CN114239379A - Transmission line geological disaster analysis method and system based on deformation detection - Google Patents
Transmission line geological disaster analysis method and system based on deformation detection Download PDFInfo
- Publication number
- CN114239379A CN114239379A CN202111340434.2A CN202111340434A CN114239379A CN 114239379 A CN114239379 A CN 114239379A CN 202111340434 A CN202111340434 A CN 202111340434A CN 114239379 A CN114239379 A CN 114239379A
- Authority
- CN
- China
- Prior art keywords
- geological
- damage
- deformation
- damage degree
- offset
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/89—Radar or analogous systems specially adapted for specific applications for mapping or imaging
- G01S13/90—Radar or analogous systems specially adapted for specific applications for mapping or imaging using synthetic aperture techniques, e.g. synthetic aperture radar [SAR] techniques
- G01S13/9021—SAR image post-processing techniques
- G01S13/9023—SAR image post-processing techniques combined with interferometric techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]
Landscapes
- Engineering & Computer Science (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Computer Networks & Wireless Communication (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Electromagnetism (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention relates to a power transmission line geological disaster analysis method and system based on deformation detection, and belongs to the technical field of remote sensing application. The method comprises the steps of obtaining tower foundation geological damage degree distribution based on comparison of a plurality of groups of remote sensing images before and after an earthquake, obtaining an earthquake deformation field diagram by adopting differential interference, marking tower foundation geological damage degree distribution, segmenting according to an offset value range and an offset direction to form a sample library, and training a plurality of artificial neural network models; the method comprises the steps of obtaining remote sensing images before and after an earthquake, obtaining a deformation field through differential interference, carrying out segmentation based on an offset value and an offset direction, selecting a corresponding artificial neural network model to obtain the foundation geological damage degree of the broken pole tower, and forming distribution of the power transmission line tower foundation geological damage condition. According to the method, the differential interference is adopted to obtain the earthquake deformation field map, the geological damage degree of the foundation of the power transmission line tower is obtained, and the rapid delineation or the accurate evaluation of the earthquake damage degree of the geological disaster range of the foundation of the power transmission line tower is realized.
Description
Technical Field
The invention belongs to the technical field of remote sensing application, and particularly relates to a power transmission line geological disaster analysis method and system based on deformation detection.
Background
The earthquake causes a large number of towers to collapse, resulting in huge life and property losses. The number and degree of damage to the tower base after the earthquake are necessary information before disaster area rescue and reconstruction.
At present, manual data reporting and field survey data are generally adopted to determine the fall-loss number of the towers. However, this method relying only on human investigation is inefficient, and has a problem that real data cannot be acquired at the first time.
The high spatial resolution remote sensing technology has the characteristics of reality, objectivity, small influence by ground communication conditions and the like, can acquire a large amount of effective data at the first time, and provides effective support for determining the fall-loss quantity of the tower in time. The tower collapse is determined to have great difficulty based on the image recognition mode, and particularly in mountainous areas, the distribution of the power transmission line towers is dispersed, the types are various, the damage degrees are different, the tower foundation geology cannot be recognized accurately, and the damage degree cannot be evaluated accurately. Therefore, how to overcome the defects of the prior art is a problem which needs to be solved urgently in the technical field of remote sensing application at present.
Disclosure of Invention
The invention aims to solve the defects of the prior art and provides a disaster analysis method and system based on deformation detection, wherein a relation between a deformation field and the geological damage degree of a tower foundation is established, a deformation field graph is obtained by adopting differential interference, the geological damage degree of the tower foundation is further obtained, and the rapid delineation of a disaster range or the accurate evaluation of the earthquake damage degree are realized.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a power transmission line geological disaster analysis method based on deformation detection comprises the following steps:
obtaining tower foundation geological damage degree distribution based on the comparison of a plurality of groups of remote sensing images before and after the earthquake, establishing a deformation field diagram by adopting differential interference, marking the tower foundation geological damage degree distribution, and segmenting according to the range of an offset value and the offset direction to form a sample library;
constructing a plurality of artificial neural network models, and respectively adopting samples with different offset value ranges and offset directions for training and packaging;
the method comprises the steps of obtaining remote sensing images before and after an earthquake to be analyzed, establishing a deformation field through differential interference, carrying out segmentation based on an offset value and an offset direction, selecting an artificial neural network model corresponding to the offset value range and the offset direction to identify the pole tower foundation geological damage degree, and forming pole tower foundation geological damage condition distribution.
