Relay protection action behavior diagnosis method and system and medium

Abstract

The invention relates to a relay protection action behavior diagnosis method and system and a medium, and the method comprises the steps: constructing a relay protection action knowledge base according to a pre-obtained model spatialization circuit drawing, a historical fault type, a protection action configuration parameter and a correct protection action sequence; acquiring current protection action data through a relay protection device and an oscillograph; matching the current protection action data with a relay protection action knowledge base to obtain a first diagnosis result, and obtaining a second diagnosis result through a random forest algorithm according to the current protection action data; and determining a diagnosis result of the relay protection action behavior based on the first diagnosis result and the second diagnosis result. Diagnosis of the current protection action behavior is realized, and the problems of low diagnosis efficiency and low diagnosis accuracy in related technologies are solved. And the current protection action behavior diagnosis efficiency and diagnosis accuracy are improved.

Description

Technical Field

[0001] The present application relates to the field of relay diagnosis technology, and in particular to a method, system and medium for diagnosing relay protection action behavior. Background Art

[0002] As the scale of power systems continues to expand, the performance of relay protection devices directly affects whether the power system can detect and isolate faults in a timely and accurate manner, prevent accidents from expanding, and ensure the continuity and safety of power supply. Relay protection technology plays a vital role in ensuring the stable operation of power systems. Traditional relay protection action diagnosis relies on the experience and skills of professionals. This process is often time-consuming and has problems with low diagnostic efficiency and accuracy. Summary of the invention

[0003] The embodiments of the present application provide a method, system and medium for diagnosing relay protection action behavior, so as to at least solve the problems of low diagnostic efficiency and low diagnostic accuracy in the related art.

[0004] In a first aspect, an embodiment of the present application provides a method for diagnosing relay protection action behavior, the method comprising:

[0005] Construct a relay protection action knowledge base based on the pre-acquired model spatialized circuit drawings, historical fault types, protection action configuration parameters and correct protection action sequence;

[0006] Obtain current protection action data through relay protection devices and recorders;

[0007] Matching the current protection action data with the relay protection action knowledge base to obtain a first diagnosis result, and obtaining a second diagnosis result based on the current protection action data through a random forest algorithm;

[0008] Based on the first diagnosis result and the second diagnosis result, a diagnosis result of the relay protection action behavior is determined.

[0009] In one embodiment, before constructing a relay protection action knowledge base based on the pre-acquired model spatialized circuit drawings, historical fault types, protection action configuration parameters, and correct protection action sequences, the method further includes:

[0010] Obtaining a power plant circuit diagram, wherein the circuit diagram includes the structure and connection method of the power plant electrical system;

[0011] converting the circuit drawing into a digital format;

[0012] The circuit drawing converted into a digital format is parsed to obtain a model spatialized circuit drawing, wherein the model spatialized circuit drawing includes drawing information of electrical equipment, circuits, connection symbols and text labels.

[0013] In one embodiment, the construction of a relay protection action knowledge base based on the pre-acquired model spatialized circuit drawings, historical fault types, protection action configuration parameters and correct protection action sequence includes:

[0014] According to the historical fault data, the historical fault data is classified to obtain the historical fault type, wherein the historical fault type includes a short circuit fault, a ground fault and a disconnection fault, and characteristic information corresponding to each fault type is recorded;

[0015] Acquire standard protection action configuration parameters, wherein the standard protection action configuration parameters include the type, setting value, action logic and protection area of ​​each protection action;

[0016] According to the standard relay protection action, determine the correct protection action sequence, wherein the correct protection action sequence includes the coordination logic of the main protection and the backup protection;

[0017] The relay protection action knowledge base is determined based on the electrical equipment, lines, connection symbols and text labels, historical fault types, protection action configuration parameters and correct protection action sequence in the model spatialized circuit drawings.

[0018] In one embodiment, matching the current protection action data with the relay protection action knowledge base to obtain a first diagnosis result includes:

[0019] Extract voltage and current waveform information from current protection action data to determine the current fault type;

[0020] Comparing the current fault type with the historical fault types in the relay protection action knowledge base, and when the current fault type matches the historical fault type, determining that the current protection action data is abnormal;

[0021] Comparing the current protection action data with the protection action configuration parameters in the relay protection action knowledge base, and when the current protection action data meets the protection action configuration parameters, determining that the current protection action data is normal;

[0022] Comparing the current protection action data with the correct protection action sequence in the relay protection action knowledge base, and when the current protection action data conforms to the correct protection action sequence, determining that the current protection action data is normal;

[0023] The current protection action data is compared with the model spatialized circuit drawing information in the relay protection action knowledge base, and when the current protection action data conforms to the model spatialized circuit drawing information, it is determined that the current protection action data is normal.

[0024] In one embodiment, obtaining a second diagnosis result according to the current protection action data by using a random forest algorithm includes:

[0025] Dividing the current protection action data into a training set and a test set, wherein the training set is used for random forest model training, and the test set is used for evaluating the performance of the random forest model;

[0026] According to the training set data, a random forest model is trained, each decision tree classifies the training samples, and the category result of the current protection action is determined through a majority voting mechanism, and the category result is used as the second diagnosis result.

[0027] In one embodiment, determining the diagnosis result of the relay protection action behavior based on the first diagnosis result and the second diagnosis result includes:

[0028] Compare the first diagnosis result with the second diagnosis result to obtain a comparison result;

[0029] If the comparison result meets the preset requirements, the first diagnosis result or the second diagnosis result is used as the diagnosis result of the relay protection action behavior.

[0030] In one embodiment, after determining that the diagnosis result of the relay protection action behavior is abnormal, the method further includes:

[0031] The voltage and current waveform data collected by the fault recorder during the fault are obtained, and the fault distance is determined through the non-linear lumped parameter model.

