Lightning activity prediction method and device for power transmission line area

Abstract

The invention belongs to the field of lightning monitoring, and discloses a lightning activity prediction method and device for a power transmission line area, and the method comprises the steps: obtaining environment data through a plurality of sensors disposed along a power transmission line, and inputting the environment data into a pre-trained lightning activity prediction model; wherein the thunder and lightning activity prediction model comprises a physical model module, a machine learning module and a result fusion module; inputting the environment data into a physical model module and a machine learning module respectively, and inputting a physical model prediction result and a machine learning prediction result output by the two modules into a result fusion module, so that the result fusion module performs weighted summation on the received prediction results; and determining a weighted summation result output by the result fusion module as a target prediction result of the lightning activity of the power transmission line area. By combining a physical model and a machine learning algorithm, the accuracy of lightning activity prediction is improved.

Description

Technical Field

[0001] The present invention belongs to the field of lightning monitoring, and in particular relates to a method and device for predicting lightning activity in a transmission line area. Background Art

[0002] In modern power systems, transmission lines, as crucial pathways for energy transmission, face threats from various natural disasters, of which lightning strikes are the most common and destructive. Lightning strikes can not only cause transmission line failures, impacting the stability of power supply, but can also cause severe damage to grid equipment, increasing maintenance costs and the risk of power outages. Therefore, developing a technology that can effectively monitor and predict lightning activity is crucial.

[0003] Among related technologies, most lightning monitoring systems can only provide single-dimensional lightning information and lack the ability to comprehensively analyze and predict the patterns of lightning activity. Many systems rely solely on data collected by sensors and fail to combine other sources (such as meteorological data) for comprehensive analysis. They lack methods to combine physical models and machine learning algorithms, resulting in one-sided and incomplete prediction results. Due to the failure to comprehensively analyze the patterns of lightning activity, the prediction results are not accurate enough, which increases the difficulty of responding to lightning risks. It is difficult for power companies to formulate effective preventive measures based on incomplete prediction results, which increases operational difficulty and risks. Summary of the Invention

[0004] In view of this, the present invention discloses a method and device for predicting lightning activity in a transmission line area, which can solve the deficiencies existing in the related art.

[0005] To achieve the above purpose, the present invention discloses the following technical solutions: According to a first aspect of the present invention, a method for predicting lightning activity in a transmission line area is proposed, comprising: Environmental data is acquired through multiple sensors installed along the power transmission line, and the environmental data is input into a pre-trained lightning activity prediction model; wherein the lightning activity prediction model includes: a physical model module, a machine learning module, and a result fusion module; Inputting the environmental data into the physical model module and the machine learning module respectively, and inputting the physical model prediction results and the machine learning prediction results output by the two modules into the result fusion module, so that the result fusion module performs a weighted summation on the received prediction results; The weighted summation result output by the result fusion module is determined as the target prediction result of lightning activity in the transmission line area.

[0006] Optionally, the physical model module predicts the probability of occurrence of lightning activity in the transmission line area based on a cloud-to-ground lightning transmission theoretical model, where the expression of the cloud-to-ground lightning transmission theoretical model is: ; in, is the prediction result of the physical model, is the temperature of the transmission line area, is the humidity in the transmission line area, is the air pressure in the transmission line area.

[0007] Optionally, the machine learning module includes a lightning location submodule and a prediction submodule; and the method further includes: Calculating lightning strike information within the data acquisition area of ​​the multiple sensors by the lightning locating submodule, wherein the lightning strike information includes the current lightning strike position and lightning strike speed; The lightning strike information is input into the prediction submodule so that the prediction submodule predicts the probability of occurrence of lightning activity in the transmission line area; wherein the prediction submodule is constructed based on a deep neural network and a long short-term memory network trained by historical lightning data.

[0008] Optionally, the calculation expression of the weighted summation of the result fusion module is: ; in, is the weighted summation result output by the result fusion module, is the prediction result of the physical model, Q is the prediction result of the machine learning, and is the optimal weight of the two prediction results determined based on the cross-validation method.

