A tree obstacle assessment method and system

Abstract

The present invention relates to the field of power system operation and maintenance technology, and in particular to a tree barrier assessment method and system, the method comprising: obtaining tree growth information in a transmission line protection area, and obtaining tree barrier growth multidimensional data based on the tree growth information; generating multiple tree barrier prediction subsets based on the tree barrier growth multidimensional data, and extracting tree barrier growth features in each of the tree barrier prediction subsets; constructing a tree barrier assessment model, the tree barrier assessment model performs tree barrier assessment on the tree growth in the transmission protection area based on the tree barrier growth features, and outputs a tree barrier assessment result. Through the present invention, the difficulty of identifying tree barriers in the transmission line area is effectively solved, and the problem of how to effectively extract data related to tree barriers from remote sensing image data is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system operation and maintenance, and in particular to a tree barrier assessment method and system. Background Art

[0002] The electric power industry is the most basic energy industry in the national economy and is closely related to daily life and work. Tree barriers refer to the problems caused by the growth of trees within the transmission line corridor, which affect power transmission equipment. These mainly include the excessive growth of trees, which causes the transmission equipment such as poles and towers to be blocked, and the insufficient safety distance between trees and power lines, which causes line short circuits or tripping.

[0003] Power line failures can cause great inconvenience and losses to people's daily life and work. On the other hand, some transmission lines and towers are located in sparsely populated jungles and deserts, and satellite technology is needed to monitor tree barriers. However, the currently collected monitoring information lacks effective extraction of data related to tree barriers to assess the interference trend of tree barriers on transmission corridors.

[0004] The information disclosed in this background technology section is only intended to deepen the understanding of the overall background technology of the present disclosure, and should not be regarded as acknowledging or suggesting in any form that the information constitutes the prior art known to those skilled in the art. Summary of the invention

[0005] The present invention provides a tree barrier assessment method and system, which can effectively solve the problems in the background technology.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is:

[0007] A tree obstacle assessment method, the method comprising:

[0008] Acquire tree growth information within the protection area of ​​the transmission line, and obtain multi-dimensional data of tree barrier growth based on the tree growth information;

[0009] Generate multiple tree barrier prediction subsets according to the tree barrier growth multidimensional data, and extract tree barrier growth features in each of the tree barrier prediction subsets;

[0010] A tree barrier assessment model is constructed, wherein the tree barrier assessment model performs a tree barrier assessment on the growth of trees in the power transmission protection area according to the tree barrier growth characteristics, and outputs a tree barrier assessment result.

[0011] Furthermore, a tree barrier assessment model is constructed, including:

[0012] Constructing a tree obstacle processing layer for the tree obstacle prediction subset, wherein the tree obstacle processing layer corresponds to the tree obstacle prediction subset one by one;

[0013] Generate a tree growth prediction sub-model according to the tree obstacle processing layer, the tree growth prediction sub-model performs a height prediction on the tree growth to obtain a height prediction result;

[0014] A tree barrier assessment model is constructed based on the tree growth prediction sub-model.

[0015] Furthermore, a tree growth prediction sub-model is generated according to the tree obstacle processing layer, including:

[0016] Based on the multi-dimensional data of tree barrier growth, a fixed array width sample is extracted by repeated sampling to generate a plurality of tree barrier prediction subsets, each of which contains different sample combinations;

[0017] Recursively construct the tree barrier processing layer according to the node splitting rule until a preset stop condition is reached, thereby completing the generation of a single tree barrier processing layer;

[0018] When each node is split, a key split point is selected based on the tree barrier growth characteristics, wherein the key split point having a large weight on the tree growth trend is preferentially selected;

[0019] The tree growth prediction sub-model is constructed corresponding to each of the tree obstacle processing layers, and each of the tree growth prediction sub-models has tree growth with different structures and feature combinations.

[0020] Furthermore, a tree obstacle processing layer is recursively constructed according to the node splitting rule, including:

[0021] Based on the tree barrier growth features in the tree barrier prediction subset, calculating the splitting gain of each of the tree barrier growth features at the current node, and selecting the tree barrier growth feature with the highest gain value as the splitting feature of the current node;

[0022] Determine the split point of the current node, and select the node with the largest split gain as the key split point by analyzing the value range of the split feature and according to the feature distribution;

[0023] Recursively calculate the feature gain and select the split point for each node to generate branch nodes until the split gain value is lower than the set minimum gain threshold, generating a predicted leaf node;

[0024] Generate a corresponding growth prediction value by integrating the tree barrier growth trend and environmental characteristic data of the samples in the predicted leaf node and combining the remaining generated leaf node results;

[0025] The tree obstacle prediction subset and the prediction leaf node are integrated to construct a tree obstacle processing layer.

