Power grid high influence weather event switching line identification and tracking method and system

Abstract

The invention relates to the technical field of power grids, and discloses a power grid high-influence weather event switching line identification and tracking method and system, and the method comprises the steps: obtaining high-influence weather event data, screening weather event data containing switching line features, manually marking the position of a switching line, and extracting an initial meteorological variable; by designing an auto-encoder and an adversarial network, a shear line recognition model is constructed, model training is carried out, and the performance of the model is analyzed and evaluated to optimize the model. According to the method, the spatial position and intensity change characteristics of the shear line in the whole process of the high-influence weather event are systematically analyzed, the early warning and coping strategy of the power grid is optimized, and the stability and the anti-risk capability of the power grid are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power grids, and in particular to a method and system for identifying and tracking shear lines of high-impact weather events in power grids. Background Art

[0002] High-impact weather events (such as heavy rainfall, strong winds, and thunderstorms) pose a major threat to the stable operation of power systems. These events may not only cause mechanical damage to transmission lines, but also cause failures in substation equipment and even lead to large-scale power outages. For example, Hurricane Harvey in the United States in 2017 caused large-scale power outages and caused serious damage to the power grid system. In these high-impact weather events, shear lines, as a typical small-scale weather system, play a key role. Shear lines refer to areas where wind speed and wind direction change significantly in the horizontal direction, usually accompanied by violent convective activities and significant weather phenomena such as thunderstorms, heavy rains, and strong winds. The rapid generation and movement characteristics of shear lines make them an important trigger for high-impact weather events.

[0003] The formation of shear lines is closely related to atmospheric dynamics and thermodynamic processes. They often appear in frontal systems, low-pressure troughs and severe convective weather conditions. The violent convective activity and significant wind shear within the shear lines lead to extreme weather phenomena such as heavy precipitation, thunderstorms and strong winds. These weather phenomena have a direct and far-reaching impact on the physical structure and operation of the power system. For example, heavy precipitation may cause the foundation of the transmission line to be unstable, thunderstorms may directly damage power equipment, and strong winds may break cables or damage towers.

[0004] Traditional meteorological forecasting models have significant deficiencies in identifying and tracking shear lines. First, existing forecasting systems have technical limitations in capturing small-scale weather systems, especially in complex terrain and changing climate conditions, which are prone to missed reports and false alarms. Second, the rapid generation and movement of shear lines make it difficult for existing forecasting models to track their changes in real time. Due to the lack of timeliness of the forecasting system, the warning information is delayed and accurate warning information cannot be provided in a timely manner, which seriously affects the emergency response capability of the power system.

[0005] High-impact weather events often require comprehensive analysis of multi-source data, but existing forecasting technologies have limitations in data fusion. Different data sources (such as radar, satellites, and ground observations) have different data formats and time resolutions, which makes comprehensive analysis difficult and affects the accuracy of shear line identification and tracking. In addition, when facing high-impact weather events, the power system lacks a response strategy based on the characteristics of small-scale weather systems. Existing power grid management systems rely more on historical experience and statistical models, lack a dynamic response mechanism for real-time weather systems, and cannot effectively reduce risks when extreme weather occurs.

[0006] Studying the changing characteristics of the spatial position, intensity, etc. of shear lines in high-impact weather events and developing corresponding identification and tracking technologies not only has important scientific value, but can also significantly enhance the early warning and response capabilities of the power system and has broad practical application prospects. Summary of the invention

[0007] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0008] In view of the above existing problems, the present invention is proposed. Therefore, the present invention provides a method for identifying and tracking shear lines of high-impact weather events in a power grid to solve the above problems.

[0009] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0010] In a first aspect, the present invention provides a method for identifying and tracking shear lines of high-impact weather events in a power grid, comprising: acquiring high-impact weather event data, screening weather event data containing shear line features, and manually marking shear line locations;

[0011] Based on the high-impact weather event data, initial meteorological variables are extracted, and a shear line recognition model is constructed by designing an autoencoder and an adversarial network;

[0012] The shear line recognition model is trained, and the performance of the model is analyzed and evaluated, and the model is optimized according to the evaluation results.