[0009] Optionally, before the environmental data is input into the lightning activity prediction model, the method further includes: The environmental data is preprocessed based on a fast Fourier transform algorithm, and a signal-to-noise ratio of the environmental data is optimized using a Kalman filter.

[0010] Optionally, the optimization formula of the Kalman filter is: ; in, is the estimated state, is the Kalman gain, is the measured value, is the observation matrix.

[0011] Optionally, the acquiring of environmental data through a plurality of sensors disposed along the transmission line includes: adjusting the working mode and the frequency of adoption of the plurality of sensors according to environmental conditions.

[0012] According to a second aspect of the present application, a lightning activity prediction device for a power transmission line area is provided, the device comprising: an acquisition unit configured to acquire environmental data through a plurality of sensors arranged along the power transmission line and input the environmental data into a pre-trained lightning activity prediction model, wherein the lightning activity prediction model comprises a physical model module, a machine learning module, and a result fusion module; a prediction unit configured to input the environmental data into the physical model module and the machine learning module respectively, and input the physical model prediction result and the machine learning prediction result output by the two modules into the result fusion module, so that the result fusion module performs weighted summation on the received prediction results; a determination unit configured to determine the weighted summation result output by the result fusion module as a target prediction result of lightning activity in the power transmission line area.

[0013] According to a third aspect of the present application, an electronic device is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor implements the steps of the method according to the first aspect by running the executable instructions.

[0014] According to a fourth aspect of the present application, a computer-readable storage medium is provided, which stores computer instructions that, when executed by a processor, implement the steps of the method according to the first aspect.

[0015] As can be seen from the above technical solutions, the lightning activity prediction method for a power transmission line area disclosed in the present application achieves the following technical effects: on the one hand, the lightning activity is predicted by combining a physical model and a machine learning algorithm, and the final prediction result is determined as the weighted summation result of the two, so that the prediction result of lightning activity is more accurate, thereby reducing the difficulty and risk of formulating effective preventive measures for power companies; on the other hand, the lightning activity is predicted based on multi-source data, such as the current lightning activity, as well as environmental data such as temperature, humidity, and air pressure, so that the prediction result is more accurate. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a flowchart of a lightning activity prediction method for a power transmission line area provided by an exemplary embodiment; Figure 2 is a schematic diagram of a lightning activity regularity monitoring and analysis system for a power transmission line area provided by an exemplary embodiment; Figure 3 is a schematic structural diagram of a device provided by an exemplary embodiment; Figure 4 A block diagram of a lightning activity prediction device for a transmission line area provided by an exemplary embodiment. DETAILED DESCRIPTION

[0017] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, like numbers in different figures represent like or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible implementations consistent with one or more embodiments of the present invention. Rather, they are merely examples of apparatuses and methods consistent with certain aspects of one or more embodiments of the present invention, as detailed in the appended claims.

[0018] It should be noted that in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in the present invention. In some other embodiments, the method may include more or fewer steps than those described in the present invention. In addition, a single step described in the present invention may be broken down into multiple steps for description in other embodiments, and multiple steps described in the present invention may be combined into a single step for description in other embodiments.

[0019] To further illustrate the present invention, the following examples are provided: In modern power systems, transmission lines, as crucial pathways for energy transmission, face threats from various natural disasters, of which lightning strikes are the most common and destructive. Lightning strikes can not only cause transmission line failures, impacting the stability of power supply, but can also cause severe damage to grid equipment, increasing maintenance costs and the risk of power outages. Therefore, developing a technology that can effectively monitor and predict lightning activity is crucial.

[0020] Among related technologies, most lightning monitoring systems can only provide single-dimensional lightning information and lack the ability to comprehensively analyze and predict the patterns of lightning activity. Many systems rely solely on data collected by sensors and fail to combine other sources (such as meteorological data) for comprehensive analysis. They lack methods to combine physical models and machine learning algorithms, resulting in one-sided and incomplete prediction results. Due to the failure to comprehensively analyze the patterns of lightning activity, the prediction results are not accurate enough, which increases the difficulty of responding to lightning risks. It is difficult for power companies to formulate effective preventive measures based on incomplete prediction results, which increases operational difficulty and risks.