[0026] Furthermore, tree growth information within the protection area of ​​the transmission line is obtained, including:

[0027] Acquire remote sensing images of the transmission line protection area through satellites, pre-process the remote sensing images, divide the remote sensing images into grids, scan them grid by grid to identify tower features, and determine the fuzzy range of the towers based on the identified grid areas;

[0028] Perform fuzzy search within the fuzzy range according to the ledger to determine the positions of all towers within the fuzzy range;

[0029] Using a ground object recognition algorithm, the longitude and latitude coordinates of the tower within the fuzzy range and the length of the tower shadow are identified from the image;

[0030] The longitude and latitude coordinates of the tree barrier and the length of the tree barrier shadow are identified from the remote sensing image according to the protection area of ​​the transmission line to identify the location of the tree barrier and obtain the tree growth information.

[0031] Further, the location of tree obstacles is identified, including:

[0032] According to the tower information recorded in the ledger, the corresponding tower height value is confirmed, and according to the tower height value and the tower shadow length, the average value of the solar altitude angle at the time of shooting is calculated;

[0033] Calculating the height of the tree barrier according to the average value of the solar altitude angle and the length of the tree barrier shadow;

[0034] According to the position of the tower body and the longitude and latitude coordinates of the tree barrier, confirm the center point of the tower and the tree barrier in the protection area of ​​the transmission line;

[0035] According to the coordinates of the center point of the tower and the tower height information and the coordinates of the center point of the tree barrier and the tree barrier height information, the vertical distance from the tree barrier to the transmission line is calculated and the location of the tree barrier is identified according to the protection area of ​​the transmission line.

[0036] Furthermore, the tree growth height is predicted, including:

[0037] Classify and identify tree barriers within the line protection area based on the remote sensing images taken by satellite, and determine the type, location and distribution characteristics of each tree barrier;

[0038] According to the multi-dimensional data of tree barrier growth, a tree allometric growth equation is constructed, and parameters are optimized according to the characteristics of different tree species;

[0039] The future growth height and growth trend of tree barriers at different times and seasons are predicted based on the tree allometric growth equation, and the potential impact of tree barriers on transmission lines is evaluated.

[0040] Furthermore, the tree allometric growth equation is constructed, including:

[0041] Based on the multi-dimensional data of tree barrier growth, characteristic variables related to tree growth trends are selected as input parameters of the tree allometric growth equation;

[0042] According to the allometric growth characteristics of trees, the tree allometric growth equation model suitable for different tree species is selected, and preliminary model parameters are set;

[0043] Using the original data set, the selected tree allometric growth equation model is tuned by maximizing the fit, and the weight of each characteristic variable and the model parameters are optimized;

[0044] Verifying the optimized tree allometric growth equation model, comparing the predicted growth height with the actual historical data, and setting an error threshold;

[0045] If the prediction error is within the set error threshold range, an optimized allometric growth equation is generated.

[0046] Furthermore, the tree barrier assessment model performs a tree barrier assessment on the growth of trees in the power transmission protection area according to the tree barrier growth characteristics, and outputs a tree barrier assessment result, including:

[0047] Based on the tree growth prediction sub-model, a tree barrier assessment model is constructed, the tree barrier growth features in the tree barrier prediction subset are used as input, and the feature importance score of each tree barrier growth feature in growth prediction is calculated through multiple trainings;

[0048] According to the feature importance scores, all the tree barrier growth features are ranked to select the key features that contribute most to tree growth prediction;

[0049] Based on the screened maximum key features, analyzing the impact on the tree barrier growth trend, and obtaining the tree barrier growth threshold;

[0050] Evaluate the tree barrier, and when the growth trend of the tree barrier reaches the preset tree barrier growth threshold, record the tree barrier information in the hidden danger database;

[0051] Tree barriers that exceed the growth threshold are marked and updated, and tree growth height prediction results and key influencing factor analysis reports are output.

[0052] A tree obstacle assessment system, the system comprising:

[0053] An information acquisition module is used to acquire tree growth information within the protection area of ​​the power transmission line, and obtain multi-dimensional data of tree barrier growth based on the tree growth information;

[0054] A feature prediction module generates a plurality of tree barrier prediction subsets according to the tree barrier growth multidimensional data, and extracts tree barrier growth features from each of the tree barrier prediction subsets;

[0055] The result evaluation module constructs a tree barrier evaluation model, wherein the tree barrier evaluation model performs a tree barrier evaluation on the tree growth in the power transmission protection area according to the tree barrier growth characteristics, and outputs a tree barrier evaluation result.

[0056] The technical solution of the present invention can achieve the following technical effects:

[0057] It effectively solves the difficulty of identifying tree barriers in the transmission line area and the problem of how to effectively extract data related to tree barriers from remote sensing image data.

