Meteorological monitoring system for power transmission line environment

Abstract

The invention discloses a power transmission line environment meteorological monitoring system, and relates to the technical field of power transmission line monitoring, and the system comprises the following steps: deploying a plurality of meteorological sensors along a power transmission line, and obtaining high-precision topographic data at the same time; carrying out cleaning and abnormal value processing on the meteorological data, carrying out grid division on the topographic data, and carrying out timestamp alignment on all the data; key feature factors are calculated from the processed topographic data; meteorological and topographic data are fused through dynamic weights, and an icing risk index is calculated; dividing risk levels, analyzing a change trend, outputting a result to a monitoring interface, and storing the result for decision support; and when a high risk is detected, an early warning notification is automatically triggered, the deicing device is controlled to be started, and meanwhile, an operation log is recorded and algorithm parameters are updated to optimize performance. The power transmission line icing risk index calculation model is constructed, the local microclimate state under the complex terrain condition can be sensed, and spatial differentiation early warning can be carried out on the icing risk.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of power transmission line monitoring, and particularly relates to a power transmission line environment and meteorological monitoring system. BACKGROUND

[0002] The icing risk monitoring of power transmission lines usually relies on the data collected by meteorological sensors (such as temperature, humidity, and wind speed sensors) deployed along the line and gives a warning by establishing an empirical model or setting a threshold. However, these existing methods have an inherent limitation: they generally regard the monitoring environment as a homogeneous or large-scale unified field, ignoring the decisive influence of complex terrain (such as mountains and valleys) on the formation of microclimate. In actual geographical environments, terrain factors such as altitude, slope, and aspect will significantly change the local temperature distribution, wind speed field, and humidity conditions, resulting in a significant deviation between the actual meteorological conditions at different terrain locations in the same area and the inferred results based on limited measurement point data. This makes it difficult for traditional methods to accurately perceive the real microclimate environment along the power transmission line, especially in areas with dramatic terrain changes, resulting in misjudgment of icing risk, either missing the real risk or generating false alarms, ultimately restricting the accuracy of the warning and the effectiveness of subsequent protective measures. Therefore, in view of the above problems, the following scheme is proposed. SUMMARY

[0003] The purpose of the present application is to provide a power transmission line environment and meteorological monitoring system that can accurately perceive the local microclimate state under complex terrain conditions and spatially differentiate the icing risk warning by introducing a dynamic weight algorithm that fuses terrain features and real-time meteorological data, thereby solving the problem of inaccurate line icing risk perception and low warning reliability caused by ignoring the shaping effect of terrain on microclimate in existing monitoring methods.

[0004] To solve the above technical problems, the present application is realized by the following technical scheme:

[0005] The present application is a power transmission line environment and meteorological monitoring system, which comprises:

[0006] Data acquisition module: responsible for real-time acquisition of meteorological data and digital elevation model data from a distributed meteorological sensor network and geographic information system;

[0007] Data preprocessing module: cleans, removes outliers, gridded processes, and spatiotemporally aligns the collected raw data to ensure the integrity and consistency of the data;

[0008] Terrain feature extraction module: extracts and calculates altitude correction factor, slope influence factor, and aspect influence factor from digital elevation model data to quantify the influence of terrain on microclimate;

[0009] Ice risk calculation module: dynamically combine meteorological data and terrain features through fusion algorithm, calculate ice risk index, use machine learning optimized weight adjustment mechanism to improve prediction accuracy in complex terrain;

[0010] Risk assessment and decision module: divide risk levels according to risk index, analyze risk change trend, generate risk assessment results and decision suggestions, provide basis for early warning;

[0011] Early warning and control module: automatically trigger multi-channel early warning notification according to risk assessment results, control deicing equipment operation, and record all operation logs and system status;

[0012] Data storage and management module: provide unified data storage, parameter configuration and historical data management services for the system, support data access and parameter update requirements of each module;

[0013] The working process of the monitoring system is as follows:

[0014] Step S1, data acquisition: deploy various meteorological sensors along the transmission line, simultaneously obtain high-precision terrain data, and receive and verify all data in real time through the data preprocessing module to ensure integrity and consistency;