[0021] In order to solve the deficiencies in the related art, the present invention proposes a method for predicting lightning activity in a transmission line area.

[0022] Figure 1 This is a flowchart of a method for predicting lightning activity in a transmission line area, provided by an exemplary embodiment. The method may include the following steps: Step 101: Acquire environmental data through multiple sensors installed along the power transmission line, and input the environmental data into a pre-trained lightning activity prediction model; wherein the lightning activity prediction model includes: a physical model module, a machine learning module, and a result fusion module; Step 102: Input the environmental data into the physical model module and the machine learning module respectively, and input the physical model prediction results and the machine learning prediction results output by the two modules into the result fusion module, so that the result fusion module performs a weighted summation on the received prediction results; Step 103: Determine the weighted sum result output by the result fusion module as the target prediction result of lightning activity in the transmission line area.

[0023] In this embodiment, on the one hand, the physical model and machine learning algorithm are combined to predict lightning activity, and the final prediction result is determined as the weighted sum of the two, so that the prediction result of lightning activity is more accurate, thereby reducing the operational difficulty and risk for power companies to formulate effective preventive measures; on the other hand, the prediction of lightning activity is based on multi-source data, such as lightning activity at the current moment, as well as environmental data such as temperature, humidity, and air pressure, so that the prediction result is more accurate.

[0024] In one embodiment, the physical model module predicts the probability of occurrence of lightning activity in the transmission line area based on a cloud-to-ground lightning transmission theoretical model. The expression of the cloud-to-ground lightning transmission theoretical model is: ; in, is the prediction result of the physical model, is the temperature of the transmission line area, is the humidity in the transmission line area, is the air pressure in the transmission line area.

[0025] In one embodiment, the machine learning module includes a lightning locating submodule and a prediction submodule; the method further includes: calculating lightning strike information within the data acquisition area of ​​the multiple sensors through the lightning locating submodule, the lightning strike information including the lightning strike position and lightning strike speed at the current moment; inputting the lightning strike information into the prediction submodule so that the prediction submodule predicts the probability of occurrence of lightning activity in the transmission line area; wherein the prediction submodule is constructed based on a deep neural network and a long short-term memory network trained by historical lightning data.

[0026] The lightning locator module can calculate the coordinates and speed of the lightning strike point based on the TDOA formula and the Doppler effect formula.

[0027] The expression of TDOA is: ; The Doppler effect formula is: ; in, is the observed frequency, is the transmitting frequency, is the velocity of the receiver relative to the medium, is the speed of the sound source relative to the medium.

[0028] The prediction submodule may include a feature extraction layer, an attention layer, a long short-term memory network layer, and a fully connected layer. The feature extraction layer is used to extract the time series feature vector of the lightning strike information input to the prediction submodule based on the dilated causal convolution of the time series convolutional network. The attention layer is used to weight the time series feature vector. The long short-term memory network layer is used to transfer the weighted time series feature vector based on the hidden state of the long short-term memory network and learn the fluctuation trend of lightning activity. The fully connected layer is used to output the probability of lightning activity occurring in the transmission line area at the next moment.

[0029] The deep neural network can be a deep residual network, which may include multiple residual blocks, each residual block including a weight layer, an activation function layer, an identity mapping and a normalization layer, and the output of each residual block is the output of the next residual block.

[0030] When training the prediction submodule, supervised learning algorithms such as support vector machine (SVM) or random forest (RF) can be used for model training, and online learning is supported to continuously update model parameters.

[0031] The specific training process can be: Acquire a training sample set, wherein each set of training samples in the training sample set includes: lightning strike information at a certain moment, and actual lightning strike probability of the transmission line area at the next moment after the moment; Inputting the training sample set into the prediction sub-model to be trained, so that the prediction sub-model outputs the predicted lightning strike probability of the transmission line area at the next moment; The actual lightning strike probability is compared with the predicted lightning strike probability, and the prediction sub-model is optimized according to the comparison result until the prediction effect of the prediction sub-model meets the expectation.