[0058] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0060] Figure 1 It is a flowchart of a tree barrier assessment method;

[0061] Figure 2 This is a schematic diagram of the tree barrier assessment model;

[0062] Figure 3 It is a structural schematic diagram of the tree barrier treatment layer;

[0063] Figure 4 Schematic diagram of the process for obtaining tree growth information;

[0064] Figure 5 Schematic diagram of the process of constructing the allometric equation for tree growth. DETAILED DESCRIPTION

[0065] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0066] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items.

[0067] Embodiment 1

[0068] like Figure 1 As shown, the present application provides a tree obstacle assessment method, the method comprising:

[0069] Obtain tree growth information within the protection area of ​​the transmission line, and obtain multi-dimensional data of tree barrier growth based on the tree growth information;

[0070] Generate multiple tree barrier prediction subsets according to the tree barrier growth multidimensional data, and extract tree barrier growth features in each tree barrier prediction subset;

[0071] A tree barrier assessment model is constructed. The tree barrier assessment model assesses the growth of trees in the transmission protection area according to the tree barrier growth characteristics and outputs the tree barrier assessment results.

[0072] Specifically, the image data of the transmission line protection area can be obtained through satellite remote sensing technology, and the growth information of the trees in the area can be extracted from the image using the object recognition algorithm, including the tree type, location, shadow length and other relevant growth characteristics. Combined with historical meteorological data and soil information, a multidimensional data set of tree growth can be constructed, including light, moisture, temperature and soil nutrients. Based on the acquired tree growth information, the features related to tree growth, such as tree height, crown width, tree age, etc., are extracted and organized into a tree barrier growth multidimensional data set. The tree barrier growth multidimensional data set is cleaned and preprocessed to ensure the accuracy and consistency of the data. Using the generated tree barrier growth multidimensional data, samples are randomly selected through repeated sampling technology to generate Multiple tree barrier prediction subsets, each of which should contain different tree barrier growth characteristics. Feature selection is performed on each subset to extract key features closely related to tree growth trends, such as environmental influencing factors and tree species characteristics. A tree barrier assessment model is constructed based on the generated multiple tree barrier prediction subsets. The growth status of trees in the transmission line protection area is evaluated by analyzing the tree barrier growth characteristics. During the model construction process, the model parameters are optimized using the cross-validation method. The tree barrier assessment model is used to conduct tree barrier assessment on trees in the transmission protection area, predict the growth trend of trees and possible impact on transmission lines in the future, and output the tree barrier assessment results, including tree growth height prediction, potential risk analysis of transmission lines and related suggestions.

[0073] The technical solution of the present invention effectively solves the difficulty of identifying tree barriers in the power transmission line area and the problem of how to effectively extract data related to tree barriers from remote sensing image data.

[0074] Further, if Figure 2 As shown in the figure, a tree barrier assessment model is constructed, including:

[0075] A tree obstacle processing layer is constructed for the tree obstacle prediction subset, and the tree obstacle processing layer corresponds to the tree obstacle prediction subset one by one;

[0076] A tree growth prediction sub-model is generated according to the tree barrier processing layer, and the tree growth prediction sub-model makes a height prediction on the tree growth to obtain a height prediction result;

[0077] A tree barrier assessment model was constructed based on the tree growth prediction sub-model.

[0078] As a preferred embodiment of the above embodiment, tree barrier growth characteristic data of each subset is extracted from the tree barrier prediction subset, including information such as tree species, tree age, and environmental variables, and a tree barrier processing layer is constructed for each tree barrier prediction subset to ensure that each tree barrier processing layer corresponds to its corresponding prediction subset one by one; within the tree barrier processing layer, a tree growth prediction submodel is constructed, firstly a training set and a test set are determined, and characteristic variables extracted from the tree barrier prediction subset are used as input data; when training the model, parameters are adjusted according to historical data of tree growth to optimize the prediction ability of the model, and the performance of the model is evaluated through cross-validation to ensure its accuracy in predicting tree heights; the constructed tree growth prediction model is used to predict the height of trees; The sub-model predicts the height of each prediction subset, inputs the corresponding tree barrier growth characteristic data into the prediction model, outputs the prediction results, and obtains the predicted height value of the tree at a specific time point in the future; based on the output results of all tree growth prediction sub-models, a tree barrier assessment model is constructed, and the prediction results of each sub-model are combined to integrate multiple height prediction values ​​using a weighted average or voting mechanism to improve the overall prediction accuracy. During subsequent analysis, the assessment model is adjusted according to different environmental conditions and time nodes to ensure its adaptability and flexibility, and finally a tree barrier assessment model is generated; the constructed tree barrier assessment model is verified, and the accuracy and reliability of the model are evaluated by comparing it with actual observation data.

[0079] Furthermore, a tree growth prediction sub-model is generated based on the tree barrier processing layer, including:

[0080] Based on the multi-dimensional data of tree barrier growth, repeated sampling is adopted to extract fixed array width samples to generate multiple tree barrier prediction subsets, each of which contains different sample combinations.

