Power transmission line icing galloping risk early warning system and early warning method

Abstract

The invention discloses a power transmission line icing galloping risk early warning system and early warning method, and belongs to the technical field of icing galloping risk early warning, and the system comprises the following modules: a historical data processing module reconstructs historical monitoring data by using a long and short term memory auto-encoder, recognizes abnormal data through reconstruction errors, and sends the abnormal data to an early warning module; removing outliers based on a 3sigma-dynamic threshold algorithm, and then marking a space-time credibility weight; the terrain compensation module constructs a micro-terrain feature vector, calculates the similarity through a Siamese network, and compensates low-confidence data; the real-time data processing module obtains the icing thickness according to the monitoring data of the current time point and the previous time point; the data acquisition and analysis module fuses historical and real-time data and analyzes a galloping state; and the risk early warning module takes the historical data with the weight and the galloping state information as input, outputs a risk value through the prediction model, and triggers early warning. According to the system, through cooperative work of all the modules, the icing galloping risk is accurately warned in real time, and the safety and reliability of power grid operation are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of icing galloping risk early warning, and particularly relates to a transmission line icing galloping risk early warning system and an early warning method. Background Art

[0002] During the power transmission process, icing galloping of transmission lines is a common natural disaster phenomenon. It is mainly due to the long-term accumulation of ice and snow on the surface of transmission lines under low-temperature and high-humidity meteorological conditions, resulting in an increase in the weight of the lines and a decrease in mechanical strength. At the same time, under the action of wind, the lines swing significantly. This phenomenon poses a serious threat to the safe and stable operation of the power grid, and may lead to faults such as conductor breakage, tower collapse, and insulator flashover, triggering large-scale power outages, which bring great inconvenience to social economy and people's lives.

[0003] Traditional monitoring means for icing galloping of transmission lines mainly rely on manual inspections and a small number of fixed monitoring devices. Manual inspections have disadvantages such as low efficiency, poor real-time performance, and being greatly affected by environmental factors, making it difficult to detect early signs of icing galloping in a timely manner. Although fixed monitoring devices can provide real-time data to a certain extent, due to their limited monitoring range and insufficient adaptability to complex terrains and meteorological conditions, they cannot comprehensively and accurately reflect the actual icing galloping situation of transmission lines.

[0004] Existing icing galloping early warning systems also have many limitations in terms of functions and performance. For example, the ability to process abnormal data is insufficient, and it is unable to effectively identify and eliminate abnormal data caused by sensor drift, communication packet loss, lightning strikes, equipment failures, etc., resulting in low data accuracy and affecting the reliability of early warning. In terms of considering terrain factors, most systems fail to fully combine complex micro-topographic features along transmission lines, such as altitude, slope, orientation, vegetation index, etc., resulting in insufficient prediction accuracy for icing galloping. In addition, there is no effective processing mechanism for the dynamics and uncertainty of real-time data, and the interpretability of the risk assessment model is poor, making it difficult to intuitively explain the basis and results of risk assessment to operation and maintenance personnel. Summary of the Invention

[0005] The purpose of the present invention is to provide a transmission line icing galloping risk early warning system, aiming to solve the problems proposed in the background art.

[0006] A transmission line icing galloping risk early warning system includes: A historical data processing module, which uses a long short-term memory autoencoder to reconstruct historical monitoring data, identifies abnormal data such as sensor drift and communication packet loss through the reconstruction error, and at the same time eliminates outliers caused by lightning strikes and equipment failures based on the 3σ-dynamic threshold algorithm, and then labels the spatio-temporal credibility weights for each piece of the reconstructed historical monitoring data; Terrain compensation module, which constructs micro-topographic feature vectors of altitude, slope, aspect, and vegetation index, calculates the similarity between the historical site and the target site through a Siamese network, processes the low-confidence reconstructed historical monitoring data in the historical data processing module, and migrates and compensates the data from high-similarity sites; Real-time data processing module, which obtains the ice thickness according to the ice thickness, meteorological conditions, and ice thickness increase rate under historical typical ice galloping monitored at the current time point and previous time points; Data acquisition and analysis module, which acquires the reconstructed historical monitoring data in the historical data processing module, and then analyzes the current status information of the ice-covered line galloping in combination with the real-time data in the real-time data processing module; Risk warning module, which takes the reconstructed historical monitoring data with spatio-temporal credibility weights output by the historical data processing module and the current status information of the ice-covered line galloping analyzed by the real-time data processing module as inputs, outputs a risk value through the constructed risk assessment and prediction model, and triggers an alarm when the risk value exceeds the preset threshold, sending a warning message containing the galloping risk level, possible galloping range, and recommended measures to the operation and maintenance personnel.

