High-voltage transmission line cable damage prediction method and system

Abstract

The invention provides a high-voltage transmission line cable damage prediction method and system, and belongs to the technical field of high-voltage transmission line detection, and the method comprises the steps: carrying out the multi-source data collaborative collection of a high-voltage transmission line cable, so as to obtain a time-space synchronous multi-source data set; preprocessing the time-space synchronized multi-source data set to obtain a normalized feature data set; on the basis of the normalized feature data set, future meteorological data are accessed, and a current state + future environment data set is obtained, so that the cable state is associated with the future meteorological data; coupling the current state and the future environment data set, and constructing a prediction model; and outputting a risk level according to prediction of the prediction model. According to the high-voltage transmission line cable damage prediction method provided by the invention, related data sets are coupled to construct the prediction model, and the cable damage risk can be effectively predicted through model operation, so that a traditional mode of depending on artificial experience for judgment is changed, and the objectivity and accuracy of prediction are improved.

Description

Technical Field

[0001] This invention belongs to the technical field of high-voltage transmission cable detection, and more specifically, it relates to a method and system for predicting damage to high-voltage transmission line cables. Background Technology

[0002] High-voltage transmission lines are constantly exposed to complex and diverse natural environments and external interference. Factors such as lightning strikes, icing, strong winds, high temperatures, corrosion, tree growth, and collisions from construction work can all lead to potential faults in cables, including insulation aging, conductor damage, and sheath breakage. Failure to detect and address these hazards in a timely manner can result in anything from line tripping and partial power outages to cable burning, line paralysis, and even major safety accidents such as fires and widespread power outages, causing severe losses to society and its daily lives. Currently, the inspection and evaluation of high-voltage transmission line cables still primarily relies on traditional manual inspections. Maintenance personnel must use on-site inspections with testing equipment, climb towers for inspection, or utilize drone aerial photography to visually observe the cable's appearance, measure key parameters, and then combine this with experience to determine the potential for damage and the extent of the damage.

[0003] The existing prediction methods have the following drawbacks: manual inspection belongs to the "post-event detection" or "current status assessment" mode. Usually, measures are only taken after obvious signs of failure are found. It is difficult to build a correlation model based on historical data, environmental parameters and other multi-dimensional information. It is impossible to predict the damage trend of cables in the future, which makes the operation and maintenance work always in a passive response state and unable to formulate targeted prevention and maintenance plans in advance. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for predicting damage to high-voltage transmission line cables, aiming to solve the problem that manual inspection cannot predict the damage trend of cables.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: Firstly, a method for predicting damage to high-voltage transmission line cables is provided, including: Multi-source collaborative data acquisition is conducted for high-voltage transmission line cables to obtain spatiotemporally synchronized multi-source datasets. Preprocess the spatiotemporally synchronized multi-source dataset to obtain a normalized feature dataset; Based on the normalized feature dataset, future meteorological data is added to obtain the current state + future environment dataset, so as to associate the cable status with the future meteorological data. Couple the current state with the future environment dataset and build a predictive model; The risk level is output based on the prediction model.

[0006] In one possible implementation, multi-source data collaborative acquisition is performed on high-voltage transmission line cables to obtain a spatiotemporally synchronized multi-source dataset, including: Acquire real-time image information, real-time stress information, and real-time resistance information of the cable with timestamps; The real-time image information, real-time force information, and real-time resistance information of the cable are spatiotemporally synchronized to obtain a spatiotemporally synchronized multi-source dataset.

[0007] In one possible implementation, the real-time image information, real-time force information, and real-time resistance information of the cable are spatiotemporally synchronized to obtain a spatiotemporally synchronized multi-source dataset, including: The timestamps of real-time image information, real-time force information, and real-time resistance information are uniformly calibrated. Divide the cable into monitoring units and determine the coordinate boundaries of the monitoring units; Real-time image information, real-time force information, and real-time resistance information are matched to the corresponding monitoring units and sorted according to timestamps to obtain a spatiotemporally synchronized multi-source dataset.

[0008] In one possible implementation, a spatiotemporally synchronized multi-source dataset is preprocessed to obtain a normalized feature dataset, including: Preprocess the spatiotemporally synchronized multi-source dataset to obtain the preprocessed dataset; Normalize the preprocessed dataset to obtain a normalized feature dataset.