[0081] Recursively construct the tree barrier processing layer according to the node splitting rule until the preset stop condition is reached, completing the generation of a single tree barrier processing layer;

[0082] When each node is split, the key split point is selected based on the tree barrier growth characteristics, among which the key split point with the greatest influence on the tree growth trend is given priority;

[0083] A tree growth prediction sub-model is constructed corresponding to each tree barrier treatment layer, and each tree growth prediction sub-model has tree growth with different structures and feature combinations.

[0084] As a preferred embodiment of the above, firstly, multidimensional data on tree barrier growth is collected, including characteristics such as tree species, tree age, soil type, and climatic conditions. A repeated sampling technique is used to randomly extract samples of a fixed array width from the multidimensional data set to generate multiple tree barrier prediction subsets, and each prediction subset should contain different sample combinations. In the process of constructing the tree barrier processing layer, recursive splitting is performed according to the set node splitting rules. The node splitting rules may include evaluating the splitting effect based on indicators such as information gain and Gini index, and setting preset stopping conditions, such as reaching the maximum depth, the number of node samples being lower than a threshold, or the information gain being lower than a predetermined value, to decide when to stop splitting, thereby completing the generation of a single tree barrier processing layer. When each node is split, a key splitting point is selected based on the tree barrier growth characteristics, and each characteristic is split. The weight of the influence of the features on the tree growth trend is calculated based on the features. The importance of the features can be evaluated by statistical analysis or machine learning algorithms, and the features that have a significant impact on the tree growth trend are given priority as split points. For each generated tree barrier treatment layer, a corresponding tree growth prediction sub-model is constructed. The model is constructed according to the structure and feature combination of the tree barrier treatment layer. Each tree growth prediction sub-model has a different structure and feature combination to capture different tree growth patterns and trends. Each tree growth prediction sub-model is used to train the training data, and the prediction performance of the model is evaluated through cross-validation. The model parameters and structure are adjusted according to the evaluation results to ensure that each model has good accuracy in predicting tree height. The results of all tree growth prediction sub-models are integrated to generate a comprehensive tree barrier assessment result.

[0085] Further, if Figure 3 As shown in the figure, the tree obstacle processing layer is recursively constructed according to the node splitting rule, including:

[0086] Based on the tree barrier growth features in the tree barrier prediction subset, the split gain of each tree barrier growth feature at the current node is calculated, and the tree barrier growth feature with the highest gain value is selected as the split feature of the current node;

[0087] Determine the split point of the current node, analyze the value range of the split feature, and select the node with the largest split gain as the key split point based on the feature distribution;

[0088] Recursively calculate the feature gain and select the split point for each node to generate branch nodes until the split gain value is lower than the set minimum gain threshold, generating a predicted leaf node;

[0089] By integrating the tree barrier growth trend and environmental characteristic data of the samples in the predicted leaf node and combining them with the remaining generated leaf node results, the corresponding growth prediction value is generated;

[0090] The tree obstacle prediction subset and the prediction leaf nodes are integrated to construct the tree obstacle processing layer.

[0091] As a preferred embodiment of the above embodiment, starting from the tree barrier growth feature in the tree barrier prediction subset, the splitting gain of the feature at the current node is evaluated one by one. The splitting gain is a measure used to evaluate the importance and effectiveness of a feature in the decision tree algorithm. It represents the information gain that can be brought by selecting a certain feature for node splitting. Specifically, the splitting gain measures how much the purity of the data set is improved by splitting the feature. The information gain, gain ratio or Gini index can be used to calculate the splitting gain of each feature; the tree barrier growth feature with the highest gain value is selected as the splitting feature of the current node to ensure that the selected feature can most effectively divide the data set into purer subsets; the selected splitting feature is analyzed to obtain the value range and distribution; according to the distribution of the feature value, the node that can maximize the splitting gain is selected as the key splitting point, and the optimal splitting point can be identified by traversing and calculating within the feature value range; for each new For the generated nodes, calculate their feature gains and select split features, generate new branch nodes based on the split features and split points, and recursively calculate feature gains and select split points for these nodes. When the split gain value is lower than the set minimum gain threshold, stop the recursion and generate a predicted leaf node; integrate the samples in each predicted leaf node, analyze its tree barrier growth trend and environmental characteristic data, and calculate the average or weighted average of the samples in the leaf node to form a growth trend prediction of the node. Integrate the results generated by each leaf node to form a more comprehensive growth prediction value; integrate the tree barrier prediction subset with the corresponding prediction leaf node results to form a complete tree barrier processing layer; evaluate the accuracy and reliability of the generated tree barrier processing layer by verifying historical data or test data, and optimize and adjust the split gain threshold, feature selection strategy or node splitting rule according to the verification results when necessary to improve the prediction performance of the model.