[0032] In one embodiment, after determining the diagnosis result of the relay protection action behavior, the method further includes:

[0033] Based on the diagnosis results of the relay protection action behavior, a fault analysis report is generated, wherein the fault analysis report includes information such as fault type, duration, fault current, fault component, fault distance and whether the protection action is correct, and based on the corresponding fault information, fault resolution suggestions are generated, including fault investigation, equipment repair, equipment replacement and protection parameter adjustment measures.

[0034] In a second aspect, an embodiment of the present application provides a relay protection action behavior diagnosis system, the system comprising a module for constructing a relay protection action knowledge base, a current protection action data module, a first diagnosis result module and a second diagnosis result module, and a relay protection action diagnosis result module, wherein:

[0035] The module for constructing a relay protection action knowledge base is used to construct a relay protection action knowledge base according to the pre-acquired model spatialized circuit drawings, historical fault types, protection action configuration parameters and correct protection action sequence;

[0036] The current protection action data module is used to obtain the current protection action data through the relay protection device and the recorder;

[0037] The first diagnosis result and the second diagnosis result module are used to match the current protection action data with the relay protection action knowledge base to obtain a first diagnosis result, and obtain a second diagnosis result according to the current protection action data through a random forest algorithm;

[0038] The relay protection action diagnosis result module is used to determine the diagnosis result of the relay protection action behavior based on the first diagnosis result and the second diagnosis result.

[0039] In a third aspect, an embodiment of the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, a method for diagnosing relay protection action behavior as described in the first aspect above is implemented.

[0040] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for diagnosing relay protection action behavior as described in the first aspect above.

[0041] The embodiments of the present application provide a relay protection action behavior diagnosis method, system and medium which have at least the following technical effects.

[0042] By building a relay protection action knowledge base based on the pre-acquired model spatial circuit diagram, historical fault types, protection action configuration parameters and correct protection action sequence; obtaining the current protection action data through the relay protection device and the recorder; matching the current protection action data with the relay protection action knowledge base to obtain the first diagnosis result, and obtaining the second diagnosis result based on the current protection action data through the random forest algorithm; determining the diagnosis result of the relay protection action behavior based on the first diagnosis result and the second diagnosis result. The current protection action behavior is diagnosed, and the problems of low diagnostic efficiency and diagnostic accuracy in related technologies are solved. The current protection action behavior diagnosis efficiency and diagnostic accuracy are improved.

[0043] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0045] Figure 1 is a flow chart of a method for diagnosing relay protection action behavior according to an embodiment of the present application;

[0046] Figure 2 is a schematic diagram of a model spatialized circuit drawing according to an exemplary embodiment;

[0047] Figure 3 is a flowchart of step S101 according to an exemplary embodiment;

[0048] Figure 4 is a structural block diagram of a relay protection action behavior system according to an embodiment of the present application;

[0049] Figure 5 is a block diagram of an electronic device according to an exemplary embodiment;

[0050] Figure 6 The diagram is a schematic diagram of a single-loop double-terminal power supply circuit according to an exemplary embodiment. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0052] Obviously, the drawings described below are only some examples or embodiments of the present application. For ordinary technicians in this field, the present application can also be applied to other similar scenarios based on these drawings without creative work. In addition, it can also be understood that although the efforts made in this development process may be complicated and lengthy, for ordinary technicians in this field related to the content disclosed in this application, some changes in design, manufacturing or production based on the technical content disclosed in this application are just conventional technical means, and should not be understood as insufficient content disclosed in this application.

[0053] Reference to "embodiments" in this application means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those of ordinary skill in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0054] Unless otherwise defined, the technical terms or scientific terms involved in this application should be understood by people with ordinary skills in the technical field to which this application belongs. The words "one", "a", "a", "the" and the like involved in this application do not indicate a quantitative limitation, and may represent the singular or plural. The terms "include", "comprise", "have" and any of their variations involved in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and / or B" can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects before and after are in an "or" relationship. The terms "first", "second", "third", etc. involved in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0055] In a first aspect, the present application provides a method for diagnosing relay protection action behavior. Figure 1 It is a flow chart of relay protection action behavior diagnosis, such as Figure 1 As shown, a relay protection action behavior diagnosis method includes:

[0056] Step S101: construct a relay protection action knowledge base based on pre-acquired model spatialized circuit drawings, historical fault types, protection action configuration parameters and correct protection action sequences.

[0057] Step S102: Obtain current protection action data through the relay protection device and the recorder.

[0058] Step S103: Match the current protection action data with the relay protection action knowledge base to obtain a first diagnosis result, and obtain a second diagnosis result based on the current protection action data through a random forest algorithm.

[0059] Step S104: Determine a diagnosis result of the relay protection action behavior based on the first diagnosis result and the second diagnosis result.

[0060] In summary, the embodiment of the present application provides a method for diagnosing relay protection action behavior, by constructing a relay protection action knowledge base based on pre-acquired model spatialized circuit drawings, historical fault types, protection action configuration parameters and correct protection action sequence; obtaining current protection action data through relay protection devices and recorders; matching the current protection action data with the relay protection action knowledge base to obtain a first diagnostic result, and obtaining a second diagnostic result based on the current protection action data through a random forest algorithm; determining the diagnostic result of the relay protection action behavior based on the first diagnostic result and the second diagnostic result. The diagnosis of the current protection action behavior is achieved, and the problems of low diagnostic efficiency and diagnostic accuracy in related technologies are solved. The diagnostic efficiency and diagnostic accuracy of the current protection action behavior are improved.