Further, it is preferable that the method further includes estimating the seismic intensity based on the obtained damage condition and outputting a seismic intensity distribution.
Further, preferably, the distribution of the degree of geological destruction of the tower foundation comprises the proportion of each damaged type, the SAR remote sensing images before and after the earthquake and the change detection results of the corresponding high-spectroscopic remote sensing images are judged, and qualitative labeling confirmation is carried out through field visual survey;
the damage types comprise collapse damage, no collapse damage, serious damage, slight damage and basic integrity, and respectively correspond to the area change rate of the front image and the back image being more than or equal to 80 percent, the area change rate of 60 percent to less than 80 percent, the area change rate of 40 percent to less than 60 percent, the area change rate of 20 percent to less than 40 percent and the area change rate of less than 20 percent; the region change rate is the root mean square error of the inclination displacement of the sequence images of the two time phases in the same region.
Further, it is preferable that the differential interference establishing the deformation field includes: and obtaining a deformation point bitmap covering the earthquake influence area for quantitative marking, and calculating the oblique displacement information of the tower area, thereby quantitatively reflecting the influence degree of the geological disaster.
Further, preferably, the artificial neural network model includes a convolutional layer, a pooling layer, an activation layer, and a full-link layer; the method comprises the steps of extracting the characteristic of the convolutional layer, extracting deformation value distribution according to a deformation contour line and extracting whether the convolutional layer is positioned in a fault zone or not according to a cross-fault section line; the pooling layer reduces the dimension of the extracted features; the activation layer adopts a Relu function; and outputting a tower foundation geological damage distribution result by the full connection layer.
Further, preferably, the dividing based on the offset value and the offset direction includes: the area is divided into a certain range of offset values and offset directions.
The invention also provides a power transmission line geological disaster analysis system based on deformation detection, which comprises an input module, a segmentation module, a selection module, a plurality of damage degree identification modules and an output module;
the input module is used for acquiring a deformation field;
the segmentation module is used for segmenting the deformation field based on the deviation value and the deviation direction;
the selection module is used for selecting a corresponding damage degree identification module to identify the geological damage degree of the tower foundation based on the offset value and the offset direction;
the output module is used for quantitatively outputting the geological damage degree of the tower foundation and combining corresponding geographic information to form a tower foundation geological damage distribution result.
Further, preferably, the damage degree identification module is constructed and trained by an artificial neural network model; obtaining tower foundation geological damage degree distribution based on comparison of a plurality of groups of remote sensing images before and after an earthquake, obtaining a deformation field graph by adopting differential interference, marking the tower foundation geological damage degree distribution, segmenting according to an offset value range and an offset direction to form a sample library, selecting a training sample from the sample library to train the artificial neural network model to meet error levels, and packaging.
Further, preferably, the artificial neural network model includes a convolutional layer, a pooling layer, an activation layer, and a full-link layer; the method comprises the steps of extracting the characteristic of the convolutional layer, extracting deformation value distribution according to a deformation contour line and extracting whether the convolutional layer is positioned in a fault zone or not according to a cross-fault section line; the pooling layer reduces the dimension of the extracted features; the activation layer adopts Relu function to carry out nonlinear calculation; outputting a tower foundation geological damage distribution result by the full connection layer;
the distribution of the geological damage degree of the tower foundation comprises the proportion of each damage type, wherein the damage types comprise damage collapse, damage non-collapse, serious damage, slight damage and basic intact;
the deformation field includes: and covering a deformation point bitmap of the earthquake influence area, and carrying out quantitative labeling based on the offset of the deformation point bitmap.