[0013] As a preferred solution of the method for identifying and tracking shear lines of high-impact weather events in power grids described in the present invention, the initial meteorological variables include the temperature, humidity, wind speed, wind direction, height field and air pressure of the ground and the preset pressure surface, and the vorticity, divergence, potential temperature and temperature horizontal gradient are calculated as modeling features.

[0014] As a preferred solution of the method for identifying and tracking shear lines of high-impact weather events in power grids according to the present invention, the construction of a shear line identification model includes:

[0015] Extracting initial meteorological variable features from the high-impact weather event data through an autoencoder architecture design, wherein the autoencoder includes an encoder and a decoder;

[0016] Through the design of adversarial network architecture, the initial meteorological variable characteristics are used as the input of the adversarial network to generate weather data samples and identify shear lines.

[0017] As a preferred solution of the method for identifying and tracking shear lines of high-impact weather events in a power grid according to the present invention, the autoencoder comprises:

[0018] The encoder compresses high-dimensional input data into a low-dimensional feature space, gradually reducing the dimension of the input data to obtain a low-dimensional feature representation;

[0019] The low-dimensional features are restored to high-dimensional input data through a decoder, and the decoder structure is symmetrical to the encoder, and the dimensions are gradually increased to reconstruct the low-dimensional features into high-dimensional data.

[0020] As a preferred solution of the method for identifying and tracking shear lines of high-impact weather events in power grids described in the present invention, the adversarial network includes:

[0021] Generate samples of real weather data from features extracted by the autoencoder through a generator, wherein the generator gradually converts low-dimensional noise or features into high-dimensional data;

[0022] The discriminator distinguishes between real data and generated data, classifies the input data, and outputs the location information of the shear line.

[0023] As a preferred solution of the method for identifying and tracking shear lines of high-impact weather events in power grids described in the present invention, the model training includes:

[0024] Autoencoder training, using mean square error to calculate the difference between high-dimensional input data and reconstructed high-dimensional data, and updating the parameters of the autoencoder to minimize the reconstruction error;

[0025] Adversarial network training, adversarial training of the generator and the discriminator through adversarial loss, shear line recognition of the discriminator through classification loss, and shear line position regression through regression loss. During the training process, the generator and the discriminator are updated alternately.

[0026] As a preferred solution of the method for identifying and tracking shear lines of high-impact weather events in power grids described in the present invention, the optimization of the model includes:

[0027] Using the validation set to evaluate the performance of the shear line recognition model, and obtaining the model evaluation result by selecting the evaluation index;

[0028] According to the model evaluation result, the shear line identification model parameters and architecture are adjusted.

[0029] In a second aspect, the present invention provides a power grid high-impact weather event shear line identification and tracking system, comprising:

[0030] The annotation module is used to obtain high-impact weather event data, filter weather event data containing shear line features, and manually annotate the shear line location;

[0031] A construction module is used to extract initial meteorological variables based on the high-impact weather event data, and to construct a shear line recognition model by designing an autoencoder and an adversarial network;

[0032] The evaluation and optimization module is used to perform model training on the shear line recognition model, analyze and evaluate the performance of the model, and optimize the model according to the evaluation results.

[0033] In a third aspect, the present invention provides an electronic device, comprising:

[0034] Memory and processor;

[0035] The memory is used to store computer executable instructions, and the processor is used to execute the computer executable instructions. When the computer executable instructions are executed by the processor, the steps of the method for identifying and tracking shear lines of high-impact weather events in the power grid are implemented.

[0036] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the method for identifying and tracking shear lines of high-impact weather events in a power grid.

[0037] Compared with the prior art, the present invention has the following beneficial effects: the present invention optimizes the early warning and response strategies of the power grid and improves the stability and risk resistance of the power grid by systematically analyzing the spatial position and intensity change characteristics of the shear line during the entire process of high-impact weather events. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0039] Figure 1 It is a schematic diagram of the overall process of a method for identifying and tracking shear lines of high-impact weather events in a power grid according to an embodiment of the present invention;

[0040] Figure 2 A schematic diagram of an autoencoder of a method for identifying and tracking shear lines of high-impact weather events in a power grid according to an embodiment of the present invention;

[0041] Figure 3A schematic diagram of a GAN architecture of a method for identifying and tracking shear lines of high-impact weather events in a power grid according to an embodiment of the present invention;

