ≤20%.

[0051] Preferably, a time window mechanism is used to perform aggregation calculations on high-frequency sampled data (such as PMU 100Hz phasors) to reduce the single-point data processing pressure and meet the accuracy requirements of the wide-area monitoring system at the same time. An anomaly detection module is established to real-time identify data storms (such as GOOSE message flooding caused by network congestion) through parallel threads and trigger traffic suppression strategies.

[0052] Preferably, it is also necessary to simulate the concurrent data scenarios of multiple devices through a digital twin platform. For example, test the mirror integrity (packet loss rate <0.001%) and processing timeliness (end-to-end delay ≤10 ms). When the network load of the station control layer ≥80%, verify that the concurrent processing accuracy rate of the protection action instructions ≥99.99%.

[0053] As a preferred solution, as Figure 2 shown, the present invention evaluates the concurrent heterogeneous data of the intelligent substation of the power system through the data section. It includes: evaluating the accuracy of heterogeneous data for the data section at the unified time point, evaluating the concurrent robustness of heterogeneous data, and evaluating the data quality of heterogeneous data.

[0054] For the subsequent data quality evaluation; compare the data measurement accuracy, control accuracy, telemetry accuracy, and data source, and evaluate the accuracy and accuracy after the concurrent throughput of a large amount of data; compare the transmitted pictures, graphics, sound, light, and electrical data, and multimedia data with the data source, and evaluate their restoration degree.

[0055] Among them, for the time synchronization check of the heterogeneous data mathematical accuracy evaluation, the Beidou / GPS clock source can be used to achieve microsecond-level time synchronization, and the consistency of the data section timestamps can be verified through satellite time synchronization technology (such as the IEC61850-9-3 standard), and the deviation needs to be ≤1 μs. Based on the digital twin model, the simulation comparison of the data at the unified time section is carried out to identify the logical correlation errors of key parameters such as sampled values (SV) and protection action signals (GOOSE).

[0056] Data calibration and anomaly detection, based on the deployment of multi-source data fusion algorithms (such as Kalman filtering) to eliminate sensor noise, ensuring the mathematical consistency of the current / voltage measurement values and the theoretical model. Identify the mutation values (such as sampled value distortion, telemetry jump) of the data section through the anomaly detection module, and mark the abnormal data nodes with a confidence level lower than 95%.

[0057] Among them, the evaluation of the concurrent robustness of heterogeneous data includes high-load stress testing and fault tolerance and recovery capabilities; high-load stress testing is to simulate the concurrent data peak (such as GOOSE message flooding) of multiple interval layer devices (measurement and control, protection devices), and verify the packet loss rate (≤0.01%) and end-to-end delay (≤5 ms) of the system under 80% bandwidth occupancy. Through the distributed computing framework (such as Apache Kafka) to partition the data stream, test the concurrent throughput balance of different priority data (such as protection instructions and historical data).

[0058] Fault tolerance and recovery capabilities are to artificially inject network jitter or device downtime events, detect the integrity recovery capabilities of data cross-sections (such as the disconnection reconnection time ≤ 3s), and the redundancy channel switching efficiency. Verify the local caching and data resume transmission mechanisms of edge computing nodes (such as intelligent terminals) in the scenario of network disconnection to ensure that critical control instructions are not lost.

[0059] Among them, the comprehensive evaluation of heterogeneous data quality includes data integrity evaluation and consistency and timeliness verification; data integrity evaluation is to count the proportion of missing data nodes in the cross-section (required ≤ 0.1%), and focus on verifying the continuity of core parameters such as protection action signals and PMU dynamic phasors. Through the mapping relationship verification between the SCD configuration file and the real-time data cross-section, identify redundant or missing data items not defined according to the model.

[0060] Consistency and timeliness verification is to compare the data cross-section version consistency between the station control layer server and the interval layer devices to ensure the semantic alignment of the model file and the real-time data (such as the logical node naming rule). Based on the time window mechanism, evaluate the data cross-section update frequency, requiring that the refresh period of real-time control data ≤ 10ms and management data ≤ 1s.