Preferably, the Relu function is adopted by the active layer, and the Relu function is utilized to perform nonlinear calculation on the characteristic vectors of the geometry and texture provided by the high-spectroscopic remote sensing image, and the polarization and the tilt displacement provided by the SAR remote sensing image.
The present invention is not limited to the specific division of the offset value and the offset direction, and the division may be performed according to the actually executable calculation amount, for example, if the calculation amount that can be carried is large, the division may be performed more finely. For another example, the offset direction is divided into 8 ranges on average in 360 °, representing 8 directions; the offset is divided into 8 ranges on average from no offset to the maximum offset.
Compared with the prior art, the invention has the beneficial effects that:
(1) according to the method, the disaster area is defined at the first time after the earthquake occurs by utilizing the characteristics that the remote sensing image is wide in coverage, is not influenced by the ground condition and can be obtained at the first time, and the earthquake damage degree is evaluated.
(2) The InSAR technology is a novel geodetic surveying technology, has the advantage of large-area coverage, and can image a disaster area and obtain deformation information of the disaster area. According to the method, the relation between the deformation field and the tower foundation geological damage degree is established, the deformation field is utilized, and based on different fault trends, deformation directions and deformation values on two sides of the fault, corresponding high-spectrum optical remote sensing images are combined to comprehensively form standard uniform regional change rates for evaluating the corresponding tower foundation geological damage degree, so that the evaluation is more accurate.
(3) According to the invention, a plurality of groups of artificial neural network models are established according to different deformation directions and deformation values, and the corresponding artificial neural network models are selected according to the deformation directions and the deformation values, so that the identification accuracy is ensured.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a flow chart of a power transmission line geological disaster analysis method based on deformation detection;
FIG. 2 is a schematic diagram of an artificial neural network model;
FIG. 3 is a schematic diagram of a power transmission line geological disaster analysis system based on deformation detection;
FIG. 4 is a field diagram of surface deformation of a study area
FIG. 5 is a schematic diagram of an image segmentation performed by a surface deformation field map according to different offset values; the circumscribed boundaries with different thicknesses correspond to different deviation values;
FIG. 6 is a schematic diagram of training the input of SAR remote sensing images and high-spectrum optical remote sensing images in a tower area by performing sample labeling based on field verification;
fig. 7 is a result diagram of tower foundation geological damage degree obtained by monitoring and identifying tower changes.
Detailed Description
The present invention will be described in further detail with reference to examples.
It will be appreciated by those skilled in the art that the following examples are illustrative of the invention only and should not be taken as limiting the scope of the invention. The examples do not specify particular techniques or conditions, and are performed according to the techniques or conditions described in the literature in the art or according to the product specifications. The materials or equipment used are not indicated by manufacturers, and all are conventional products available by purchase. Moreover, in the following description, descriptions of well-known structures and techniques are omitted so as to not unnecessarily obscure the concepts of the present invention.
As shown in fig. 1, a power transmission line geological disaster analysis method based on deformation detection includes:
obtaining tower foundation geological damage degree distribution based on the comparison of a plurality of groups of remote sensing images before and after the earthquake, establishing a deformation field diagram by adopting differential interference, marking the tower foundation geological damage degree distribution, and segmenting according to the range of an offset value and the offset direction to form a sample library;
constructing a plurality of artificial neural network models, and respectively adopting samples with different offset value ranges and offset directions for training and packaging;
the method comprises the steps of obtaining remote sensing images before and after an earthquake to be analyzed, establishing a deformation field through differential interference, carrying out segmentation based on an offset value and an offset direction, selecting an artificial neural network model corresponding to the offset value range and the offset direction to identify the pole tower foundation geological damage degree, and forming pole tower foundation geological damage condition distribution.