[0042] Figure 4 A schematic diagram of shear line tracking of a method for identifying and tracking shear lines of high-impact weather events in a power grid according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0044] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0045] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0046] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0047] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0048] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0049] Example 1

[0050] Reference Figure 1-Figure 3 , is an embodiment of the present invention, and provides a method for identifying and tracking shear lines of high-impact weather events in a power grid, such as Figure 1 As shown, including:

[0051] S101, obtaining high-impact weather event data, screening weather event data containing shear line features, and manually marking the shear line positions;

[0052] S102, based on high-impact weather event data, extract initial meteorological variables, and build a shear line recognition model by designing autoencoders and adversarial networks;

[0053] S103, training the shear line recognition model, analyzing and evaluating the performance of the model, and optimizing the model according to the evaluation results.

[0054] Specifically, in step S101, first, weather event instances containing obvious shear line features are screened out from high-impact weather event data. This selection process focuses on events with complex meteorological conditions and obvious shear line features to ensure the representativeness and diversity of the data; then, professional meteorological analysis tools such as geographic information system (GIS) software are used for manual annotation by experienced meteorological experts. In this process, experts will intuitively mark the precise location of the shear line on the weather image to ensure that each mark accurately reflects the actual location of the shear line.

[0055] It should be noted that in the development of the shear line recognition model, this step involves a precise manual labeling process in order to construct an effective training data set. These labeled data will serve as the basis for subsequent machine learning model training and are the key to establishing an efficient recognition system.

[0056] Preferably, in step S102, the initial meteorological variables include temperature, humidity, wind speed, wind direction, height field and air pressure of the ground and a preset pressure surface, and vorticity, divergence, potential temperature and temperature horizontal gradient are calculated as modeling features, wherein the preset pressure surface is preferably a pressure surface of 1000hPa-500hPa.

[0057] It should be noted that in the development process of shear line identification model, feature screening is a key step. The performance and efficiency of the model can be improved by screening out the most relevant features. The main purpose of feature screening is to select those variables that are most useful for shear line identification and tracking from a large number of meteorological variables. This process can adopt a variety of methods, including statistical analysis, correlation analysis and machine learning algorithms.

[0058] Preferably, constructing the shear line recognition model includes:

[0059] The initial meteorological variable features are extracted from the high-impact weather event data through the design of the autoencoder architecture, which includes an encoder and a decoder;

[0060] Through the adversarial network architecture design, the initial meteorological variable characteristics are used as the input of the adversarial network to generate weather data samples and identify shear lines.

[0061] Preferably, Figure 2 As shown, the autoencoder includes:

[0062] The encoder compresses high-dimensional input data into a low-dimensional feature space, gradually reducing the dimension of the input data to obtain a low-dimensional feature representation;

[0063] The decoder is used to restore low-dimensional features to high-dimensional input data. The decoder structure is symmetrical to the encoder, and the dimensions are gradually increased to reconstruct low-dimensional features into high-dimensional data.

[0064] Specifically, an autoencoder is used to extract effective features from the data. The encoder compresses the high-dimensional input data into a low-dimensional feature space, and gradually reduces the dimension of the input data through several layers of neural networks (such as linear layers and activation functions), and finally obtains a compact feature representation. The formula is as follows:

[0065] h=f(x)=RELU(Wx+b)

[0066] Among them, x represents input data, W represents weight matrix, b represents bias vector, and h represents low-dimensional feature representation.

[0067] The decoder restores low-dimensional features to high-dimensional input data. The structure of the decoder is symmetrical to that of the encoder. It gradually increases the dimensions through several layers of neural networks and reconstructs low-dimensional features into high-dimensional data similar to the input data, expressed as:

[0068] x'=g(h)=RELU(W'h+b')

[0069] Among them, h represents the low-dimensional feature representation, W′ represents the weight matrix, x′ represents the reconstructed high-dimensional data, b′ represents the bias vector, and ReLU (Rectified Linear Unit) is a nonlinear transformation function that enables the neural network to approximate complex functions.

