Transmission line tree obstacle hazard detection method, system and medium considering wind deviation effect

Abstract

The present invention relates to the technical field of power systems, and in particular to a method, system, and medium for detecting tree barrier hazards on power transmission lines that consider the effects of wind deviation. The method comprises: predicting tree growth based on an original multi-source data set to obtain a tree growth prediction result; performing a three-dimensional wind deviation effect analysis based on the tree growth prediction result and the original multi-source data set to obtain a tree wind deviation quantification result; calculating the risk time of trees in the vertical, horizontal, and oblique directions based on the tree wind deviation quantification result; evaluating the comprehensive risk index between trees and power transmission lines based on the risk time to obtain the degree of tree barrier hazard; generating a tree pruning strategy based on the degree of tree barrier hazard with the optimization goal of minimizing the total amount of pruning and maximizing the safety margin, and pruning transmission line risk trees according to the tree pruning strategy. The present invention effectively reduces the risk of transmission line failures caused by tree barriers by fusing multi-source data to perform tree growth prediction and wind deviation effect quantification.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a method, system and medium for detecting tree obstacle hazards in power transmission lines taking into account the influence of wind deviation. Background Art

[0002] Transmission lines, as an important part of the power system, are widely distributed in complex terrain environments such as mountains, hills, forests and lakes. The growth of trees along these areas poses a potential security threat to the safe and stable operation of transmission lines. When trees grow too high, they can easily invade the safe distance range of transmission lines, which in turn causes line discharge, short circuit tripping and other faults in the power system, and may even cause forest fires, causing serious damage.

[0003] Currently, power operation and maintenance departments mainly use technical means such as manual inspections, drone inspections, and remote sensing monitoring to manage trees on transmission lines, and combine experience to formulate felling plans. However, these traditional methods have exposed many problems in actual applications and are difficult to meet the efficiency, accuracy, and economy requirements of modern smart grids for transmission line safety management. Traditional manual inspections rely on operation and maintenance personnel to walk or drive along the lines to inspect the growth of trees. However, since transmission lines often pass through complex terrain and transportation is extremely inconvenient in some areas, the inspection work is not only arduous but also has a long cycle. At the same time, due to terrain and weather factors, some areas are difficult to reach, and there are blind spots for inspections, which makes it difficult to detect safety hazards in a timely manner. In addition, manual inspections mainly rely on experience and judgment, and estimate the height of trees and their distance from the conductors by visual inspection. This method is highly subjective and cannot obtain accurate data. It is difficult to effectively predict the future growth trend of trees and it is difficult to ensure the rationality of the inspection frequency.

[0004] With the rapid development of artificial intelligence technology, intelligent inspection methods such as drone inspections and remote sensing monitoring have gradually been applied to tree management along transmission lines. These technologies can obtain tree growth information along transmission lines, which has improved inspection efficiency to a certain extent. However, these technologies still have obvious limitations. Drone inspections and remote sensing monitoring often rely on regular aerial photography or satellite imaging, with long data update cycles and inability to achieve real-time monitoring, which can easily lead to data lags and affect the accuracy of operation and maintenance decisions. In addition, drone inspections are significantly restricted by weather conditions and are difficult to operate normally in severe weather such as strong winds, rain, snow, fog and haze. The imaging quality of remote sensing monitoring is also easily affected by factors such as cloud cover and lighting conditions, resulting in unstable data or large errors. More importantly, current drone inspections and remote sensing monitoring mainly focus on obtaining the current tree height and spatial distribution. There is a lack of tree growth prediction models, and it is impossible to accurately predict whether trees will invade the safe distance in the future. This makes felling plans have a lag and it is difficult to effectively ensure the safety of transmission lines. Summary of the Invention

[0005] To solve the above technical problems, the present invention provides a method, system and medium for detecting tree obstacle hazards on power transmission lines taking into account the influence of wind deviation.

[0006] In a first aspect, the present invention provides a method for detecting tree obstacle hazards on power transmission lines taking into account the influence of wind deviation, the method comprising the following steps:

[0007] Collect tree growth characteristic data, meteorological environment monitoring data, and transmission line conductor parameters in the target area within the transmission line corridor to form an original multi-source data set;

[0008] Performing tree growth dynamics prediction based on the original multi-source data set to obtain tree growth prediction results for each tree in the target area;

[0009] Performing a three-dimensional wind deflection effect analysis based on the tree growth prediction result and the original multi-source data set to obtain a quantitative result of tree wind deflection;

[0010] Calculate the risk time of the tree in the vertical, horizontal and oblique directions respectively according to the tree growth prediction result and the tree wind deviation quantification result;

[0011] Evaluate the comprehensive risk index between trees and transmission lines based on the risk time to obtain the degree of tree barrier hazard;

[0012] With the optimization goals of minimizing the total amount of pruning and maximizing the safety margin, a non-dominated sorting genetic algorithm is used to generate a tree pruning strategy based on the hazard level of the tree obstacle, and the risky trees on the transmission line are pruned according to the tree pruning strategy.

[0013] In a further embodiment, the step of performing tree growth dynamics prediction based on the original multi-source dataset to obtain tree growth prediction results for each tree in the target area includes:

[0014] Preprocessing the original multi-source data set to extract tree growth-related features, conductor dynamic features, and geographic features, and generate a standardized feature matrix;

[0015] Obtain the measured data of tree heights and use the average value of the measured data of tree heights as the benchmark prediction value;

[0016] Calculating the prediction error and error change rate between the reference prediction value and the actual tree height data;

[0017] Traversing each feature in the standardized feature matrix, calculating the information gain of each split point according to the prediction error and the error change rate, selecting the split point with the largest information gain as the optimal split point of the current node, and recursively constructing an extreme gradient boosting tree;

[0018] The normalized feature matrix is ​​subjected to tree growth dynamic prediction by using the extreme gradient boosting tree to obtain tree growth prediction results for each tree in the target area.

[0019] In a further embodiment, the step of calculating the prediction error and the error change rate between the baseline prediction value and the actual tree height data includes:

[0020] The prediction error and the error change rate between the reference prediction value and the actual tree height data are calculated based on the mean square error loss function.

