Power system equipment fault early warning method and system

Abstract

The invention relates to the technical field of power, and discloses a power system equipment fault early warning method and system, and the method comprises the steps: obtaining the fault information of power equipment of a provincial power grid system as regional sample data; performing data preprocessing on the regional sample data to eliminate interference data in the regional sample data; training a teacher model by using the regional sample data to obtain a trained teacher model; constructing a relation constraint between the teacher model and the student model, and constructing a local student model based on the relation constraint; acquiring local sample data, and training the local student model based on the regional sample data and the local sample data; and performing fault early warning on local power system equipment based on the trained local student model. According to the method, data of a provincial power grid system is utilized to train a teacher network to carry out teacher model training. A student network is adopted in a regional power grid, so that the system is more suitable for a local environment and identifies special faults.

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and in particular to a method and system for early warning of equipment failure in an electric power system. Background Art

[0002] my country's power system adopts a hierarchical management model. As an important part of the national power grid, the provincial power grid undertakes the functions of regional power resource allocation and backbone grid construction. Provincial power grid companies achieve optimal allocation of power resources within the province through the main grid of 220kV and above voltage levels, and form regional interconnected power grids with surrounding provinces. Regional power grids cover multiple provinces, using 500kV and above ultra-high voltage power grids as a link to build inter-provincial power transmission channels, achieving a wider range of power surplus and shortage adjustments and clean energy consumption.

[0003] In the process of implementing the embodiments of the present application, it is found that there are at least the following problems in the related technology: The power grid system is distributed in various regions. The weather and service life of the power grid are different in each region. For power equipment such as power cables and transformers, the faults that are prone to occur vary greatly due to different environments. If the same fault analysis method is used for all regions, the adaptability is poor and it is easy to ignore the faults that are unique to the region. If a system is designed independently for each region, it will take a long time.

[0004] It should be noted that the information disclosed in the above background technology section is only used to enhance the understanding of the background of the present application, and therefore may include information that does not constitute the prior art known to ordinary technicians in the field. Summary of the invention

[0005] In order to provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. The summary is not an extensive review, nor is it intended to identify key / critical elements or delineate the scope of protection of these embodiments, but rather serves as a prelude to the detailed description that follows.

[0006] The embodiments of the present application provide a method and system for early warning of power system equipment failures to improve the accuracy of power grid system failure prediction and its adaptability to the environment.

[0007] In some embodiments, the power system equipment fault warning method includes: obtaining fault information of power equipment in the provincial power grid system as regional sample data; performing data preprocessing on the regional sample data to exclude interference data in the regional sample data; training a teacher model using the regional sample data to obtain a trained teacher model; constructing a relationship constraint between the teacher model and the student model, and constructing a local student model based on the relationship constraint; obtaining local sample data, and training the local student model based on the regional sample data and the local sample data; and performing fault warning on local power system equipment based on the trained local student model.

[0008] Optionally, performing data preprocessing on the regional sample data includes: performing SMOTE oversampling on the regional sample data; and performing filtering processing on the oversampled regional sample data.

[0009] Optionally, performing SMOTE oversampling on the regional sample data includes: calculating the k-nearest neighbor density Wi of each minority class sample; assigning sampling probabilities according to the reciprocal of the density; and adjusting the weight sensitivity according to an exponential decay factor.

[0010] Optionally, the regional teacher model includes a main loss function and a secondary loss function, the main loss function is a cross-entropy loss function, and the secondary loss function is dynamically selected from a loss function library.

[0011] Optionally, the selection method of the secondary loss function includes: selecting an initial loss function from the loss function library according to a preset rule; calculating the weighted value of the initial loss function and the main loss function, and when the convergence value of the weighted value is less than a preset convergence for N consecutive times, selecting other loss functions from the loss function library for retraining.

[0012] Optionally, obtaining local sample data includes: obtaining local initial samples; calculating the similarity of each initial sample in the regional samples, and if the similarity is greater than a similarity threshold, deleting the initial sample; and determining the local initial samples after deletion as local sample data.