[0015] Step S2, data preprocessing: clean and process abnormal values of meteorological data, divide terrain data into grids, and align all data timestamps to provide standardized input for subsequent analysis;

[0016] Step S3, terrain feature extraction: calculate key feature factors from processed terrain data, such as altitude, slope and slope direction, to quantify the effect of terrain on local microclimate;

[0017] Step S4, risk index calculation: dynamically weight meteorological and terrain data, and apply nonlinear formula to calculate ice risk index;

[0018] Step S5, risk assessment: divide risk levels according to ice risk index and analyze change trend, output results to monitoring interface, and store for decision support;

[0019] Step S6, early warning and control: automatically trigger early warning notification when high risk is detected, control deicing device to start, record operation logs and update algorithm parameters to optimize performance.

[0020] Further, the step S1, data acquisition specifically includes the following steps:

[0021] Step S11: Deploy multiple weather sensor nodes along the power transmission line, each node including a temperature sensor, a humidity sensor, a wind speed sensor, and a wind direction sensor; the sensors are connected to a data preprocessing module in a wireless manner;

[0022] Step S12: Obtain digital elevation model (DEM) data for the area where the power transmission line is located, with a resolution of no less than 10 meters x 10 meters and covering several hundred meters on both sides of the power transmission line; the DEM data is pre-stored through a geographic information system (GIS) and updated in real time to reflect changes in the terrain;

[0023] Step S13: The central processing unit receives all sensor data and DEM data through a communication module (such as 4G / 5G or satellite link) and performs preliminary verification to ensure data integrity and consistency; for example, check if the sensor data is within a reasonable range (temperature: -50℃ to 50℃, humidity: 0% to 100%, wind speed: 0m / s to 60m / s), and invalid data is marked and excluded.

[0024] Further, the step S2, the data preprocessing specifically includes the following steps:

[0025] Step S21: Clean the collected weather data, use the sliding window method (window size is 5 sampling points) to detect and eliminate outliers, outliers are defined as points deviating from the window mean by more than 3 times the standard deviation, and linear interpolation method is used to fill in missing values;

[0026] Step S22: Grid processing of DEM data, divide DEM data into grid cells corresponding to sensor node positions, each grid cell size is 10 meters x 10 meters, extract the elevation value (in meters), slope value (in degrees) and aspect value (in degrees, 0° represents north, 90° represents east) of each grid cell;

[0027] Step S23: Align the time stamp of weather data and terrain data to ensure that all data correspond to the same time point (accurate to seconds); for example, for each sensor node, associate its weather data with the terrain data of the nearest grid cell.

[0028] Further, the step S3, the terrain feature extraction specific steps are:

[0029] From the pre-processed DEM data, calculate the terrain feature factors of each sensor node position, including:

[0030] Elevation correction factor E, defined as In the formula, H node is the node elevation, H base is the regional reference elevation (take the lowest point of the line), H max and Hmin are the highest and lowest altitudes in the region;

[0031] Slope impact factor S, defined as where θ is the slope value, θ0 is the reference slope (empirical value), σ θ is the slope standard deviation;

[0032] Aspect impact factor A, defined as where is the aspect value, is the current wind direction value (converted to degrees, 0° represents north);

[0033] These factors are used to quantify the impact of terrain on microclimate, for example, high altitude and specific aspect may increase the risk of icing.

[0034] Further, the step S4, the risk index calculation specific steps are:

[0035] Calculate the icing risk index R for each sensor node by the formula, and take the average of all nodes as the overall icing risk index R avg ; The formula dynamically integrates meteorological and terrain data, and the weights are optimized by machine learning models trained by historical data, as follows:

[0036]

[0037] where T is the current temperature (unit: ℃), T wet is the wet-bulb temperature (calculated from temperature T and humidity H), T min is the empirical minimum icing temperature; W is the current wind speed (unit: m / s), φ wind is the wind direction, φ line is the transmission line orientation (in degrees, obtained from GIS data), W max is the maximum reference wind speed; E, S, A are terrain feature factors obtained from step S3; α, β, Y are dynamic weight coefficients updated by online learning models: where ΔT, ΔW, ΔE are the deviations of the current values from the historical mean values, and k1, k2, k3 are training parameters (initial values are set based on historical icing event data);

[0038] This formula is different from the linear weighting of the prior art, and more accurately reflects the icing risk under complex terrain through nonlinear terms and dynamic weights.