[0021] In a further embodiment, the tree windage quantification results include horizontal windage values, vertical windage values, and maximum horizontal swing values ​​of the wire under different wind speeds and wind directions; and the step of performing a three-dimensional windage effect analysis based on the tree growth prediction results and the original multi-source dataset to obtain the tree windage quantification results includes:

[0022] Extracting tree mechanical parameters, crown morphological characteristics, and conductor mechanical parameters from the original multi-source data set; the tree mechanical parameters include tree elastic modulus and tree cross-sectional inertia moment; the crown morphological characteristics include tree crown wind-exposed area and tree crown center of gravity height; the conductor mechanical parameters include transmission line conductor tension;

[0023] Calculating the wind moment exerted on the tree at different wind speeds based on the tree's windward area and instantaneous wind speed, and calculating the tree's restoring moment based on the tree's elastic modulus, the tree's cross-sectional moment of inertia, and the tree's inclination angle;

[0024] The tree tilt angle corresponding to the moment when the wind moment and the tree restoring moment are equal is used as the maximum safe wind deflection angle of the tree;

[0025] Calculate the horizontal deviation of the tree toward the wire direction according to the maximum safe wind deviation angle of the tree and the tree growth prediction result to obtain the horizontal wind deviation;

[0026] Calculating the center of gravity offset of the tree under wind action based on the wind-exposed area of ​​the tree canopy and the height of the center of gravity of the tree canopy to obtain the vertical wind deflection;

[0027] The maximum allowable horizontal swing amplitude of the transmission line conductor is calculated based on the transmission line conductor tension and real-time wind speed monitoring data, and the maximum horizontal swing amount of the conductor is obtained.

[0028] In a further embodiment, the risk time includes vertical risk time, horizontal risk time, and oblique risk time; and the step of calculating the risk time of the tree in the vertical, horizontal, and oblique directions based on the tree growth prediction result and the tree wind deviation quantification result includes:

[0029] Obtaining the hanging height data of the transmission line conductor, and obtaining the predicted tree height according to the tree growth prediction result;

[0030] Calculating a vertical safety height threshold based on the transmission line conductor suspension height data and a preset conductor safety height margin, and calculating an effective tree height based on the tree predicted height and the vertical wind deflection;

[0031] Comparing the effective height of the tree with the vertical safety height threshold, finding all vertical threat time points at which the effective height of the tree is not less than the vertical safety height threshold, and selecting the minimum time value from the vertical threat time points as the vertical risk time;

[0032] determining a horizontal safety distance threshold according to a difference between the horizontal wind deviation and the maximum horizontal swing of the conductor;

[0033] When the horizontal safety distance threshold is not greater than the preset conductor safety horizontal margin, the shortest time for the tree to horizontally deviate to the danger zone under the action of wind deviation is calculated based on the maximum safe wind deviation angle of the tree and the maximum horizontal swing of the conductor to obtain the horizontal risk time;

[0034] Calculating the equivalent height of the tree's oblique projection according to the predicted tree height and the cosine value of the oblique growth angle, and calculating the oblique safety distance according to the horizontal offset of the tree's oblique growth and the equivalent height of the tree's oblique projection;

[0035] The earliest time point at which the oblique safety distance is not greater than the preset wire safety oblique margin for the first time is found as the oblique risk time when the tree enters the danger zone in an oblique growth manner.

[0036] In a further embodiment, the step of evaluating the comprehensive risk index between trees and power transmission lines based on the risk time to obtain the degree of tree barrier hazard comprises:

[0037] Comparing the vertical risk time, the horizontal risk time, and the oblique risk time with a preset critical response period, respectively, and classifying trees with risk times in at least two directions less than the critical response period as emergency risk trees;

[0038] The comprehensive risk index of emergency risk trees is calculated by the weighted sum of the inverse of risk time to obtain the degree of hazard of tree obstacles.

[0039] In a further embodiment, the step of generating a tree pruning strategy using a non-dominated sorting genetic algorithm based on the degree of hazard of the tree obstacle with minimizing the total amount of pruning and maximizing the safety margin as the optimization goal includes:

[0040] Determine the planned pruning trees based on the emergency risk trees in the target area, and take the cumulative sum of the square of the DBH multiplied by the tree height of all planned pruning trees as the total pruning amount;

[0041] Calculate the safety margin between the transmission line and surrounding trees based on the minimum safe distance increment between the pruned trees and the transmission line conductors and the degree of tree barrier hazards;

[0042] Randomly generate the initial population; each individual in the initial population represents a tree pruning strategy;

[0043] Calculating the fitness value of each individual in the initial population based on the safety margin and the total amount of pruning;

[0044] Perform non-dominated sorting on the individuals in the initial population and divide them into multiple non-dominated levels according to their fitness values. Use tournament selection to select individuals from the non-dominated levels to form the next generation population.

[0045] Perform crossover and mutation operations on individuals in the next generation population until a preset maximum number of iterations is reached, stop the iterative process, output a non-dominated optimal solution set, and select the optimal tree pruning strategy from the non-dominated optimal solution set.

[0046] In a further embodiment, the tree pruning strategy includes a tree pruning priority and a tree pruning degree.

[0047] In a second aspect, the present invention provides a system for detecting tree obstacles on power transmission lines that takes into account the influence of wind deviation, the system comprising:

[0048] The data acquisition module is used to collect tree growth characteristic data, meteorological environment monitoring data and transmission line conductor parameters in the target area within the transmission line corridor to form an original multi-source data set;

[0049] a growth prediction module, configured to perform tree growth dynamics prediction based on the original multi-source dataset to obtain tree growth prediction results for each tree in the target area;

[0050] a windage quantification module, configured to perform a three-dimensional windage effect analysis based on the tree growth prediction result and the original multi-source data set to obtain a tree windage quantification result;

[0051] A risk analysis module, configured to calculate the risk time of the tree in the vertical, horizontal and oblique directions respectively according to the tree growth prediction result and the tree wind deviation quantification result;

[0052] a hidden danger detection module, configured to evaluate the comprehensive risk index between trees and transmission lines based on the risk time, and obtain the degree of hazard of tree barrier hidden dangers;

[0053] The strategy generation module is used to generate a tree pruning strategy based on the degree of hazard of the tree obstacle hazards using a non-dominated sorting genetic algorithm with the optimization goal of minimizing the total amount of pruning and maximizing the safety margin, and prune the risky trees on the transmission line according to the tree pruning strategy.

[0054] In a third aspect, the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0055] The present invention provides a method, system and medium for detecting tree barrier hazards on power transmission lines that take into account the influence of wind deviation. The method collects tree growth characteristic data, meteorological environment monitoring data and transmission line conductor parameters in a target area within a transmission line corridor to form an original multi-source data set; performs tree growth dynamic prediction based on the original multi-source data set to obtain tree growth prediction results for each tree in the target area; performs three-dimensional wind deviation effect analysis based on the tree growth prediction results and the original multi-source data set to obtain tree wind deviation quantification results; calculates the risk time of trees in the vertical, horizontal and oblique directions based on the tree growth prediction results and the tree wind deviation quantification results; evaluates the comprehensive risk index between trees and transmission lines based on the risk time to obtain the degree of hazard of tree barrier hazards; uses minimizing the total amount of pruning and maximizing the safety margin as optimization goals, generates a tree pruning strategy based on the degree of hazard of tree barrier hazards using a non-dominated sorting genetic algorithm, and prunes risky trees on the transmission line according to the tree pruning strategy. Compared with existing technologies, this method predicts tree growth and quantifies wind deviation effects by integrating multi-source data, accurately assesses the degree of tree barrier hazards and generates optimal tree pruning strategies, effectively reducing the risk of transmission line failures caused by tree barriers and ensuring the safe and stable operation of transmission lines. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a flow chart of a method for detecting tree obstacles in power transmission lines taking into account the influence of wind deviation, provided by an embodiment of the present invention;

[0057] Figure 2 This is an example diagram of the height distribution of bamboo at different time points provided by an embodiment of the present invention;

[0058] Figure 3This is a block diagram of a transmission line tree obstacle hazard detection system that takes into account the influence of wind deviation, provided by an embodiment of the present invention.

