A method and system for evaluating the operation of relay protection equipment

Abstract

This invention provides a method and system for evaluating the operation of relay protection equipment, relating to the power system field. The method includes: using the current sensing data of the relay protection equipment as input, generating a predicted state value for the relay protection equipment based on a preset digital twin model; verifying the consistency between the predicted state value and the sensing data of the relay protection equipment at the next moment and outputting a verification report; determining the health state probability and network security risk value of the relay protection equipment based on the verification report; and determining the reliability of the relay protection equipment based on the health state probability and network security risk value. This method allows for real-time sensing of the dynamic changes of the relay protection equipment and state prediction, realistically reproducing various conditions of the relay protection equipment in actual operation, improving the realism and accuracy of the state prediction. Furthermore, by incorporating the sensing data of the relay protection equipment at the next moment to correct the state prediction value, it ensures the reliable operation of the power system.

Description

Technical Field

[0001] This invention relates to the field of power systems, and in particular to a method and system for evaluating the operation of relay protection equipment. Background Technology

[0002] Relay protection equipment is an important component of power systems, used to detect faults and quickly disconnect circuits to protect equipment and personnel. Relay protection devices typically consist of relays, measuring elements, logic decision elements, and execution elements. They monitor parameters such as current, voltage, and frequency of the power system in real time. When abnormal conditions such as short circuits or overloads occur, relay protection equipment can quickly trigger circuit breakers to trip, isolating the faulty area.

[0003] Evaluation of relay protection equipment is a crucial aspect of power system reliability management. It aims to comprehensively assess equipment performance, operational status, and management level to ensure reliable operation during power system faults. While related technologies utilize remote control systems, some relay protection functions rely on actual field conditions. Remote simulation testing may not fully reproduce real-world fault scenarios, leading to evaluation results that deviate from actual performance. In particular, the simulation of electrical quantities under fault scenarios is distorted. Because it fails to consider the dynamic changes in the relay equipment's operating status and environment, the evaluated electrical quantities are significantly overestimated, masking equipment faults and resulting in delayed maintenance and potential safety hazards. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a method and system for evaluating the operation of relay protection equipment, which solves the problem that remote simulation testing in the prior art may not be able to fully reproduce the real fault scenario, resulting in the evaluation results deviating from the actual performance.

[0005] Firstly, this application provides a method for evaluating the operation of relay protection equipment, including:

[0006] Using the current sensing data of the relay protection device as input, a state prediction value of the relay protection device is generated based on a preset digital twin model;

[0007] The consistency between the predicted state value and the sensing data of the relay protection device at the next moment is checked, and a check report is output.

[0008] The health status probability and network security risk value of the relay protection equipment are determined based on the calibration report.

[0009] The reliability of the relay protection device is determined based on the health status probability and the network security risk value.

[0010] In one embodiment, before generating the state prediction value of the relay protection device based on the preset digital twin model, the method further includes:

[0011] Based on the static data of the relay protection device, the device model library, and the corresponding image library, a knowledge graph of the relay protection device is determined;

[0012] The knowledge graph is semantically encoded to obtain the ontology model of the relay protection device;

[0013] Using historical perception data and historical state data as input, the ontology model is trained to obtain the digital twin model.

[0014] In one embodiment, the device model library includes physical and mechanistic models of relay protection devices; the image library includes device images of the relay protection devices; and the step of semantically encoding the knowledge graph to obtain the ontology model of the relay protection devices specifically includes:

[0015] The static data of the relay protection device is semantically matched with the physical parameters of the physical model to obtain a first semantic correspondence; the embedding vector of the static data and the physical parameters is determined according to the first semantic correspondence and denoted as the first embedding vector;

[0016] The static data of the relay protection device is semantically matched with the mechanism parameters of the mechanism model to obtain a second semantic correspondence; the embedding vector of the static data and the physical parameters is determined according to the second semantic correspondence and denoted as the second embedding vector;

[0017] The static data is mapped to a low-dimensional semantic space using word embedding to obtain a distribution vector of the static data, which is denoted as the first distribution vector. Feature extraction and semantic encoding are performed on the device image to obtain an encoded device image. The encoded device image is mapped to the low-dimensional semantic space corresponding to the static data to obtain a distribution vector of the device image, which is denoted as the second distribution vector. A semantic association matrix between the first distribution vector and the second distribution vector is determined, and an embedding vector between the static data and the device image is determined based on the semantic association matrix, which is denoted as the third embedding vector.

[0018] The first embedding vector, the second embedding vector, and the third embedding vector are used as real-valued weights to determine the target weighted group; the target weighted group is added to the knowledge graph to obtain the ontology model.

[0019] In one embodiment, training the ontology model and obtaining the digital twin model using historical perception data and historical state data as input specifically includes:

[0020] Based on the historical perception data, the fault operation scenario corresponding to the ontology model is determined from the preset fault database;

[0021] Using the historical perception data as input, the operating status data of the ontology model under the fault operation scenario is determined, and the ontology model is updated based on the similarity between the operating status data and the historical status data, and a new ontology model is generated.