[0009] In one possible implementation, the spatiotemporally synchronized multi-source dataset is preprocessed to obtain a preprocessed dataset, including: Noise and background interference in real-time image information are eliminated to enhance the feature recognition of defect areas and extract feature vectors of defect areas. Eliminate noise and outliers in real-time force information and calculate the effective force value; The real-time resistance information is smoothed, and the relative rate of change of resistance is calculated.

[0010] In one possible implementation, based on the normalized feature dataset, future meteorological data is integrated to obtain a current state + future environment dataset, which facilitates the correlation between cable status and future meteorological data, including: Acquire high temporal resolution meteorological data covering the transmission line corridor area for the next 7 days; Acquire topographic data of the transmission line corridor area and fuse meteorological data with topographic data to obtain fused data; The fused data is mapped to each monitoring unit to facilitate spatial matching between the fused data and the normalized feature dataset; By correlating the fused data with the time series of the normalized feature dataset, a dataset of the current state plus the future environment is obtained.

[0011] In one possible implementation, the fused data is correlated with the time series of a normalized feature dataset to obtain a current state + future environment dataset, including: Extract timestamp sequences from the normalized feature dataset; Extract the fused data for the next 7 days corresponding to each timestamp, and construct a time series data pair of normalized feature data + fused data for the next 7 days; Correct the time deviation between the normalized feature data and the fused data for the next 7 days to obtain a dataset of current state + future environment.

[0012] In one possible implementation, the current state and future environment datasets are coupled together to build a predictive model, including: Redundant and highly correlated features in the current state + future environment dataset are removed to obtain core features, thereby reducing the computational complexity of the model. Feature weights are dynamically allocated according to different meteorological scenarios in order to highlight the contribution of key risk factors; A damage risk assessment model is constructed based on core features and feature weights.

[0013] In one possible implementation, a damage risk assessment model is constructed based on core features and feature weights, including: Construct an improved LSTM model; Meteorological time series correlation factors are embedded in the gating units of the LSTM model, and the weights of the forget gate and input gate are modified so that the LSTM model can capture the interaction patterns between state and meteorological time series. The cable material tolerance parameter is embedded as a hard constraint in the model loss calculation process.

[0014] Secondly, a high-voltage transmission line cable damage prediction system is provided, applied to the high-voltage transmission line cable damage prediction method as described in the first aspect, including: The data acquisition unit is used to perform multi-source collaborative data acquisition for high-voltage transmission line cables in order to obtain a spatiotemporally synchronized multi-source dataset. The data preprocessing unit is used to preprocess spatiotemporally synchronized multi-source datasets to obtain normalized feature datasets. The future meteorological data access unit is used to access future meteorological data on the basis of the normalized feature dataset to obtain the current status + future environment dataset, so as to associate the cable status with the future meteorological data. The predictive model building unit is used to couple the current state and future environment datasets and build a predictive model. The risk level output unit is used to output the risk level based on the prediction of the prediction model.

[0015] The beneficial effects of the high-voltage transmission line cable damage prediction method provided by the present invention are as follows: Compared with the prior art, the high-voltage transmission line cable damage prediction method of the present invention can comprehensively acquire cable-related data through multi-source data collaborative acquisition, breaking the limitations of traditional single data acquisition, making the perception of cable status more three-dimensional and accurate, and providing rich and comprehensive basic information for subsequent prediction.

[0016] Preprocessing the collected multi-source datasets removes interference factors, standardizes data formats, improves data quality, and prevents data clutter or errors from affecting subsequent analysis results, ensuring that subsequent model construction is based on a reliable data foundation. Integrating future meteorological data and correlating it with current cable status data combines the cable's own condition with changes in the external environment. This move moves beyond relying solely on current status assessments and allows for proactive consideration of future environmental impacts on cables, providing crucial information for predicting future cable damage.