[0059] Explanation of the accompanying reference numerals: 101, data acquisition module; 102, growth prediction module; 103, wind deviation quantification module; 104, risk analysis module; 105, hidden danger detection module; 106, strategy generation module. DETAILED DESCRIPTION

[0060] The following describes the embodiments of the present invention in detail with reference to the accompanying drawings. The embodiments are provided for illustrative purposes only and are not to be construed as limiting the present invention. The accompanying drawings are provided for reference and illustration only and do not constitute a limitation on the scope of protection of the present invention. Many changes may be made to the present invention without departing from the spirit and scope of the present invention.

[0061] refer to Figure 1 The embodiment of the present invention provides a method for detecting tree obstacle hazards in power transmission lines taking into account the influence of wind deviation. Figure 1 As shown, the method includes the following steps:

[0062] S1. Collect tree growth characteristic data, meteorological environment monitoring data, and transmission line conductor parameters in the target area within the transmission line corridor to form an original multi-source dataset.

[0063] Specifically, this embodiment can collect tree growth characteristic data, meteorological environment monitoring data, and transmission line conductor parameters in the target area within the transmission line corridor through drone laser point cloud scanning, meteorological monitoring stations, or satellite remote sensing images to construct an original multi-source data set. The tree growth characteristic data may include tree species classification identification, breast diameter time series data, and annual ring growth rate. The meteorological environment monitoring data may include wind speed time series data, precipitation distribution, and temperature change curves. The transmission line conductor parameters may include the suspension height, horizontal spacing, and tension coefficient of the transmission line conductor. It should be noted that since the trees in the transmission corridor are unevenly distributed and the initial heights of the surrounding trees are not uniform, to solve this problem, this embodiment can assume that before making a prediction, the trees around the transmission line have been processed accordingly, wherein the trees vertically below the line have been pruned to a safe height, and the trees horizontal to the line have also been processed to a safe distance.

[0064] S2. Perform tree growth dynamics prediction based on the original multi-source dataset to obtain tree growth prediction results for each tree in the target area.

[0065] In some embodiments, the step of performing tree growth dynamics prediction based on the original multi-source dataset to obtain tree growth prediction results for each tree in the target area includes:

[0066] Preprocessing the original multi-source data set to extract tree growth-related features, conductor dynamic features, and geographic features, and generate a standardized feature matrix;

[0067] Obtain the measured data of tree heights and use the average value of the measured data of tree heights as the benchmark prediction value;

[0068] Calculating the prediction error and error change rate between the reference prediction value and the actual tree height data;

[0069] Traversing each feature in the standardized feature matrix, calculating the information gain of each split point according to the prediction error and the error change rate, selecting the split point with the largest information gain as the optimal split point of the current node, and recursively constructing an extreme gradient boosting tree;

[0070] The normalized feature matrix is ​​subjected to tree growth dynamic prediction by using the extreme gradient boosting tree to obtain tree growth prediction results for each tree in the target area.

[0071] Specifically, this embodiment performs spatiotemporal alignment, missing value interpolation, and normalization on the original multi-source dataset, extracts tree growth-related features, conductor dynamic features, and geographic features, and generates a standardized feature matrix. In this embodiment, the row dimension of the standardized feature matrix corresponds to individual tree samples, and the column dimension includes three types of features: tree growth-related features, conductor dynamic features, and geographic features. Tree growth-related features may include data such as tree species growth coefficients and environmental response factors, conductor dynamic features may include data such as swing response coefficients, and geographic features may include data such as slope influence factors and soil moisture indexes. Next, this embodiment uses an XGBoost model based on a gradient boosting decision tree to construct a tree growth prediction model, and preferentially uses a mean square error loss function as the objective function of the regression task. This embodiment initializes the training set target value to the mean of the measured tree height data as the baseline prediction value, and calculates the first-order gradient (residual direction) and second-order gradient (error change rate) of each sample based on the mean square error loss function in each iteration.

[0072] For the node splitting process of each decision tree, this embodiment traverses each feature and splitting threshold in the standardized feature matrix, and obtains the information gain of each splitting point based on the ratio of the sum of the squares of the first-order gradients of the left child node and the right child node to the sum of the second-order gradients, minus the corresponding ratio of the parent node, and selects the splitting point with the largest information gain as the optimal splitting point of the current node. When the number of node samples is lower than the preset sample number threshold or the information gain is less than the minimum split improvement, the splitting is stopped and a leaf node is generated. For each leaf node, this embodiment calculates the node weight value based on the ratio of the sum of the first-order gradients of the leaf node to the sum of the second-order gradients. The node weight value represents the contribution of the node to the final prediction value. Then, this embodiment adds the prediction values ​​of all current decision trees to the baseline prediction value, updates the overall model output, and recalculates the first-order gradient and second-order gradient based on the updated baseline prediction value as the training target of the next decision tree. The above steps are repeated until the preset number of decision trees is reached to construct an extreme gradient boosting tree. Finally, this embodiment inputs the standardized feature matrix into the trained XGBoost model, obtains the predicted height value of each tree sample by traversing all decision trees, and obtains the tree growth prediction result of each tree in the target area.

[0073] S3. Perform a three-dimensional windage effect analysis based on the tree growth prediction results and the original multi-source data set to obtain a quantitative result of tree windage.

[0074] In some embodiments, the tree windage quantification results include horizontal windage values, vertical windage values, and maximum horizontal sway values ​​of the wire under different wind speeds and wind directions; and the step of performing a three-dimensional windage effect analysis based on the tree growth prediction results and the original multi-source dataset to obtain the tree windage quantification results includes:

[0075] Extracting tree mechanical parameters, crown morphological characteristics, and conductor mechanical parameters from the original multi-source data set; the tree mechanical parameters include tree elastic modulus and tree cross-sectional inertia moment; the crown morphological characteristics include tree crown wind-exposed area and tree crown center of gravity height; the conductor mechanical parameters include transmission line conductor tension;

[0076] Calculating the wind moment exerted on the tree at different wind speeds based on the tree's windward area and instantaneous wind speed, and calculating the tree's restoring moment based on the tree's elastic modulus, the tree's cross-sectional moment of inertia, and the tree's inclination angle;

[0077] The tree tilt angle corresponding to the moment when the wind moment and the tree restoring moment are equal is used as the maximum safe wind deflection angle of the tree;

[0078] Calculate the horizontal deviation of the tree toward the wire direction according to the maximum safe wind deviation angle of the tree and the tree growth prediction result to obtain the horizontal wind deviation;

[0079] Calculating the center of gravity offset of the tree under wind action based on the wind-exposed area of ​​the tree canopy and the height of the center of gravity of the tree canopy to obtain the vertical wind deflection;

[0080] The maximum allowable horizontal swing amplitude of the transmission line conductor is calculated based on the transmission line conductor tension and real-time wind speed monitoring data, and the maximum horizontal swing amount of the conductor is obtained.