[0032] In one embodiment, the calculation expression of the weighted summation of the result fusion module is: ; in, is the weighted summation result output by the result fusion module, is the prediction result of the physical model, Q is the prediction result of the machine learning, and is the optimal weight of the two prediction results determined based on the cross-validation method.

[0033] In one embodiment, before the environmental data is input into the lightning activity prediction model, the method further includes: preprocessing the environmental data based on a fast Fourier transform algorithm, and optimizing the signal-to-noise ratio of the environmental data according to a Kalman filter.

[0034] Furthermore, the optimization formula of the Kalman filter is: ; in, is the estimated state, is the Kalman gain, is the measured value, is the observation matrix.

[0035] In one embodiment, the acquiring of environmental data through a plurality of sensors disposed along the power transmission line includes: adjusting the working modes and the operating frequencies of the plurality of sensors according to environmental conditions.

[0036] Correspondingly, such as Figure 2 As shown, the present invention also proposes a system for monitoring and analyzing lightning activity patterns in a transmission line area, comprising: Data Acquisition Module: The data acquisition module consists of multiple sensor units distributed along the transmission line, including temperature sensors, humidity sensors, air pressure sensors, and electromagnetic field strength sensors. These sensors acquire environmental parameters in real time, and the data acquisition module features a built-in adaptive adjustment mechanism that automatically adjusts the sensor's operating mode and sampling frequency based on current environmental conditions (such as temperature and humidity) and lightning activity intensity. Each sensor unit also features a self-test function, regularly checking its operating status and reporting it to the system management module via wireless communication, ensuring high reliability during data acquisition.

[0037] Data processing module: responsible for receiving data from the data acquisition module and preprocessing it based on the fast Fourier transform (FFT) algorithm, while applying the Kalman filter to optimize the signal-to-noise ratio. ,in is the estimated state, is the Kalman gain, is the measured value, is the observation matrix to optimize the signal-to-noise ratio.

[0038] Lightning location module: uses time difference of arrival (TDOA) method combined with Doppler effect to calculate the location coordinates and velocity of lightning occurrence. Based on TDOA formula and Doppler effect formula to calculate the lightning point coordinates and velocity.

[0039] Data analysis module: includes machine learning unit, uses supervised learning algorithm such as support vector machine (SVM) or random forest (RF) for model training, and supports online learning to continuously update model parameters. In addition, the feature extraction unit extracts key features from raw data, and the prediction unit makes predictions based on the trained model for future lightning activity.

[0040] System management module: coordinates the workflow of the above modules, and supports remote configuration and update. Intelligent scheduling unit dynamically adjusts the work priority of each module according to the grid load condition, automatic update unit can automatically download and install the latest software version, fault diagnosis unit is used to monitor the system running state and provides diagnostic information when fault occurs, user interface unit allows operators to intuitively view the system running status and lightning activity prediction results through graphical interface.

[0041] Lightning activity law prediction module: combines physical model and machine learning algorithm to predict the law of future lightning activity. Physical model unit is based on cloud-to-ground lightning transmission theory model, calculates the lightning occurrence probability under given environmental conditions where represents temperature, represents humidity, represents air pressure; calibration unit adjusts model parameters regularly according to new observation data to improve prediction accuracy.

[0042] Machine learning unit uses deep neural network (DNN) to train historical lightning data, and uses long short-term memory (LSTM) network to capture long-term dependence relationship in time series data. Fusion unit fuses the prediction results of physical model and machine learning model through weighted average formula to generate the final lightning activity prediction report.

[0043] Emergency response module: when predicting high-risk lightning activity, automatically sends warning information to relevant departments, and provides emergency response measures suggestions to help reduce the impact of lightning on power system.

[0044] The adaptive adjustment of the data acquisition module in this transmission line area lightning activity monitoring and analysis system improves data acquisition accuracy and efficiency. This mechanism automatically adjusts the sensor's operating mode and sampling frequency based on real-time environmental conditions (such as temperature and humidity) and lightning activity intensity, ensuring high-quality data acquisition in diverse environments. A fuzzy logic controller automatically switches sensor operating modes and dynamically adjusts the sampling frequency, enabling the system to maintain efficient operation in diverse and complex environments. The FFT algorithm and Kalman filter in the data processing module significantly improve data processing speed and accuracy. The FFT algorithm rapidly processes large amounts of data, while the Kalman filter further optimizes the signal-to-noise ratio, reduces noise interference, and improves data quality. The lightning activity prediction module provides more accurate lightning activity forecasts. By combining physical models with machine learning algorithms, the system enables a more comprehensive analysis of lightning activity patterns, thereby improving forecast accuracy.