[0070] Preferably, Figure 3 As shown, the adversarial network includes:

[0071] The generator generates samples of real weather data from the features extracted by the autoencoder, and the generator gradually converts low-dimensional noise or features into high-dimensional data;

[0072] The discriminator distinguishes between real data and generated data, classifies the input data, and outputs the location information of the shear line.

[0073] Specifically, Generative Adversarial Networks (GAN) uses the features extracted by the autoencoder as the input of GAN to generate realistic weather data samples and identify shear lines.

[0074] The generator is responsible for generating samples similar to real weather data from random noise or features extracted by the autoencoder. The structure of the generator usually includes several layers of neural networks, which gradually convert low-dimensional noise or features into high-dimensional data.

[0075] The discriminator is responsible for distinguishing between real data and generated data. In a semi-supervised setting, it is also necessary to classify the input data (identify whether it contains shear lines) and output the location information of the shear lines.

[0076] The structure of the adversarial network includes:

[0077] Feature extraction layer: extracts the features of input data through several layers of neural networks;

[0078] Adversarial layer: outputs true or false discrimination results, usually using a single neuron output (activated by Sigmoid);

[0079] Classification layer: outputs the result of whether the shear line is contained;

[0080] Regression layer: outputs the position of the shear line, such as coordinate information.

[0081] Preferably, in step S103, model training includes:

[0082] Autoencoder training, using mean square error to calculate the difference between high-dimensional input data and reconstructed high-dimensional data, and updating the parameters of the autoencoder to minimize the reconstruction error;

[0083] Adversarial network training, the generator and discriminator are trained adversarially through adversarial loss, the discriminator is trained to identify the shear line through classification loss, and the shear line position is regressed through regression loss. During the training process, the generator and discriminator are updated alternately.

[0084] Specifically, in the training of the autoencoder, unlabeled data is used to train the autoencoder so that it can effectively compress and reconstruct the input data. During the training process, the parameters of the autoencoder are repeatedly updated to reduce the difference between the input and output.

[0085] GAN training is an alternating training. During the training process, the generator and the discriminator are updated alternately. The discriminator is trained using real data and generated data so that it can distinguish between true and false data, and at the same time classifies and predicts the position of shear lines, and updates the discriminator parameters to minimize classification and regression errors; the generator is trained to deceive the discriminator through the data generated by the generator, making it difficult for the discriminator to distinguish between true and false data, and updates the generator parameters to maximize the adversarial loss, while ensuring that the generated data can reasonably contain shear line features.

[0086] Preferably, optimizing the model includes:

[0087] Use the validation set to evaluate the performance of the shear line recognition model, and obtain the model evaluation results by selecting evaluation indicators;

[0088] According to the model evaluation results, the shear line identification model parameters and architecture are adjusted.

[0089] Specifically, the model evaluation process includes: using the validation set to evaluate the performance of the model, especially the accuracy of shear line recognition and the accuracy of position prediction, and indicators such as accuracy, recall, F1 score and mean square error (MSE) can be used to evaluate the performance of the model.

[0090] The model optimization process includes: adjusting the model parameters and architecture according to the evaluation results, such as adjusting the number of network layers, the number of neurons, the learning rate, etc., and using different activation functions and regularization techniques to improve the generalization ability and robustness of the model.

[0091] It should be noted that through this semi-supervised learning method that combines autoencoders and GANs, we can make full use of a large amount of unlabeled data to extract effective features from them, and with the assistance of a small amount of labeled data, we can achieve accurate identification and position prediction of shear lines.

[0092] The method for identifying and tracking shear lines of high-impact weather events in a power grid provided by the present invention optimizes the early warning and response strategies of the power grid and improves the stability and risk resistance of the power grid by systematically analyzing the spatial position and intensity change characteristics of the shear lines throughout the entire process of high-impact weather events.

[0093] The above is a schematic scheme of a method for identifying and tracking shear lines of high-impact weather events in a power grid in this embodiment. It should be noted that the technical scheme of the shear line identification and tracking system for high-impact weather events in the power grid and the technical scheme of the shear line identification and tracking method for high-impact weather events in the power grid belong to the same concept. For details not described in detail in the technical scheme of the shear line identification and tracking system for high-impact weather events in this embodiment, please refer to the description of the technical scheme of the shear line identification and tracking method for high-impact weather events in the power grid.