[0013] Optionally, it further includes: encoding the local sample data and the sample data of other regions in the region using a feature extraction network to generate high-dimensional feature vectors; calculating the matching degree between the feature vectors, and when the matching degree is higher than a matching threshold, directly using the student model of this region as the local student model.

[0014] In some embodiments, the power system equipment fault warning system includes: an acquisition module, configured to acquire fault information of power equipment in a provincial power grid system as regional sample data; a preprocessing module, configured to perform data preprocessing on the regional sample data to exclude interference data in the regional sample data; a first training module, configured to train a teacher model using the regional sample data to obtain a trained teacher model; a construction module, configured to construct a relationship constraint between the teacher model and the student model, and construct a local student model based on the relationship constraint; a second training module, configured to acquire local sample data, and train the local student model based on the regional sample data and the local sample data; and a warning module, configured to perform fault warning on local power system equipment based on the trained local student model.

[0015] In some embodiments, there is also provided an electronic device, including a processor and a memory, where the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the above-mentioned power system equipment fault warning method.

[0016] In some embodiments, there is also provided a computer-readable storage medium, where the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are called and executed by a processor, the computer-executable instructions cause the processor to implement the above-mentioned power system equipment fault warning method.

[0017] The fault warning method and system for power system equipment provided by the embodiments of the present application can achieve the following technical effects: Use the data of the provincial power grid system to train the teacher network, collect the fault data of power system equipment in all regions of the province, and perform model teacher model training. In the regional power grid, a student network can be used to quickly identify basic faults, and continuously improve the student network model for local fault information. Thus, while ensuring the ability to identify basic faults, the system is more suitable for the local environment and can identify special faults.

[0018] The above general description and the following description are only exemplary and explanatory, and are not used to limit the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] One or more embodiments are exemplarily illustrated by corresponding drawings. These exemplary illustrations and the drawings do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings are shown as similar elements. The drawings do not constitute a scale limitation, and among them: Figure 1 It is a flowchart of a method for fault warning of power system equipment according to an embodiment of the present application; Figure 2Schematic diagram of a power system equipment fault warning system according to an embodiment of the present application; Figure 3 Schematic diagram of an electronic device provided by an embodiment of the present application. Detailed implementation manners

[0020] In order to more fully understand the features and technical content of the embodiments of the present application, the implementation of the embodiments of the present application will be described in detail below with reference to the accompanying drawings. The accompanying drawings are only for reference and illustration purposes and are not used to limit the embodiments of the present application. In the following technical description, for the sake of explanation, numerous details are provided to give a full understanding of the disclosed embodiments. However, one or more embodiments may still be implemented without these details. In other cases, well-known structures and devices may be shown in a simplified manner to simplify the drawings.

[0021] In the description of the embodiments of the present application, the terms "first", "second", etc. in the specification and claims of the embodiments of the present application and the above drawings are used to distinguish similar objects and do not necessarily describe a specific order or sequence. It should be understood that such data may be interchanged under appropriate circumstances so as to implement the embodiments of the present application described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion.

[0022] Unless otherwise specified, the term "plurality" means two or more.

[0023] In the embodiments of the present application, the character " / " indicates that the objects before and after are in an "or" relationship. For example, A / B means: A or B.

[0024] The term "and / or" is an associative relationship describing objects and indicates that three relationships may exist. For example, A and / or B means: A or B, or, A and B these three relationships.

[0025] Combined with Figure 1 As shown, it is a flowchart of a power system equipment fault warning method provided by an embodiment of the present application. The method includes the following steps: S101: Obtain the fault information of the power equipment in the provincial power grid system as regional sample data; S102: Perform data preprocessing on the regional sample data to exclude interference data in the regional sample data; S103: Use the regional sample data to train a teacher model to obtain a trained teacher model; S104: Construct a relationship constraint between the teacher model and the student model, and construct a local student model based on the relationship constraint; S105: Obtain local sample data, and train the local student model based on the regional sample data and the local sample data; S106: Perform fault warning on local power system equipment based on the trained local student model.