[0017] By coupling relevant datasets to construct a predictive model, a scientific analytical framework can be established by integrating multiple factors. Through model calculations, the risk of cable damage can be effectively predicted, changing the traditional method of relying on manual experience and improving the objectivity and accuracy of predictions. Based on the risk level output by the model, different degrees of cable damage risk can be clearly presented, allowing maintenance personnel to intuitively understand the cable risk status and formulate targeted prevention and maintenance plans in advance. This transforms passive response to faults into proactive prevention, reducing safety accidents and losses caused by cable faults. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of the structure of the high-voltage transmission line cable damage prediction method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the main steps of step S100 provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the main steps of step S120 provided in an embodiment of the present invention; Figure 4This is a schematic diagram of the main steps of step S200 provided in an embodiment of the present invention; Figure 5 This is a schematic diagram of the main steps of step S210 provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the main steps of step S300 provided in an embodiment of the present invention; Figure 7 This is a schematic diagram of the main steps of step S340 provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of the main steps of step S400 provided in an embodiment of the present invention; Figure 9 A schematic diagram of the main steps of step S430 provided in an embodiment of the present invention. Detailed Implementation

[0020] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0021] Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without inventive effort are within the scope of protection of this application.

[0022] It should be further noted that the accompanying drawings and embodiments of the present invention mainly describe the concept of the present invention. Based on this concept, some specific forms and arrangements of connection relationships, positional relationships, power mechanisms, power supply systems, hydraulic systems and control systems may not be fully described. However, under the premise that those skilled in the art understand the concept of the present invention, they can implement the above-mentioned specific forms and arrangements in a well-known manner.

[0023] When a component is referred to as being "fixed to" or "set on" another component, it can be directly on or indirectly on that other component. When a component is referred to as being "connected to" another component, it can be directly connected to or indirectly connected to that other component.

[0024] In the description of this invention, "a plurality of" means two or more, and "several" means one or more, unless otherwise explicitly specified.

[0025] The directional terms "inner" and "outer" refer to the inner and outer contours relative to the outline of each component itself; the term "length"... "Width", "Top", "Bottom", "Front", "Back", "Left", "Right", "Vertical" The terms "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the purpose of facilitating the description of the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting the present invention.

[0026] For ease of description, spatial relative terms such as "above," "on top of," "on the upper surface of," and "above" are used here to describe the spatial positional relationship between a device or feature and other devices or features, as shown in the figure. It should be understood that spatial relative terms are intended to... The invention includes different orientations of the device in use or operation, in addition to those described in the figures. For example, if a device in the figures is inverted, a device described as "above" or "on top of" other devices or structures will be positioned as "below" or "under" other devices or structures. Thus, the exemplary term "above" can include both "above" and "below". The device may also be positioned in other different ways, and the spatial relative descriptions used herein are interpreted accordingly. The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the invention, "a plurality of" means two or more, and "a number" means one or more, unless otherwise explicitly specified.

[0027] Reference Figures 1 to 9 The present invention will now describe the method and system for predicting damage to high-voltage transmission line cables.

[0028] S100. Perform multi-source collaborative data acquisition for high-voltage transmission line cables to obtain a spatiotemporally synchronized multi-source dataset.

[0029] In one possible implementation, S100. Multi-source data collaborative acquisition is performed on high-voltage transmission line cables to obtain a spatiotemporally synchronized multi-source dataset, including: S110. Obtain real-time image information, real-time stress information, and real-time resistance information of the cable with timestamps.

[0030] Drones are used to conduct regular inspections along power transmission lines and capture real-time images of the cables. The drones are equipped with a dual-mode BeiDou / GPS positioning module. Data on the power transmission line corridor is imported into the drone's GIS system, and a zigzag inspection route is planned. The route is kept 8-10 meters parallel to the cables, and the flight altitude is 5-8 meters above the cables, ensuring that the cables occupy at least 80% of the image. Sampling is performed every 2 hours, with GPS coordinates and UTC timestamps (t) recorded simultaneously during data acquisition.

[0031] Real-time stress information of the cable is obtained by installing fiber optic tension sensors at 1 / 3 and 2 / 3 of the cable span, as well as a vibration sensor at the midpoint of the span.

[0032] By embedding high-precision four-wire resistance sensors inside the cable intermediate joints and terminal joints, real-time resistance information of the cable's conductive circuit is collected once per hour.

[0033] S120. Real-time image information, real-time force information and real-time resistance information of the cable are spatiotemporally synchronized to obtain a spatiotemporally synchronized multi-source dataset.