[0081] Specifically, this embodiment extracts tree mechanical parameters, crown morphological characteristics, and conductor mechanical parameters from the original multi-source data set. The tree mechanical parameters may include tree elastic modulus and tree cross-sectional inertia moment, etc. The tree elastic modulus reflects the ability of trees to resist deformation when subjected to force. The tree cross-sectional inertia moment is used to describe the geometric characteristics of the tree cross-section and affects the bending stiffness of the tree; the crown morphological characteristics may include the wind-exposed area of ​​the tree crown and the height of the center of gravity of the tree crown, etc., wherein the wind-exposed area of ​​the tree crown reflects the projected area of ​​the tree crown in the wind direction, and the height of the center of gravity of the tree crown represents the vertical height of the center of gravity of the tree crown; the conductor mechanical parameters may include transmission line conductor tension, etc. The transmission line conductor tension represents the tension of the conductor in the operating state. Based on the fluid mechanics flow theory, this embodiment calculates the wind force on the tree according to the product relationship between the windward area and the square of the instantaneous wind speed, and obtains the wind torque on the tree under different wind speeds. The calculation formula of the wind torque is:

[0082]

[0083] Where, is the wind torque; is the air resistance coefficient; is the air density; A is the windward area of ​​the tree; is the instantaneous wind speed; L is the height of the tree's center of gravity.

[0084] In this embodiment, the bending resistance of trees is calculated based on the material mechanics bending deformation formula and the product relationship between the elastic modulus and the cross-sectional inertia moment, and the tree restoring moment under wind force is obtained. The calculation formula of the tree restoring moment is:

[0085]

[0086] Where, is the tree restoring moment; E is the tree elastic modulus; I is the tree section inertia moment; The angle of inclination of the tree.

[0087] This embodiment uses an iterative method to find the tree tilt angle when the wind torque is equal to the tree restoring torque, and defines the tree tilt angle corresponding to the wind torque and the tree restoring torque being equal as the tree's maximum safe wind deflection angle. The maximum safe wind deflection angle of the tree indicates that under a specific wind speed, when the tree tilts to this angle, its own elastic restoring torque can just balance the wind torque. When the angle exceeds the maximum safe wind deflection angle of the tree, it is determined that the structure is unstable. Then, based on the maximum safe wind deflection angle of the tree and the tree growth prediction results, the horizontal deviation of the tree affected by the wind deflection during its growth process is calculated, that is, the horizontal wind deflection of the tree toward the wire direction. The calculation formula for the horizontal wind deflection is:

[0088]

[0089] Where, is the horizontal wind deviation; Predicting tree heights for tree growth prediction results; The tree tilt angle when the wind moment is equal to the tree restoring moment, that is, the maximum safe wind deflection angle of the tree; is the tangent function.

[0090] Next, this embodiment calculates the canopy deformation under wind load based on the distribution of the canopy's wind-exposed area and the center of gravity height data. Specifically, this embodiment divides the canopy into units along the tree's center of gravity height direction, and calculates the canopy vertex deformation of each canopy unit based on the product of the canopy unit's wind-exposed area and the square of the wind speed. The canopy vertex deformation variable is added to the predicted tree height to obtain the vertical wind deviation, which reflects the comprehensive displacement of the tree in the height direction under the action of wind.

[0091] Regarding the maximum horizontal swing of the conductor, this embodiment uses a conductor swing calculation formula to calculate the maximum swing amplitude of the transmission line conductor under different wind speeds based on the transmission line conductor tension and real-time wind speed monitoring data. By analyzing the relationship between the maximum conductor swing amplitude data and wind speed, a threshold for the maximum horizontal swing amplitude allowed for the conductor under various wind speed conditions is determined. This threshold is the maximum horizontal swing amplitude of the conductor. For example, within the normal operating wind speed range, the maximum horizontal swing amplitude of the conductor is set by calculation and verification to not exceed a certain proportion of the conductor span to ensure that the conductor swing does not collide with trees or other obstacles, thereby ensuring the safe operation of the transmission line. The conductor swing calculation formula is specifically as follows:

[0092]

[0093] Where, is the maximum swing amplitude of the transmission line conductor under different wind speeds; is the wind force acting on the conductor; is the conductor spacing.

[0094] S4. Calculate the risk time of the tree in the vertical, horizontal and oblique directions respectively according to the tree growth prediction result and the tree wind deflection quantification result.

[0095] In some embodiments, the risk time includes vertical risk time, horizontal risk time, and oblique risk time; and the step of calculating the risk time of the tree in the vertical, horizontal, and oblique directions based on the tree growth prediction result and the tree wind deviation quantification result includes:

[0096] Obtaining the hanging height data of the transmission line conductor, and obtaining the predicted tree height according to the tree growth prediction result;

[0097] Calculating a vertical safety height threshold based on the transmission line conductor suspension height data and a preset conductor safety height margin, and calculating an effective tree height based on the tree predicted height and the vertical wind deflection;

[0098] Comparing the effective height of the tree with the vertical safety height threshold, finding all vertical threat time points at which the effective height of the tree is not less than the vertical safety height threshold, and selecting the minimum time value from the vertical threat time points as the vertical risk time;

[0099] determining a horizontal safety distance threshold according to a difference between the horizontal wind deviation and the maximum horizontal swing of the conductor;

[0100] When the horizontal safety distance threshold is not greater than the preset conductor safety horizontal margin, the shortest time for the tree to horizontally deviate to the danger zone under the action of wind deviation is calculated based on the maximum safe wind deviation angle of the tree and the maximum horizontal swing of the conductor to obtain the horizontal risk time;

[0101] Calculating the equivalent height of the tree's oblique projection according to the predicted tree height and the cosine value of the oblique growth angle, and calculating the oblique safety distance according to the horizontal offset of the tree's oblique growth and the equivalent height of the tree's oblique projection;

[0102] The earliest time point at which the oblique safety distance is not greater than the preset wire safety oblique margin for the first time is found as the oblique risk time when the tree enters the danger zone in an oblique growth manner.