[0022] When the similarity meets the preset update conditions, the corresponding ontology model is used as the digital twin model.

[0023] In one embodiment, the sensing data includes real-time electrical quantities, real-time mechanical quantities, and real-time network security data; the state prediction value includes electrical prediction quantities, mechanical prediction quantities, and network security prediction data; the consistency check between the state prediction value and the sensing data of the relay protection device at the next moment specifically includes:

[0024] Perform correlation checks and / or parameter deviation rate checks on the waveforms of the real-time electrical quantities and the predicted electrical quantities at the next time step; and / or

[0025] Perform synchronization checks and / or contact stroke error checks on the real-time mechanical quantity and the predicted mechanical quantity's action time at the next moment; and / or

[0026] The real-time network security data at the next moment will be compared with the network security prediction data to perform a communication link reliability check.

[0027] In one embodiment, the calibration report includes the electrical quantity deviation, mechanical quantity deviation, functional logic deviation, communication bit error rate, transmission delay, number of illegal messages, number of encryption failures, and message compliance rate of the relay protection device; determining the health status probability and network security risk value of the relay protection device based on the calibration report specifically includes:

[0028] Based on the electrical quantity deviation, the mechanical quantity deviation, and the functional logic deviation, the health status probability of the relay protection device is determined by a preset Bayesian classification model.

[0029] Based on preset network security rules, the risk weight of each of the following is determined: the communication error rate, the transmission delay, the number of illegal messages, the number of encryption failures, and the message compliance rate. The network security risk value is then determined by weighted summation based on the risk weights.

[0030] In one embodiment, determining the reliability of the relay protection device based on the health status probability and the network security risk value specifically includes:

[0031] The health status probability is converted into a basic health score of the relay protection device by weighted summation, and the basic health score is converted into a health correction score based on the predicted state transition function.

[0032] The basic network security score is obtained by subtracting the network security risk value from the preset maximum security score, and the basic network security score is converted into a network security correction score based on a preset penalty function.

[0033] A reliability score is obtained by weighted summation of the health correction score and the network security correction score.

[0034] In one embodiment, the expression for the state transition function is:

[0035]

[0036] in, Correcting the health score, To score for basic health, The probability of transitioning from state i to state j This is the trend sensitivity coefficient.

[0037] In one embodiment, the expression for the penalty function is:

[0038]

[0039]

[0040] in, Correcting the score for cybersecurity For basic cybersecurity scores, This is the penalty coefficient; The probability that the relay protection equipment is in a fault warning state. This represents the cybersecurity risk value.

[0041] In a second aspect, this application provides an operation evaluation system for relay protection equipment, characterized in that it includes a processor and a memory; wherein the memory stores a computer program, the computer program being loaded by the processor and executed as described in any one of the first aspects, an operation evaluation method for relay protection equipment.

[0042] In the operation evaluation method and system for relay protection equipment in this embodiment, the digital twin model can perceive the dynamic changes of the relay protection equipment in real time and predict its status, realistically reproducing various conditions of the relay protection equipment in actual operation, improving the authenticity and accuracy of the status prediction of the relay protection equipment. At the same time, the status prediction value is corrected by combining the perception data of the relay protection equipment at the next moment, which can effectively avoid the distortion of digital twin simulation from masking the real faults of the relay protection equipment. Furthermore, the reliability of the relay protection equipment is evaluated by combining the health status of the relay protection equipment and network security, effectively eliminating safety hazards and ensuring the reliable operation of the power system. Attached Figure Description

[0043] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a flowchart illustrating an operational evaluation method for relay protection equipment, provided as an embodiment.

[0045] Figure 2 This is a schematic diagram of the structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0046] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are merely some, not all, of the embodiments of the present invention. Based on the description of the present invention, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of the present invention.

[0047] In the description of this invention, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms according to the specific circumstances.

[0048] The terms “upper,” “lower,” “left,” “right,” “front,” “back,” “top,” “bottom,” “inner,” and “outer,” etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of the invention is in use. They are only for the convenience of description and simplification, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the present invention.

[0049] The terms “first,” “second,” “third,” etc., are used merely to distinguish elements with similar properties, not to indicate or imply relative importance or a specific order.

[0050] The terms “include,” “comprising,” or any other variation thereof are intended to cover non-exclusive inclusion, which includes not only the elements listed but also other elements not expressly listed.

[0051] like Figure 1 As shown, this embodiment provides a method for evaluating the operation of relay protection equipment, including:

[0052] Step S10: Using the current sensing data of the relay protection device as input, generate the state prediction value of the relay protection device based on the preset digital twin model;

[0053] Step S20: Perform a consistency check between the predicted state value and the sensing data of the relay protection device at the next moment and output a check report;

[0054] Step S30: Determine the health status probability and network security risk value of the relay protection equipment based on the calibration report;

[0055] Step S40: Determine the reliability of the relay protection device based on the health status probability and the network security risk value.