[0061] The evaluation of heterogeneous data cross-sections in intelligent substations needs to integrate time synchronization, distributed architecture, and digital twin technologies. Through mathematical modeling verification, high-concurrency stress testing, and multi-dimensional quality analysis, ensure the accuracy, reliability, and real-time performance of data in the new power system. The evaluation results can provide core data support for the optimal dispatching and fault defense of the digital and intelligent strong grid.

[0062] As Figure 3 shown, the present invention also specifically provides: Data quality evaluation analyzes factors such as the health status of the power system intelligent substation system, system trends, familial defects, and equipment operation by measuring accuracy, control accuracy, telemetry accuracy, and multimedia data restoration in the concurrent heterogeneous data results. According to the analysis results, conduct early warning evaluation of the system state, perform abnormal diagnosis and analysis and display of the power automation system, count the defects of system modules in the integrated monitoring system, and sense and predict the risks that the power system intelligent substation system may pose to the power grid. Data quality evaluation can also evaluate the equipment quality of in-station station control equipment, secondary equipment, network equipment, and data networks.

[0063] As a further preferred embodiment, in the data quality evaluation of intelligent substations, the following is the mathematical expression framework based on the multi-factor comprehensive evaluation model. Combining the analytic hierarchy process (AHP) and the weighted fusion method, conduct data quality evaluation.

[0064] 1. Definition of factor importance weights First, weights are assigned to the core evaluation metrics (measurement accuracy, control accuracy, telemetry accuracy, multimedia data restoration) to reflect their contribution to the upper-level objectives (such as system health status, risk prediction). The Analytic Hierarchy Process (AHP) is used to determine the weights, and the steps are as follows: Step 1: Construct the judgment matrix Assume that the importance of the four core indicators is compared pairwise to form a judgment matrix A :

[0065] Among them, a ij represents the indicator i relative to the indicator j importance (for example, using the 1-9 scale method).

[0066] Step 2: Calculate the weight vector The weight vector is obtained by the eigenvalue method or the geometric mean method W = w 1, w 2, w 3, w 4], satisfying:

[0067] 2. Influence relationship of factors on system health status System health status H can be regarded as a comprehensive function of each evaluation index, considering the non-linear superposition effect: H = α × f (Measurement accuracy)+ β × g (Control accuracy)+ γ × h (Telemetry accuracy)+ δ × k (Multimedia restoration degree) Among them: α , β , γ , δ are weight coefficients (determined by AHP, such as α = w 1).

[0068] f , g , h , k is a normalization function that maps the original index to the interval [0, 1], for example:

[0069] 3. Association between Familial Defects and Equipment Operating Status Familial defects (caused by problems in batches of similar equipment) can quantify their relationship with the operating status of sub - equipment through cluster analysis and probability models:

[0070] Defect similarity: Calculated based on the cosine similarity of equipment models and failure modes.

[0071] Status deviation: The Euclidean distance between the current status of the equipment and the health benchmark.

[0072] 4. Dynamic Model for System Trend Prediction System trends (such as performance degradation) can establish ARIMA or LSTM prediction models through time - series analysis combined with evaluation metrics:

[0073] Where: is the predicted value of the system trend at the next moment.

[0074] is the regression coefficient, reflecting the influence intensity of each factor on the trend.

[0075] is the noise term.

[0076] 5. Joint Probability Model for Risk Perception System risk R can be modeled as a multi - factor joint probability distribution:

[0077] represents the conditional probability of a single indicator anomaly causing risk. For example: P (Control failure ∣ Control accuracy < θ ) = Total control times in history / Number of control failures.

[0078] 6. Unified Expression for Multi - dimensional Quality Analysis, Comprehensive Data Quality Score Q can be expressed as:

[0079] I i : Score of core indicators (such as measurement accuracy, etc.).

[0080] D j : Score of extended dimensions (such as data consistency, communication reliability).

[0081] λ : Influence coefficient of the extended dimension on the overall quality (to be determined by regression analysis).