[0061] In one embodiment, before constructing a relay protection action knowledge base based on the pre-acquired model spatialized circuit drawings, historical fault types, protection action configuration parameters, and correct protection action sequences, the method further includes:

[0062] Obtain the power plant circuit diagram, which includes the structure and connection method of the power plant electrical system;

[0063] Convert circuit drawings into digital format;

[0064] The circuit drawing converted into a digital format is parsed to obtain a model spatialized circuit drawing, wherein the model spatialized circuit drawing includes drawing information of electrical equipment, lines, connection symbols and text labels.

[0065] Optionally, first, obtain the circuit drawing data of the power plant, including information such as the main electrical equipment, line connections, and electrical parameters. These drawings record the structure and connection method of the power plant's electrical system. Then, convert the paper drawings into a digital format, such as an image file or a vector file, by scanning or manual input. Such digital processing can make circuit drawings easier to be processed and edited by computers. Subsequently, the digitized drawings are parsed and analyzed through image recognition technology or other image processing algorithms. This includes identifying elements such as electrical equipment, lines, electrical connection symbols, text labels, and extracting their location and attribute information. Next, the circuit drawings converted into a digital format are modeled spatially, and the drawings are converted into a usable knowledge reserve for intelligent analysis.

[0066] In order to spatialize the model of the circuit drawings after conversion into digital format, a sequence feature extraction model based on the recurrent neural network encoding and decoding framework is proposed. The recurrent neural network is an extension of the feedforward neural network. The recurrent neural network adds the connection edges of adjacent original nodes (also called loop edges), introducing the concept of time to the model. At any time i, the state value h of the hidden layer of the node connected by the loop edge (i) Will be affected by the value x of the current input node (i) and the hidden layer state value h at the previous moment (i-1) impact.

[0067] The feed-forward step from input to output can be represented by the following set of equations:

[0068] h (i) =tanh(Ux (i) +Wh (i-1) )

[0069]

[0070] Among them, U represents the weight matrix connecting the input layer and the hidden layer. W represents the weight matrix of the hidden layer at the current time point and the hidden layer at the adjacent time point, that is, the weight matrix of the loop edge, and V is the weight matrix connecting the hidden layer and the output layer. The recurrent neural network is a deep neural network. Each time point will form a network topology graph, and the network topology weights at each time point are shared. The weights of traditional neural networks are trained and updated using the back propagation algorithm, and the recurrent neural network in this embodiment introduces a time-based back propagation algorithm on this basis.

[0071] As the length of the original sequence increases, the number of network cycles increases and the network deepens, resulting in gradient vanishing or gradient exploding problems. For the gradient exploding problem, the gradient clipping strategy is used to solve it. That is, if the sum of the squares of the gradients of each weight in one iteration is greater than a certain threshold, then in order to avoid the weight change value being too large, a scaling factor (threshold divided by the sum of squares) is calculated based on the threshold and the sum of squares, and all gradients are multiplied by this factor. The gradient vanishing problem is difficult to solve. By improving the structure of the hidden layer of the recurrent neural network, the LSTM and GRU network structures are selected to alleviate and solve the gradient vanishing problem.

[0072] The YOLO network algorithm is used to divide the image into 7×7 grids and the image intersection over union (IOU) is calculated. IOU is the ratio of the intersection of the bounding box and the ground truth to the union of the two. The calculation formulas for confidence and IOU are as follows:

[0073]

[0074] P(Object) is the probability that the bounding box contains the target electrical component. If the predicted box and the true value overlap 100%, P(Object) = 1, and the confidence is 1; on the contrary, if there is no predicted box in the grid, the predicted box and the true value have no overlapping parts, P(Object) = 0, and the confidence is 0; BB gt is the target real area annotated based on the training samples, BB pt is the bounding box of the predicted target electrical component, and area is the area of ​​the specified image region. YOLO examines the entire image during training to learn the contextual information of the given category of its target electrical component and its background. There are 24 convolutional layers and 2 fully connected layers in the network structure of YOLO, which reduces the feature space generated by the previous layer by interspersing 1×1 convolutional layers.

[0075] The YOLO detection algorithm treats the detection of target objects directly as a regression problem of position coordinates and confidence scores. Therefore, the YOLO detection algorithm can predict the categories and bounding box positions of multiple target objects in real time at one time. Unlike traditional target detection algorithms that use sliding windows and Faster R-CNN series algorithms to extract candidate areas, the YOLO detection algorithm directly uses the entire image as input into the network model for training or detection, which greatly improves the training and detection speed of the network model.

[0076] Compared with other convolutional neural network structures that use sliding classifiers, YOLO is a unified network that can simultaneously predict the location, size, and category of a target. It is a real-time target detection system based on a deep convolutional neural network. Since the YOLO network has an end-to-end feature, from data input to result output, the entire training and detection process is completed within the network model, so it not only ensures accuracy, but also shows a faster detection speed.

[0077] By analyzing the identification area of ​​the primary wiring diagram, analyzing the wiring form and operation status of the interval, the calculation and analysis model is determined. The analysis model has four types according to the different on-site wiring forms: single mother, double mother, double mother and double segment, and 3 / 2 wiring. The embodiment of this application uses the double mother wiring form. Figure 2 1 is a schematic diagram of a model spatialized circuit drawing according to an exemplary embodiment, that is, a circuit drawing after paper drawings are parsed and converted into a digital format. The direction of the arrow is the direction of the current.