[0092] Further, if Figure 4 As shown, tree growth information within the protection area of ​​the transmission line is obtained, including:

[0093] Obtain remote sensing images of the transmission line protection area through satellites, pre-process the remote sensing images, divide the remote sensing images into grids, scan them grid by grid to identify the tower features, and determine the fuzzy range of the tower based on the identified grid areas;

[0094] Perform fuzzy search within the fuzzy range according to the records to determine the locations of all towers within the fuzzy range;

[0095] Using the object recognition algorithm, the longitude and latitude coordinates of the tower within the fuzzy range and the length of the tower shadow are identified from the image;

[0096] According to the protection area of ​​the transmission line, the latitude and longitude coordinates of the tree barrier and the length of the tree barrier shadow are identified from the remote sensing image to identify the location of the tree barrier and obtain the tree growth information.

[0097] As a preferred embodiment of the above, a remote sensing image of the power transmission line protection area that can clearly display the characteristics of trees and poles and towers in the area is obtained through satellite images, and the obtained remote sensing image is preprocessed by radiation correction, geometric correction and noise removal to improve the image quality and the accuracy of subsequent analysis; the preprocessed remote sensing image is divided into a number of uniform grids, and the pole tower features in the grid are scanned and identified grid by grid, and the outline and position of the pole tower are identified by using image processing technology (such as edge detection, morphological analysis, etc.); according to the ledger data, the identified pole tower position is fuzzy searched to determine the fuzzy range of each pole tower, including spatial query of the ledger information to find the pole towers intersecting with the identified grid area. Tower position; Apply the object recognition algorithm to extract the longitude and latitude coordinates of the tower within the blurred range from the remote sensing image, and calculate the shadow length of the tower; In the same remote sensing image, use the object recognition algorithm to identify the longitude and latitude coordinates of the tree barrier and the shadow length of the tree barrier, and effectively locate the position of the tree barrier by analyzing the characteristics of the trees in the image, and extract the relevant tree growth information by comparing with historical data or existing growth models based on the position and shadow length of the tower and tree barrier; Integrate the acquired growth information of the tower and tree barrier to form a tree growth database, use data analysis tools to perform statistical analysis on the integrated growth information, evaluate the potential impact of tree barriers on transmission lines, and support subsequent decision-making.

[0098] Specifically, identifying the location of tree obstacles includes:

[0099] According to the tower information recorded in the ledger, confirm the corresponding tower height value, and calculate the average solar altitude angle at the time of shooting based on the tower height value and the tower shadow length;

[0100] Calculate the height of the tree barrier based on the average solar altitude angle and the length of the tree barrier shadow;

[0101] According to the location of the tower body and the longitude and latitude coordinates of the tree barrier, determine the center point of the tree barrier within the protection area of ​​the tower and the transmission line;

[0102] According to the coordinates of the center point of the tower and the tower height information as well as the coordinates of the center point of the tree barrier and the tree barrier height information, the vertical distance from the tree barrier to the transmission line is calculated and the location of the tree barrier is identified according to the protection area of ​​the transmission line.

[0103] As a preferred embodiment of the above embodiment, the height value corresponding to the identified tower is extracted from the ledger, and the extracted tower height is verified as necessary; the tower shadow length and tower height value at the time of shooting are collected, and the trigonometric function formula is applied to calculate the solar altitude angle at the time of shooting in combination with the tower height and shadow length:

[0104] ;

[0105] If there are multiple shooting data, calculate the solar altitude angle at all time points and take the average value. Use the calculated solar altitude angle and the shadow length of the tree barrier to calculate the actual height of the tree barrier using the same trigonometric function formula:

[0106] ;

[0107] According to the coordinates of the main position of the tower and the longitude and latitude coordinates of the tree barrier, the center point of the tower and the tree barrier is confirmed. The midpoint formula is usually used to calculate the center point coordinates. If necessary, the longitude and latitude coordinates are converted into the local plane coordinate system for subsequent calculations. The tower height and tree barrier height information are combined with the height information of the transmission line to determine whether the vertical distance is within the protection area. According to the calculated vertical distance and the protection area requirements, it is determined whether the tree barrier poses a threat to the transmission line. If the vertical distance is less than the set threshold, it is identified as affecting the tree barrier and corresponding measures need to be taken.

[0108] To go further, the tree growth height is predicted, including:

[0109] Based on the remote sensing images taken by satellite, the tree barriers within the line protection area are classified and identified, and the type, location and distribution characteristics of each tree barrier are determined;

[0110] Based on the multi-dimensional data of tree barrier growth, the tree allometric growth equation is constructed, and the parameters are optimized according to the characteristics of different tree species;

[0111] The future growth height and growth trend of tree barriers at different times and seasons are predicted based on the tree allometric equation, and the potential impact of tree barriers on transmission lines is evaluated.