[0082] Among them, the weight is dynamically adjusted: In practical applications, the weight w i can be adaptively updated according to real-time data or environmental changes (such as based on the entropy weight method). Nonlinear relationship modeling: For complex coupling relationships, neural networks or decision trees can be introduced to replace the linear combination. Verification method: The rationality of the model needs to be verified through Monte Carlo simulation or historical data backtesting. Example: Suppose the evaluation results of a certain substation are as follows: Measurement accuracy I 1 = 0.95, weight w 1 = 0.3; Control accuracy I 2 = 0.88, weight w 2 = 0.25; Telemetry accuracy I 3 = 0.92, weight w ¾ = 0.2; Multimedia restoration degree I 4 = 0.78, weight w 4 = 0.25; Then the comprehensive quality score: Q = 0.3×0.95 + 0.25×0.88 + 0.2×0.92 + 0.25×0.78 = 0.882 The scoring results can be used for horizontal comparison or to trigger an alarm (such as setting the threshold to Q <0.85 for alarm).

[0083] In view of the characteristics of large data throughput and many types of heterogeneous data in the intelligent substation of the power system, the present invention analyzes the possible weak points that may occur in the intelligent substation of the power system. Starting from the heterogeneous database, under the condition of running a large amount of heterogeneous data in the intelligent substation of the power system, the health status, familial defects, system aging trend, and operation conditions of each device of the system are evaluated.

[0084] Such as Figure 4 shown, the second object of the present invention is to provide a multi-source heterogeneous data quality evaluation system for an intelligent substation of a power system, including: a trimming module 100, a mirroring module 200, a concurrency module 300, and an evaluation module 400. The multi-source heterogeneous data quality evaluation system for an intelligent substation of the power system of the present invention is based on the multi-source heterogeneous data quality evaluation method for an intelligent substation of a power system.

[0085] The clipping module 100 is used to access the substation layer devices, the station control layer devices, and the data transmission module, obtain the model files accessed by the data communication gateway and the substation layer devices, and clip the model files. The mirroring module 200 is used to access the heterogeneous data transmission module of the station control layer, receive the heterogeneous data streams of the station control layer, and perform mirroring processing on all the heterogeneous data streams. The concurrency module 300 is used to perform data integration on the clipped model files and the mirrored heterogeneous data streams, and perform data concurrency after data integration. The evaluation module 400 is used to evaluate the concurrent data through the data section, including: evaluating the accuracy of heterogeneous data, the robustness of heterogeneous data concurrency, and the data quality of heterogeneous data for the data section at the same time point.

[0086] Among them, the data quality evaluation of the heterogeneous data includes: Comparing the data measurement accuracy, control accuracy, telemetry signal accuracy, and data source, and evaluating the accuracy and precision after a large amount of data throughput concurrency; comparing the transmitted pictures, graphics, audio-visual data, and multimedia data with the data source, and evaluating the restoration degree.

[0087] Analyze the health status, system trends, family defects, and operation factors of individual devices of the power system intelligent substation system through the measurement accuracy, control accuracy, telemetry accuracy, and multimedia data in the concurrent heterogeneous data results.

[0088] Perform early warning evaluation on the system state according to the analysis results, perform abnormal diagnosis analysis and display on the power automation system, count the defects of system modules in the integrated monitoring system, and perceive and predict the risks that the power system intelligent substation system may pose to the power grid.

[0089] Evaluate the equipment quality of the in-station station control equipment, secondary equipment, network equipment, and data network.

[0090] As Figure 5 shown, the third object of the embodiment of the present invention is to provide an electronic device, including a memory 701, a processor 702, and a computer program stored in the memory 701 and executable on the processor. When the processor executes the computer program, the above-mentioned power system intelligent substation multi-source heterogeneous data quality evaluation method is implemented. It also includes a communication interface 703 and a bus 704.

[0091] The fourth object of the embodiment of the present invention is to provide a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the above-mentioned power system intelligent substation multi-source heterogeneous data quality evaluation method is implemented.

[0092] The fifth objective of the embodiments of the present invention is to provide a computer program product, which includes computer instructions that direct a computer to execute the above-mentioned method for evaluating the quality of multi-source heterogeneous data in an intelligent substation of a power system.

[0093] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured article including an instruction device, and the instruction device implements the processes Figure 1 one process or multiple processes and / or blocks Figure 1 the functions specified in one block or multiple blocks.

[0094] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0095] The present invention can be implemented in the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can be implemented in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, readable storage media, optical memory, etc.) containing computer-usable program code.