[0078] Based on the parsed circuit drawing data, models of electrical components are established. These models can be basic component models (such as resistors, capacitors, inductors, etc.) or models of specific equipment (such as transformers, generators, etc.). The model can include information such as component parameters, dimensions, characteristic curves, etc. Next, the connection relationship between components is established based on the connecting lines and connection symbols in the circuit drawing. These connection relationships include the connection relationship of nodes, the connection path of the line, and the special conditions at the connection point. Subsequently, according to the actual situation, the parameter values ​​and equipment characteristics of the electrical components are entered. These parameters may include resistance values, capacitance values, inductance values, etc., as well as the rated power and rated voltage of the equipment. Finally, the obtained model spatialized circuit drawings are verified and optimized. The accuracy and reliability of the model spatialized circuit drawings are verified by performing simulation calculations and comparing actual data, and the model is adjusted and optimized as needed.

[0079] Through the modeling of power plant circuit drawings, tasks such as power system analysis, short circuit calculation, and fault analysis can be more conveniently performed. At the same time, the modeled circuit drawings can be integrated with other software tools to better support the operation and management of the power system. The key nodes of line sequence changes obtained through intelligent analysis of the drawings are compared with the actual situation, anomaly detection and identification are performed, and online diagnosis and alarm are performed on the secondary circuit of the relay protection.

[0080] Figure 3 is a flowchart of step S101 according to an exemplary embodiment. Figure 3 As shown, step S101, based on the pre-acquired model spatialized circuit drawings, historical fault types, protection action configuration parameters and correct protection action sequence, construct a relay protection action knowledge base, specifically including the following steps:

[0081] Step S1011: classify the historical fault data according to the historical fault data, obtain the historical fault type, wherein the historical fault type includes short circuit fault, ground fault and disconnection fault, and record the characteristic information corresponding to each fault type;

[0082] Step S1012: acquiring standard protection action configuration parameters, wherein the standard protection action configuration parameters include the type, setting value, action logic and protection area of ​​each protection action;

[0083] Step S1013: Determine a correct protection action sequence according to standard relay protection actions, where the correct protection action sequence includes the coordination logic of the main protection and the backup protection;

[0084] Step S1014: Determine a relay protection action knowledge base based on the electrical equipment, lines, connection symbols and text labels, historical fault types, protection action configuration parameters and correct protection action sequence in the model spatialized circuit drawing.

[0085] Optionally, firstly, by analyzing the historical fault data, classify it into short circuit fault, ground fault and disconnection fault, and record the characteristic information of each fault, such as fault current, voltage, duration, etc. This step is based on data mining and pattern recognition technology, using the rules in the historical data to identify and classify the fault type. Subsequently, collect and organize the type of each protection action (such as differential protection, overcurrent protection), setting value (such as protection setting), action logic (such as delay setting) and protection area (such as line, transformer). Ensure the accuracy and effectiveness of the protection configuration. Next, according to the operation standard of the power system and the principle of relay protection, determine the coordination logic of the main protection and backup protection. This step is based on the theory and practical experience of the power system to ensure that the protection system can act correctly in a predetermined order and logic under various fault conditions. Finally, the information in the spatialized circuit drawing of the comprehensive model (electrical equipment, lines, connection symbols and text labels), historical fault types, protection action configuration parameters and correct protection action sequence are constructed to build a relay protection action knowledge base.

[0086] Step S101: organize the scattered information into an orderly knowledge base to provide decision support for the intelligent diagnosis system, realize intelligent and automated diagnosis and analysis, and improve operation efficiency.

[0087] In one embodiment, step S101, obtaining current protection action data through a relay protection device and a recorder, specifically includes:

[0088] The recorder is a special device used to record the transient changes of voltage and current in the power system. When a fault occurs, the recorder will capture detailed waveform data, including voltage and current changes before and after the fault, as well as parameters such as frequency and phase. The current protection action data includes protection action events: including action information such as tripping, resetting, and reclosing of the protection device, as well as related timestamps. Waveform data: recorded current and voltage waveforms, which can be used to analyze the nature, severity, and location of the fault point. Switching information: state change information of protection devices and circuit breakers, such as the tripping and closing positions of circuit breakers. Protection configuration parameters: setting values ​​of protection devices, such as fixed values, delays, etc., which are crucial to determine the correctness of protection actions. When a fault occurs in the power system, the relay protection device will detect the abnormality and immediately initiate protection actions, such as tripping and alarming. At the same time, it will record the current and voltage waveforms when the fault occurs, and determine the type, time, and duration of the fault based on the current and voltage waveforms when the fault occurs.

[0089] Step S101: Acquire current protection action data through relay protection devices and recorders, so as to quickly and accurately locate the cause of the fault and determine the correctness of the protection action.

[0090] In one embodiment, the step S103 of matching the current protection action data with the relay protection action knowledge base to obtain the first diagnosis result includes:

[0091] Extract voltage and current waveform information from current protection action data to determine the current fault type;

[0092] Compare the current fault type with the historical fault types in the relay protection action knowledge base. When the current fault type matches the historical fault type, determine that the current protection action data is abnormal;

[0093] Compare the current protection action data with the protection action configuration parameters in the relay protection action knowledge base, and when the current protection action data meets the protection action configuration parameters, determine that the current protection action data is normal;

[0094] Compare the current protection action data with the correct protection action sequence in the relay protection action knowledge base, and when the current protection action data conforms to the correct protection action sequence, determine that the current protection action data is normal;

[0095] The current protection action data is compared with the model spatialized circuit drawing information in the relay protection action knowledge base. When the current protection action data conforms to the model spatialized circuit drawing information, it is determined that the current protection action data is normal.