[0056] In the operation evaluation method for relay protection equipment in this embodiment, the digital twin model can perceive the dynamic changes of the relay protection equipment in real time and make state predictions, realistically reproducing various conditions of the relay protection equipment in actual operation, improving the authenticity and accuracy of the state prediction of the relay protection equipment. At the same time, the state prediction value is corrected by combining the perception data of the relay protection equipment at the next moment, which can effectively avoid the distortion of digital twin simulation from masking the real faults of the relay protection equipment. Furthermore, the reliability of the relay protection equipment is evaluated by combining the health status of the relay protection equipment and network security, effectively eliminating safety hazards and ensuring the reliable operation of the power system.

[0057] Step S10: Using the current sensing data of the relay protection device as input, generate the state prediction value of the relay protection device based on the preset digital twin model.

[0058] Sensing data refers to the real-time collection of equipment operating parameters and network security data by various sensors and network detection units deployed on relay protection equipment. Equipment operating parameters include electrical quantities (parameters reflecting the electrical characteristics of the equipment, such as current, voltage, power, frequency, and harmonic content) and mechanical quantities (operating parameters of the equipment's mechanical components, such as the travel, pressure, and action time of relay contacts, the opening and closing speed of circuit breakers, and the amplitude of mechanical vibration).

[0059] Digital twin models can accurately simulate various extreme fault scenarios and simulate the impact of relay protection equipment aging (such as relay contact wear and sensor accuracy degradation) on protection performance based on the current sensing data of relay protection equipment. Through real-time data driving, they can accurately reflect the operating characteristics of relay protection equipment, avoid errors caused by environmental factors during testing, discover potential defects in advance, and carry out timely safety maintenance of relay protection equipment.

[0060] Before generating the state prediction value of the relay protection device based on the preset digital twin model, the method further includes:

[0061] Step S101: Determine the knowledge graph of the relay protection device based on the static data, device model library, and corresponding image library of the relay protection device;

[0062] Step S102: Semantically encode the knowledge graph to obtain the ontology model of the relay protection device;

[0063] Step S103: Using historical perception data and historical state data as input, train the ontology model and obtain the digital twin model.

[0064] In step S101, static data refers to the basic data of the relay protection equipment during the design, manufacturing, and deployment stages, including basic equipment parameters, equipment manufacturing data, and power grid-related data. Basic equipment parameters include physical attributes such as rated voltage, rated current, rated frequency, dimensions, weight, and installation method; electrical parameters such as operating thresholds (overcurrent / overvoltage / zero-sequence current settings), return coefficients, operating time constants, and power consumption characteristics; and industry standards such as IEC 61850 and GB / T 14285. Equipment manufacturing data includes schematic diagrams such as secondary circuit wiring diagrams, logic control diagrams, and chip-level hardware architecture diagrams; material properties of core components (such as the silver alloy composition of relay contacts and FR-4 parameters of the PCB substrate); temperature resistance performance indicators; and manufacturer information such as equipment model, serial number, design life, warranty period, and firmware version number. The grid-related data includes access system parameters such as the voltage level (10kV / 110kV) and protection range (line / transformer / bus protection type), as well as network topology data such as the associated switchgear number, CT / PT ratio, and coordination relationship of adjacent protection devices.

[0065] The equipment model library contains standardized models of the geometric, physical, and logical characteristics of relay protection equipment and its associated systems, typically including physical models and mechanistic models. Physical models include the three-dimensional geometry of physical components such as relay housings, terminal blocks, and displays, and integrate finite element structures of parameters such as material mechanical properties (elastic modulus, Poisson's ratio), thermal conductivity, and electromagnetic shielding effectiveness. Mechanistic models integrate electrical characteristic units, dynamic characteristic units, logic protection units, and communication protocol units. Electrical characteristic units are based on Kirchhoff's laws and include equivalent circuit models, encompassing parameters such as coil inductance, contact resistance, and capacitor filtering circuit parameters. Dynamic characteristic units contain differential equations describing the operation of relay protection equipment (such as relay pull-in time-current curves and A / D conversion delay rules for sampling circuits). Logic protection units contain protection criterion execution logic rules described in state machine form. Communication protocol units integrate IEC 61850 communication service mapping protocol, GOOSE message transmission delay protocol, Modbus data interaction timing protocol, etc.

[0066] The image library contains equipment images of relay protection devices. The equipment images specifically include two-dimensional engineering drawings (such as circuit diagrams of the secondary circuit of a relay) and three-dimensional renderings (such as the internal armature movement mechanism of a relay), fault waveform diagrams (such as voltage / current waveform curves, phase vector diagrams, and spectrum analysis diagrams), and other visual graphics.