[0096] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, as well as the combination of processes and / or blocks in the flowchart and / or block diagram. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0097] Obviously, the described embodiments are only partial embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0098] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: it is still possible to modify the specific implementation manners of the present invention or make equivalent substitutions, and any modification or equivalent substitution that does not depart from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. A method for evaluating the quality of multi - heterogeneous data in an intelligent substation of a power system, characterized in that, Including: Accessing the interval layer devices, substation control layer devices, and data transmission module, obtaining the model files accessed by the data communication network shutdown device and interval layer devices, and trimming the model files; Accessing the substation control layer heterogeneous data transmission module, receiving the heterogeneous data streams of the substation control layer, and mirroring all the heterogeneous data streams; Integrating the trimmed model files and the mirrored heterogeneous data streams, and performing data concurrency after data integration; Evaluating the concurrent data through the data section, including: evaluating the accuracy of heterogeneous data, the robustness of heterogeneous data concurrency, and the data quality of heterogeneous data for the data section at the same time point; 2. The method for evaluating the quality of multi-source heterogeneous data in an intelligent substation of a power system according to claim 1, wherein The device models accessed by the data communication network shutdown device and interval layer devices are designed to maximize the same level of the substation, and the data nodes of the device models trim the model files according to the maximum load-bearing capacity of the device performance.

3. The method for evaluating the quality of multi-source heterogeneous data in an intelligent substation of a power system according to claim 1, wherein The heterogeneous data streams include image files, audio files, video files, and miscellaneous irregular files.

4. The method for evaluating the quality of multi-source heterogeneous data in an intelligent substation of a power system according to claim 1, characterized in that The data quality evaluation of the heterogeneous data includes: Comparing the data measurement accuracy, control accuracy, telemetry accuracy, and data sources, and evaluating the accuracy and accuracy after a large amount of data throughput concurrency; comparing the transmitted pictures, graphics, sound, light, and electrical data, and multimedia data with the data sources, and evaluating the restoration degree.

5. The method for evaluating the quality of multi-source heterogeneous data in an intelligent substation of a power system according to claim 4, wherein, The data quality evaluation of the heterogeneous data also includes: Analyzing the health status, system trends, family defects, and operation factors of individual devices of the intelligent substation system of the power system through the measurement accuracy, control accuracy, telemetry accuracy, and multimedia data in the results of concurrent heterogeneous data.

6. The intelligent substation multi-source heterogeneous data quality assessment method for a power system according to claim 4, characterized in that The data quality evaluation of the heterogeneous data also includes: Conducting early warning evaluation on the system status according to the analysis results, analyzing and displaying the anomalies of the power automation system, counting the defects of system modules in the integrated monitoring system, and perceiving and predicting the risks that the intelligent substation system of the power system may pose to the power grid.

7. The method for evaluating the quality of multi - heterogeneous data of an intelligent substation in a power system according to claim 4, wherein The data quality evaluation of the heterogeneous data also includes: Evaluating the device quality of the substation control devices, secondary devices, network devices, and data networks in the station.

8. An intelligent substation multi-source heterogeneous data quality evaluation system for a power system, characterized in that, Including: A trimming module, which is used to access the interval layer devices, substation control layer devices, and data transmission module, obtain the model files accessed by the data communication network shutdown device and interval layer devices, and trim the model files; A mirroring module, which is used to access the substation control layer heterogeneous data transmission module, receive the heterogeneous data streams of the substation control layer, and mirror all the heterogeneous data streams; A concurrency module, which is used to integrate the trimmed model files and the mirrored heterogeneous data streams, and perform data concurrency after data integration; An evaluation module, which is used to evaluate the concurrent data through the data section, including: evaluating the accuracy of heterogeneous data, the robustness of heterogeneous data concurrency, and the data quality of heterogeneous data for the data section at the same time point.

9. An electronic device, characterized in that, Including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the method for evaluating the quality of multi-source heterogeneous data of the intelligent substation of the power system according to any one of claims 1-6.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the method for evaluating the quality of multi-source heterogeneous data in an intelligent substation of a power system according to any one of claims 1-6.

11. A computer program product, the computer program product comprising computer instructions, characterized in that, The computer instructions direct the computer to execute the method for evaluating the quality of multi-source heterogeneous data in an intelligent substation of a power system according to any one of claims 1-6.

Images