[0096] Optionally, voltage and current waveform information is extracted from the current protection action data, and the current fault type, such as short circuit, grounding or disconnection fault, is identified by waveform characteristics (such as waveform distortion, peak value, frequency change, etc.). The identified current fault type is compared with the historical fault type in the relay protection action knowledge base. If the current fault type is consistent with the abnormal type recorded in the historical database, the current protection action data is judged to be abnormal. The protection action type, action setting, action time, etc. recorded in the current protection action data are compared with the standard protection action configuration parameters in the knowledge base. If the current data is consistent with the configuration parameters, it means that the protection action is executed within the expected parameter range. Otherwise, it means that there is a problem with the protection action configuration or the fault condition exceeds expectations. The main protection and backup protection should act in a predetermined logical order. If the current protection action data follows the correct protection action sequence defined in the knowledge base, this indicates that the protection system is effective and selective in fault handling. A sequence that does not meet expectations may indicate an error in the protection system configuration or the complexity of the fault. Compare the current protection action data with the information on electrical equipment, line connections, protection areas, etc. in the model spatialized circuit drawings to confirm whether the protection action is consistent with the actual system layout. If the protection action data is consistent with the drawing information, it indicates that the protection action is performed in the correct physical location and electrical environment.

[0097] The diagnosis process in step S103 not only improves the efficiency and accuracy of fault diagnosis, but also promotes the intelligent operation and maintenance of the power system, and provides strong technical support for the stable operation of the power system.

[0098] In one embodiment, the step S103 of obtaining the second diagnosis result according to the current protection action data by using a random forest algorithm includes:

[0099] The current protection action data is divided into a training set and a test set. The training set is used for random forest model training, and the test set is used to evaluate the performance of the random forest model.

[0100] According to the training set data, the random forest model is trained. Each decision tree classifies the training samples. The category result of the current protection action is determined through the majority voting mechanism, and the category result is used as the second diagnostic result.

[0101] Optionally, first, the collected current protection action data set is divided into a training set and a test set. The training set is used to train the random forest model, while the test set is used to evaluate the performance of the model. A portion of features is randomly extracted from the training set and used to build each decision tree. The purpose of this is to reduce the correlation between decision trees and thus improve the generalization ability of the model. Using the training set data, multiple decision trees are built based on randomly extracted features. Each tree is trained independently, using different features and data subsets. For a new protection action data (test data), each tree will give a classification result. The random forest model will adopt a majority voting mechanism, that is, the results given by most decision trees are used as the final classification result, which is the second diagnostic result.

[0102] In step S103, by applying the random forest algorithm to the diagnosis of relay protection action data, not only the accuracy of diagnosis can be improved, but also the robustness of the model can be enhanced, and feature importance analysis can be provided, which helps power system maintenance personnel to quickly and accurately locate and solve problems.

[0103] In one embodiment, step S104, based on the first diagnosis result and the second diagnosis result, determines the diagnosis result of the relay protection action behavior. Specifically includes:

[0104] Compare the first diagnosis result with the second diagnosis result to obtain a comparison result;

[0105] If the comparison result meets the preset requirements, the first diagnosis result or the second diagnosis result is used as the diagnosis result of the relay protection action behavior.

[0106] Optionally, first, the diagnostic result based on the expert knowledge base (first diagnostic result) is compared with the diagnostic result based on the random forest algorithm (second diagnostic result). The purpose of this step is to verify the consistency and reliability of the two independent diagnostic methods. Check whether the two diagnostic results meet the preset requirements. The preset requirements include a consistency threshold for the diagnostic results. For example, if the similarity of the two results is within the threshold range, they are considered to be consistent, and vice versa. If the comparison results show that the two diagnostic results are consistent or meet the preset conditions, one of the results can be used as the final diagnostic result of the relay protection action behavior. If there is a significant difference between the two results, further analysis of the cause may be required, which may be a data quality problem, model bias, or uncertainty in a specific fault situation. In this case, manual intervention may be required for review, or a combination of more information sources and diagnostic techniques may be used to make a final decision.

[0107] Step S104: The first diagnosis result is combined with the second diagnosis result, thereby improving the comprehensiveness and reliability of the diagnosis.

[0108] In one embodiment, after determining that the diagnosis result of the relay protection action behavior is abnormal, the method further includes:

[0109] The voltage and current waveform data collected by the fault recorder during the fault are obtained, and the fault distance is determined through the non-linear lumped parameter model.

[0110] Optionally, the line adopts a non-linear lumped parameter model. Assume that the line is uniform, the resistance, inductance, and distributed capacitance to the ground per unit length are r, l, and c respectively, and the total length of the line is L. Assume that a ground short circuit occurs at point F at the busbar name on the M side, and the transition resistance is Rg. The voltage and current time domain signals on both sides measured from busbars M and N are uM iM and uN iN respectively. The voltage time domain signal at the fault point is uF, and the current time domain signal flowing through the transition resistance is iF.

[0111] The KCL equation for the fault point F is:

[0112] i1+i4-i3-i5-i F =0 (10-1)

[0113]

[0114] The KVL equations for the loops of the non-type lumped parameter model are:

[0115]

[0116] The expression of fault distance can be written from formula (10-1):

[0117]

[0118] Substituting formula (10-4) into formula (10-2) and (10-3), we can obtain:

[0119]

[0120] Formula (10-5) contains two equations, corresponding to two unknowns uF and iF. Since the unknown Ge has been eliminated by X, the solution process is not affected by the fault location result X. Substituting the voltage and current time domain signals on both sides of the line into formula (10-5), the unknowns uF and iF can be obtained, and the transition resistance is:

[0121] Rg=uF / iF

[0122] B. Impedance Variance

[0123] The impedance is calculated every half cycle, and the impedance value is subjected to variance calculation.

[0124] Single-phase grounding impedance:

[0125]

[0126] Where: Z A , Z B , Z C Indicates the measured impedance of each phase; is the phase voltage at the beginning of the line; is the phase current at the beginning of the line; z1 is the positive sequence impedance per unit length of the line, z0 is the zero sequence impedance per unit length of the line; K is called the zero sequence compensation coefficient:

[0127] Phase-to-phase and three-phase short-circuit impedance:

[0128]

[0129] The impedance value x is taken every half cycle, and then the impedance variance during the fault duration of phase A, phase B, phase C, phase AB, phase AC, and phase BC is calculated respectively.