[0007] Furthermore, the calculation of the marked spatio-temporal credibility weights is as follows: Temporal credibility weight: W t =e -λ*∣tcurrent-trecord∣ Where: λ is the time decay factor, representing the decay rate of the credibility of historical data over time; t current is the current warning judgment time, and t record is the recording time of the historical data; Spatial credibility weight: W s =1 / (1 + α * Δd) Where: α is the spatial decay coefficient, representing the decay rate of the credibility of historical data over the spatial distance; Δd is the spatial distance between the current monitoring point and the historical recording point; Comprehensive credibility weight: W total =W t ×W s ×W device , and the W device is the equipment weight.

[0008] Furthermore, the micro-topographic feature vector also includes slope curvature and vegetation cover type. The Siamese network includes two sub-networks with shared weights, which are respectively used to extract the micro-topographic feature vectors of the historical site and the target site and calculate the similarity.

[0009] Furthermore, the risk warning module includes: A dual-input fusion unit that receives the reconstructed historical monitoring data with spatio-temporal credibility weights output by the historical data processing module and the current state information of the ice-covered line galloping output by the real-time data processing module; A spatio-temporal weighted training unit that constructs a risk assessment prediction model and assigns training sample weights to the reconstructed historical monitoring data according to the credibility weights; A real-time risk inference unit that inputs the current state information into the trained spatio-temporal weighted LSTM prediction model, outputs a risk probability distribution, and generates a risk level in combination with dynamic warning rules; A credibility feedback unit that calculates the credibility matching degree between the real-time warning result and the historical similar scenario.

[0010] Furthermore, the dynamic warning rules dynamically adjust the threshold according to the real-time meteorological conditions and the grid operation status. Specifically, real-time meteorological data and grid operation parameters are collected, and the threshold of risk assessment is adjusted according to the preset rule engine. When encountering severe meteorological conditions or high-load grid operation, the threshold is appropriately reduced to improve the sensitivity of warning.

[0011] Furthermore, the terrain compensation module is connected to the vegetation growth model and the geological change monitoring system to obtain the dynamic change information of the terrain in real time. According to this information, the characteristic values such as the vegetation index and soil type in the micro-topography feature vector are updated regularly, and then the historical data is dynamically corrected.

[0012] Furthermore, the vegetation index calculation unit of the terrain compensation module recalculates the vegetation index based on the data provided by the vegetation growth model and combines the vegetation spectral information obtained by satellite remote sensing, and matches the calculated vegetation index with other micro-topography feature data of the terrain compensation module in time and space.

[0013] Furthermore, the terrain compensation module uses the updated vegetation index and combines the model of the influence of vegetation on ice-covered galloping to correct the historical ice thickness, wind speed and other data.

[0014] Compared with the prior art, the beneficial effects of the present invention are: This solution combines a long short-term memory autoencoder and a 3σ-dynamic threshold algorithm to efficiently process historical data, remove anomalies, and accurately reconstruct it, providing a reliable basis for subsequent analysis. At the same time, a micro-topography compensation module for features such as slope curvature and vegetation cover type is introduced, and the Siamese network is used to accurately calculate the similarity of sites, realizing accurate data migration compensation and significantly improving the prediction accuracy of ice galloping. In addition, the real-time data processing module comprehensively uses multi-source data to track the change of ice thickness in real time to ensure the timeliness of early warning. The risk early warning module combines a spatio-temporal weighted LSTM prediction model and dynamic early warning rules to achieve accurate risk assessment and early warning, and enhances the credibility of the results through credibility feedback. Each module of the system works closely together to build a unified data management platform to optimize interactions, and cooperate with a visual interface to improve operation and maintenance efficiency. Overall, it greatly improves the accuracy, reliability, and practicality of the ice galloping risk early warning of transmission lines, providing a strong guarantee for the safe and stable operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 FIG. is an architecture diagram of a transmission line ice galloping risk early warning system provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0016] The embodiments described below by referring to the accompanying drawings are exemplary and are only used to explain the present invention and should not be construed as limiting the present invention. Embodiment