[0103] Specifically, this embodiment uses the tree predicted height output by the XGBoost model Added to the vertical wind deflection, the equivalent effective height of the tree considering wind-induced deformation is obtained. , and set the vertical safety height threshold The suspension height data of the transmission line conductor and preset conductor safety height margin For example, the default conductor height margin Set to 1m, this embodiment compares the effective height of the tree with the vertical safety height threshold to determine all The time point is recorded as the vertical threat time point, and the minimum time value is selected from all vertical threat time points as the vertical risk time .

[0104] For the horizontal risk time, this embodiment determines the horizontal safety distance threshold based on the difference between the horizontal wind deviation and the maximum horizontal swing of the conductor. The calculation formula of the horizontal safety distance threshold is:

[0105]

[0106] Where, is the horizontal safety distance threshold; is the maximum horizontal swing of the conductor.

[0107] When the horizontal safety distance threshold is not greater than the preset conductor safety margin, the tree is considered to have entered the danger zone, for example, In the most unfavorable case, the horizontal offset of the tree is exactly equal to the preset horizontal safety margin of the conductor. The time required for the tree to grow to the horizontal threat distance is determined according to the maximum safe wind deviation angle of the tree and the maximum horizontal swing of the conductor. From the time required for the tree to grow to the horizontal threat distance, the earliest time point when the tree is horizontally offset to the danger zone under the action of wind deviation is found, and the horizontal risk time is obtained. The horizontal risk time can be expressed as:

[0108]

[0109] Where, is the horizontal risk time; t is the time; The predicted height of the tree at time t; is the preset conductor safety margin; in this embodiment, the horizontal risk time The goal of the formula is to find the time t such that the horizontal displacement of the tree due to wind deflection reaches or exceeds the horizontal displacement of the wire plus a safety margin.

[0110] For the oblique risk time, this embodiment calculates the equivalent height of the oblique projection of the tree based on the predicted tree height and the cosine value of the oblique growth angle of the tree. The calculation formula for the equivalent height of the oblique projection of the tree is:

[0111]

[0112] Where, is the equivalent height of the tree's oblique projection, which represents the effective height of the tree under oblique growth conditions; Predicting height for trees; The angle at which trees grow obliquely.

[0113] This embodiment calculates the oblique safety distance based on the horizontal wind deviation and the equivalent height of the tree's oblique projection, and finds the earliest time point at which the oblique safety distance is no greater than the preset wire safety oblique margin for the first time. This time point is used as the oblique risk time when the tree enters the danger zone in an oblique growth manner. Assuming that the preset wire safety oblique margin is 1 meter, when the oblique safety distance is no greater than 1 meter for the first time, this earliest time point is determined to be the oblique risk time. The oblique risk time can be expressed as:

[0114]

[0115]

[0116]

[0117] Where, is the oblique risk time; It is the oblique safety distance between trees and conductors; is the horizontal offset of the oblique growth of the tree, which represents the horizontal offset of the tree when it grows obliquely.

[0118] S5. Evaluate the comprehensive risk index between the trees and the transmission line based on the risk time to obtain the degree of hazard of the tree barrier.

[0119] In some embodiments, the step of evaluating the comprehensive risk index between trees and power transmission lines based on the risk time to obtain the degree of tree barrier hazard comprises:

[0120] Comparing the vertical risk time, the horizontal risk time, and the oblique risk time with a preset critical response period, respectively, and classifying trees with risk times in at least two directions less than the critical response period as emergency risk trees;

[0121] The comprehensive risk index of emergency risk trees is calculated by the weighted sum of the inverse of risk time to obtain the degree of hazard of tree obstacles.

[0122] Specifically, this embodiment pre-sets a critical response period. The critical response period refers to the time window in which the power system can effectively take measures, such as pruning or removing trees, to prevent tree-line conflicts. For example, if the critical response period is determined to be 30 days, if a tree poses a risk of conflict with a transmission line within 30 days, it needs to be handled in advance. Then, this embodiment compares the vertical risk time, horizontal risk time, and diagonal risk time with the critical response period. If the risk time in at least two directions is less than the critical response period, the corresponding tree is marked as an emergency risk tree. At the same time, this embodiment assigns weights to the risk time in each of the three directions based on their impact on the hazard level of the tree barrier. For each emergency risk tree, this embodiment multiplies the inverse of the risk time in each direction by the corresponding weight and sums them to obtain a comprehensive risk index. The tree barrier hazard level is graded based on the size of the comprehensive risk index R. This embodiment can rank emergency risk trees based on the size of the comprehensive risk index. The larger the comprehensive risk index, the higher the hazard level of the tree barrier hazard to the transmission line, and the more priority treatment is required.

[0123] S6. With minimizing the total amount of pruning and maximizing the safety margin as the optimization goals, based on the degree of hazard of the tree obstacle, a non-dominated sorting genetic algorithm is used to generate a tree pruning strategy, and the risky trees on the transmission line are pruned according to the tree pruning strategy.

[0124] In some embodiments, the step of generating a tree pruning strategy using a non-dominated sorting genetic algorithm based on the degree of hazard of the tree obstacle with minimizing the total amount of pruning and maximizing the safety margin as the optimization goal includes:

[0125] Determine the planned pruning trees based on the emergency risk trees in the target area, and take the cumulative sum of the square of the DBH multiplied by the tree height of all planned pruning trees as the total pruning amount;

[0126] Calculate the safety margin between the transmission line and surrounding trees based on the minimum safe distance increment between the pruned trees and the transmission line conductors and the degree of tree barrier hazards;

[0127] An initial population is randomly generated; each individual in the initial population represents a tree pruning strategy; the tree pruning strategy includes a tree pruning priority and a tree pruning degree;

[0128] Calculating the fitness value of each individual in the initial population based on the safety margin and the total amount of pruning;

[0129] Perform non-dominated sorting on the individuals in the initial population and divide them into multiple non-dominated levels according to their fitness values. Use tournament selection to select individuals from the non-dominated levels to form the next generation population.

[0130] Perform crossover and mutation operations on individuals in the next generation population until a preset maximum number of iterations is reached, stop the iterative process, output a non-dominated optimal solution set, and select the optimal tree pruning strategy from the non-dominated optimal solution set.

[0131] Specifically, this embodiment identifies emergency risk trees within the target area as trees planned for pruning. For each tree planned for pruning, its diameter at breast height (DBH) (trunk diameter) and tree height are obtained. The product of the square of the DBH and the tree height of each tree is calculated to obtain the pruning amount for each tree. The pruning amounts of all trees planned for pruning are accumulated to obtain the total pruning amount. For each tree planned for pruning, the tree's morphology after pruning is simulated according to the pruning strategy (pruning priority and pruning degree). The minimum post-pruning distance between the pruned tree and the transmission line conductor is calculated, and the minimum pre-pruning distance between the tree and the conductor before pruning is obtained. The minimum post-pruning distance is subtracted from the minimum pre-pruning distance to obtain the minimum safety distance increment. Next, this embodiment calculates the safety margin between the transmission line and surrounding trees in combination with the tree barrier hazard level (comprehensive risk index). The safety margin can be obtained by weighted summation of the tree barrier hazard level and the minimum safety distance increment. The safety margin reflects the safety level between the transmission line and surrounding trees after pruning. The higher the safety margin, the lower the tree-line conflict risk.