[0130] Average:

[0131] variance:

[0132] Figure 6 is a schematic diagram of a single-loop double-terminal power supply circuit according to an exemplary embodiment. Figure 6 As shown, the fault distance is calculated, taking a single-loop double-terminal power line as an example.

[0133] In the figure, the total length of the line is L, and a fault occurs at a distance of M, Lf from the busbar. The transition resistance is RF The device is installed on the bus M side. The task of fault distance measurement is to calculate the fault distance Lf through the known quantities (the three-phase voltage and current of the fault line measured by the device). Assuming that the distributed capacitance of the measured transmission line can be ignored, the line section from the fault point to the installation location of the device can be represented by a resistance-inductance series circuit.

[0134] Take phase A ground fault as an example. is the A-phase bus voltage, Measure the current for phase A, is the zero sequence current, is the fault component of the measured current of phase A, R F is the transition resistance, P is the percentage of fault distance, Z L is the positive sequence impedance of the entire line length, then the line impedance between the busbar and the fault point during a fault is: Z L P, the voltage drop across the transition resistor is I f R F , the following formula holds:

[0135] U A =(I A +K N I0)Z L P+L F R F

[0136] Where U A ,I A , K N (Zero sequence current compensation coefficient )、I0、Z L are known quantities, fault distance percentage P, I f , R F is an unknown quantity, the key is to find the percentage of fault distance. f It is the sum of the fault current at this end and the fault current at the other end, so it cannot be obtained directly. What can be obtained is the fault current at this end According to the analysis of each sequence network after the fault, Where DA is the current distribution coefficient, Z A , Z B are the impedance of the local system and the impedance of the opposite system respectively. So when a fault occurs, the following equation holds true:

[0137] U A =(I A +K N I0)Z L P+(L FA / D A )R F (10-6)

[0138] Among them, I FA is the fault current component at this end. There are two unknown quantities in formula (10-6), namely, the fault distance percentage P and the transition resistance R F D A Substituting the expression into formula (10-6), we can get a quadratic equation for the fault distance percentage P:

[0139] U A =(I A +K N I0)Z L P+I FA (Z L +Z A +Z B )R F / ((1-P)Z L +Z B )

[0140] Simplifying this formula, we can get the following formula:

[0141] P 2 -K1P+K2-K3R F =0 (10-7)

[0142] The expressions of coefficients K1, K2, and K3 in the equation are:

[0143]

[0144]

[0145] Expand the equation into real and imaginary parts, eliminating the transition resistance R F We can get:

[0146]

[0147] Solving this equation gives two roots. The root greater than 0 and less than 1 is the actual fault distance percentage. Multiplying this root by the total line length gives the actual fault distance.

[0148] By determining the fault distance, it provides technical support for the rapid restoration of the power system. It is worth noting that the accuracy of fault distance measurement is affected by many factors, including data quality, accuracy of line parameters, and complexity of algorithms. Therefore, in practical applications, multiple distance measurement methods and technologies are usually combined to improve the accuracy and reliability of distance measurement.

[0149] In one embodiment, fault analysis can be performed on the collected protection and recording data, including: drawing curves according to sampling points, drawing vector diagrams, drawing impedance trajectories, calculating sequence components, harmonic analysis, etc. Channels can be added and deleted, waveforms can be enlarged and reduced, waveform display values ​​can be switched, and analysis reports can be printed out. Recording data can be imported and exported, channels and time periods can be selected, and distance measurement functions are provided. It supports the display of dual cursors, realizing cursor display and hiding functions; it supports the correct display of variable frequency segmented recording files; it supports the formula editor for vector synthesis calculations.

[0150] In one embodiment, after determining the diagnosis result of the relay protection action behavior, the method further includes:

[0151] Record the solutions and measures taken for each fault event, including troubleshooting, equipment repair or replacement, protection parameter adjustment, etc. These solutions can be used as a reference for handling similar fault events in the future.

[0152] In one embodiment, after determining the diagnosis result of the relay protection action behavior, the method further includes:

[0153] Generate corresponding reports and statistics based on archived fault event data, such as fault frequency, fault type distribution, equipment failure rate, etc. These reports and statistics can provide information about system health and equipment operation status for decision making and improvement measures.

[0154] In one embodiment, after determining the diagnosis result of the relay protection action behavior, the method further includes:

[0155] According to the correlation between primary and secondary equipment, the received real-time protection events, fault characteristics, protection recording files and recorder recording file information are automatically screened and filtered, and the real protection tripping information, protection recording and recorder recording files related to a single fault are selected and sorted to form an original action report.

[0156] In one embodiment, after determining the diagnosis result of the relay protection action behavior, the method further includes:

[0157] Based on the diagnosis results of the relay protection action behavior, a fault analysis report is generated, which includes information such as fault type, duration, fault current, fault component, fault distance and whether the protection action is correct. Based on the corresponding fault information, fault resolution suggestions are generated, including fault investigation, equipment repair, equipment replacement and protection parameter adjustment measures.

[0158] In summary, the embodiment of the present application provides a method for diagnosing relay protection action behavior, by constructing a relay protection action knowledge base based on pre-acquired model spatialized circuit drawings, historical fault types, protection action configuration parameters and correct protection action sequence; obtaining current protection action data through relay protection devices and recorders; matching the current protection action data with the relay protection action knowledge base to obtain a first diagnostic result, and obtaining a second diagnostic result based on the current protection action data through a random forest algorithm; determining the diagnostic result of the relay protection action behavior based on the first diagnostic result and the second diagnostic result. The diagnosis of the current protection action behavior is achieved, and the problems of low diagnostic efficiency and diagnostic accuracy in related technologies are solved. The diagnostic efficiency and diagnostic accuracy of the current protection action behavior are improved.