[0034] In one possible implementation, S120. The real-time image information, real-time force information, and real-time resistance information of the cable are spatiotemporally synchronized to obtain a spatiotemporally synchronized multi-source dataset, including: S121. The timestamps of real-time image information, real-time force information, and real-time resistance information are uniformly calibrated.

[0035] Using a GPS-BeiDou time synchronization module, the timestamps of the drone, tension sensor, and resistance sensor are uniformly calibrated to UTC time, with a synchronization error of no more than 1ms.

[0036] S122. Divide the monitoring units along the cable route and determine the coordinate boundaries of the monitoring units.

[0037] Using 50-meter intervals as monitoring units, the center point coordinates of each unit and the angle between the cable route and true north are determined through a GIS system. The unit coordinate boundaries are then calculated using the following formula:

[0038] in, , , and To monitor the latitude and longitude coordinates of the unit boundary; and The latitude and longitude coordinates of the center point of the monitoring unit; The angle between the cable's direction and due north.

[0039] S123. Match the real-time image information, real-time force information, and real-time resistance information to the corresponding monitoring units and sort them according to the timestamp to obtain a spatiotemporally synchronized multi-source dataset.

[0040] Based on GPS coordinates, the data from each sensor is matched to the corresponding monitoring unit, and a four-dimensional dataset of "time-monitoring unit-image / tension / vibration / resistance" is generated by sorting by timestamp. This four-dimensional dataset is then used as a multi-source dataset for spatiotemporal synchronization.

[0041] S200. Preprocess the spatiotemporally synchronized multi-source dataset to obtain a normalized feature dataset.

[0042] In one possible implementation, S200. preprocesses the spatiotemporally synchronized multi-source dataset to obtain a normalized feature dataset, including: S210. Preprocess the spatiotemporally synchronized multi-source dataset to obtain the preprocessed dataset.

[0043] In one possible implementation, S210. Preprocessing the spatiotemporally synchronized multi-source dataset to obtain a preprocessed dataset, including: S211. Eliminate noise and background interference in real-time image information to enhance the feature recognition of defect areas and extract feature vectors of defect areas.

[0044] The following formula is used for 8×8 block adaptive median filtering: the median is calculated for each pixel block and noise points are replaced. The filtering window is dynamically adjusted (from 3×3 to 7×7) based on the noise intensity within the block:

[0045] in, These are the filtered pixel values; These are the original pixel values; To calculate the median of the original pixel values ​​in a 3×3 window; To calculate the median of the original pixel values ​​in a 7x7 window; and These represent the minimum and maximum grayscale values ​​of the pixels within the block, respectively.

[0046] Histogram equalization is performed on the denoised image, and the gray value distribution is adjusted by the cumulative distribution function to improve the gray value difference between defects and the background.

[0047] The Canny edge detection algorithm is used to extract the cable outline, and the Hough linear transformation is used to locate the cable area, while background elements such as sky and vegetation are removed.

[0048] A 12-dimensional feature vector is extracted from the segmented defect region, including defect area, mean gray value, shape factor, eccentricity, etc.

[0049] S212. Eliminate noise and outliers in real-time force information and calculate the effective force value.

[0050] Kalman filtering of real-time force information is performed using the following formula:

[0051]

[0052] in, for The estimated optimal state at time t; for Prior estimates at time points; Kalman gain; for Time-based observations; for Time estimation error covariance; It is the identity matrix.

[0053] The effective value of vibration data within 1 minute is calculated using the following formula:

[0054] in, The valid values ​​of vibration data within 1 minute; This represents the total number of vibration data samples collected within one minute. For the first One vibration data sample value.

[0055] Calculate the valid value of tension data within 5 minutes using the following formula:

[0056] in, The arithmetic mean of tension data over a 5-minute period; This represents the total number of tension data samples collected within 5 minutes. For the first One sample value of tension data.

[0057] S213. Smooth the real-time resistance information and calculate the relative rate of change of resistance.

[0058] The real-time resistance information is smoothed using the following calculation formula:

[0059] in, This is the standard resistance value after smoothing with a 5-point moving average. The first in the corrected standard resistance value sequence One data point; and The first The first two and first one corrected standard resistance values ​​for each data point; and The first The last one and the last two corrected standard resistance values ​​for each data point.