[0067] The knowledge graph of relay protection equipment is constructed using "entity-relationship-attribute" triples to build an intelligent knowledge base. This intelligent knowledge base consists of three parts: entity type, relationship definition, and construction method. For example, the entity type of the equipment is a relay, its relationship is defined as "relay-connection-contact group," and the construction method is "extracting the connection relationship between the relay and the contact group from technical manuals and fault reports using NLP technology." The knowledge graph of relay protection equipment can be constructed using "entity-relationship-attribute" triples.

[0068] In step S102, the semantic encoding of the knowledge graph to obtain the ontology model of the relay protection device specifically includes:

[0069] Step S1021: Semantically match the static data of the relay protection device with the physical parameters of the physical model to obtain a first semantic correspondence; determine the embedding vector of the static data and the physical parameters according to the first semantic correspondence and record it as the first embedding vector;

[0070] Step S1022: Semantically match the static data of the relay protection device with the mechanism parameters of the mechanism model to obtain a second semantic correspondence; determine the embedding vector of the static data and the physical parameters according to the second semantic correspondence and record it as the second embedding vector;

[0071] Step S1023: Map the static data to a low-dimensional semantic space using word embedding to obtain the distribution vector of the static data and denote it as the first distribution vector; perform feature extraction and semantic encoding on the device image to obtain an encoded device image; map the encoded device image to the low-dimensional semantic space corresponding to the static data to obtain the distribution vector of the device image and denote it as the second distribution vector; determine the semantic association matrix between the first distribution vector and the second distribution vector, and determine the embedding vector of the static data and the device image based on the semantic association matrix and denote it as the third embedding vector;

[0072] Step S1024: Use the first embedding vector, the second embedding vector, and the third embedding vector as real-valued weights to determine the target weighted group; add the target weighted group to the knowledge graph to obtain the ontology model.

[0073] In step S1021, the physical model parameters in the physical model are associated with the static data in the knowledge graph through semantic mapping rules. By querying synonyms and hyponyms defined in the semantic dictionary and integrating them into the knowledge graph, the first semantic correspondence between static data and physical model parameters can be established. For example, although the terms "rated current" in static data and "rated current value" in the physical model are slightly different, they have the same essential meaning. By using natural language processing technology combined with the semantic dictionary, the text of static data and physical parameters is segmented, part-of-speech tagging is performed, and named entity recognition is performed to associate concepts with the same semantic meaning, thereby obtaining the first semantic correspondence.

[0074] Embedded vectors are a way to convert textual information into numerical vector representations so that computers can process and analyze it. Based on the first semantic correspondence, word embedding models (such as Word2Vec, GloVe, etc.) are used to convert each word in the static data and physical parameters into a low-dimensional vector. For example, for the word "rated current," after processing by a word embedding model, a fixed-length vector is obtained. Then, these vectors are combined to obtain the embedded vector of the static data and physical parameters, i.e., the first embedded vector.

[0075] In step S1022, the mechanism model describes the working principle and internal mechanism of the relay protection device, and its mechanism parameters include operating time, operating threshold, protection logic, etc. Similarly, by semantically matching the static data with the mechanism parameters of the mechanism model, the correspondence between them can be found. For example, the "operating current setting value" in the static data and the "overcurrent protection operating current threshold" in the mechanism model are related concepts, and a second semantic correspondence is established through semantic matching. Similar to the method for determining the first embedding vector, based on the second semantic correspondence, the static data and mechanism parameters are converted into a second embedding vector using a word embedding model.

[0076] In step S103, the textual information in the static data is mapped to a low-dimensional semantic space using word embeddings. The word embedding model represents each word as a vector, and then obtains the distribution vector of the entire static data, i.e., the first distribution vector, by averaging and summing these vectors. This distribution vector reflects the semantic features of the static data and exhibits a certain distribution pattern in the low-dimensional space.

[0077] Computer vision techniques (such as convolutional neural networks, CNNs) can be used to extract features such as color, texture, and shape from images. These features are then semantically encoded to convert the image features into semantic information for text description (e.g., converting an image of a device into a text description such as "The casing of the relay protection device is square, and there are flashing indicator lights"). Next, word embedding processing is performed on these text descriptions, mapping them to the same low-dimensional semantic space as the static data to obtain the distribution vector of the device image, i.e., the second distribution vector.

[0078] The semantic association matrix between the first and second distribution vectors can be constructed by calculating the cosine similarity between the two vectors. Each element in the semantic association matrix represents the degree of semantic association between the static data and the device image in a certain dimension. Based on the semantic association matrix, and combined with the original vector information of the static data and the device image, the embedding vector of the static data and the device image, i.e., the third embedding vector, is determined.