Preferably, the method further comprises evaluating seismic intensity based on the obtained damage condition and outputting seismic intensity distribution.
Preferably, the distribution of the geological damage degree of the tower foundation comprises the proportion of each damage type, the SAR remote sensing images before and after the earthquake and the change detection results of the corresponding high-spectroscopic remote sensing images are judged, and qualitative labeling confirmation is carried out through field visual investigation;
the damage types comprise collapse damage, no collapse damage, serious damage, slight damage and basic integrity, and respectively correspond to the area change rate of the front image and the back image being more than or equal to 80 percent, the area change rate of 60 percent to less than 80 percent, the area change rate of 40 percent to less than 60 percent, the area change rate of 20 percent to less than 40 percent and the area change rate of less than 20 percent; the region change rate is the root mean square error of the inclination displacement of the sequence images of the two time phases in the same region.
Preferably, the differential interference establishing the deformation field comprises: and obtaining a deformation point bitmap covering the earthquake affected area for quantitative marking, and calculating the oblique displacement information of the tower area.
Preferably, the artificial neural network model comprises a convolutional layer, a pooling layer, an activation layer and a full-connection layer; the method comprises the steps of extracting the characteristic of the convolutional layer, extracting deformation value distribution according to a deformation contour line and extracting whether the convolutional layer is positioned in a fault zone or not according to a cross-fault section line; the pooling layer reduces the dimension of the extracted features; the activation layer adopts a Relu function; and outputting a tower foundation geological damage distribution result by the full connection layer.
Preferably, the dividing based on the offset value and the offset direction includes: the area is divided into a certain range of offset values and offset directions.
As shown in fig. 3, a power transmission line geological disaster analysis system based on deformation detection includes an input module, a segmentation module, a selection module, a plurality of damage degree identification modules, and an output module;
the input module is used for acquiring a deformation field;
the segmentation module is used for segmenting the deformation field based on the deviation value and the deviation direction;
the selection module is used for selecting a corresponding damage degree identification module to identify the geological damage degree of the tower foundation based on the offset value and the offset direction;
the output module is used for quantitatively outputting the geological damage degree of the tower foundation and combining corresponding geographic information to form a tower foundation geological damage distribution result.
Preferably, the damage degree identification module is constructed and trained by an artificial neural network model; obtaining tower foundation geological damage degree distribution based on comparison of a plurality of groups of remote sensing images before and after an earthquake, obtaining a deformation field graph by adopting differential interference, marking the tower foundation geological damage degree distribution, segmenting according to an offset value range and an offset direction to form a sample library, selecting a training sample from the sample library to train the artificial neural network model to meet error levels, and packaging.
Preferably, the artificial neural network model comprises a convolutional layer, a pooling layer, an activation layer and a full-connection layer; the method comprises the steps of extracting the characteristic of the convolutional layer, extracting deformation value distribution according to a deformation contour line and extracting whether the convolutional layer is positioned in a fault zone or not according to a cross-fault section line; the pooling layer reduces the dimension of the extracted features; the activation layer adopts Relu function to carry out nonlinear calculation; outputting a tower foundation geological damage distribution result by the full connection layer;
the distribution of the geological damage degree of the tower foundation comprises the proportion of each damage type, wherein the damage types comprise damage collapse, damage non-collapse, serious damage, slight damage and basic intact;
the deformation field includes: and covering a deformation point bitmap of the earthquake influence area, and carrying out quantitative labeling based on the offset of the deformation point bitmap.
Specifically, the invention provides a power transmission line geological disaster analysis method based on deformation detection, which comprises the following steps in combination with fig. 1:
(1) obtaining the geological damage degree distribution of the tower foundation based on the comparison of a plurality of groups of remote sensing images before and after the earthquake, obtaining the topographic phase difference of the area by adopting differential interference, establishing a deformation field diagram, segmenting according to the range and the migration direction of the deviant, forming a sample database by combining the corresponding high-beam optical remote sensing images, and marking the geological damage degree of the tower foundation.