[0017] An embodiment of the present invention first provides a transmission line ice galloping risk early warning system, which can be applied to electronic devices, such as computer terminals, specifically ordinary computers, quantum computers, etc.

[0018] See Figure 1 , Figure 1 FIG. is a schematic diagram of a transmission line ice galloping risk early warning system provided by an embodiment of the present invention, including: A historical data processing module, which uses a long short-term memory autoencoder to reconstruct historical monitoring data, identifies abnormal data such as sensor drift and communication packet loss through reconstruction errors, and at the same time eliminates outliers caused by lightning strikes and equipment failures based on the 3σ-dynamic threshold algorithm, and then labels the spatio-temporal credibility weights for each piece of the reconstructed historical monitoring data; A terrain compensation module, which constructs a micro-topography feature vector of altitude, slope, orientation, and vegetation index, calculates the similarity between the historical site and the target site through the Siamese network, and migrates and compensates the data of the low-confidence reconstructed historical monitoring data in the historical data processing module from the high-similarity sites; A real-time data processing module, which obtains the ice thickness according to the ice thickness, meteorological conditions, and the ice thickness increase rate under historical typical ice galloping monitored at the current time point and previous time points. A data acquisition and analysis module, which acquires the reconstructed historical monitoring data in the historical data processing module, and then analyzes the current state information of the ice-covered line galloping in combination with the real-time data analysis in the real-time data processing module. A risk warning module, which takes the reconstructed historical monitoring data with spatio-temporal credibility weights output by the historical data processing module and the current state information of the ice-covered line galloping analyzed by the real-time data processing module as inputs, and outputs a risk value through a constructed risk assessment and prediction model. When the risk value exceeds the preset threshold, an alarm is triggered, and a warning message containing the galloping risk level, possible galloping range, and recommended measures is sent to the operation and maintenance personnel.

[0019] Specifically, the calculation of the marked spatio-temporal credibility weights is as follows: Temporal credibility weight: W t =e -λ*∣tcurrent-trecord∣ Where: λ is the time decay factor, representing the decay rate of the historical data credibility over time; t current is the current warning judgment moment, and t record is the recording moment of the historical data; Spatial credibility weight: W s =1 / (1 + α * Δd) Where: α is the spatial decay coefficient, representing the decay rate of the historical data credibility over the spatial distance; Δd is the spatial distance between the current monitoring point and the historical recording point; Comprehensive credibility weight: W total =W t ×W s ×W device where W device is the equipment weight.

[0020] Furthermore, the historical data processing module uses a long short-term memory autoencoder to encode and decode the historical monitoring data, reconstruct the data sequence, calculate the reconstruction error, and identify abnormal data according to the error magnitude. The 3σ-dynamic threshold algorithm dynamically determines the threshold range according to the mean and standard deviation of the historical data, eliminates the outliers outside the range, and calculates the spatio-temporal credibility weights through formulas, comprehensively considering factors such as the time decay factor, spatial decay coefficient, and equipment reliability, to assign accurate weight values to each reconstructed historical monitoring data.

[0021] Specifically, the micro-topography feature vector is constructed based on Geographic Information System (GIS) data and field measurement data to comprehensively describe the topographic features along the transmission line. The Siamese network contains two sub-networks with shared weights, which respectively extract the features of the micro-topography feature vectors of the historical site and the target site, calculate the similarity, select appropriate compensation data from the high-similarity sites according to the similarity, and correct the reconstructed historical monitoring data with low confidence to improve the accuracy and reliability of the data.