[0079] In step S1024, the first, second, and third embedding vectors are used as real-valued weights, and the three embedding vectors are fused by weighted summation to obtain the target weighted group. Since the knowledge graph already contains information such as the entities, relationships, and attributes of the relay protection equipment, the addition of the target weighted group allows the knowledge graph to more accurately represent the semantic information and inherent relationships of the equipment. The constructed ontology model of the relay protection equipment can be used for knowledge reasoning, fault diagnosis, and state prediction, providing strong support for remote simulation testing and operation and maintenance management of the relay protection equipment. It is understandable that the weight of each embedding vector can be adjusted according to its importance in the model.

[0080] After obtaining the ontological model of the relay protection device, since the ontological model is built based on the static data of the relay protection device, in order to make the digital twin model more reflective of the recent operating status of the relay protection device, it is necessary to train the ontological model by combining the recent historical sensing data and historical status data of the relay protection device, so that the fault diagnosis and status prediction of the digital twin model can fit the recent status of the relay protection device.

[0081] Step S103: Using historical perception data and historical state data as input, train the ontology model and obtain the digital twin model.

[0082] Historical sensing data is typically collected from the relay protection equipment for the most recent week (or month). The sensing data is collected in real time by various sensors and network detection units deployed on the relay protection equipment to collect equipment operating parameters and network security data. Equipment operating parameters include electrical quantities (parameters reflecting the electrical characteristics of the equipment, such as current, voltage, power, frequency, and harmonic content) and mechanical quantities (operating parameters of the equipment's mechanical components, such as the travel, pressure, and action time of relay contacts, the opening and closing speed of circuit breakers, and the amplitude of mechanical vibration).

[0083] Historical status data is a tagged record of the equipment's historical operating status, including both normal operating status and abnormal statuses such as faults, defects, and warnings. Typically, status data includes three main types: operating status tags (e.g., normal operation, overload operation, abnormal temperature rise, protection malfunction / failure to operate, etc.), fault attributes (e.g., coil aging, poor contact, logic chip failure, fault occurrence time and repair records, etc.), and manually verified data (e.g., fault diagnosis conclusions, insulation resistance test values, operational characteristic verification data, etc.).

[0084] The process of training the ontology model and obtaining the digital twin model using historical perception data and historical state data as input specifically includes:

[0085] Step S1031: Based on the historical perception data, determine the fault operation scenario corresponding to the ontology model from the preset fault database;

[0086] Step S1032: Using the historical perception data as input, determine the operating status data of the ontology model under the fault operation scenario, and update the ontology model based on the similarity between the operating status data and the historical status data and generate a new ontology model.

[0087] Step S1033: When the similarity meets the preset update conditions, the corresponding ontology model is used as the digital twin model.

[0088] In step S1031, the fault database contains fault scenarios such as internal faults (e.g., protection device hardware failure, logic chip malfunction), external faults (e.g., transmission line short circuit, grounding fault), and communication faults (e.g., SV message interruption, GOOSE link malfunction). By matching the preprocessed historical sensing data with the feature vectors in the fault database using a pattern recognition algorithm (e.g., Support Vector Machine (SVM), Random Forest), the fault operation scenario corresponding to the current data can be identified.

[0089] In step S1032, the ontology model predicts the operational state data for the next moment based on the historical perception data from the previous moment. Then, it compares the historical state data for the next moment with the corresponding operational state data using similarity metrics (such as root mean square error or cosine similarity). If the similarity is below a threshold, it indicates a deviation in the model simulation. In this case, the trainable parameters in the ontology model (such as the error compensation coefficient in the mechanism model) need to be adjusted using the backpropagation algorithm; or manual intervention can be used to correct the knowledge graph of the ontology model (such as supplementing missing fault logic rules), and then semantic encoding can be performed again.

[0090] In step S1033, when the similarity is within the maximum threshold range or its value converges, it can be determined that the update condition has been met and the update of the ontology model is terminated. Taking the root mean square error (RMSE) as an example, if the RMSE of all key state parameters is less than the threshold of 3% in N consecutive iterations and the accuracy of the logical criterion is greater than 99.5%, the update of the ontology model is terminated, and the ontology model at this time is used as the digital twin model.

[0091] This embodiment uses historical data to drive the model to traverse a preset fault database and calibrates the model parameters based on historical state data to ensure that the digital twin model is highly consistent with the real equipment in terms of timing response and logic output. Furthermore, it continuously iterates the model using actual operating data to solve the characteristic deviation problem caused by long-term operation of relay protection equipment and improve the reliability of remote testing.

[0092] Step S20: Perform consistency verification between the predicted state value and the sensing data of the relay protection device at the next moment and output a verification report.

[0093] The consistency check between the predicted state value and the sensing data of the relay protection device at the next moment specifically includes: performing correlation check and / or parameter deviation rate check on the waveforms of the real-time electrical quantity and the predicted electrical quantity at the next moment; and / or performing synchronization check and / or contact travel error check on the action time of the real-time mechanical quantity and the predicted mechanical quantity at the next moment; and / or performing communication link reliability check on the real-time network security data and the predicted network security data at the next moment.