[0026] The fault warning method and system for power system equipment proposed in the embodiments of this application utilize the data of the provincial power grid system to train the teacher network, collect the fault data of power system equipment in all regions of the province, and perform model teacher model training. In the regional power grid, the student network can be used to quickly identify basic faults and continuously improve the student network model for local fault information. Thus, while ensuring the ability to identify basic faults, the system is more suitable for the local environment and can identify special faults.

[0027] Specifically, it is necessary to set up a general server in the provincial power grid system to collect the fault information of all power equipment within the provincial power grid area. The fault information includes fault type, occurrence time, weather conditions before and after the occurrence, geographical characteristics, solution methods, etc. The power system equipment in the embodiments of this application includes but is not limited to power cables and transformers. To comprehensively obtain the fault sample data of the regional power grid, it is first necessary to establish a unified data acquisition platform in the provincial power grid system. This platform takes the provincial general server as the core node and realizes interconnection and interoperability with the monitoring systems of power supply companies in each city through standardized interfaces. The general server will collect in real time the fault information of various power facilities such as substations, transmission lines, and distribution equipment within its jurisdiction, including basic data such as fault types (such as short circuit, grounding, overload, etc.), fault occurrence time accurate to the millisecond level, and abnormal records of equipment operation parameters. At the same time, it will automatically associate with the real-time weather data of the meteorological department and record environmental parameters such as temperature, humidity, wind speed, and rainfall before and after the fault occurs.

[0028] During the data acquisition process, the system will pay special attention to the correlation and integration of multi-dimensional information. In addition to the fault information of the equipment body, it will also collect geographical feature data such as the geographical coordinates, altitude, and surrounding vegetation coverage of the fault point, and perform spatial overlay analysis with the GIS geographical information system. For the disposal process of each fault, the operation and maintenance personnel need to enter detailed fault phenomenon descriptions, on-site inspection situations, solution methods adopted (such as equipment replacement, parameter adjustment, isolation operation, etc.), and the final power supply restoration time through a mobile terminal to form a complete closed-loop record of fault handling. The system will also automatically associate archive data such as the historical maintenance records and family defect information of the equipment to provide background support for subsequent analysis.

[0029] To ensure data quality, the provincial general server will deploy a data verification mechanism to clean and standardize the collected information through technical means such as logical rule judgment and data integrity check. The system will establish a fault sample database, classify and store it according to dimensions such as voltage level, equipment type, and fault nature, and regularly generate a data quality assessment report. At the same time, the platform supports data exchange with the power grid systems of surrounding provinces, realizes the sharing and complementarity of fault samples at the regional power grid level, and provides a more comprehensive sample basis for subsequent big data analysis. All data collection and processing processes follow the data security specifications of the power industry to ensure information security and privacy protection.

[0030] Furthermore, taking the transformer as an example, the fault types in the fault information include the five gas concentrations (such as H2, CH4, C2H2, etc.) with the dissolved gas analysis (DGA) data in oil as the core characteristic quantities. Taking the cable as an example, the fault types in the fault information include the collection of time-domain / frequency-domain characteristics such as three-phase current, voltage harmonics, and partial discharge signals.

[0031] In some embodiments, the data preprocessing of the regional sample data in step S202 includes: performing SMOTE oversampling on the regional sample data; and performing filtering processing on the oversampled regional sample data.

[0032] SMOTE (Synthetic Minority Over-sampling Technique) is a classic algorithm for dealing with class imbalance problems. Its core idea is to generate synthetic samples by interpolating in the feature space rather than simply replicating minority class samples. The embodiments of the present application adopt an improved SMOTE algorithm to oversample minority class samples. Specifically, first, calculate the k-nearest neighbor density Wi of each minority class sample; next, allocate the following sampling probabilities according to the reciprocal of the density.

[0033]

[0034] Finally, adjust the weight sensitivity according to the exponential decay factor:

[0035] Among them, the parameter α>0 is used to represent the influence intensity of density on probability (the larger α is, the more attention is paid to sparse regions). The embodiments of the present application improve the SMOTE algorithm, dynamically adjust the sampling weight according to the local density of the samples, and thus generate more samples in sparse regions. The diversity of the samples is improved.