[0039] Further, the step S5, the risk assessment step specifically includes the following steps:

[0040] Step S51: According to R avgThe icing risk is divided into three levels: low, medium and high according to the R avg When R < 0.3, the risk level is "low"; when 0.3 ≤ R avg When R < 0.6, the risk level is "medium"; when R avg When R ≥ 0.6, the risk level is "high".

[0041] Step S52: Calculate the risk trend, compare the current R avg With the average value in the past 1 hour, when the trend rises more than the threshold value, it is marked as "risk increase"; the result is output to the monitoring interface, including the risk index, level and trend, and recorded to the database for subsequent analysis.

[0042] Further, the step S6, the early warning and control specifically includes the following steps:

[0043] Step S61: When the risk level is "high" or "risk increase", trigger automatic early warning: notify the operation and maintenance personnel through SMS, email or monitoring platform alarm; the early warning information includes the specific risk position (based on the sensor node ID), risk index and recommended measures (such as starting the deicing device);

[0044] When the risk level is high, and after a period of time, start the heating deicing device (such as electric heating wire) on the transmission line, and adjust the power based on R avg value (power p = k × R avg , wherein k is a proportional coefficient);

[0045] Step S62: all operations are recorded to the log, and the weight parameters of the machine learning model are updated regularly (for example, retrain k1, k2, k3 using new data every week).

[0046] The present application has the following beneficial effects:

[0047] 1、The present application overcomes the defect of insufficient monitoring accuracy in complex terrain of the traditional method by dynamically integrating the terrain feature factor into the risk assessment model; the traditional method only relies on meteorological sensor data, ignores the shaping effect of terrain on microclimate, and causes deviation between the monitoring result and the actual microenvironment of the line; the present method quantifies the influence of terrain on meteorological parameters through algorithm, realizes more fine description of the real environment along the line, makes the perception of icing and other meteorological risks no longer limited to discrete measuring points, but extended to the entire line corridor, significantly improves the dimension and reliability of state perception, and provides a more solid data foundation for subsequent risk decision-making.

[0048] 2、The dynamic weight adjustment mechanism adopted by the application enables the system to adaptively adjust the contribution degree of each influencing factor according to the changes of real-time weather and terrain conditions; this changes the rigidity and hysteresis problem of the traditional fixed threshold or static weight model when facing the nonlinear and rapid change of microclimate; instead of mechanically responding to a single exceeding parameter, the system can comprehensively judge the trend risk under the coupling action of multiple factors, thereby realizing more sensitive and forward-looking capture of potential weather risks; this adaptability enables the monitoring system to better match the complex and changeable actual application scenarios, improving the timeliness and accuracy of risk warning.

[0049] 3、The application enables the operation and maintenance work to focus on the most critical parts by accurately identifying high-risk sections, such as prioritizing the inspection or starting the deicing device of the tower under specific terrain conditions; this decision support based on accurate positioning avoids indiscriminate line operation, ensures line safety while significantly reducing the consumption of manpower, material resources and energy, and improves the overall economy and intelligent level of transmission line operation and management.

[0050] Of course, implementing any product of the application does not necessarily require all the advantages described above. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the application, the following will briefly introduce the drawings needed to be used in the description of the embodiments. Obviously, the drawings in the following description only some embodiments of the application, and for those skilled in the art, other drawings can be obtained without creative labor on the basis of these drawings.

[0052] Figure 1 The flowchart of the transmission line environment weather monitoring system of the application. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only some embodiments of the application, not all embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application.

