Method for monitoring and evaluating state of transmission line assisted by airborne laser radarTechnical Field
The invention relates to the technical field of laser radars, in particular to an airborne laser radar-assisted power transmission line state monitoring and evaluating method.
Background
Traditional transmission line inspection mainly relies on the manual work, and inspection personnel need walk along transmission line passageway or inspect with the help of means of transportation, and this kind of mode intensity of labour is big, inefficiency to receive geographical environment and climate condition's restriction seriously, inspection personnel can be difficult to reach certain key positions of transmission line, thereby can't in time discover potential hidden danger, and under bad weather, manual inspection often can't normally develop moreover, this just can lead to the untimely and incomplete to transmission line state information acquisition.
The laser radar (LiDAR) is an active optical remote sensing technology, has the characteristics of high precision and high resolution, can quickly acquire a three-dimensional model of a large-area landform and an object, can quickly fly along a power transmission line corridor and cover a longer line section once compared with ground monitoring, greatly improves the monitoring efficiency, can acquire a large amount of power transmission line data in a short time, and can flexibly adjust the flight route and the height by an airborne platform so as to adapt to the monitoring requirements of different terrains and power transmission line directions.
Because the salt fog concentration in the air of the offshore transmission line is higher, the offshore transmission line is more easily corroded compared with the onshore transmission line, so that operation and maintenance personnel need to have a specific judgment on the corrosion degree of the transmission line to overhaul the offshore transmission line.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides an airborne laser radar-assisted power transmission line state monitoring and evaluating method, which can effectively solve the problems in the prior art.
The invention provides an airborne laser radar-assisted power transmission line state monitoring and evaluating method, which comprises the following steps of 1, utilizing airborne laser radar equipment to conduct three-dimensional scanning on a marine power transmission line and surrounding environments thereof to obtain point cloud data and airplane positions, 2, transmitting the collected point cloud data to a ground data processing center for preprocessing, 3, extracting geometric features of the power transmission line based on the preprocessed point cloud data, analyzing distance changes between adjacent points in the point cloud data, calculating roughness of the surface of the power transmission line, finally utilizing intensity information of the laser radar to obtain current marine salt fog concentration, 4, taking the extracted geometric features, roughness and salt fog concentration of the power transmission line as input, establishing a corrosion identification model, judging corrosion degree of the power transmission line, 5, applying the trained corrosion identification model to monitoring of the marine power transmission line, collecting the point cloud data of the power transmission line in real time through the laser radar, extracting features and inputting the features into the corrosion identification model, and achieving real-time identification and monitoring of the corrosion degree of the power transmission line.
Preferably, the step 2 specifically includes the steps of step 2.1, performing noise reduction processing on the point cloud data, and for each point in the point cloud dataCalculate its local densityDefining a radius r, counting the number of points within the radius r to represent the local densityThe formula is: Wherein X is an indication function,Is taken as a pointTo adjacent pointsDistance between whenWhen X=1, otherwise X=0, find all local densities greater thanPoints of (2)Form a collection,Calculate the pointTo a collectionDistance of one point in (2)Point thenTo a local density greater thanAverage distance of points of (2)The calculation formula of (2) is as follows: Wherein m is a setThe number of midpoints, defining denoising indexes: wherein, the method comprises the steps of,Is the maximum value of the local density of all points in the point cloud, and sets a denoising threshold valueWhen (when)When it is, then pointStep 2.2, carrying out direct filtering processing on the noise-reduced point cloud data, wherein each point in the noise-reduced point cloud dataExtracting the coordinate value of the X-ray tube on the Z axisSetting a threshold range in the Z-axis directionJudgingWhether a threshold condition is met, namely: If the condition is satisfied, reserving the point in the point cloud dataAdding it into the new filtered point cloud data set, if the condition is not satisfied, discarding the points in the point cloud dataAnd after traversing all points, the points in the new point cloud data set which are reserved after the through filtering are the power transmission line point clouds in the threshold range.