[0036] Furthermore, due to the particularity of the power industry, its interference signals are strong, so anti-interference processing is performed on the collected initial data. Specifically, Kalman filtering processing is performed on the current signal.

[0037] In some examples, in the method provided by the embodiments of the present application, a dynamic master-slave loss function is adopted to construct a teacher model. Among them, the master loss function uses the cross-entropy loss function, and the slave loss function is dynamically selected from a loss function library. In this way, the loss function can be dynamically adjusted. The specific selection method is that when the loss value does not converge significantly for N consecutive times, another loss function is selected from the loss function library and retrained N times. The loss function library may include, but is not limited to, contrastive loss for enhancing the aggregation of similar fault features, or Focal Loss for alleviating the problem of imbalance between easy and difficult samples.

[0038] The above embodiments adopt a dynamic master-slave loss function mechanism, which ensures basic convergence through the cross-entropy master loss function. At the same time, auxiliary objectives are dynamically selected from the loss function library (such as using Focal Loss to solve class imbalance or KL divergence to achieve knowledge distillation), which can adaptively adjust the optimization direction and significantly improve the generalization ability of the model under different data distributions. This design not only enhances the training stability through multi-objective collaborative optimization (such as balancing accuracy and robustness), but also reduces the dependence on manual hyperparameter tuning. Especially in the teacher-student framework, it can flexibly match feature differences and effectively improve the knowledge transfer efficiency.

[0039] In some examples, step S204 in the above embodiments may specifically include: constructing a relationship constraint between the teacher network and the student network, and determining an initial local student network based on the relationship constraint.

[0040] First, construct a knowledge distillation loss function, which can be expressed as:

[0041] Among them, Ft is the feature set of the teacher model, Fs is the feature set of the student model, and is a similarity function. Lm is a correlation function between the teacher model and the student model. Based on the knowledge distillation loss function, a student model is constructed. The above knowledge distillation loss function can construct an identity relationship between the teacher network and the student network, so that the student network can continuously learn the relevant knowledge of the teacher network.

[0042] In some examples, the above method may further include: obtaining local initial samples; calculating the similarity of each initial sample in the regional samples, and if the similarity is greater than the similarity threshold, deleting the initial sample; determining the deleted local initial samples as local sample data.

[0043] Specifically, a feature extraction network can be used to encode local sample data and sample data from other regions within the area, generating high-dimensional feature vectors; calculate the matching degree between the feature vectors, and when the matching degree is higher than the matching threshold, directly use the student model of this area as the local student model.

[0044] The above example can achieve cross-regional model migration through feature similarity measurement. For example, the following cosine similarity formula is used for feature matching:

[0045] where represents the feature extraction network, S and T are local samples and regional samples, is the threshold.

[0046] This formula eliminates the feature scale difference through normalization processing and can effectively evaluate the distribution similarity. It can be further extended to weighted multi-level feature matching.

[0047] In some feasible embodiments, the above method may further include: using a feature extraction network to encode local sample data and sample data from other regions within the area, generating high-dimensional feature vectors; calculating the matching degree between the feature vectors, and when the matching degree is higher than the matching threshold, directly use the student model of this area as the local student model.

[0048] Specifically, the above matching degree calculation can use a feature extraction network (such as a Siamese network or a contrast learning model) to encode the fault data of the local area and other regions, generating high-dimensional feature vectors. The matching degree of fault features in different regions is measured by cosine similarity or Euclidean distance to quantify their distribution similarity.

[0049] Furthermore, if the fault feature of a certain area has a matching degree with the local area exceeding the preset threshold (such as 0.85), it indicates that the fault modes of the two are highly similar. The pre-trained student network model of this area can be directly used as the initial local student model to avoid training from scratch. Otherwise, it is necessary to initialize independently or select the regional model with the highest matching degree for fine-tuning.