[0112] As a preferred embodiment of the above, the remote sensing images acquired by the satellite are geometrically corrected, radiometrically corrected and noise removed, the remote sensing images are classified, the types of tree barriers in the line protection area are identified, the type, location and spatial distribution characteristics of each tree barrier are output, and the classification results are verified through field surveys or existing data; multidimensional data related to the growth of tree barriers are collected, including growth height, age, tree species, climate conditions, soil characteristics, etc., and a tree allometric growth equation is constructed based on the collected data to describe the growth law of trees under different conditions. According to the growth characteristics of different tree species, the model parameters are adjusted, and the prediction ability of the model is evaluated using methods such as cross-validation; the constructed tree allometric growth equation is used in combination with historical growth data to perform time series analysis to predict the growth height of tree barriers at different time points in the future, and the model input is adjusted according to climate change and seasonal factors to evaluate the growth trend of trees in different seasons. The predicted tree barrier growth height is trend analyzed to evaluate the potential impact on the transmission line and determine the possible risk area; the growth height, type and distribution characteristics of the tree barriers are comprehensively analyzed to analyze the potential impact on the transmission line, evaluate the threat to power safety, and propose corresponding management suggestions and maintenance measures based on the impact assessment results.

[0113] Further, if Figure 5 As shown, the tree allometric growth equation is constructed, including:

[0114] Based on the multidimensional data of tree barrier growth, characteristic variables related to tree growth trends are screened as input parameters of the tree allometric growth equation;

[0115] According to the allometric growth characteristics of trees, select the tree allometric growth equation model suitable for different tree species and set the preliminary model parameters;

[0116] Using the original data set, the parameters of the selected tree allometric growth equation model were tuned by maximizing the fit, and the weight of each characteristic variable and model parameters were optimized;

[0117] Verify the optimized tree allometric growth equation model, compare the predicted growth height with the actual historical data, and set the error threshold;

[0118] If the prediction error is within the set error threshold, the optimized allometric equation is generated.

[0119] As a preferred embodiment of the above embodiment, multi-dimensional data of tree barrier growth, including tree species, age, environmental factors, growth height, etc., are collected, and characteristic variables highly correlated with tree growth trends are screened out by statistical analysis methods. The screened characteristic variables are used as input parameters to lay the foundation for the subsequent construction of allometric growth equations; according to the growth characteristics of different tree species, a suitable allometric growth equation model is selected, and preliminary model parameters are set to provide a basis for parameter tuning; using the original data set, the least squares method or other optimization algorithms are used to maximize the model fit, adjust the weight and model parameters of each characteristic variable, and the gradient descent method, genetic algorithm, etc. can be used for parameter optimization to find the optimal parameter combination; the tuned tree allometric growth equation model is applied to historical data to predict the growth height, and compared with the actual measured growth height, and the prediction error threshold is set according to the actual application requirements; if the prediction error is within the set error threshold range, the model is considered valid, and the optimized tree allometric growth equation is generated, and the optimized tree allometric growth equation and its parameters are saved for subsequent applications to ensure dynamic monitoring and prediction of tree growth.

[0120] Furthermore, the tree barrier assessment model assesses the growth of trees in the transmission protection area according to the tree barrier growth characteristics and outputs the tree barrier assessment results, including:

[0121] Based on the tree growth prediction sub-model, a tree barrier assessment model is constructed. The tree barrier growth features in the tree barrier prediction subset are used as input. The feature importance score of each tree barrier growth feature in growth prediction is calculated through multiple trainings.

[0122] According to the feature importance score, all tree barrier growth features are ranked to select the key features that contribute most to tree growth prediction;

[0123] Based on the largest key features screened out, the impact on the growth trend of tree barriers is analyzed to obtain the tree barrier growth threshold;

[0124] Evaluate the tree obstacles, and when the growth trend of the tree obstacles reaches the preset tree obstacle growth threshold, record the tree obstacle information in the hidden danger database;

[0125] Tree barriers that exceed the growth threshold are marked and updated, and tree growth height prediction results and key influencing factor analysis reports are output.