[0159] In a second aspect, an embodiment of the present application provides a system for diagnosing relay protection action behavior. Figure 4 FIG. 1 is a block diagram of a system for diagnosing relay protection action behavior according to an exemplary embodiment. Figure 4 As shown, the system includes a module 410 for constructing a relay protection action knowledge base, a current protection action data module 420, a first diagnosis result and a second diagnosis result module 430, and a relay protection action diagnosis result module 440, wherein:

[0160] A relay protection action knowledge base building module 410 is used to build a relay protection action knowledge base according to the pre-acquired model spatialized circuit drawings, historical fault types, protection action configuration parameters and correct protection action sequence;

[0161] The current protection action data module 420 is used to obtain the current protection action data through the relay protection device and the recorder;

[0162] The first diagnosis result and the second diagnosis result module 430 are used to match the current protection action data with the relay protection action knowledge base to obtain the first diagnosis result, and obtain the second diagnosis result according to the current protection action data through the random forest algorithm;

[0163] The relay protection action diagnosis result module 440 is used to determine the diagnosis result of the relay protection action behavior based on the first diagnosis result and the second diagnosis result.

[0164] In summary, the embodiment of the present application provides a method for diagnosing relay protection action behavior, which solves the problems of low diagnostic efficiency and low diagnostic accuracy in related technologies by constructing a relay protection action knowledge base module 410, a current protection action data module 420, a first diagnostic result and a second diagnostic result module 430, and a diagnostic result module 440 of a relay protection action. Specifically, a relay protection action knowledge base is constructed according to a pre-acquired model spatial circuit drawing, a historical fault type, a protection action configuration parameter, and a correct protection action sequence; the current protection action data is obtained through a relay protection device and a recorder; the current protection action data is matched with the relay protection action knowledge base to obtain a first diagnostic result, and the second diagnostic result is obtained through a random forest algorithm according to the current protection action data; based on the first diagnostic result and the second diagnostic result, the diagnostic result of the relay protection action behavior is determined. The diagnosis of the current protection action behavior is realized, and the problems of low diagnostic efficiency and low diagnostic accuracy in related technologies are solved. The diagnostic efficiency and diagnostic accuracy of the current protection action behavior are improved.

[0165] It should be noted that the system for diagnosing the action behavior of relay protection provided in this embodiment is used to implement the above-mentioned implementation mode, and the descriptions that have been made will not be repeated. As used above, the terms "module", "unit", "subunit", etc. can implement a combination of software and / or hardware for a predetermined function. Although the device described in the above embodiment is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0166] In a third aspect, an embodiment of the present application provides an electronic device, Figure 5 FIG. 1 is a block diagram of an electronic device according to an exemplary embodiment. Figure 5 As shown, the electronic device may include a processor 51 and a memory 52 storing computer program instructions.

[0167] Specifically, the processor 51 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0168] Among them, the memory 52 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 52 may include a hard disk drive (HDD), a floppy disk drive, a solid state drive (SSD), a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory 52 may include a removable or non-removable (or fixed) medium. Where appropriate, the memory 52 may be inside or outside the data processing device. In a specific embodiment, the memory 52 is a non-volatile memory. In a specific embodiment, the memory 52 includes a read-only memory (ROM) and a random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (Programmable Read-Only Memory, PROM for short), an erasable PROM (Erasable ProgrammableRead-Only Memory, EPROM for short), an electrically erasable PROM (Electrically Erasable ProgrammableRead-Only Memory, EEPROM for short), an electrically alterable ROM (Electrically Alterable Read-Only Memory, EAROM for short) or a flash memory (FLASH) or a combination of two or more of these. Under appropriate circumstances, the RAM can be a static random access memory (SRAM) or a dynamic random access memory (DRAM), wherein the DRAM can be a fast page mode dynamic random access memory (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0169] The memory 52 may be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor 51 .

[0170] The processor 51 implements any one of the relay protection action behavior diagnosis methods in the above embodiments by reading and executing computer program instructions stored in the memory 52 .

[0171] In one embodiment, a device for diagnosing relay protection action behavior may further include a communication interface 53 and a bus 50. Figure 5 As shown, the processor 51, the memory 52, and the communication interface 53 are connected via a bus 50 and communicate with each other.

[0172] The communication interface 53 is used to implement communication between the modules, devices, units and / or equipment in the embodiment of the present application. The communication port 53 can also implement data communication with other components such as: external devices, image / data acquisition equipment, databases, external storage, and image / data processing workstations.

[0173] The bus 50 includes hardware, software or both, and couples the components of a device for relay protection action behavior diagnosis to each other. The bus 50 includes but is not limited to at least one of the following: data bus, address bus, control bus, expansion bus, and local bus. By way of example and not limitation, bus 50 may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an InfiniBand interconnect, a Low Pin Count (LPC) bus, a memory bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses or a combination of two or more of these. Where appropriate, bus 50 may include one or more buses. Although embodiments of the present application describe and illustrate a particular bus, the present application contemplates any suitable bus or interconnect.

[0174] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a program stored thereon, and when the program is executed by a processor, a relay protection action behavior diagnosis method provided in the first aspect is implemented.

[0175] The readable storage medium may include but is not limited to: a portable disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical storage device, a magnetic storage device or any suitable combination of the above.