[0094] Correlation verification is a quantitative analysis of the similarity between real-time electrical quantities and the electrical quantity waveforms predicted by digital twin models, determining their consistency in morphology and trend. The overall waveform similarity can be calculated using the Pearson correlation coefficient, with a similarity value ranging from -1 to 1. The closer the absolute value is to 1, the stronger the correlation and the higher the degree of consistency. It is commonly used for consistency checks of waveform data such as voltage, current, and power.

[0095] Parameter deviation rate calibration assesses the degree of deviation of the state prediction values ​​by calculating the percentage difference between the sensed data and the state prediction values ​​predicted by the digital twin model. It is commonly used for consistency calibration of data such as protection action thresholds, sampling frequency, and signal delay time.

[0096] Synchronization verification targets the mechanical action time of relay protection equipment, verifying whether the real-time mechanical quantities are consistent with the action timing predicted by the model. It is commonly used for consistency checks of data such as circuit breaker opening and closing times and relay contact response times.

[0097] Contact stroke error calibration uses displacement sensors to monitor contact stroke or mechanism displacement in real time, verifies the displacement stroke of mechanical components, compares real-time stroke data with predicted theoretical stroke, and detects wear, jamming, or deformation of mechanical components. It is commonly used for consistency calibration of mechanical structures such as relay contacts and circuit breaker operating mechanisms.

[0098] Communication link reliability verification includes verification of basic indicators such as bit error rate, packet loss rate, and transmission delay, compliance verification, and anti-interference capability verification, assessing the stability, integrity, and anti-interference capability of the communication link. It is commonly used for consistency verification of communication data between protection devices and instrument transformers, switches, and back-end systems.

[0099] The calibration report includes basic information about the relay protection equipment, as well as the calibration purpose, calibration method, calibration results, and handling suggestions for abnormal calibration items for each type of calibration item.

[0100] Step S30: Determine the health status probability and network security risk value of the relay protection equipment based on the calibration report.

[0101] The calibration report includes the electrical quantity deviations (typically data from correlation and parameter deviation rate calibrations) of the relay protection equipment, mechanical quantity deviations (typically data from contact travel error calibrations), functional logic deviations (typically data from synchronization calibrations), communication error rate, transmission delay, number of illegal messages, number of encryption failures, and message compliance rate. By extracting and analyzing the calibration results of various items in the calibration report, the health status probability and network security risk value of the relay protection equipment can be assessed.

[0102] The determination of the health status probability and network security risk value of the relay protection equipment based on the calibration report specifically includes:

[0103] Step S301: Based on the electrical quantity deviation, the mechanical quantity deviation, and the functional logic deviation, determine the health status probability of the relay protection device using a preset Bayesian classification model;

[0104] Step S302: Based on preset network security rules, determine the risk weight of each of the following: communication error rate, transmission delay, number of illegal messages, number of encryption failures, and message compliance rate; and determine the network security risk value based on the risk weights by weighted summation.

[0105] In step S301, the calculation formula for the Bayesian classification model is as follows:

[0106]

[0107] in For the kth health state, This refers to electrical quantity deviation; This is for mechanical quantity deviation; This is due to a functional logic deviation. The prior probability of a healthy state (obtained from statistical analysis of historical operating data of relay protection equipment); For relay protection equipment to be in good condition Under these conditions, deviation characteristics appear. The probability, The health status is as follows: , and The probability (set according to the factory standard of the relay protection equipment), of which For a healthy state, In a state of alert, This indicates a fault condition.

[0108] In step S302, the network security rules set corresponding scoring levels for each of the following: communication error rate, transmission delay, number of illegal messages, number of encryption failures, and message compliance rate. Each scoring level has a different risk weight. The corresponding network security risk value can be determined by weighted summation of each risk weight.

[0109] Step S40: Determine the reliability of the relay protection device based on the health status probability and the network security risk value.

[0110] Because the "basic health score" only reflects a static assessment of the current or historical state of relay protection equipment, while the health status of relay protection equipment (such as aging and wear) is a gradual process with a certain time-series correlation, and the "cybersecurity risk value" is obtained linearly through weighted summation, the impact of cybersecurity events has non-linear characteristics (e.g., a single encryption failure may have no impact, but high-frequency encryption failures may lead to a sharp increase in the risk of data leakage), it is necessary to modify the basic health score and cybersecurity risk value to adapt to the dynamic evolution of the health status of relay protection equipment and cybersecurity, and to avoid assessment bias caused by relying solely on current data.

[0111] The determination of the reliability of the relay protection device based on the health status probability and the network security risk value specifically includes:

[0112] Step S401: Convert the health state probability into a basic health score of the relay protection device by weighted summation, and convert the basic health score into a health correction score based on the predicted state transition function;

[0113] Step S402: Subtract the network security risk value from the preset maximum security score to obtain the network security basic score, and convert the network security basic score into a network security correction score based on a preset penalty function;

[0114] Step S403: Obtain the reliability score by weighted summation based on the health correction score and the network security correction score.