Preferably, in the step 3, based on the preprocessed point cloud data, the geometric characteristics of the power transmission line are extracted, and the method specifically comprises the steps of performing self-adaptive space division on the point cloud data of the power transmission line, dividing the point cloud into a plurality of local areas by using a density clustering algorithm according to the density distribution of the point cloud data, wherein the divided areas are collected asFor each regionCalculating the point cloud densityAnd average spacingComputing cloud densityThe formula of (2) is: whereinIs the number of points within the region,Is the volume of the region, average spacingBy calculating the average value of the distances between points in the region, the points are setAndTwo points within the regionAndEuclidean distance betweenThe method comprises the following steps: Average distanceThe method comprises the following steps: wherein, the method comprises the steps of,Representing slaveSelecting 2 combinations of elements, extracting local areaIs characterized in that in each local areaIn the method, a local geometric diagram is built, points in the area are used as nodes of the diagram, edge weights between the two points are Euclidean distances between the nodes, and a Laplacian matrix of the diagram is calculatedIts elementsIs defined as whenWhen (1): wherein, the method comprises the steps of,Is the edge weight between the node i and the node j, namely the Euclidean distance between the two points, whenAnd node i and node j have edges connected: When (1)And node i and node j are connected without edges: A Laplace matrixPerforming feature decomposition to obtain feature valuesAnd feature vectorSelecting the smallest non-zero characteristic valueCorresponding feature vectorIs provided withLocal diameter estimate within a regionThe method comprises the following steps: wherein, the method comprises the steps of,As the difference value of the feature vector elements,Is the average distance, estimated value according to the diameter of each local areaAnd carrying out global diameter fusion, wherein the global diameter D is as follows: wherein, the method comprises the steps of,Is the weight of the region.
Preferably, in the step 3, the distance change between adjacent points in the point cloud data is analyzed, and the roughness of the surface of the power transmission line is calculated, and the method specifically comprises the steps of setting the point cloud data asRepresenting the coordinate set of the surface and surrounding points of the power transmission line, wherein n is the total number of the points, and calculating each point by adopting a radius neighborhood methodWith adjacent pointsThe distance between the two is expressed as the formula: wherein, the method comprises the steps of,Is taken as a pointWith adjacent pointsDistance between them, roughness of the surface of the transmission lineThe calculation formula is as follows: Wherein n is the total number of points, k is the number of adjacent points for each point,Is a dotWith adjacent pointsThe amount of change in the distance between them.
Preferably, the step 3 utilizes the intensity information of the laser radar to obtain the current sea salt fog concentration, and specifically comprises the steps of scanning a sea area by the laser radar, wherein the received echo intensity is I, the measuring distance is d, and the measuring angle isForming an original data setWherein n represents different measuring points, and obtains information of offshore wind speed v, humidity h and temperature t, and a relation model of salt fog concentration and laser radar intensity is established, which is as follows: wherein I is the echo intensity,Is the laser radar echo intensity without salt fog,Is the attenuation coefficient of salt fog to laser,For salt fog concentration, d is the measured distance, f is the correction coefficient, and the point is calculatedSubstituting the salt fog concentration and the laser radar intensity into a relation model to obtain: wherein, the method comprises the steps of,For the concentration of the salt fog,The wind speed, the humidity and the temperature corresponding to the measuring point are respectively obtained.
Preferably, in the step 4, the corrosion identification model is established based on a neural network algorithm, and the geometric characteristic data Z and the roughness data of the transmission line are calculatedAnd salt fog concentration dataAs input, let the geometric feature data beRoughness data isSalt fog concentration data isNormalizing the geometric feature data, the roughness data and the salt spray concentration data, and adopting a linear normalization formula for the geometric feature data: wherein, the method comprises the steps of,As the minimum value in the geometric feature data,Is the maximum value in the geometric feature data,The same normalization method is carried out on the roughness data and the salt fog concentration data to obtain normalized roughness data asNormalized salt spray concentration data isCombining the normalized geometric features, roughness and salt spray concentration data into feature vectorsCalculating a risk value H of the corrosion degree by using a convolutional neural network, wherein the risk value H is expressed as: Wherein W is a weight coefficient, F is an output function,Is a convolutional neural network model of feature vector X.
Preferably, a risk value threshold is setAndWhen (when)When the transmission line is judged to be slightly corroded, whenWhen the transmission line is determined to be moderately corroded, whenAnd when the power transmission line is severely corroded, judging.