[0126] As a preferred embodiment of the above embodiment, features related to tree growth are extracted from the tree barrier prediction subset to form the input of the tree growth prediction sub-model, and the model is trained multiple times using a machine learning algorithm; after the training is completed, the importance score of each tree barrier growth feature in the model is calculated. A model-based feature importance assessment method can be adopted, such as feature importance based on tree model. The calculated feature importance scores are sorted to form feature importance scores. According to the feature importance scores, all tree barrier growth features are sorted to screen out the key features that contribute most to tree growth prediction. These features may include tree species, soil type, climatic conditions, etc. Based on the screened key features, their impact on the growth trend of tree barriers is deeply analyzed to determine the threshold of tree barrier growth. According to the analysis results, a tree barrier growth threshold is set as the basis for evaluating the growth trend of tree barriers. The tree barriers in the protection area of ​​the transmission line are monitored in real time, their growth trends are evaluated, and compared with the set growth threshold. When the growth trend of the tree barrier reaches or exceeds the preset growth threshold, the relevant tree barrier information is recorded in the hidden danger library. The data can come from remote sensing image analysis, on-site monitoring, meteorological data, historical records, and expert evaluations. The information in the hidden danger library can be classified according to multiple standards, such as tree species, tree barrier location, potential impact, treatment status, etc., for rapid retrieval and classification. Analysis, each record should include the unique identification of the tree barrier, geographic coordinates, relevant tower information, predicted growth height, recording time, treatment suggestions and other fields. The hidden danger database should set up regular updates and real-time monitoring mechanisms to ensure the accuracy and timeliness of information. For new monitoring results, especially tree barriers with high risk levels, their status needs to be updated immediately. Using the data in the hidden danger database, regular risk assessments should be conducted. The hidden danger database can be combined with the monitoring system to establish an early warning mechanism. When the growth trend of the tree barrier reaches the preset safety critical value, an alarm will be issued and corresponding emergency treatment measures will be recommended. The hidden danger database provides a basis for historical data analysis, identifies tree barrier growth patterns and environmental changes, and thus provides a basis for future tree growth prediction and management decisions; based on the evaluation results, the predicted results of tree growth height are output and provided to relevant managers for decision-making reference, and a key influencing factor analysis report is generated, detailing the influencing factors and their weights on tree barrier growth, to help formulate more accurate management strategies; based on the results of tree barrier growth monitoring and evaluation, the tree barrier evaluation model is continuously optimized, and input features and model parameters are adjusted.

[0127] Embodiment 2

[0128] Based on the same inventive concept as the tree barrier assessment method in the above embodiment, the present invention further provides a tree barrier assessment system, the system comprising:

[0129] An information acquisition module obtains tree growth information within the protection area of ​​the transmission line and obtains multi-dimensional data of tree barrier growth based on the tree growth information;

[0130] A feature prediction module generates multiple tree barrier prediction subsets based on the tree barrier growth multidimensional data, and extracts tree barrier growth features in each tree barrier prediction subset;

[0131] The result evaluation module constructs a tree barrier evaluation model. The tree barrier evaluation model evaluates the growth of trees in the transmission protection area according to the growth characteristics of the tree barriers and outputs the tree barrier evaluation results.

[0132] The above adjustment system in the present invention can effectively implement a tree barrier assessment method, and the technical effects that can be achieved are as described in the above embodiments, which will not be repeated here.

[0133] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application as defined therein, and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A tree obstacle assessment method, characterized in that: The method comprises: Acquire tree growth information within the protection area of ​​the transmission line, and obtain multi-dimensional data of tree barrier growth based on the tree growth information; Generate multiple tree barrier prediction subsets according to the tree barrier growth multidimensional data, and extract tree barrier growth features in each of the tree barrier prediction subsets; Constructing a tree barrier assessment model, wherein the tree barrier assessment model performs a tree barrier assessment on the growth of trees in the power transmission protection area according to the tree barrier growth characteristics, and outputs a tree barrier assessment result; Construct a tree barrier assessment model, including: Constructing a tree obstacle processing layer for the tree obstacle prediction subset, wherein the tree obstacle processing layer corresponds to the tree obstacle prediction subset one by one; Generate a tree growth prediction sub-model according to the tree obstacle processing layer, the tree growth prediction sub-model performs a height prediction on the tree growth to obtain a height prediction result; constructing a tree barrier assessment model based on the tree growth prediction sub-model; A tree growth prediction sub-model is generated according to the tree barrier processing layer, including: Based on the multi-dimensional data of tree barrier growth, a fixed array width sample is extracted by repeated sampling to generate a plurality of tree barrier prediction subsets, each of which contains different sample combinations; Recursively construct the tree barrier processing layer according to the node splitting rule until a preset stop condition is reached, thereby completing the generation of a single tree barrier processing layer; When each node is split, a key split point is selected based on the tree barrier growth characteristics, wherein the key split point with a large weight on the tree growth trend is selected; Constructing the tree growth prediction sub-model corresponding to each of the tree obstacle processing layers, and each of the tree growth prediction sub-models has tree growth with different structures and feature combinations; The tree obstacle processing layer is recursively constructed according to the node splitting rules, including: Based on the tree barrier growth features in the tree barrier prediction subset, calculating the splitting gain of each of the tree barrier growth features at the current node, and selecting the tree barrier growth feature with the highest gain value as the splitting feature of the current node; Determine the split point of the current node, and select the node with the largest split gain as the key split point by analyzing the value range of the split feature and according to the feature distribution; Recursively calculate the feature gain and select the split point for each node to generate branch nodes until the split gain value is lower than the set minimum gain threshold, generating a predicted leaf node; Generate a corresponding growth prediction value by integrating the tree barrier growth trend and environmental characteristic data of the samples in the predicted leaf node and combining the remaining generated leaf node results; The tree obstacle prediction subset and the prediction leaf node are integrated to construct a tree obstacle processing layer.