[0054] Please refer to Figure 1 The application is a transmission line environment weather monitoring system, and the monitoring system comprises:

[0055] Data acquisition module: responsible for real-time acquisition of weather data and digital elevation model data from distributed weather sensor network and geographic information system;

[0056] Data preprocessing module: The original data collected is cleaned, outliers are removed, gridding processing and spatio-temporal alignment are performed to ensure the integrity and consistency of the data;

[0057] Terrain feature extraction module: Extract and calculate the altitude correction factor, slope influence factor and aspect influence factor from the digital elevation model data, quantify the influence of terrain on microclimate;

[0058] Icing risk calculation module: Through the fusion algorithm, meteorological data and terrain features are dynamically combined to calculate the icing risk index, and the weight adjustment mechanism optimized by machine learning is used to improve the prediction accuracy in complex terrain;

[0059] Risk assessment and decision module: According to the risk index, the risk level is divided, the risk change trend is analyzed, and the risk assessment results and decision suggestions are generated to provide the basis for early warning;

[0060] Early warning and control module: According to the risk assessment results, automatic multi-channel early warning notification is triggered, the deicing equipment is controlled to operate, and all operation logs and system status are recorded;

[0061] Data storage and management module: Provide unified data storage, parameter configuration and historical data management services for the system, support data access and parameter update requirements of each module;

[0062] The workflow of the monitoring system is as follows:

[0063] Step S1, data collection: Deploy multiple weather sensors along the transmission line, obtain high-precision terrain data, and receive and verify all data in real time through the data preprocessing module to ensure integrity and consistency;

[0064] Step S1, data collection specifically includes the following steps:

[0065] Step S11: Deploy multiple weather sensor nodes along the transmission line, each node including a temperature sensor, a humidity sensor, a wind speed sensor and a wind direction sensor; The sensors are connected to the data preprocessing module in a wireless manner;

[0066] Step S12: Obtain the digital elevation model (DEM) data of the area where the transmission line is located, the DEM data resolution is not less than 10 meters x 10 meters, covering both sides of the transmission line hundreds of meters; DEM data is pre-stored through geographic information system (GIS) and updated in real time to reflect the changes in the terrain;

[0067] Step S13: The central processing unit receives all sensor data and DEM data through the communication module (such as 4G / 5G or satellite link) and performs preliminary verification to ensure data integrity and consistency; for example, check if the sensor data is within a reasonable range (temperature: -50℃ to 50℃, humidity: 0% to 100%, wind speed: 0m / s to 60m / s), invalid data is marked and excluded.

[0068] Step S2, data preprocessing: clean and outlier processing on weather data, gridding division on terrain data, and aligning all data timestamps to provide standardized input for subsequent analysis;

[0069] Step S2, data preprocessing specifically includes the following steps:

[0070] Step S21: clean the collected weather data, use the sliding window method (window size is 5 sampling points) to detect and eliminate outliers, outliers are defined as points deviating from the window mean by more than 3 times the standard deviation, and linear interpolation method is used to fill in missing values;

[0071] Step S22: gridding processing on DEM data, divide DEM data into grid cells corresponding to sensor node positions, each grid cell size is 10m x 10m, extract the elevation value (in meters), slope value (in degrees) and aspect value (in degrees, 0° represents north, 90° represents east) of each grid cell;

[0072] Step S23: align the weather data and terrain data timestamps to ensure that all data correspond to the same time point (accurate to seconds); for example, for each sensor node, associate its weather data with the terrain data of the nearest grid cell.

[0073] Step S3, terrain feature extraction: calculate key feature factors such as elevation, slope and aspect from processed terrain data to quantify the effect of terrain on local microclimate;

[0074] Step S3, terrain feature extraction specifically includes the following steps:

[0075] From the preprocessed DEM data, calculate the terrain feature factors of each sensor node position, including:

[0076] Elevation correction factor E, defined as where H node is the node elevation, H base is the regional reference elevation (take the lowest point of the line), H max and H min are the highest and lowest elevations of the region;

[0077] Slope impact factor S, defined as where θ is the slope value, θ0 is the reference slope (empirical value), σ θ is the slope standard deviation;

[0078] Aspect impact factor A, defined as where is the aspect value, is the current wind direction value (converted to degrees, 0° represents north);

[0079] These factors are used to quantify the impact of terrain on microclimate, for example, high altitude and specific aspect may increase the risk of icing.