[0115] In step S401, the categories of health status include (Health status) (Warning status) and (Fault status). The formula for the weighted summation method is: ,in The base scores for health status, early warning status, and fault status can be adaptively adjusted based on historical fault detection results and factory settings.

[0116] Then, a hidden Markov chain is used to introduce a state transition trend to adjust the health baseline score. The expression for the state transition function is:

[0117]

[0118] in, Correcting the health score, To score for basic health, The probability of transitioning from state i to state j This is the trend sensitivity coefficient (with a value range of 0.1-0.2).

[0119] In step S402, firstly through the formula The cybersecurity risk value (SR) is converted into a security baseline score (SCS), and then the security baseline score is corrected through a coupling effect. The expression for the penalty function is:

[0120]

[0121]

[0122] in, Correcting the score for cybersecurity For basic cybersecurity scores, This is the penalty coefficient; The probability that the relay protection equipment is in a fault warning state. This represents the cybersecurity risk value.

[0123] In step S403, the weighted summation formula for the reliability score is: , and These are the weighting coefficients for the health correction score and the cybersecurity correction score, respectively. Its value can be adaptively set according to requirements.

[0124] When the reliability score deviates significantly from actual historical experience, a hidden Markov chain can be further incorporated, using the reliability score as an observed variable to update the state transition matrix. :

[0125]

[0126] in, The reliability sensitivity coefficient typically ranges from 0.3. The lower the score, the higher the probability of transitioning to a failure state.

[0127] In summary, the operation evaluation method for relay protection equipment in this embodiment achieves two main benefits. First, by establishing a digital twin model, the dynamic changes of the relay protection equipment are perceived in real time and its status is predicted. This realistically reproduces various conditions of the relay protection equipment during actual operation, improving the authenticity and accuracy of the status prediction. Furthermore, by combining the perceived data of the relay protection equipment at the next moment with the status prediction value, the distortion of the digital twin simulation can be effectively avoided from masking the actual faults of the relay protection equipment, ensuring the reliable operation of the power system. Second, by combining the health status probability with the network security risk value, a multi-dimensional reliability assessment system is constructed. Even if the physical state of the equipment is normal, a network security warning will be triggered, preventing the protection function from failing due to network attacks. This aligns with the smart grid's requirement for "information-physical" integrated security.

[0128] Based on the same inventive concept as the above embodiments, this embodiment also provides an operation evaluation system for relay protection equipment, which further includes a processor and a memory; wherein, the memory stores a computer program, which is used by the processor to load and execute the operation evaluation method for relay protection equipment as described above.

[0129] like Figure 2 As shown, based on the same inventive concept as the above embodiments, this embodiment also provides a computer-readable storage medium storing instructions for loading and executing by a processor the above-described method for evaluating the operation of relay protection equipment.

[0130] The embodiments of the mobile terminal and computer-readable storage medium provided in this application include all the technical features of the embodiments of the above control method. The extended and explanatory content of the specification is basically the same as that of the embodiments of the above method, and will not be repeated here.

[0131] This application also provides a computer program product, which includes computer program code. When the computer program code is run on a computer, it causes the computer to perform the methods described in the various possible implementations above.

[0132] This application also provides a chip, including a memory and a processor. The memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that a device with the chip installed performs the methods described in the various possible implementations above.

[0133] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0134] In this application, the same or similar terms, concepts, technical solutions and / or application scenario descriptions are generally described in detail only when they appear for the first time. When they appear again, they are generally not repeated for the sake of brevity. When understanding the technical solutions and other contents of this application, the same or similar terms, concepts, technical solutions and / or application scenario descriptions that are not described in detail later can be referred to their previous relevant detailed descriptions.

[0135] In this application, the descriptions of the various embodiments have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0136] The technical features of the present application can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the present application.

[0137] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in the above-mentioned storage medium and includes several instructions to cause a terminal device to execute the methods of each embodiment of this application. The above are only preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made based on the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