[0094] The power grid high-impact weather event shear line identification and tracking system in this embodiment includes:

[0095] The annotation module is used to obtain high-impact weather event data, filter weather event data containing shear line features, and manually annotate the shear line location;

[0096] A construction module is used to extract initial meteorological variables based on high-impact weather event data and construct a shear line recognition model by designing autoencoders and adversarial networks;

[0097] The evaluation and optimization module is used to train the shear line recognition model, analyze and evaluate the performance of the model, and optimize the model based on the evaluation results.

[0098] This embodiment also provides an electronic device, which is applicable to the situation of identifying and tracking shear lines of high-impact weather events in power grids, including:

[0099] Memory and processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the method for identifying and tracking shear lines of high-impact weather events in power grids as proposed in the above embodiments.

[0100] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for identifying and tracking shear lines of high-impact weather events in a power grid as proposed in the above embodiment is implemented.

[0101] The storage medium proposed in this embodiment and the method for identifying and tracking shear lines of high-impact weather events in power grids proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0102] Through the above description of the implementation methods, the technicians in the relevant field can clearly understand that the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ReadOnly, Memory, ROM), random access memory (RandomAccess Memory, RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform the methods of various embodiments of the present invention.

[0103] Example 2

[0104] Reference Figure 4 , which is an embodiment of the present invention, provides a method for identifying and tracking shear lines of high-impact weather events in a power grid. In order to verify its beneficial effects, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0105] During a high-impact weather event, a shear line passes through a certain area, accompanied by severe convective weather and thunderstorm activities. We will use the constructed model to identify and track the shear line and analyze its potential impact on the power grid.

[0106] The specific implementation steps are as follows:

[0107] Step 1: Data preparation

[0108] The meteorological data of the event are extracted from the high-impact weather event database, including: temperature, humidity, wind speed, wind direction, air pressure, etc. on the ground and in the high altitude (1000hPa-500hPa). The time span covers the entire process of the shear line crossing, and the data interval is 1 hour.

[0109] Step 2: Feature extraction

[0110] The autoencoder is used to extract effective features, and the input high-dimensional meteorological data is compressed into a low-dimensional feature space through the trained autoencoder. These low-dimensional features include key meteorological variables such as vorticity, divergence, temperature gradient, etc.

[0111] Step 3: Shear line identification

[0112] The trained GAN model is used for shear line recognition. The generator generates realistic meteorological data samples, including potential shear line features. The discriminator classifies the input data, identifies whether it contains shear lines, and outputs the location information of the shear lines.

[0113] Step 4: Track the shear line

[0114] Track the identified shear line, and use the model to predict the position change of the shear line through time series data; record the movement path and intensity change of the shear line.

[0115] Step 5: Result Analysis

[0116] Analyze the recognition and tracking results: Use real-time data to evaluate the recognition accuracy and position prediction accuracy of the model. Analyze the spatial position and intensity change characteristics of the shear line.

[0117] The example results are as follows:

[0118] Assume that during a shear line transit in June 2024, our model successfully identified and tracked the shear line, and the specific results are as follows:

[0119] Identification results:

[0120] At 08:00 on June 1, 2024, the model identified a shear line located at 35 degrees north latitude, 120 degrees east longitude, and a height of 500hPa;

[0121] The shear line characteristics are significant, the vorticity is as high as 12×10^-5s^-1, and the horizontal temperature gradient reaches 5℃ / 100km.

[0122] Tracking results:

[0123] From 08:00 on June 1, 2024 to 20:00 on June 1, 2024, the model continuously tracks the shear line position;

[0124] The shear line moves from 35 degrees north latitude, 120 degrees east longitude to 34 degrees north latitude, 125 degrees east longitude, with a moving speed of about 30 km / h;

[0125] The intensity gradually weakened, the vorticity dropped to 8×10^-5s^-1, and the horizontal temperature gradient dropped to 3℃ / 100km.

[0126] Analysis of impact on power grid:

[0127] During the passage of the shear line, there will be strong convection and thunderstorm activities, posing a potential threat to the stability of the power grid. Based on the shear line path predicted by the model, power grid monitoring and emergency measures are strengthened in the affected areas in advance, effectively reducing the risks caused by weather changes.