Preferably, the maintenance of different degrees is carried out according to the risk value, the monitoring period is shortened when the power transmission line is judged to be slightly corroded, the corrosion change is closely concerned, the comprehensive anti-corrosion treatment is carried out on the power transmission line when the power transmission line is judged to be moderately corroded, the anti-corrosion protection film is formed by adopting the hot galvanizing and anti-corrosion paint spraying modes, and the power transmission line is replaced when the power transmission line is judged to be severely corroded.
Compared with the prior art, the technical scheme has the advantages that the corrosion degree of the offshore power transmission line can be analyzed by combining the geometric characteristics, the roughness and the salt fog concentration, the condition of the power transmission line can be known more accurately and comprehensively, the regional division and the characteristic extraction can be performed in a self-adaptive mode according to the self-characteristics of the point cloud data through the self-adaptive local geometric analysis algorithm, the diameter characteristics of the power transmission line can be extracted more accurately, and the method is particularly suitable for the conditions that the distribution of the point cloud data is uneven or the power transmission line has a complex structure.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings which are required in the description of the embodiments or the prior art will be briefly described below, it being obvious that the drawings in the description below are only some embodiments of the invention and that other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments, and all other embodiments obtained by those skilled in the art without making any inventive effort based on the embodiments of the present invention are within the scope of protection of the present invention.
The invention is further described below with reference to examples.
The method comprises the steps of 1, carrying out three-dimensional scanning on an offshore power transmission line and surrounding environments of the offshore power transmission line by using airborne laser radar equipment to obtain point cloud data and airplane positions, 2, transmitting the collected point cloud data to a ground data processing center for preprocessing, 3, extracting geometric features of the power transmission line based on the preprocessed point cloud data, analyzing distance changes between adjacent points in the point cloud data, calculating roughness of the surface of the power transmission line, finally obtaining the current offshore salt fog concentration by using intensity information of the laser radar, 4, establishing a corrosion identification model by taking the extracted geometric features, roughness and salt fog concentration of the power transmission line as input, judging the corrosion degree of the power transmission line, 5, applying the trained corrosion identification model to the offshore power transmission line monitoring, collecting the point cloud data of the power transmission line in real time by using the laser radar, extracting features and inputting the features into the corrosion identification model, and realizing real-time identification and monitoring of the corrosion degree of the power transmission line, and completing closed loop.
Preferably, the step 2 specifically includes the steps of step 2.1, performing noise reduction processing on the point cloud data, and for each point in the point cloud dataCalculate its local densityDefining a radius r, counting the number of points within the radius r to represent the local densityThe formula is: Wherein X is an indication function,Is taken as a pointTo adjacent pointsDistance between whenWhen X=1, otherwise X=0, find all local densities greater thanPoints of (2)Form a collection,Calculate the pointTo a collectionDistance of one point in (2)Point thenTo a local density greater thanAverage distance of points of (2)The calculation formula of (2) is as follows: Wherein m is a setThe number of midpoints, defining denoising indexes: wherein, the method comprises the steps of,Is the maximum value of the local density of all points in the point cloud, and sets a denoising threshold valueWhen (when)When it is, then pointStep 2.2, carrying out direct filtering processing on the noise-reduced point cloud data, wherein each point in the noise-reduced point cloud dataExtracting the coordinate value of the X-ray tube on the Z axisSetting a threshold range in the Z-axis directionJudgingWhether a threshold condition is met, namely: If the condition is satisfied, reserving the point in the point cloud dataAdding it into the new filtered point cloud data set, if the condition is not satisfied, discarding the points in the point cloud dataAnd after traversing all points, the points in the new point cloud data set which are reserved after the through filtering are the power transmission line point clouds in the threshold range.