[0060] The relative rate of change of resistance is calculated using the following formula:

[0061] in, for Time relative to The relative rate of change of resistance at time t; for Standard resistance value at 20℃ after time smoothing; for Standard resistance value at 20℃ after time smoothing.

[0062] S220. Perform normalization on the preprocessed dataset to obtain a normalized feature dataset.

[0063] Normalization is performed using the following formula:

[0064] in, These are the normalized data values; The original data value; This is the minimum value in the historical data of this parameter; This is the maximum value in the historical data of this parameter.

[0065] S300. Based on the normalized feature dataset, access future meteorological data to obtain the current state + future environment dataset, so as to associate the cable status with the future meteorological data.

[0066] In one possible implementation, S300, based on the normalized feature dataset, accesses future meteorological data to obtain a current state + future environment dataset, so as to associate the cable status with the future meteorological data, including: S310. Acquire high temporal resolution meteorological data covering the transmission line corridor area for the next 7 days.

[0067] S320. Obtain topographic data of the transmission line corridor area and fuse meteorological data with topographic data to obtain fused data.

[0068] ① Connect to the WRF mesoscale meteorological forecasting system and set up a three-layer nested grid, with the inner grid completely covering the transmission line corridor; ② Forecast parameter settings: output hourly meteorological data for the next 1-7 days, including temperature, wind speed, wind direction, precipitation, air humidity and icing thickness, with a forecast lead time step of 1 hour.

[0069] Acquire NCEP global topographic field data, extract topographic elevation data of transmission line corridors, and generate elevation raster maps. Correct the WRF output data based on the topographic elevation. Associate the corrected meteorological data with the topographic elevation data to form a fused "meteorological parameter-topographic parameter" data set.

[0070] S330. Map the fused data to each monitoring unit to facilitate spatial matching of the fused data with the normalized feature dataset.

[0071] Using the coordinates of the center point of the monitoring unit as the target point, meteorological data from four surrounding WRF grid points are selected as input; the meteorological parameter values ​​of the target point are calculated using a bilinear interpolation formula to achieve downscaling from 3km grid data to a 50m monitoring unit; the interpolated meteorological data are smoothed using a 3×3 window to eliminate abrupt changes between adjacent units.

[0072] S340. Correlate the fused data with the time series of the normalized feature dataset to obtain the current state + future environment dataset.

[0073] In one possible implementation, S340. The fused data is correlated with the time series of the normalized feature dataset to obtain a current state + future environment dataset, including: S341. Extract the timestamp sequence of the normalized feature dataset.

[0074] S342. Extract the fused data for the next 7 days corresponding to each timestamp, and construct a time series data pair of normalized feature data + fused data for the next 7 days.

[0075] S343. Correct the time deviation between the normalized feature data and the fused data for the next 7 days to obtain the current state + future environment dataset.

[0076] Time difference correction is performed using the following formula:

[0077] in, This is the timestamp for the synchronized meteorological data; This is the timestamp of the original meteorological data; This is due to time deviation.

[0078] S400. Couple the current state and future environment datasets and build a prediction model.

[0079] In one possible implementation, S400 couples the current state with the future environment dataset and constructs a predictive model, including: S410. Remove redundant and highly correlated features from the current state + future environment dataset to obtain core features, thereby reducing the computational complexity of the model.

[0080] The preprocessed 12-dimensional image features, 2-dimensional force features, 1-dimensional resistance features (ΔR%*), and 5-dimensional meteorological features are integrated to form a 20-dimensional initial feature set. The variance contribution rate of each feature is calculated, and features with a variance contribution rate of not less than 5% are retained. The Pearson correlation coefficient matrix of the filtered features is calculated, and redundant features with a correlation coefficient of not less than 0.8 are removed, finally retaining 10-12 core features.

[0081] S420. Dynamically allocate feature weights according to different meteorological scenarios in order to highlight the contribution of key risk factors.

[0082] An attention mechanism module based on MLP is built, which takes the core features as input and calculates the scores of each feature using the following formula:

[0083] in, For the first The score values ​​of each core feature; This is the weight matrix for the attention module; For the first One core feature; This is the bias term for the attention module.