[0080] Step S4, risk index calculation: by dynamically weighting meteorological and terrain data, applying a nonlinear formula to calculate the icing risk index;

[0081] Step S4, the specific steps of risk index calculation are:

[0082] Calculate the icing risk index R for each sensor node by the formula, and take the average of all nodes as the overall icing risk index R avg ; the formula dynamically fuses meteorological and terrain data, and the weights are optimized by a machine learning model trained by historical data, and the formula is as follows:

[0083]

[0084] where T is the current temperature (unit: ℃), T wet is the wet-bulb temperature (calculated from temperature T and humidity H), T min is the empirical minimum icing temperature; W is the current wind speed (unit: m / s), φ wind is the wind direction, φ line is the orientation of the transmission line (in degrees, obtained from GIS data), W max is the maximum reference wind speed; E, S, A are terrain feature factors, obtained from step S3; α, β, Y are dynamic weight coefficients, updated by an online learning model: where ΔT, ΔW, ΔE are the deviations of the current values from the historical mean values, and k1, k2, k3 are training parameters (initial values are set based on historical icing event data);

[0085] This formula is different from the linear weighting of the prior art, and more accurately reflects the icing risk under complex terrain through nonlinear terms and dynamic weights.

[0086] Step S5, risk assessment: divide the risk level according to the icing risk index and analyze the trend, output the results to the monitoring interface, and store them for decision support;

[0087] Step S5, the risk assessment step specifically includes the following steps:

[0088] Step S51: According to R avg value, the icing risk is divided into three levels: low, medium and high. When R avg <0.3, the risk level is "low"; when 0.3≤R avg <0.6, the risk level is "medium"; when R avg ≥0.6, the risk level is "high";

[0089] Step S52: Calculate the risk trend, compare the current R avg with the average value in the past 1 hour, when the trend rises more than the threshold, it is marked as "risk increase"; the results are output to the monitoring interface, including risk index, level and trend, and recorded to the database for subsequent analysis.

[0090] Step S6, warning and control: automatically trigger warning notification when high risk is detected, and control the deicing device to start, while recording operation log and updating algorithm parameters to optimize performance.

[0091] Step S6, warning and control specifically includes the following steps:

[0092] Step S61: When the risk level is "high" or "risk increase", trigger automatic warning: notify operation and maintenance personnel through SMS, email or monitoring platform alarm; warning information includes specific risk location (based on sensor node ID), risk index and recommended measures (such as starting deicing device);

[0093] When the risk level is high and lasts for a period of time, start the heating deicing device (such as electric heating wire) on the transmission line, and adjust the power based on R avg value (power p=k×R avg , where k is the proportional coefficient);

[0094] Step S62: All operations are recorded to the log, and the weight parameters of the machine learning model are updated regularly (for example, retrain k1, k2, k3 every week using new data).

[0095] In the description of the present specification, the description of the terms "one embodiment", "example", "specific example" and the like means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms does not necessarily refer to the same embodiment or example. Moreover, the described specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0096] The preferred embodiments of the application disclosed above are only to facilitate the elucidation of the application. The preferred embodiments do not describe all the details of the application and limit the application to the specific embodiments described. Obviously, many modifications and variations can be made in light of the teachings above. The description is chosen and described in order to best explain the principles of the application and its practical application to thereby enable others skilled in the art to best utilize the application and get the best results from the application. The application is only limited by the claims and their full scope and equivalents.

Claims

1. A power line environmental meteorological monitoring system, characterized in that, The monitoring system comprises: a data acquisition module: responsible for real-time acquisition of meteorological data from a distributed meteorological sensor network and geographic information system and digital elevation model data; a data preprocessing module: for cleaning, outlier removal, gridding processing and spatio-temporal alignment of the collected raw data, to ensure the integrity and consistency of the data; a terrain feature extraction module: for extracting and calculating elevation correction factors, slope influence factors and aspect influence factors from the digital elevation model data, to quantify the influence of terrain on microclimate; an icing risk calculation module: for dynamically combining meteorological data and terrain features through a fusion algorithm to calculate an icing risk index, and using a machine learning optimized weight adjustment mechanism to improve prediction accuracy in complex terrain; a risk assessment and decision module: for dividing risk levels according to the risk index, analyzing risk trends, generating risk assessment results and decision recommendations, and providing a basis for early warning; an early warning and control module: for automatically triggering multi-channel early warning notifications according to risk assessment results, controlling deicing equipment operation, and recording all operation logs and system status; a data storage and management module: for providing unified data storage, parameter configuration and historical data management services to support data access and parameter update requirements of each module; The working process of the monitoring system is as follows: Step S1, data acquisition: deploying multiple meteorological sensors along the power transmission line, simultaneously acquiring high-precision terrain data, and receiving and verifying all data in real time through the data preprocessing module to ensure integrity and consistency; Step S2, data preprocessing: cleaning and outlier processing of meteorological data, gridding division of terrain data, and timestamp alignment of all data to provide standardized input for subsequent analysis; Step S3, terrain feature extraction: calculating key feature factors from processed terrain data to quantify the effect of terrain on local microclimate; Step S4, risk index calculation: dynamically weighting meteorological and terrain data and applying a nonlinear formula to calculate the icing risk index; Step S5, risk assessment: dividing risk levels according to the icing risk index and analyzing trends, outputting results to the monitoring interface and storing them for decision support; Step S6, early warning and control: automatically triggering early warning notifications when high risk is detected and controlling deicing devices to start, while recording operation logs and updating algorithm parameters to optimize performance.