[0050] Based on the above method embodiments, the embodiments of the present invention also provide a power system equipment fault warning system. Referring to Figure 2 shown, this system 200 includes: An acquisition module 201, configured to acquire the fault information of power equipment in the provincial power grid system as regional sample data; A preprocessing module 202, configured to perform data preprocessing on the regional sample data to exclude interference data in the regional sample data; A first training module 203, configured to use the regional sample data to train a teacher model to obtain a trained teacher model; A building module 204 for constructing a relationship constraint between a teacher model and a student model, and constructing a local student model based on the relationship constraint; A second training module 205 for obtaining local sample data and training the local student model based on the regional sample data and the local sample data; An early warning module 206 for performing fault early warning on local power system equipment based on the trained local student model.

[0051] The fault early warning system for power system equipment proposed in the embodiments of the present application uses the data of the provincial power grid system to train the teacher network, collects the fault data of power system equipment in all regions of the province, and performs model teacher model training. In the regional power grid, a student network can be used to quickly identify basic faults, and continuously improve the student network model for local fault information. Thus, while ensuring the ability to identify basic faults, the system is more suitable for the local environment and can identify special faults.

[0052] Embodiments of the present invention also provide an electronic device, as Figure 3 shown, which is a schematic structural diagram of the electronic device. Among them, the electronic device includes a processor 301 and a memory 302. The memory 302 stores computer-executable instructions that can be executed by the processor 301, and the processor 301 executes the computer-executable instructions to implement the above neural network training method and target recognition method.

[0053] In Figure 3 the shown embodiment, the electronic device further includes a bus 303 and a communication interface 304, wherein the processor 301, the communication interface 304, and the memory 302 are connected through the bus 303.

[0054] Among them, the memory 302 may include high-speed random access memory (RAM), and may also include non-volatile memory, such as at least one disk memory. The communication connection between this system network element and at least one other network element is realized through at least one communication interface 304 (which can be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. can be used. The bus 303 can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, an EISA (Extended Industry Standard Architecture) bus, etc. The said bus 403 can be divided into an address bus, a data bus, a control bus, etc. For the sake of convenience of representation, Figure 3 only a bidirectional arrow is used in Figure 3 , but it does not mean that there is only one bus or one type of bus.

[0055] The processor 301 may be an integrated circuit chip with signal processing capabilities. In the implementation process, the steps of the above method can be completed by the integrated logic circuit in hardware or instructions in software form in the processor 301. The above-mentioned processor 301 can be a general-purpose processor, including a central processing unit (CPU for short), a network processor (NP for short), etc.; it can also be a digital signal processor (DSP for short), an application specific integrated circuit (ASIC for short), a field programmable gate array (FPGA for short) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in combination with the embodiments of the present invention can be directly embodied as being executed and completed by a hardware decoding processor, or executed and completed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the art such as random access memory, flash memory, read-only memory, programmable read-only memory or electrically erasable programmable memory, register, etc. This storage medium is located in the memory, and the processor 301 reads the information in the memory and combines its hardware to complete the steps of the neural network training method and the target recognition method in the foregoing embodiments.

[0056] An embodiment of the present invention further provides a computer-readable storage medium storing computer-executable instructions, which, when called and executed by a processor, cause the processor to implement the above neural network training method and target recognition method. For specific implementation, reference may be made to the foregoing method embodiments, which will not be elaborated herein.

[0057] The above description and drawings fully illustrate the embodiments of the present application, enabling those skilled in the art to practice them. Other embodiments may include structural, logical, electrical, process, and other changes. Embodiments merely represent possible variations. Unless explicitly required, individual components and functions are optional, and the order of operations may vary. Parts and features of some embodiments may be included in or replace parts and features of other embodiments. Moreover, the terms used in this application are only for describing embodiments and do not limit the claims. As used in the description of the embodiments and claims, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to include the plural forms as well. Similarly, as used in this application, the term "and / or" refers to any and all possible combinations including one or more of the associated listed items. Additionally, when used in this application, the term "comprise" and its variants "comprises" and / or "comprising" etc. mean the presence of the stated features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or groups thereof. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of additional identical elements in the process, method, or device comprising the element. In this document, each embodiment may focus on the differences from other embodiments, and the same or similar parts among the embodiments may be referred to each other. For the methods, products, etc. disclosed in the embodiments, if they correspond to the method parts disclosed in the embodiments, the relevant parts may refer to the description of the method parts.