[0176] In a possible implementation, the present invention can also be implemented in the form of a program product, which includes program code. When the program product runs on a terminal device, the program code is used to enable the terminal device to execute the steps of a relay protection action behavior diagnosis method provided in the first aspect.

[0177] Among them, the program code for executing the present invention can be written in any combination of one or more programming languages, and the program code can be executed completely on the user device, partially on the user device, as an independent software package, partially on the user device and partially on a remote device, or completely on the remote device.

[0178] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, 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 this specification.

[0179] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A method for diagnosing relay protection action behavior, characterized in that: The method comprises: Construct a relay protection action knowledge base based on the pre-acquired model spatial circuit drawings, historical fault types, protection action configuration parameters and correct protection action sequence; Obtain current protection action data through relay protection devices and recorders; Matching the current protection action data with the relay protection action knowledge base to obtain a first diagnosis result, and obtaining a second diagnosis result based on the current protection action data through a random forest algorithm; Based on the first diagnosis result and the second diagnosis result, a diagnosis result of the relay protection action behavior is determined.

2. The method according to claim 1, characterized in that The step of matching the current protection action data with the relay protection action knowledge base to obtain a first diagnosis result includes: Extract voltage and current waveform information from the current protection action data to determine the current fault type; compare the current fault type with the historical fault type in the relay protection action knowledge base, and when the current fault type meets the historical fault type, determine that the current protection action data is abnormal; Comparing the current protection action data with the protection action configuration parameters in the relay protection action knowledge base, and when the current protection action data meets the protection action configuration parameters, determining that the current protection action data is normal; Comparing the current protection action data with the correct protection action sequence in the relay protection action knowledge base, and when the current protection action data conforms to the correct protection action sequence, determining that the current protection action data is normal; The current protection action data is compared with the model spatialized circuit drawing information in the relay protection action knowledge base, and when the current protection action data conforms to the model spatialized circuit drawing information, it is determined that the current protection action data is normal.

3. The method according to claim 1, characterized in that The obtaining a second diagnosis result according to the current protection action data by using a random forest algorithm includes: Dividing the current protection action data into a training set and a test set, wherein the training set is used for random forest model training, and the test set is used for evaluating the performance of the random forest model; According to the training set data, a random forest model is trained, each decision tree classifies the training samples, and the category result of the current protection action is determined through a majority voting mechanism, and the category result is used as the second diagnosis result.

4. The method according to claim 1, characterized in that: The relay protection action knowledge base is constructed based on the pre-acquired model spatialized circuit drawings, historical fault types, protection action configuration parameters and correct protection action sequence, including: According to the historical fault data, the historical fault data is classified to obtain the historical fault type, wherein the historical fault type includes a short circuit fault, a ground fault and a disconnection fault, and characteristic information corresponding to each fault type is recorded; Acquire standard protection action configuration parameters, wherein the standard protection action configuration parameters include the type, setting value, action logic and protection area of ​​each protection action; According to the standard relay protection action, determine the correct protection action sequence, wherein the correct protection action sequence includes the coordination logic of the main protection and the backup protection; The relay protection action knowledge base is determined based on the electrical equipment, lines, connection symbols and text labels, historical fault types, protection action configuration parameters and correct protection action sequence in the model spatialized circuit drawings.

5. The method according to claim 1, characterized in that Before constructing a relay protection action knowledge base based on the pre-acquired model spatialized circuit drawings, historical fault types, protection action configuration parameters, and correct protection action sequences, the method further includes: Obtaining a power plant circuit diagram, wherein the circuit diagram includes the structure and connection method of the power plant electrical system; converting the circuit drawing into a digital format; The circuit drawing converted into a digital format is parsed to obtain a model spatialized circuit drawing, wherein the model spatialized circuit drawing includes drawing information of electrical equipment, circuits, connection symbols and text labels.

6. The method according to claim 1, characterized in that The step of determining the diagnosis result of the relay protection action behavior based on the first diagnosis result and the second diagnosis result includes: Compare the first diagnosis result with the second diagnosis result to obtain a comparison result; If the comparison result meets the preset requirements, the first diagnosis result or the second diagnosis result is used as the diagnosis result of the relay protection action behavior.

7. The method according to claim 1, characterized in that After determining that the diagnosis result of the relay protection action behavior is abnormal, the method further includes: The voltage and current waveform data collected by the fault recorder during the fault are obtained, and the fault distance is determined through the non-linear lumped parameter model.

8. The method according to claim 1, characterized in that After determining the diagnosis result of the relay protection action behavior, the method further includes: Based on the diagnosis results of the relay protection action behavior, a fault analysis report is generated, wherein the fault analysis report includes information such as fault type, duration, fault current, fault component, fault distance and whether the protection action is correct, and based on the corresponding fault information, fault resolution suggestions are generated, including fault investigation, equipment repair, equipment replacement and protection parameter adjustment measures.

9. A relay protection action behavior diagnosis system, characterized in that: The system includes a module for constructing a relay protection action knowledge base, a current protection action data module, a first diagnosis result module and a second diagnosis result module, and a relay protection action diagnosis result module, wherein: The module for constructing a relay protection action knowledge base is used to construct a relay protection action knowledge base according to the pre-acquired model spatialized circuit drawings, historical fault types, protection action configuration parameters and correct protection action sequence; The current protection action data module is used to obtain the current protection action data through the relay protection device and the recorder; The first diagnosis result and the second diagnosis result module are used to match the current protection action data with the relay protection action knowledge base to obtain a first diagnosis result, and obtain a second diagnosis result according to the current protection action data through a random forest algorithm; The relay protection action diagnosis result module is used to determine the diagnosis result of the relay protection action behavior based on the first diagnosis result and the second diagnosis result.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, a relay protection action behavior diagnosis method according to any one of claims 1 to 8 is implemented.

Images