2. A system for monitoring the weather conditions in the environment of a power transmission line according to claim 1, characterized in that, The step S1, data acquisition specifically includes the following steps: Step S11: deploying multiple meteorological sensor nodes along the power transmission line, each node including a temperature sensor, a humidity sensor, a wind speed sensor and a wind direction sensor; Step S12: acquiring DEM data for the area where the power transmission line is located; DEM data is pre-stored through a geographic information system and updated in real time to reflect terrain changes; Step S13: the central processing unit receives all sensor data and DEM data through the communication module and performs preliminary verification to ensure data integrity and consistency.

3. The system for monitoring weather conditions along a power line route according to claim 1, wherein The step S2, data preprocessing specifically includes the following steps: Step S21: clean the collected meteorological data, detect and remove outliers using the sliding window method, and fill in missing values using linear interpolation; Step S22: grid the DEM data, divide the DEM data into grid cells corresponding to the sensor node positions, and extract the elevation, slope, and aspect values of each grid cell; Step S23: align the timestamps of the weather data and the terrain data to ensure that all data correspond to the same time point.

4. The system for monitoring weather conditions along a power line route according to claim 1, wherein The step S3, the terrain feature extraction specific steps are: From the preprocessed DEM data, calculate the terrain feature factors of each sensor node position, including: An elevation correction factor E is defined as where H node is the node elevation, H base is the zone reference elevation, H max and H min are the zone maximum and minimum elevations; The slope influence factor S is defined as where θ is the slope value, θ0is the reference slope, σ θ is the slope standard deviation; The slope direction influence factor A is defined as wherein is the slope direction value, is the current wind direction value.

5. The power line environmental weather monitoring system of claim 1, wherein, The step S4, the risk index calculation specific steps are: The icing risk index R of each sensor node is calculated by a formula, and the average value of all nodes is taken as the overall icing risk index R avg ; the formula dynamically fuses meteorological and terrain data, and the weight is optimized by a machine learning model trained by historical data, and the formula is as follows: where T is the current temperature, T wet is the wet-bulb temperature, T min is the empirical minimum icing temperature; W is the current wind speed, is the wind direction, is the transmission line orientation, W max is the maximum reference wind speed; E, S, and A are all terrain feature factors; and α, β, and Y are all dynamic weight coefficients.

6. The system for monitoring weather conditions along a power line route according to claim 1, wherein The step S5, the risk assessment step specifically includes the following steps: Step S51: According to R avg The icing risk is divided into three levels of low, medium and high according to the value. Step S52: Calculate risk trend, compare current R avg With past average, when trend rises above threshold, then flag "risk increasing"; output results to monitoring interface, including risk index, grade and trend, and record to database for later analysis.

7. The system for monitoring weather conditions along a power line route according to claim 1, wherein The step S6, the early warning and control specifically includes the following steps: Step S61: when the risk level is "high" or "risk increased", trigger automatic early warning: notify the operation and maintenance personnel through SMS, email or monitoring platform alarm; the early warning information includes the specific risk position, risk index and recommended measures; When the risk level is high, and for a certain period of time, the heating de-icing devices on the power transmission lines are activated, and the power is adjusted based on R avg values; Step S62: all operations are recorded to the log, and the weight parameters of the machine learning model are updated regularly.

Images