[0022] Specifically, the real-time data processing module receives the monitoring data at the current and previous time points, including ice thickness, meteorological conditions, etc. Combining with the ice thickness increase rate under historical typical ice galloping, according to the preset calculation model and algorithm, calculate the current ice thickness, and perform preliminary processing on the calculated ice thickness data, such as data format conversion, unit unification, etc., to make it conform to the data specifications within the system.

[0023] Furthermore, the data acquisition and analysis module obtains data from the historical data processing module and the real-time data processing module, adopts data fusion algorithms, such as Kalman filtering, weighted average, etc., to organically combine historical data and real-time data, and uses data analysis algorithms, such as time-frequency analysis, feature extraction, etc., to extract the key features in the data and generate the current state information of the ice-covered line galloping.

[0024] Specifically, the risk warning module integrates historical data and current state information with spatio-temporal credibility weights, inputs them into the spatio-temporal weighted LSTM prediction model, calculates the risk probability distribution, combines with dynamic warning rules, dynamically adjusts the threshold according to real-time meteorological conditions and the power grid operation state, generates the final risk level, and the credibility feedback unit calculates the credibility matching degree between the real-time warning result and the historical similar scenario to ensure the credibility of the warning result.

[0025] Furthermore, the real-time monitoring device in the real-time data processing module collects the current operation data of the transmission line, including ice thickness, meteorological conditions (temperature, wind speed, humidity, etc.), and transmits the data to the real-time data processing module through a wireless or optical fiber communication link. This module has efficient data reception capabilities to ensure the real-time and integrity of the data. Embodiment

[0026] The embodiment of the present invention also provides an operation method for a transmission line ice galloping risk warning system.

[0027] Specifically, in this embodiment, the above operation method performs the following steps: S1. When the system starts, the configuration management module sends loading instructions to each module. Each module reads the parameters it needs from the configuration file or database, such as the historical data path, model hyperparameters, communication ports, etc., to prepare for subsequent operation. The historical data processing module sends a connection request to the database or data warehouse to establish a data channel for reading the original historical monitoring data. The real-time data processing module connects to the data output port of the front-end monitoring device to prepare for receiving real-time monitoring data. The terrain compensation module loads the micro-topographic feature data to construct an initial micro-topographic feature vector. The risk warning module initializes the risk assessment and prediction model, including the spatio-temporal weighted LSTM prediction model, etc., and at the same time loads the pre-trained model parameters to ensure that the model is in a runnable state. The risk warning module initializes the risk assessment and prediction model, including the spatio-temporal weighted LSTM prediction model, etc., and at the same time loads the pre-trained model parameters to ensure that the model is in a runnable state. The real-time data processing module calculates the current ice thickness based on the monitoring data at the current and previous time points, combined with the ice thickness increase rate under historical typical ice galloping. S2. The historical data processing module reads the original historical monitoring data, including vibration frequency, ice thickness, wind speed, etc., from the database or data file according to the configured path and format. The read data is preliminarily cleaned, such as removing obviously incorrect records, unifying the data format, etc., to ensure the basic quality of the data. The preprocessed data is input into the long short-term memory autoencoder, and the autoencoder encodes and decodes the data to reconstruct the historical monitoring data. By comparing the reconstruction error between the reconstructed data and the original data, abnormal data such as sensor drift and communication packet loss is identified and marked. Based on the 3σ - dynamic threshold algorithm, the mean and standard deviation of the historical monitoring data are calculated to determine the dynamic threshold range, and the outliers caused by lightning strikes and equipment failures are removed to obtain a clean historical monitoring data set. A spatio-temporal credibility weight is assigned to each clean historical monitoring data. S3. The terrain compensation module constructs a micro-topographic feature vector containing elements such as altitude, slope, orientation, vegetation index, etc., based on the geographic information system (GIS) data and field measurement data to comprehensively describe the topographic features along the transmission line. The Siamese network is used to calculate the similarity between the historical site and the target site. The Siamese network contains two sub-networks with shared weights, which respectively extract the features of the micro-topographic feature vectors of the historical site and the target site, and then calculate their similarity to provide a basis for data compensation. According to the similarity, the compensated data is migrated from the high-similarity site to the low-confidence reconstructed historical monitoring data in the historical data processing module. In this way, the deviation of the monitoring data caused by terrain differences is corrected to improve the accuracy and reliability of the data. S4. The real-time data processing module receives the monitoring data at the current time point and previous time points from the front-end monitoring devices, including ice coating thickness, meteorological conditions, etc. The monitoring devices transmit the data to the real-time data processing module through a wireless communication network or an optical fiber communication link; S5. The data acquisition and analysis module obtains the reconstructed historical monitoring data with spatio-temporal credibility weights from the historical data processing module and simultaneously obtains the real-time monitoring data from the real-time data processing module. These two parts of data are fused using data fusion algorithms such as Kalman filtering and weighted average to organically combine the historical data and real-time data, improving the integrity and accuracy of the data. The fused data will be used to extract the current state information of the galloping of the ice-covered line. Data analysis algorithms such as time-frequency analysis and feature extraction are used to extract key features in the data, such as galloping amplitude, frequency, phase, etc., to generate the current state information of the galloping of the ice-covered line, providing real-time data support for risk warning; S6. The risk warning module receives the current state information of the galloping of the ice-covered line output by the data acquisition and analysis module and the reconstructed historical monitoring data with spatio-temporal credibility weights output by the historical data processing module. These data are integrated and preprocessed to meet the input requirements of the risk assessment and prediction model, and the integrated data is input into the spatio-temporal weighted LSTM prediction model. The model calculates the risk probability distribution based on the characteristics of the historical data and real-time data. Combining with dynamic warning rules, the threshold is dynamically adjusted according to the real-time meteorological conditions and the grid operation status to generate the final risk level. The credibility feedback unit calculates the credibility matching degree between the real-time warning result and the historical similar scenarios. If the risk value exceeds the preset threshold, the warning mechanism is triggered to send a warning message to the operation and maintenance personnel. The warning message includes the galloping risk level, possible galloping range, recommended measures, etc., and is notified to the operation and maintenance personnel in a timely manner through methods such as text messages, emails, or system interface displays.