The method comprises the steps of utilizing an SAR interferometry technology to monitor earth surface deformation and evaluate earthquake damage indexes, obtaining a deformation field through image registration, noise filtering and phase unwrapping processing in the interferometry processing, and extracting the space dynamic change characteristics of the deformation field through quantitative analysis of offset information of each deformation point.
The differential interferometry utilizes two SAR complex images observed in the same area to carry out interference processing, and earth surface elevation information and deformation information are obtained through phase information. InSAR can be divided into single-pass and repeat-pass modes according to imaging time. The single track interference means that two pairs of antennas are loaded on the same airborne or satellite-borne platform, wherein one antenna transmits signals, and the two pairs of antennas receive ground echo signals and perform interference processing by using the acquired data. The repeated orbit interference means that the same sensor or similar sensors image the ground twice according to parallel orbits, and the obtained data is used for interference processing. The spatial distance between the SAR systems at two imaging times is called the spatial baseline distance, and the time interval is called the time baseline. Under the condition of neglecting noise, the atmospheric conditions are basically consistent when imaging is carried out twice, and deformation information of the ground target point can be obtained by removing the flat ground phase and the terrain phase. According to the method, time sequence InSAR images before and after an earthquake are adopted, and deformation monitoring points obtained through processing have three-dimensional position coordinates, deformation directions and speed information. Deformation refers to deformation of the radar Line of Sight (LOS), positive values represent movement toward the radar, negative values represent movement away from the radar, and for convenience of expression, the deformation is respectively expressed as lifting and sinking. Green indicates stability, settlement to the left and elevation to the right of the scale, the darker the color, the greater the deformation. As shown in fig. 4.
And quantitatively evaluating the tower foundation geological damage degree distribution, including the proportion of each damage type. The change detection results of the remote sensing images before and after geology are judged, and the area change rate is used as an index, namely the root mean square error of the inclined displacement and the like of the two images in the same area. The tilt direction is a direction away from the satellite radar or a direction close to the satellite radar. By counting the root mean square error (unit cm,
) The ratio of the rate of change of the area is obtained by dividing by a threshold value (e.g., 10 cm) of the ground. (X, Y) is coordinates after the displacement, (X ', Y') is coordinates before the displacement, and the types of damage include collapse with damage, collapse without damage, severe damage, slight damage, and substantial integrity, with a regional rate of change of 80% or more, 60% or more and less than 80% or less, 40% or less and less than 60% or 20% or less, respectively, corresponding to the preceding and succeeding imagesThe domain change rate is less than 40 percent, and the region change rate is less than 20 percent.
Artificial neural networks require a large number of training samples. According to the method, a geological deformation field map is segmented according to the range of an offset value and the offset direction (the offset direction is far away from the satellite radar or close to the satellite radar, and the offset value is usually in the order of 1 mm), and a sample library is formed by combining corresponding high-spectroscopic remote sensing images. And determining the damage degree of the tower monomer by adopting manual visual interpretation and combining the issued information of the image corresponding to the geological deformation field map, and correcting the damage degree by combining the issued information, such as the images shown in the figures 5 and 6.
Furthermore, in order to ensure the richness of the sample and the comprehensiveness of the coverage, multiple geological deformation field diagrams in different areas can be established and labeled to form a sample library.
(2) And constructing a plurality of artificial neural network models, and respectively adopting samples with different deviation value ranges and deviation directions for training and packaging.
In order to obtain more accurate evaluation results, the deviation value is divided into a plurality of ranges, different sample training is respectively adopted for different deviation value ranges and deviation directions, and the different sample training is used for identifying different deviation value ranges and deviation directions in use.