[0045] Figure 3 This is a schematic structural diagram of a device provided by an exemplary embodiment. Figure 3 At the hardware level, the device includes a processor 302, an internal bus 304, a network interface 306, a memory 308, and a non-volatile memory 310. Of course, it may also include hardware required for other functions. One or more embodiments of the present invention can be implemented based on software, such as the processor 302 reading the corresponding computer program from the non-volatile memory 310 into the memory 308 and then running it. Of course, in addition to software implementation, one or more embodiments of the present invention do not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc. In other words, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0046] Please refer to Figure 4 , a lightning activity prediction device for transmission line areas can be applied to Figure 4 In the device shown, to implement the technical solution of the present invention, the device may include: An acquisition unit 401 is configured to acquire environmental data from a plurality of sensors disposed along the transmission line and input the environmental data into a pre-trained lightning activity prediction model; wherein the lightning activity prediction model includes: a physical model module, a machine learning module, and a result fusion module; The prediction unit 402 is configured to input the environmental data into the physical model module and the machine learning module respectively, and input the physical model prediction results and the machine learning prediction results output by the two modules into the result fusion module, so that the result fusion module performs a weighted summation on the received prediction results; The determination unit 403 is configured to determine the weighted sum result output by the result fusion module as a target prediction result of lightning activity in the transmission line area.

[0047] Optionally, the physical model module predicts the probability of occurrence of lightning activity in the transmission line area based on a cloud-to-ground lightning transmission theoretical model, where the expression of the cloud-to-ground lightning transmission theoretical model is: ; in, is the prediction result of the physical model, is the temperature of the transmission line area, is the humidity in the transmission line area, is the air pressure in the transmission line area.

[0048] Optionally, the machine learning module includes a lightning location submodule and a prediction submodule; the device further includes: A calculation unit 404 is configured to calculate lightning strike information within the data acquisition area of ​​the multiple sensors through the lightning locating submodule, wherein the lightning strike information includes a current lightning strike position and a lightning strike speed; An input unit 405 is used to input the lightning strike information into the prediction submodule so that the prediction submodule predicts the probability of occurrence of lightning activity in the transmission line area; wherein the prediction submodule is constructed based on a deep neural network and a long short-term memory network trained by historical lightning data.

[0049] Optionally, the calculation expression of the weighted summation of the result fusion module is: ; in, is the weighted summation result output by the result fusion module, is the prediction result of the physical model, Q is the prediction result of the machine learning, and is the optimal weight of the two prediction results determined based on the cross-validation method.

[0050] Optionally, before the environmental data is input into the lightning activity prediction model, the device further includes: The processing unit 406 is configured to pre-process the environmental data based on a fast Fourier transform algorithm and optimize the signal-to-noise ratio of the environmental data according to a Kalman filter.

[0051] Furthermore, the optimization formula of the Kalman filter is: ; in, is the estimated state, is the Kalman gain, is the measured value, is the observation matrix.

[0052] Optionally, the acquiring unit 401 is specifically configured to: The operating modes and frequencies of the multiple sensors are adjusted according to environmental conditions.

[0053] The systems, devices, modules, or units described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer, which may be in the form of a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email transceiver, game console, tablet computer, wearable device, or any combination of these devices.

[0054] In a typical configuration, a computer includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0055] Memory may include non-permanent storage in a computer-readable medium, in the form of random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0056] Computer-readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage, quantum memory, graphene-based storage media or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0057] For the computer-readable medium (or computer-readable storage medium) as described above or in any other form, computer instructions may be stored thereon, which, when executed by a processor, implement one or more of the above-mentioned embodiments, thereby realizing the technical solution of the present invention.