[0028] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present invention is not limited by the described action sequence because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0029] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0030] In several embodiments provided by the present invention, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the above division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling, direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of devices or units can be in electrical or other forms.

[0031] The units described above as separate components may or may not be physically separated. The components displayed as units may or may 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 achieve the purpose of the solution of this embodiment.

[0032] In addition, each functional unit in various embodiments of the present invention 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. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0033] If the above integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. The aforementioned memory includes: various media such as USB flash drives, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), mobile hard disks, magnetic disks or optical discs that can store program codes.

[0034] The above has introduced the embodiments of the present invention in detail. Specific examples are used in this article to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A risk early warning system for icing galloping of transmission lines, characterized in that, The system includes a historical data processing module that uses a long short-term memory autoencoder to reconstruct historical monitoring data, identifies abnormal data such as sensor drift and communication packet loss through reconstruction errors, and at the same time eliminates outliers caused by lightning strikes and equipment failures based on the 3σ-dynamic threshold algorithm, and then annotates spatio-temporal credibility weights for each piece of the reconstructed historical monitoring data; a terrain compensation module that constructs micro-topographic feature vectors of altitude, slope, aspect, and vegetation index, calculates the similarity between the historical site and the target site through a Siamese network, and migrates and compensates data from high-similarity sites for the low-confidence reconstructed historical monitoring data in the historical data processing module; a real-time data processing module that obtains the ice coating thickness based on the ice coating thickness, meteorological conditions, and ice coating thickening rate under historical typical ice coating galloping monitored at the current time point and previous time points; a data acquisition and analysis module that acquires the reconstructed historical monitoring data in the historical data processing module, and then analyzes the current state information of the ice coating line galloping in combination with the real-time data in the real-time data processing module; a risk warning module that takes the reconstructed historical monitoring data with spatio-temporal credibility weights output by the historical data processing module and the current state information of the ice coating line galloping analyzed by the real-time data processing module as inputs, outputs a risk value through the constructed risk assessment prediction model, and when the risk value exceeds the preset threshold, triggers an alarm and sends an alarm message containing the galloping risk level, possible galloping range, and recommended measures to the operation and maintenance personnel.