[0138] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0139] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method for evaluating the operation of relay protection equipment, characterized in that, include: Using the current sensing data of the relay protection device as input, a state prediction value of the relay protection device is generated based on a preset digital twin model; The consistency between the predicted state value and the sensing data of the relay protection device at the next moment is checked, and a check report is output. The health status probability and network security risk value of the relay protection equipment are determined based on the calibration report. The reliability of the relay protection device is determined based on the health status probability and the network security risk value. Before generating the state prediction value of the relay protection device based on the preset digital twin model, the method further includes: Based on the static data of the relay protection device, the device model library, and the corresponding image library, a knowledge graph of the relay protection device is determined; The knowledge graph is semantically encoded to obtain the ontology model of the relay protection device; Using historical perception data and historical state data as input, the ontology model is trained and the digital twin model is obtained; The equipment model library includes physical and mechanistic models of relay protection equipment; the image library includes equipment images of the relay protection equipment; the semantic encoding of the knowledge graph to obtain the ontology model of the relay protection equipment specifically includes: The static data of the relay protection device is semantically matched with the physical parameters of the physical model to obtain a first semantic correspondence; the embedding vector of the static data and the physical parameters is determined according to the first semantic correspondence and denoted as the first embedding vector; The static data of the relay protection device is semantically matched with the mechanism parameters of the mechanism model to obtain a second semantic correspondence; the embedding vector of the static data and the physical parameters is determined according to the second semantic correspondence and denoted as the second embedding vector; The static data is mapped to a low-dimensional semantic space using word embedding to obtain a distribution vector of the static data, which is denoted as the first distribution vector. Feature extraction and semantic encoding are performed on the device image to obtain an encoded device image. The encoded device image is mapped to the low-dimensional semantic space corresponding to the static data to obtain a distribution vector of the device image, which is denoted as the second distribution vector. A semantic association matrix between the first distribution vector and the second distribution vector is determined, and an embedding vector between the static data and the device image is determined based on the semantic association matrix, which is denoted as the third embedding vector. The first embedding vector, the second embedding vector, and the third embedding vector are used as real-valued weights to determine the target weighted group; the target weighted group is added to the knowledge graph to obtain the ontology model.

2. The method for evaluating the operation of relay protection equipment according to claim 1, characterized in that, The process of training the ontology model and obtaining the digital twin model using historical perception data and historical state data as input specifically includes: Based on the historical perception data, the fault operation scenario corresponding to the ontology model is determined from the preset fault database; Using the historical perception data as input, the operating status data of the ontology model under the fault operation scenario is determined, and the ontology model is updated based on the similarity between the operating status data and the historical status data, and a new ontology model is generated. When the similarity meets the preset update conditions, the corresponding ontology model is used as the digital twin model.

3. The method for evaluating the operation of relay protection equipment according to claim 1, characterized in that, The sensing data includes real-time electrical quantities, real-time mechanical quantities, and real-time network security data; the state prediction values ​​include electrical prediction quantities, mechanical prediction quantities, and network security prediction data. The consistency check between the predicted state value and the sensing data of the relay protection device at the next moment specifically includes: Perform correlation and / or parameter deviation rate checks on the waveforms of the real-time electrical quantity and the predicted electrical quantity at the next moment; and / or Perform synchronization checks and / or contact stroke error checks on the timing of the real-time mechanical quantity and the predicted mechanical quantity at the next moment; and / or The real-time network security data at the next moment will be compared with the network security prediction data to perform a communication link reliability check.

4. The method for evaluating the operation of relay protection equipment according to claim 1, characterized in that, The calibration report includes the electrical quantity deviation, mechanical quantity deviation, functional logic deviation, communication bit error rate, transmission delay, number of illegal messages, number of encryption failures, and message compliance rate of the relay protection equipment. The determination of the health status probability and network security risk value of the relay protection equipment based on the calibration report specifically includes: Based on the electrical quantity deviation, the mechanical quantity deviation, and the functional logic deviation, the health status probability of the relay protection device is determined by a preset Bayesian classification model. Based on preset network security rules, the risk weight of each of the following is determined: the communication error rate, the transmission delay, the number of illegal messages, the number of encryption failures, and the message compliance rate. The network security risk value is then determined by weighted summation based on the risk weights.

5. A method for evaluating the operation of relay protection equipment according to any one of claims 1-4, characterized in that, The determination of the reliability of the relay protection device based on the health status probability and the network security risk value specifically includes: The health status probability is converted into a basic health score of the relay protection device by weighted summation, and the basic health score is converted into a health correction score based on the predicted state transition function. The basic network security score is obtained by subtracting the network security risk value from the preset maximum security score, and the basic network security score is converted into a network security correction score based on a preset penalty function. A reliability score is obtained by weighted summation of the health correction score and the network security correction score.

6. The method for evaluating the operation of relay protection equipment according to claim 5, characterized in that, The expression for the state transition function is: in, Correcting the health score, To score for basic health, The probability of transitioning from state i to state j This is the trend sensitivity coefficient.

7. The method for evaluating the operation of relay protection equipment according to claim 5, characterized in that, The expression for the penalty function is: in, Correcting the score for cybersecurity For basic cybersecurity scores, This is the penalty coefficient; The probability that the relay protection equipment is in a fault warning state. This represents the cybersecurity risk value.

8. An operation evaluation system for relay protection equipment, characterized in that, It includes a processor and a memory; wherein the memory stores a computer program for being loaded by the processor and executed as described in any one of claims 1-7, a method for evaluating the operation of relay protection equipment.

Images