[0132] Then, this embodiment sets the size of the initial population to N, and each individual code represents a tree pruning strategy. The tree pruning strategy includes at least a tree pruning priority and a tree pruning degree. This embodiment can assign a priority number to each tree planned for pruning, where the smaller the number, the higher the priority. N groups of different priority rankings are randomly generated. For each tree, the pruning degree is set to a value between 0 (no pruning) and 1 (complete pruning). N groups of different pruning degree combinations are randomly generated. At the same time, for each individual in the initial population, this embodiment uses a weighted method to perform a weighted summation of the inverse of the total pruning amount and the safety margin, and calculates the fitness value of each individual in the population. The fitness value reflects the comprehensive performance of its pruning strategy under the two optimization objectives. The higher the fitness value, the better the pruning strategy, that is, while minimizing the pruning amount, the safety margin is better improved.

[0133] Perform non-dominated sorting on the individuals in the initial population, traverse all individuals in the initial population, compare the fitness values ​​of every two individuals, for individual A and individual B, if the safety margin of individual A is not less than that of individual B and the total amount of pruning of individual A is not greater than that of individual B, and at least one indicator is better than that of individual B, then individual A is considered to dominate individual B. According to the dominance relationship, the individuals are divided into multiple non-dominated levels. This embodiment adopts the tournament selection method to select individuals from the non-dominated level to form the next generation population. Each time, several individuals are randomly selected, and the individuals with the best fitness value are selected from the selected individuals to enter the next generation population. This process is repeated until the size of the next generation population reaches N. The individuals in the next generation population are cross-operated. For example, for the pruning priority part, a crossover point can be randomly selected to exchange parts of the two individuals. The trees are sorted by priority; for the pruning degree, a new pruning degree combination can be generated by arithmetic crossover or other methods. Then, this embodiment performs a mutation operation on the individuals after crossover. For the pruning priority, the priority number of a tree can be randomly changed; for the pruning degree, the pruning degree value can be randomly adjusted within a certain range, and the above iterative process is repeated until the preset maximum number of iterations is reached. After the iteration, a non-dominated optimal solution set is obtained, and the tree pruning strategy with the best fitness value is selected from the non-dominated optimal solution set as the final optimal tree pruning strategy. The strategy includes the optimal tree pruning priority and tree pruning degree. This embodiment uses a non-dominated sorting genetic algorithm to generate a tree pruning strategy that minimizes the total amount of pruning and maximizes the safety margin, providing a scientific basis for tree barrier treatment of transmission lines.

[0134] In order to verify the transmission line tree obstacle hazard detection method considering the influence of wind deviation proposed in this embodiment, this embodiment accurately predicts the time when the top of the bamboo enters the safe distance range of the transmission line due to the influence of wind deviation by considering the influence of wind deviation, and then formulates a scientific pruning or early warning plan to ensure the safe operation of the transmission line. To achieve this goal, this embodiment selects a 35kV transmission line corridor area in a certain area as a pilot area. The pilot area is mainly hilly. There are a large number of bamboos in the growth stage distributed along the pilot area. The local annual dominant wind direction is southeast-southeast, and the annual maximum wind speed can reach 10m / s. This natural condition makes the bamboo extremely susceptible to the influence of wind deviation during its growth, which poses a potential threat to the safety of the transmission line. In the specific implementation process, this embodiment dynamically predicts the growth of bamboo in the pilot area to obtain the tree height distribution at different time points. Figure 2 The growth trend of bamboo in the next ten months is presented. At the same time, Table 1 lists the predicted time when some bamboos in the pilot area will enter the safe distance range of transmission lines and the corresponding risk assessment (five typical bamboos in the area are selected as examples for demonstration). Table 1 is as follows:

[0135] Table 1

[0136]

[0137] The above-mentioned forecast data indicate that some bamboos will enter the safe distance range of the transmission line in a short period of no more than two months. Therefore, to ensure the safe operation of the transmission line, the operation and maintenance personnel should complete the pruning of the relevant bamboos within a time window of no more than two months to effectively prevent the bamboos from invading the safe distance range of the transmission line due to wind deflection, thereby ensuring the stable operation of the transmission line.

[0138] An embodiment of the present invention provides a method for detecting tree barrier hazards on power transmission lines that considers the effects of windage. The method includes collecting tree growth characteristic data, meteorological environment monitoring data, and transmission line conductor parameters in a target area within a transmission line corridor to form an original multi-source dataset; dynamically predicting tree growth based on the original multi-source dataset to obtain a tree growth prediction result for each tree in the target area; performing a three-dimensional windage effect analysis based on the tree growth prediction results and the original multi-source dataset to obtain a tree windage quantification result; calculating the risk time of trees in the vertical, horizontal, and diagonal directions based on the tree growth prediction results and the tree windage quantification results; evaluating a comprehensive risk index between the tree and the transmission line based on the risk time to obtain the degree of tree barrier hazard; and generating a tree pruning strategy based on the degree of tree barrier hazard using a non-dominated sorting genetic algorithm with the optimization goals of minimizing the total amount of pruning and maximizing the safety margin. The method then prunes risky trees on the transmission line according to the tree pruning strategy. Compared with existing technologies, this method integrates multi-source data to perform tree growth prediction and windage effect quantification, accurately assesses the degree of tree barrier hazard, and generates an optimal tree pruning strategy, effectively reducing the risk of transmission line failure caused by tree barriers and ensuring the safe and stable operation of the transmission line.

[0139] It should be noted that the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of this application.

[0140] In one embodiment, Figure 3 As shown, an embodiment of the present invention provides a transmission line tree obstacle hazard detection system considering the influence of wind deviation, the system comprising:

[0141] The data acquisition module 101 is used to collect tree growth characteristic data, meteorological environment monitoring data and transmission line conductor parameters in the target area within the transmission line corridor to form an original multi-source data set;

[0142] A growth prediction module 102 is configured to perform tree growth dynamic prediction based on the original multi-source dataset to obtain a tree growth prediction result for each tree in the target area;

[0143] A wind deviation quantification module 103 is configured to perform a three-dimensional wind deviation effect analysis based on the tree growth prediction result and the original multi-source data set to obtain a tree wind deviation quantification result;

[0144] The risk analysis module 104 is configured to calculate the risk time of the tree in the vertical, horizontal and oblique directions respectively according to the tree growth prediction result and the tree wind deviation quantification result;

[0145] The hidden danger detection module 105 is used to evaluate the comprehensive risk index between the tree and the transmission line according to the risk time to obtain the degree of hazard of the tree obstacle;

[0146] The strategy generation module 106 is used to minimize the total amount of pruning and maximize the safety margin as the optimization goal, generate a tree pruning strategy based on the hazard level of the tree obstacle using a non-dominated sorting genetic algorithm, and prune the risky trees on the transmission line according to the tree pruning strategy.