2. The ice galloping risk early warning system for a transmission line according to claim 1, characterized in that The calculation of the marked spatio-temporal credibility weight is divided into: the time credibility weight: W t =e -λ*∣tcurrent-trecord∣ Where: λ is the time decay factor, representing the decay rate of the historical data credibility over time; t current is the current warning judgment time, t record is the recording time of historical data; Spatial credibility weight: W s = 1 / (1 + α * Δd) Where: α is the spatial decay coefficient, representing the decay rate of the historical data credibility over the spatial distance; Δd is the spatial distance between the current monitoring point and the historical recording point; Comprehensive credibility weight: W total =W t ×W s ×W device , where the W device is the device weight.

3. The ice galloping risk early warning system for a transmission line according to claim 2, characterized in that, The micro-topographic feature vector further includes slope curvature and vegetation cover type. The Siamese network includes two sub-networks with shared weights, which are respectively used to extract the micro-topographic feature vectors of the historical site and the target site and calculate the similarity.

4. The ice-shedding galloping risk early warning system for a transmission line according to claim 1, characterized in that, The risk warning module includes: a dual-input fusion unit that receives the reconstructed historical monitoring data with spatio-temporal credibility weights output by the historical data processing module and the current state information of the ice coating line galloping output by the real-time data processing module; a spatio-temporal weighted training unit that constructs a risk assessment prediction model and assigns training sample weights to the reconstructed historical monitoring data according to the credibility weights; a real-time risk inference unit that inputs the current state information into the trained spatio-temporal weighted LSTM prediction model, outputs a risk probability distribution, and generates a risk level in combination with the dynamic warning rule; a credibility feedback unit that calculates the credibility matching degree between the real-time warning result and the historical similar scenario.

5. The icing galloping risk early warning system for a transmission line according to claim 1, characterized in that, The dynamic warning rule dynamically adjusts the threshold according to the real-time meteorological conditions and the power grid operation state. Specifically, it collects real-time meteorological data and power grid operation parameters, and adjusts the threshold of risk assessment according to the preset rule engine. When encountering severe meteorological conditions or high-load operation of the power grid, the threshold is appropriately reduced to improve the sensitivity of the warning.

6. The ice galloping risk early warning system for a transmission line according to claim 1, characterized in that The terrain compensation module is connected to the vegetation growth model and the geological change monitoring system to obtain real-time dynamic change information of the terrain. According to this information, the characteristic values such as vegetation index and soil type in the micro-topography feature vector are updated regularly, and then the historical data is dynamically corrected.

7. The ice accretion galloping risk early warning system for a transmission line according to claim 6, wherein The vegetation index calculation unit of the terrain compensation module recalculates the vegetation index based on the data provided by the vegetation growth model and combines the vegetation spectral information obtained by satellite remote sensing, and matches the calculated vegetation index with other micro-topography feature data of the terrain compensation module in time and space.

8. A transmission line icing galloping risk early warning system according to claim 7, characterized in that, The terrain compensation module uses the updated vegetation index and combines it with the model of the influence of vegetation on galloping icing to correct the historical data such as icing thickness and wind speed.

9. A risk early warning method for icing galloping of transmission lines, characterized in that, It includes: The historical data processing module uses the long short-term memory autoencoder to reconstruct the historical monitoring data, identifies and eliminates abnormal data, then eliminates outliers based on the 3σ-dynamic threshold algorithm, and finally labels the spatio-temporal credibility weights for the reconstructed data; The terrain compensation module constructs a micro-topography feature vector, calculates the similarity between the historical site and the target site through the Siamese network, and migrates the compensation data from the high-similarity site to the low-confidence reconstructed historical monitoring data; For real-time data processing, the module calculates the current icing thickness according to the monitoring data at the current and previous time points and combines the icing thickening rate under historical typical galloping icing. The data acquisition and analysis module acquires the data of the historical data processing module and the real-time data processing module, and fuses and analyzes to obtain the current state information of the galloping of the icing line; The risk warning module receives relevant data, outputs a risk value through the risk assessment prediction model. If the risk value exceeds the preset threshold, an alarm is triggered and a warning message is sent to the operation and maintenance personnel.

Images