Preferably, in the step 3, based on the preprocessed point cloud data, the geometric characteristics of the power transmission line are extracted, and the method specifically comprises the steps of performing self-adaptive space division on the point cloud data of the power transmission line, dividing the point cloud into a plurality of local areas by using a density clustering algorithm according to the density distribution of the point cloud data, wherein the divided areas are collected asFor each regionCalculating the point cloud densityAnd average spacingComputing cloud densityThe formula of (2) is: whereinIs the number of points within the region,Is the volume of the region, average spacingBy calculating the average value of the distances between points in the region, the points are setAndTwo points within the regionAndEuclidean distance betweenThe method comprises the following steps: Average distanceThe method comprises the following steps: wherein, the method comprises the steps of,Representing slaveSelecting 2 combinations of elements, extracting local areaIs characterized in that in each local areaIn the method, a local geometric diagram is built, points in the area are used as nodes of the diagram, edge weights between the two points are Euclidean distances between the nodes, and a Laplacian matrix of the diagram is calculatedIts elementsIs defined as whenWhen (1): wherein, the method comprises the steps of,Is the edge weight between the node i and the node j, namely the Euclidean distance between the two points, whenAnd node i and node j have edges connected: When (1)And node i and node j are connected without edges: A Laplace matrixPerforming feature decomposition to obtain feature valuesAnd feature vectorSelecting the smallest non-zero characteristic valueCorresponding feature vectorIs provided withLocal diameter estimate within a regionThe method comprises the following steps: wherein, the method comprises the steps of,As the difference value of the feature vector elements,Is the average distance, estimated value according to the diameter of each local areaAnd carrying out global diameter fusion, wherein the global diameter D is as follows: wherein, the method comprises the steps of,Is the weight of the region.
Preferably, in the step 3, the distance change between adjacent points in the point cloud data is analyzed, and the roughness of the surface of the power transmission line is calculated, and the method specifically comprises the steps of setting the point cloud data asRepresenting the coordinate set of the surface and surrounding points of the power transmission line, wherein n is the total number of the points, and calculating each point by adopting a radius neighborhood methodWith adjacent pointsThe distance between the two is expressed as the formula: wherein, the method comprises the steps of,Is taken as a pointWith adjacent pointsDistance between them, roughness of the surface of the transmission lineThe calculation formula is as follows: Wherein n is the total number of points, k is the number of adjacent points for each point,Is a dotWith adjacent pointsThe amount of change in the distance between them.
Preferably, the step 3 utilizes the intensity information of the laser radar to obtain the current sea salt fog concentration, and specifically comprises the steps of scanning a sea area by the laser radar, wherein the received echo intensity is I, the measuring distance is d, and the measuring angle isForming an original data setWherein n represents different measuring points, and obtains information of offshore wind speed v, humidity h and temperature t, and a relation model of salt fog concentration and laser radar intensity is established, which is as follows: wherein I is the echo intensity,Is the laser radar echo intensity without salt fog,Is the attenuation coefficient of salt fog to laser,For salt fog concentration, d is the measured distance, f is the correction coefficient, and the point is calculatedSubstituting the salt fog concentration and the laser radar intensity into a relation model to obtain: wherein, the method comprises the steps of,For the concentration of the salt fog,The wind speed, the humidity and the temperature corresponding to the measuring point are respectively obtained.
Preferably, in the step 4, the corrosion identification model is established based on a neural network algorithm, and the geometric characteristic data Z and the roughness data of the transmission line are calculatedAnd salt fog concentration dataAs input, let the geometric feature data beRoughness data isSalt fog concentration data isNormalizing the geometric feature data, the roughness data and the salt spray concentration data, and adopting a linear normalization formula for the geometric feature data: wherein, the method comprises the steps of,As the minimum value in the geometric feature data,Is the maximum value in the geometric feature data,The same normalization method is carried out on the roughness data and the salt fog concentration data to obtain normalized roughness data asNormalized salt spray concentration data isCombining the normalized geometric features, roughness and salt spray concentration data into feature vectorsCalculating a risk value H of the corrosion degree by using a convolutional neural network, wherein the risk value H is expressed as: Wherein W is a weight coefficient, F is an output function,Is a convolutional neural network model of feature vector X.
Preferably, a risk value threshold is setAndWhen (when)When the transmission line is judged to be slightly corroded, whenWhen the transmission line is determined to be moderately corroded, whenAnd when the power transmission line is severely corroded, judging.
Preferably, the maintenance of different degrees is carried out according to the risk value, the monitoring period is shortened when the power transmission line is judged to be slightly corroded, the corrosion change is closely concerned, the comprehensive anti-corrosion treatment is carried out on the power transmission line when the power transmission line is judged to be moderately corroded, the anti-corrosion protection film is formed by adopting the hot galvanizing and anti-corrosion paint spraying modes, and the power transmission line is replaced when the power transmission line is judged to be severely corroded.
Although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that the foregoing embodiments may be modified or equivalents may be substituted for some of the features thereof, and that the modification or substitution does not depart from the spirit and scope of the embodiments.