[0147] Regarding the specific definition of a transmission line tree barrier hazard detection system that takes into account the influence of wind deviation, please refer to the above-mentioned definition of a transmission line tree barrier hazard detection method that takes into account the influence of wind deviation, which will not be repeated here. A person of ordinary skill in the art will appreciate that the various modules and steps described in conjunction with the embodiments disclosed in this application can be implemented in hardware, software, or a combination of both. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0148] An embodiment of the present invention provides a transmission line tree barrier hazard detection system that takes into account the influence of wind deviation. The system collects tree growth characteristic data, meteorological environment monitoring data and transmission line conductor parameters in a target area within a transmission line corridor through a data acquisition module to form an original multi-source data set; a growth prediction module performs dynamic tree growth prediction based on the original multi-source data set to obtain tree growth prediction results for each tree in the target area; a wind deviation quantification module performs three-dimensional wind deviation effect analysis based on the tree growth prediction results and the original multi-source data set to obtain tree wind deviation quantification results; a risk analysis module calculates the risk time of trees in the vertical, horizontal and oblique directions based on the tree growth prediction results and the tree wind deviation quantification results; a hazard detection module evaluates the comprehensive risk index between trees and transmission lines based on the risk time to obtain the degree of tree barrier hazard hazard; a strategy generation module uses a non-dominated sorting genetic algorithm to generate a tree pruning strategy based on the degree of tree barrier hazard hazard, with minimizing the total amount of pruning and maximizing the safety margin as optimization goals, and prunes risky trees on the transmission line according to the tree pruning strategy. Compared with existing technologies, this system predicts tree growth and quantifies wind deviation effects by integrating multi-source data, accurately assesses the degree of tree barrier hazards and generates optimal tree pruning strategies, effectively reducing the risk of transmission line failures caused by tree barriers and ensuring the safe and stable operation of transmission lines.

[0149] In one embodiment, the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the above method are implemented.

[0150] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., an SSD).

[0151] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing related hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods.

[0152] The above-described embodiments merely represent several preferred implementations of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent. It should be noted that a person skilled in the art could make several improvements and substitutions without departing from the technical principles of the present invention, and these improvements and substitutions should also be considered within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be based on the scope of protection of the claims.

Claims

1. A method for detecting tree obstacles on power transmission lines considering the influence of wind deviation, characterized in that: The following steps are involved: Collect tree growth characteristic data, meteorological environment monitoring data, and transmission line conductor parameters in the target area within the transmission line corridor to form an original multi-source data set; Performing tree growth dynamics prediction based on the original multi-source data set to obtain tree growth prediction results for each tree in the target area; A three-dimensional wind deflection effect analysis is performed based on the tree growth prediction result and the original multi-source data set to obtain a quantitative result of the tree wind deflection, which includes the horizontal wind deflection amount, vertical wind deflection amount and maximum horizontal swing amount of the conductor of the tree at different wind speeds and wind directions, including: extracting tree mechanical parameters, crown morphological characteristics and conductor mechanical parameters from the original multi-source data set; the tree mechanical parameters include the tree elastic modulus and the tree cross-sectional inertia moment, the crown morphological characteristics include the wind-exposed area of ​​the tree crown and the height of the center of gravity of the tree crown; the conductor mechanical parameters include the tension of the transmission line conductor; the wind torque exerted on the tree at different wind speeds is calculated according to the tree windward area and the instantaneous wind speed, and the tree restoring torque is calculated according to the tree elastic modulus, the tree cross-sectional inertia moment and the tree inclination angle; the tree inclination angle corresponding to the equality of the wind torque and the tree restoring torque is used as the maximum safe wind deflection angle of the tree; Calculate the horizontal deviation of the tree toward the wire direction according to the maximum safe wind deviation angle of the tree and the tree growth prediction result to obtain the horizontal wind deviation; Calculating the center of gravity offset of the tree under wind action based on the wind-exposed area of ​​the tree canopy and the height of the center of gravity of the tree canopy to obtain the vertical wind deflection; The maximum horizontal swing amplitude allowed for the transmission line conductor is calculated based on the transmission line conductor tension and real-time wind speed monitoring data, and the maximum horizontal swing amount of the conductor is obtained; Calculating the risk time of trees in the vertical, horizontal and oblique directions respectively according to the tree growth prediction result and the tree wind deviation quantification result, including: obtaining the hanging height data of the transmission line conductor, and obtaining the predicted tree height according to the tree growth prediction result; calculating the vertical direction safety height threshold according to the hanging height data of the transmission line conductor and the preset conductor safety height margin, and calculating the effective height of the tree according to the tree predicted height and the vertical wind deviation; comparing the effective height of the tree with the vertical direction safety height threshold, finding all vertical threat time points at which the effective height of the tree is not less than the vertical direction safety height threshold, and selecting the minimum time value from the vertical threat time points as the vertical risk time; Determining a horizontal safety distance threshold based on the difference between the horizontal wind deviation and the maximum horizontal swing of the conductor; and when the horizontal safety distance threshold is not greater than a preset conductor safety horizontal margin, calculating the shortest time for the tree to horizontally deviate to the danger zone under the influence of wind deviation based on the maximum safe wind deviation angle of the tree and the maximum horizontal swing of the conductor to obtain a horizontal risk time; Calculate the equivalent height of the tree's oblique projection based on the predicted tree height and the cosine value of the oblique growth angle, and calculate the oblique safety distance based on the tree's oblique growth horizontal offset and the equivalent height of the tree's oblique projection; find the earliest time point at which the oblique safety distance is not greater than the preset conductor safety oblique margin for the first time, and use it as the oblique risk time of the tree entering the danger zone in an oblique growth manner; Evaluate the comprehensive risk index between trees and transmission lines based on the risk time to obtain the degree of tree barrier hazard; With the optimization goals of minimizing the total amount of pruning and maximizing the safety margin, a non-dominated sorting genetic algorithm is used to generate a tree pruning strategy based on the hazard level of the tree obstacle, and the risky trees on the transmission line are pruned according to the tree pruning strategy.