[0058] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner can depend on the specific application and design constraints of the technical solution. The technician can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the embodiments of this application. Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0059] In the embodiments disclosed herein, the disclosed methods, products (including but not limited to devices, equipment, etc.) can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units can be only a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Additionally, the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms. The units described as separate components can be or can not be physically separated, and the components displayed as units can be or can not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to implement this embodiment. Additionally, in the embodiments of this application, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0060] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a part of code, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions noted in the blocks may occur in a different order than noted in the accompanying drawings. For example, two consecutive blocks may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description, and sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. Each block in the block diagram and / or flowchart, as well as combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

Claims

1. A method for warning of power system equipment failures, characterized in that, The method includes: Obtaining the fault information of the power equipment in the provincial power grid system as regional sample data; Performing data preprocessing on the regional sample data to exclude the interference data in the regional sample data; Training the teacher model with the regional sample data to obtain a trained teacher model; Constructing the relationship constraint between the teacher model and the student model, and constructing a local student model based on the relationship constraint; Obtaining local sample data, and training the local student model based on the regional sample data and the local sample data; Performing fault warning on the local power system equipment based on the trained local student model.

2. The power system equipment fault warning method according to claim 1, wherein Performing data preprocessing on the regional sample data includes: Performing SMOTE oversampling on the regional sample data; Performing filtering processing on the oversampled regional sample data.

3. The power system equipment fault warning method according to claim 2, characterized in that Performing SMOTE oversampling on the regional sample data includes: Calculating the k-nearest neighbor density of each minority class sample; Allocating sampling probabilities according to the reciprocal of the k-nearest neighbor density; Adjusting the weight sensitivity according to the sampling probability and the exponential decay factor.

4. The power system equipment fault warning method according to claim 1, wherein, The regional teacher model includes a main loss function and a secondary loss function. The main loss function is a cross-entropy loss function, and the secondary loss function is dynamically selected from the loss function library.

5. The method for warning of power system equipment faults according to claim 4, characterized in that The selection method of the secondary loss function includes: Selecting an initial loss function from the loss function library according to a preset rule; Calculating the weighted value of the initial loss function and the main loss function. When the convergence value of the weighted value is less than the preset convergence for N consecutive times, other loss functions are selected from the loss function library for retraining.

6. The power system equipment fault warning method according to claim 1, characterized in that, Obtaining local sample data includes: Obtaining local initial samples; Calculating the similarity of each initial sample in the regional samples. If the similarity is greater than the similarity threshold, the initial sample is deleted; Determining the deleted local initial samples as the local sample data.

7. The method for warning of power system equipment failures according to claim 1, characterized in that It also includes: Using a feature extraction network to encode the local sample data and the sample data of other regions within the region to generate high-dimensional feature vectors; Calculating the matching degree between the high-dimensional feature vectors. When the matching degree is higher than the matching threshold, directly using the student model of this region as the local student model.

8. A power system equipment fault warning system, characterized in that, It includes: An acquisition module, used to obtain the fault information of the power equipment in the provincial power grid system as regional sample data; A preprocessing module, used to perform data preprocessing on the regional sample data to exclude the interference data in the regional sample data; A first training module, used to train the teacher model with the regional sample data to obtain a trained teacher model; A construction module, used to construct the relationship constraint between the teacher model and the student model, and construct a local student model based on the relationship constraint; A second training module, used to obtain local sample data and train the local student model based on the regional sample data and the local sample data; A warning module, used to perform fault warning on the local power system equipment based on the trained local student model.

9. An electronic device, characterized in that, It includes a processor and a memory, and the memory stores computer-executable instructions that can be executed by the processor. The processor executes the computer-executable instructions to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions cause the processor to implement the method according to any one of claims 1 to 7.

Images