2. The tree obstacle assessment method according to claim 1, characterized in that: Obtain information on tree growth within the transmission line protection area, including: Acquire remote sensing images of the transmission line protection area through satellites, pre-process the remote sensing images, divide the remote sensing images into grids, scan them grid by grid to identify tower features, and determine the fuzzy range of the towers based on the identified grid areas; Perform fuzzy search within the fuzzy range according to the ledger to determine the positions of all towers within the fuzzy range; Using a ground object recognition algorithm, the longitude and latitude coordinates of the tower within the fuzzy range and the length of the tower shadow are identified from the image; The longitude and latitude coordinates of the tree barrier and the length of the tree barrier shadow are identified from the remote sensing image according to the protection area of ​​the transmission line to identify the location of the tree barrier and obtain the tree growth information.

3. The tree obstacle assessment method according to claim 2, characterized in that: Identify tree obstacle locations, including: According to the tower information recorded in the ledger, the corresponding tower height value is confirmed, and according to the tower height value and the tower shadow length, the average value of the solar altitude angle at the time of shooting is calculated; Calculating the height of the tree barrier according to the average value of the solar altitude angle and the length of the tree barrier shadow; According to the position of the tower body and the longitude and latitude coordinates of the tree barrier, confirm the center point of the tower and the tree barrier in the protection area of ​​the transmission line; According to the coordinates of the center point of the tower and the tower height information and the coordinates of the center point of the tree barrier and the tree barrier height information, the vertical distance from the tree barrier to the transmission line is calculated and the location of the tree barrier is identified according to the protection area of ​​the transmission line.

4. The tree obstacle assessment method according to claim 2, characterized in that: Predict tree growth height, including: Classify and identify tree barriers within the line protection area based on the remote sensing images taken by satellite, and determine the type, location and distribution characteristics of each tree barrier; According to the multi-dimensional data of tree barrier growth, a tree allometric growth equation is constructed, and parameters are optimized according to the characteristics of different tree species; The future growth height and growth trend of tree barriers at different times and seasons are predicted based on the tree allometric growth equation, and the potential impact of tree barriers on transmission lines is evaluated.

5. The tree obstacle assessment method according to claim 4, characterized in that: Construct the tree allometric growth equation, including: Based on the multi-dimensional data of tree barrier growth, characteristic variables related to tree growth trends are selected as input parameters of the tree allometric growth equation; According to the allometric growth characteristics of trees, the tree allometric growth equation model suitable for different tree species is selected, and preliminary model parameters are set; Using the original data set, the selected tree allometric growth equation model is tuned by maximizing the degree of fit, and the weight of each characteristic variable and the model parameters are optimized; Verifying the optimized tree allometric growth equation model, comparing the predicted growth height with actual historical data, and setting an error threshold; If the prediction error is within the set error threshold range, an optimized allometric growth equation is generated.

6. The tree obstacle assessment method according to claim 1, characterized in that: The tree barrier assessment model performs a tree barrier assessment on the growth of trees in the power transmission protection area according to the tree barrier growth characteristics, and outputs a tree barrier assessment result, including: Based on the tree growth prediction sub-model, a tree barrier assessment model is constructed, the tree barrier growth features in the tree barrier prediction subset are used as input, and the feature importance score of each tree barrier growth feature in growth prediction is calculated through multiple trainings; According to the feature importance scores, all the tree barrier growth features are ranked to select the key features that contribute most to tree growth prediction; Based on the screened maximum key features, analyzing the impact on the tree barrier growth trend, and obtaining the tree barrier growth threshold; Evaluate the tree barrier, and when the growth trend of the tree barrier reaches the preset tree barrier growth threshold, record the tree barrier information in the hidden danger database; Tree barriers that exceed the growth threshold are marked and updated, and tree growth height prediction results and key influencing factor analysis reports are output.

7. A tree obstacle assessment system, characterized in that: Using the tree obstacle assessment method as claimed in claim 1, the system comprises: An information acquisition module is used to acquire tree growth information within the protection area of ​​the power transmission line, and obtain multi-dimensional data of tree barrier growth based on the tree growth information; A feature prediction module generates a plurality of tree barrier prediction subsets according to the tree barrier growth multidimensional data, and extracts tree barrier growth features from each of the tree barrier prediction subsets; The result evaluation module constructs a tree barrier evaluation model, wherein the tree barrier evaluation model performs a tree barrier evaluation on the tree growth in the power transmission protection area according to the tree barrier growth characteristics, and outputs a tree barrier evaluation result.

Images