2. A method for detecting tree obstacles in power transmission lines considering the influence of wind deviation according to claim 1, characterized in that: The step of performing tree growth dynamic prediction based on the original multi-source data set to obtain tree growth prediction results for each tree in the target area includes: Preprocessing the original multi-source data set to extract tree growth-related features, conductor dynamic features, and geographic features, and generate a standardized feature matrix; Obtain the measured data of tree heights and use the average value of the measured data of tree heights as the benchmark prediction value; Calculating the prediction error and error change rate between the reference prediction value and the actual tree height data; Traversing each feature in the standardized feature matrix, calculating the information gain of each split point according to the prediction error and the error change rate, selecting the split point with the largest information gain as the optimal split point of the current node, and recursively constructing an extreme gradient boosting tree; The normalized feature matrix is ​​subjected to tree growth dynamic prediction by using the extreme gradient boosting tree to obtain tree growth prediction results for each tree in the target area.

3. A method for detecting tree obstacles in power transmission lines considering the influence of wind deviation according to claim 2, characterized in that: The step of calculating the prediction error and the error change rate between the reference prediction value and the tree height measured data comprises: The prediction error and the error change rate between the reference prediction value and the actual tree height data are calculated based on the mean square error loss function.

4. The method for detecting tree obstacles on power transmission lines considering the influence of wind deviation according to claim 1, characterized in that: The step of evaluating the comprehensive risk index between trees and transmission lines according to the risk time to obtain the degree of tree barrier hazard comprises: Comparing the vertical risk time, the horizontal risk time, and the oblique risk time with a preset critical response period, respectively, and classifying trees with risk times in at least two directions less than the critical response period as emergency risk trees; The comprehensive risk index of emergency risk trees is calculated by the weighted sum of the inverse of risk time to obtain the degree of hazard of tree obstacles.

5. The method for detecting tree obstacles in power transmission lines considering the influence of wind deviation according to claim 1, characterized in that: The steps of generating a tree pruning strategy using a non-dominated sorting genetic algorithm based on the degree of hazard of tree obstacles with the optimization goal of minimizing the total amount of pruning and maximizing the safety margin include: Determine the planned pruning trees based on the emergency risk trees in the target area, and take the cumulative sum of the square of the DBH multiplied by the tree height of all planned pruning trees as the total pruning amount; Calculate the safety margin between the transmission line and surrounding trees based on the minimum safe distance increment between the pruned trees and the transmission line conductors and the degree of tree barrier hazards; Randomly generate the initial population; each individual in the initial population represents a tree pruning strategy; Calculating the fitness value of each individual in the initial population based on the safety margin and the total amount of pruning; Perform non-dominated sorting on the individuals in the initial population and divide them into multiple non-dominated levels according to their fitness values. Use tournament selection to select individuals from the non-dominated levels to form the next generation population. Perform crossover and mutation operations on individuals in the next generation population until a preset maximum number of iterations is reached, stop the iterative process, output a non-dominated optimal solution set, and select the optimal tree pruning strategy from the non-dominated optimal solution set.

6. The method for detecting tree obstacles on power transmission lines considering the influence of wind deviation according to claim 5, characterized in that: The tree pruning strategy includes a tree pruning priority and a tree pruning degree.

7. A power transmission line tree obstacle hazard detection system considering the influence of wind deviation, characterized in that: The system comprises: The data acquisition module is used to collect tree growth characteristic data, meteorological environment monitoring data and transmission line conductor parameters in the target area within the transmission line corridor to form an original multi-source data set; a growth prediction module, configured to perform tree growth dynamics prediction based on the original multi-source dataset to obtain tree growth prediction results for each tree in the target area; A wind deflection quantification module is used to perform a three-dimensional wind deflection effect analysis based on the tree growth prediction result and the original multi-source data set to obtain a tree wind deflection quantification result, wherein the tree wind deflection quantification result includes the horizontal wind deflection amount, vertical wind deflection amount and maximum horizontal swing amount of the conductor under different wind speeds and wind directions, including: extracting tree mechanical parameters, crown morphological characteristics and conductor mechanical parameters from the original multi-source data set; the tree mechanical parameters include the tree elastic modulus and the tree cross-sectional inertia moment, the crown morphological characteristics include the wind-exposed area of ​​the tree crown and the height of the center of gravity of the tree crown; the conductor mechanical parameters include the tension of the transmission line conductor; the wind torque exerted on the tree at different wind speeds is calculated according to the tree windward area and instantaneous wind speed, and the tree restoring torque is calculated according to the tree elastic modulus, the tree cross-sectional inertia moment and the tree inclination angle; the tree inclination angle corresponding to the equality of the wind torque and the tree restoring torque is used as the maximum safe wind deflection angle of the tree; Calculate the horizontal deviation of the tree toward the wire direction according to the maximum safe wind deviation angle of the tree and the tree growth prediction result to obtain the horizontal wind deviation; Calculating the center of gravity offset of the tree under wind action based on the wind-exposed area of ​​the tree canopy and the height of the center of gravity of the tree canopy to obtain the vertical wind deflection; The maximum horizontal swing amplitude allowed for the transmission line conductor is calculated based on the transmission line conductor tension and real-time wind speed monitoring data, and the maximum horizontal swing amount of the conductor is obtained; a risk analysis module for respectively calculating the risk time of trees in the vertical, horizontal and oblique directions based on the tree growth prediction result and the tree wind deviation quantification result, comprising: obtaining the hanging height data of the transmission line conductor, and obtaining the predicted tree height based on the tree growth prediction result; calculating a vertical direction safety height threshold based on the hanging height data of the transmission line conductor and a preset conductor safety height margin, and calculating the effective height of the tree based on the tree predicted height and the vertical wind deviation; comparing the effective height of the tree with the vertical direction safety height threshold, finding all vertical threat time points at which the effective height of the tree is not less than the vertical direction safety height threshold, and selecting the minimum time value from the vertical threat time points as the vertical risk time; Determining a horizontal safety distance threshold based on the difference between the horizontal wind deviation and the maximum horizontal swing of the conductor; and when the horizontal safety distance threshold is not greater than a preset conductor safety horizontal margin, calculating the shortest time for the tree to horizontally deviate to the danger zone under the influence of wind deviation based on the maximum safe wind deviation angle of the tree and the maximum horizontal swing of the conductor to obtain a horizontal risk time; Calculate the equivalent height of the tree's oblique projection based on the predicted tree height and the cosine value of the oblique growth angle, and calculate the oblique safety distance based on the tree's oblique growth horizontal offset and the equivalent height of the tree's oblique projection; find the earliest time point at which the oblique safety distance is not greater than the preset conductor safety oblique margin for the first time, and use it as the oblique risk time of the tree entering the danger zone in an oblique growth manner; a hidden danger detection module, configured to evaluate the comprehensive risk index between trees and transmission lines based on the risk time, and obtain the degree of hazard of tree barrier hidden dangers; The strategy generation module is used to minimize the total amount of pruning and maximize the safety margin as the optimization goals, generate a tree pruning strategy based on the degree of hazard of the tree obstacle hazards using a non-dominated sorting genetic algorithm, and prune the risky trees on the transmission line according to the tree pruning strategy.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed, the method according to any one of claims 1 to 6 is implemented.

Images