[0084] The attention weights are calculated using the following formula:

[0085] in, For the first Attention weights for each core feature; For calculation functions; For the first The score of each core feature.

[0086] The weights are dynamically adjusted based on the weather scenario.

[0087] The weighted fusion feature vector is calculated using the following formula:

[0088] in, This is a weighted fusion feature vector; For the first Attention weights for each core feature; For the first One core feature.

[0089] S430. Construct a damage risk assessment model based on core features and feature weights.

[0090] In one possible implementation, S430. A damage risk assessment model is constructed based on core features and feature weights, including: S431. Construct an improved LSTM model.

[0091] Model structure design: The input layer receives a weighted fused feature vector (10-12 dimensions) and meteorological time series data for the next 168 hours; the hidden layer is a 3-layer improved LSTM (64 neurons per layer); a dropout layer (dropout=0.3) is added to suppress overfitting; a batch normalization layer is introduced to accelerate convergence; the output layer outputs the damage probability value in the 0-1 interval through the Sigmoid activation function.

[0092] S432. Embed meteorological time series correlation factors in the gating units of the LSTM model and modify the weights of the forget gate and the input gate so that the LSTM model can capture the interaction patterns between state and meteorological time series.

[0093] S433. The cable material tolerance parameter is embedded as a hard constraint in the model loss calculation process.

[0094] S500. Output the risk level based on the prediction of the prediction model.

[0095] The damage probability values ​​output by the model in step four are aligned with the operation and maintenance standards to classify four risk levels: low risk (≤10%), medium risk (10%-30%), high risk (30%-60%), and extremely high risk (>60%). The typical damage types corresponding to each level are also clearly defined, such as mechanical fracture and conductive failure corresponding to extremely high risk.

[0096] Secondly, a high-voltage transmission line cable damage prediction system is provided, applied to the high-voltage transmission line cable damage prediction method as described in the first aspect, including: The data acquisition unit is used to perform multi-source collaborative data acquisition for high-voltage transmission line cables in order to obtain a spatiotemporally synchronized multi-source dataset. The data preprocessing unit is used to preprocess spatiotemporally synchronized multi-source datasets to obtain normalized feature datasets. The future meteorological data access unit is used to access future meteorological data on the basis of the normalized feature dataset to obtain the current status + future environment dataset, so as to associate the cable status with the future meteorological data. The predictive model building unit is used to couple the current state and future environment datasets and build a predictive model. The risk level output unit is used to output the risk level based on the prediction of the prediction model.

[0097] The beneficial effects of the high-voltage transmission line cable damage prediction method provided by this invention are as follows: Compared with the prior art, the high-voltage transmission line cable damage prediction method of this invention... The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0098] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this application. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0099] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps described in these embodiments do not limit the scope of this application. It should also be understood that, for ease of description, the dimensions of the various parts shown in the drawings are not drawn to actual scale. Techniques, methods, and devices known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and devices should be considered part of the specification. In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values. It should be noted that similar reference numerals and letters in the following drawings denote similar items; therefore, once an item is defined in one drawing, it need not be further discussed in subsequent drawings.

Claims

1. A method of predicting damage to a high voltage power line cable, characterized by, include: Multi-source collaborative data acquisition is conducted for high-voltage transmission line cables to obtain spatiotemporally synchronized multi-source datasets. Preprocess the spatiotemporally synchronized multi-source dataset to obtain a normalized feature dataset; Based on the normalized feature dataset, future meteorological data is added to obtain the current state + future environment dataset, so as to associate the cable status with the future meteorological data. Couple the current state with the future environment dataset and build a predictive model; The risk level is output based on the prediction model.

2. The method for predicting damage to high-voltage transmission line cables as described in claim 1, characterized in that, Multi-source collaborative data acquisition is performed on high-voltage transmission line cables to obtain spatiotemporally synchronized multi-source datasets, including: Acquire real-time image information, real-time stress information, and real-time resistance information of the cable with timestamps; The real-time image information, real-time force information, and real-time resistance information of the cable are spatiotemporally synchronized to obtain a spatiotemporally synchronized multi-source dataset.