[0128] The model performance evaluation results are:

[0129] Accuracy: 90%

[0130] Recall: 85%

[0131] F1 score: 0.875

[0132] Mean square error (MSE): 0.05° (position prediction accuracy)

[0133] The experimental results verify the effectiveness and practicability of the shear line identification and tracking technology in the present invention. The model can accurately identify and track the spatial position and intensity changes of the shear line, and provide important reference for power grid early warning and response strategies, thereby improving the stability and risk resistance of the power grid.

[0134] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for identifying and tracking shear lines of high-impact weather events in a power grid, characterized in that: include: Obtain high-impact weather event data, filter weather event data containing shear line features, and manually mark the shear line locations; Based on the high-impact weather event data, initial meteorological variables are extracted, and a shear line recognition model is constructed by designing an autoencoder and an adversarial network; The shear line recognition model is trained, and the performance of the model is analyzed and evaluated, and the model is optimized according to the evaluation results.

2. The method for identifying and tracking shear lines of high-impact weather events in a power grid according to claim 1, characterized in that: The initial meteorological variables include temperature, humidity, wind speed, wind direction, height field and air pressure of the ground and the preset pressure surface, and vorticity, divergence, potential temperature and temperature horizontal gradient are calculated as modeling features.

3. The method for identifying and tracking shear lines of high-impact weather events in a power grid according to claim 1, characterized in that: The constructing of the shear line recognition model comprises: Extracting initial meteorological variable features from the high-impact weather event data through an autoencoder architecture design, wherein the autoencoder includes an encoder and a decoder; Through the design of adversarial network architecture, the initial meteorological variable characteristics are used as the input of the adversarial network to generate weather data samples and identify shear lines.

4. The method for identifying and tracking shear lines of high-impact weather events in a power grid according to claim 3, characterized in that: The autoencoder comprises: The encoder compresses high-dimensional input data into a low-dimensional feature space, gradually reducing the dimension of the input data to obtain a low-dimensional feature representation; The low-dimensional features are restored to high-dimensional input data through a decoder, and the decoder structure is symmetrical to the encoder, and the dimensions are gradually increased to reconstruct the low-dimensional features into high-dimensional data.

5. The method for identifying and tracking shear lines of high-impact weather events in a power grid according to claim 3, characterized in that: The adversarial network comprises: Generate samples of real weather data from features extracted by the autoencoder through a generator, wherein the generator gradually converts low-dimensional noise or features into high-dimensional data; The discriminator distinguishes between real data and generated data, classifies the input data, and outputs the location information of the shear line.

6. The method for identifying and tracking shear lines of high-impact weather events in a power grid according to claim 5, characterized in that: The model training includes: Autoencoder training, using mean square error to calculate the difference between high-dimensional input data and reconstructed high-dimensional data, and updating the parameters of the autoencoder to minimize the reconstruction error; Adversarial network training, adversarial training of the generator and the discriminator through adversarial loss, shear line recognition of the discriminator through classification loss, and shear line position regression through regression loss. During the training process, the generator and the discriminator are updated alternately.

7. The method for identifying and tracking shear lines of high-impact weather events in a power grid according to any one of claims 3 to 6, characterized in that: The optimization of the model comprises: Using the validation set to evaluate the performance of the shear line recognition model, and obtaining the model evaluation result by selecting the evaluation index; According to the model evaluation result, the shear line identification model parameters and architecture are adjusted.

8. A power grid high-impact weather event shear line identification and tracking system, characterized in that: include, The annotation module is used to obtain high-impact weather event data, filter weather event data containing shear line features, and manually annotate the shear line location; A construction module is used to extract initial meteorological variables based on the high-impact weather event data, and to construct a shear line recognition model by designing an autoencoder and an adversarial network; The evaluation and optimization module is used to perform model training on the shear line recognition model, analyze and evaluate the performance of the model, and optimize the model according to the evaluation results.

9. An electronic device, characterized in that: include: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the method for identifying and tracking shear lines of high-impact weather events in power grids as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: It stores computer executable instructions, which, when executed by a processor, can implement the steps of the method for identifying and tracking shear lines of high-impact weather events in power grids as described in any one of claims 1 to 7.

Images