[0058] The present invention further provides a computer program that, when executed by a processor, implements one or more of the aforementioned embodiments, thereby realizing the technical solution of the present invention. The computer program may be recorded on the aforementioned or any other form of computer-readable medium, and the present invention is not limited thereto.

[0059] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0060] The foregoing description describes specific embodiments of the present invention. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0061] The terms used in one or more embodiments of the present invention are for the purpose of describing specific embodiments only and are not intended to limit one or more embodiments of the present invention. The singular forms "a", "an", "the" and "the" used in one or more embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0062] It should be understood that although the terms first, second, third, etc. may be used to describe various information in one or more embodiments of the present invention, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of one or more embodiments of the present invention, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining."

[0063] The above description is merely a preferred embodiment of one or more embodiments of the present invention and is not intended to limit one or more embodiments of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of the present invention shall be included in the scope of protection of one or more embodiments of the present invention.

Claims

1. A method for predicting lightning activity in a transmission line area, characterized in that: include: Environmental data is acquired through multiple sensors installed along the power transmission line, and the environmental data is input into a pre-trained lightning activity prediction model; wherein the lightning activity prediction model includes: a physical model module, a machine learning module, and a result fusion module; Inputting the environmental data into the physical model module and the machine learning module respectively, and inputting the physical model prediction results and the machine learning prediction results output by the two modules into the result fusion module, so that the result fusion module performs a weighted summation on the received prediction results; The weighted summation result output by the result fusion module is determined as the target prediction result of lightning activity in the transmission line area.

2. The method according to claim 1, characterized in that The physical model module predicts the probability of lightning activity in the transmission line area based on the cloud-to-ground lightning transmission theoretical model. The expression of the cloud-to-ground lightning transmission theoretical model is: ； in, is the prediction result of the physical model, is the temperature of the transmission line area, is the humidity in the transmission line area, is the air pressure in the transmission line area.

3. The method according to claim 1, characterized in that The machine learning module includes a lightning location submodule and a prediction submodule; the method further includes: Calculating lightning strike information within the data acquisition area of ​​the multiple sensors by the lightning locating submodule, wherein the lightning strike information includes the current lightning strike position and lightning strike speed; The lightning strike information is input into the prediction submodule so that the prediction submodule predicts the probability of occurrence of lightning activity in the transmission line area; wherein the prediction submodule is constructed based on a deep neural network and a long short-term memory network trained by historical lightning data.

4. The method according to claim 1, wherein The calculation expression of the weighted summation of the result fusion module is: ； in, is the weighted summation result output by the result fusion module, is the prediction result of the physical model, Q is the prediction result of the machine learning, and is the optimal weight of the two prediction results determined based on the cross-validation method.

5. The method according to claim 1, wherein Before the environmental data is input into the lightning activity prediction model, the method further includes: The environmental data is preprocessed based on a fast Fourier transform algorithm, and a signal-to-noise ratio of the environmental data is optimized using a Kalman filter.

6. The method according to claim 5, characterized in that The optimization formula of the Kalman filter is: ； in, is the estimated state, is the Kalman gain, is the measured value, is the observation matrix.

7. The method according to claim 1, characterized in that The method of obtaining environmental data by using multiple sensors disposed along the power transmission line includes: The operating modes and frequencies of the multiple sensors are adjusted according to environmental conditions.

8. A lightning activity prediction device for a transmission line area, characterized in that: The device comprises: An acquisition unit acquires environmental data through multiple sensors installed along the power transmission line and inputs the environmental data into a pre-trained lightning activity prediction model; wherein the lightning activity prediction model includes: a physical model module, a machine learning module, and a result fusion module; Prediction unit: inputs the environmental data into the physical model module and the machine learning module respectively, and inputs the physical model prediction results and the machine learning prediction results output by the two modules into the result fusion module, so that the result fusion module performs a weighted summation on the received prediction results; A determination unit is configured to determine the weighted summation result output by the result fusion module as a target prediction result of lightning activity in the transmission line area.

9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor implements the steps of the method according to any one of claims 1 to 7 by running the executable instructions.

10. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the instruction is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images