3. The method for predicting damage to high-voltage transmission line cables as described in claim 2, characterized in that, The real-time image information, real-time stress information, and real-time resistance information of the cable are spatiotemporally synchronized to obtain a spatiotemporally synchronized multi-source dataset, including: The timestamps of real-time image information, real-time force information, and real-time resistance information are uniformly calibrated. Divide the cable into monitoring units and determine the coordinate boundaries of the monitoring units; Real-time image information, real-time force information, and real-time resistance information are matched to the corresponding monitoring units and sorted according to timestamps to obtain a spatiotemporally synchronized multi-source dataset.

4. The method for predicting damage to high-voltage transmission line cables as described in claim 1, characterized in that, Preprocessing of spatiotemporally synchronized multi-source datasets to obtain normalized feature datasets includes: Preprocess the spatiotemporally synchronized multi-source dataset to obtain the preprocessed dataset; Normalize the preprocessed dataset to obtain a normalized feature dataset.

5. The method for predicting damage to high-voltage transmission line cables as described in claim 4, characterized in that, Preprocessing of spatiotemporally synchronized multi-source datasets yields preprocessed datasets, including: Noise and background interference in real-time image information are eliminated to enhance the feature recognition of defect areas and extract feature vectors of defect areas. Eliminate noise and outliers in real-time force information and calculate the effective force value; The real-time resistance information is smoothed, and the relative rate of change of resistance is calculated.

6. The method for predicting damage to high-voltage transmission line cables as described in claim 3, characterized in that, Based on the normalized feature dataset, future meteorological data is incorporated to obtain a current state + future environment dataset, which facilitates the correlation between cable status and future meteorological data, including: Acquire high temporal resolution meteorological data covering the transmission line corridor area for the next 7 days; Acquire topographic data of the transmission line corridor area and fuse meteorological data with topographic data to obtain fused data; The fused data is mapped to each monitoring unit to facilitate spatial matching between the fused data and the normalized feature dataset; By correlating the fused data with the time series of the normalized feature dataset, a dataset of the current state plus the future environment is obtained.

7. The method for predicting damage to high-voltage transmission line cables as described in claim 6, characterized in that, By correlating the fused data with the time series of the normalized feature dataset, a current state + future environment dataset is obtained, including: Extract timestamp sequences from the normalized feature dataset; Extract the fused data for the next 7 days corresponding to each timestamp, and construct a time series data pair of normalized feature data + fused data for the next 7 days; Correct the time deviation between the normalized feature data and the fused data for the next 7 days to obtain a dataset of current state + future environment.

8. The method for predicting damage to high-voltage transmission line cables as described in claim 1, characterized in that, The current state and future environment datasets are coupled together to build a predictive model, including: Redundant and highly correlated features in the current state + future environment dataset are removed to obtain core features, thereby reducing the computational complexity of the model. Feature weights are dynamically allocated according to different meteorological scenarios in order to highlight the contribution of key risk factors; A damage risk assessment model is constructed based on core features and feature weights.

9. The method for predicting damage to high-voltage transmission line cables as described in claim 8, characterized in that, A damage risk assessment model is constructed based on core features and feature weights, including: Construct an improved LSTM model; Meteorological time series correlation factors are embedded in the gating units of the LSTM model, and the weights of the forget gate and input gate are modified so that the LSTM model can capture the interaction patterns between state and meteorological time series. The cable material tolerance parameter is embedded as a hard constraint in the model loss calculation process.

10. A high-voltage transmission line cable damage prediction system, applied to the high-voltage transmission line cable damage prediction method as described in any one of claims 1 to 9, characterized in that, include: The data acquisition unit is used to perform multi-source collaborative data acquisition for high-voltage transmission line cables in order to obtain a spatiotemporally synchronized multi-source dataset. The data preprocessing unit is used to preprocess spatiotemporally synchronized multi-source datasets to obtain normalized feature datasets. The future meteorological data access unit is used to access future meteorological data on the basis of the normalized feature dataset to obtain the current status + future environment dataset, so as to associate the cable status with the future meteorological data. The predictive model building unit is used to couple the current state and future environment datasets and build a predictive model. The risk level output unit is used to output the risk level based on the prediction of the prediction model.

Images