An artificial neural network model, which is combined with fig. 2, and includes a convolutional layer, a pooling layer, an activation layer, and a full connection layer; the input to the convolutional layer is a geological deformation field map segmented into a single offset and direction. Extracting characteristics including the geometry and texture of a high-spectroscopic remote sensing image and the polarization and oblique displacement characteristics of the SAR remote sensing image, extracting deformation value distribution according to a deformation contour line, extracting whether the SAR remote sensing image is positioned in a fault zone according to a cross-fault section line, and obtaining characteristics 1 to M after extraction; the pooling layer performs dimensionality reduction on the extracted features, as shown in fig. 2, the features 1 to M are reduced to nodes a21 to a2n by a weight W1n, the nodes a21 to a2n are reduced to nodes a31 to a3n by a weight W2n, and the extracted features are reduced to output nodes by a weight W3 n; the activation layer adopts a Relu function to carry out nonlinear calculation on the required characteristic vector; and outputting a tower foundation geological damage distribution result by the full connection layer. And continuously updating W1n, W2n and W3n in the training process to finish classification.
Each artificial neural network model is obtained by training, and the training comprises the following steps:
selecting training samples and verification samples from the corresponding sub-sample library; training in a set round is carried out by adopting a training sample; and (5) verifying by adopting a verification sample, judging whether the error grade is accepted, finishing training if the error grade is accepted, and otherwise, performing the set round of training. The set number of rounds is, for example, 50.
And packaging the trained artificial neural network for field identification.
(3) The method comprises the steps of obtaining remote sensing images before and after an earthquake, obtaining a deformation field through differential interference, carrying out segmentation based on an offset value and an offset direction, selecting a corresponding artificial neural network model to obtain the foundation geological damage degree of the broken pole tower by combining high-spectroscopic remote sensing images, and forming distribution of foundation geological damage conditions of the pole tower.
The segmenting based on the deviant value and the deviant direction comprises the steps of segmenting the area to have a single deviant value and a single deviant direction, further segmenting the area to be recognizable size if the size recognized by the artificial neural network model is exceeded, and amplifying the area to meet the recognition size if the size of the image is too small. And a single deviation value and a deviation direction are input into a corresponding artificial neural network model for identification, the damage degree is marked, and the marking accuracy is ensured. And marking each part to form the distribution of the geological damage condition of the tower foundation.
A plurality of artificial neural networks are adopted for identification, so that on one hand, the accuracy is higher, and on the other hand, the parallel identification efficiency is higher.
The invention provides a power transmission line geological disaster analysis system based on deformation detection, which is combined with the graph 3 and comprises an input module, a segmentation module, a selection module, a plurality of damage degree identification modules and an output module;
the input module acquires a geological deformation field;
the segmentation module segments the geological deformation field based on the offset value and the offset direction;
the selection module selects a corresponding damage degree identification module to identify the geological damage degree of the tower foundation based on the offset value and the offset direction;
the damage degree identification module is constructed and trained by an artificial neural network model; obtaining tower foundation geological damage degree distribution based on comparison of a plurality of groups of remote sensing images before and after an earthquake, obtaining a deformation field graph by adopting differential interference, marking the tower foundation geological damage degree distribution, segmenting according to an offset value range and an offset direction to form a sample library, selecting a training sample from the sample library to train the artificial neural network model to meet error levels, and packaging. The method comprises the steps of segmenting an image based on two indexes of an offset value and an offset direction, correspondingly inputting SAR remote sensing images with different damage degree labels and high-spectrum optical remote sensing images into a neural network as training samples, and outputting the training samples to obtain a tower foundation geological damage distribution diagram through the network, wherein the diagram is shown in fig. 7.
And the output module corresponds the geological damage degree of the tower foundation to the geographical position to form the geological damage condition distribution of the tower foundation.
1) The geological deformation field map, namely a ground surface deformation field distribution map obtained by a D-InSAR technology, is obtained by preprocessing two complex SAR images before and after geology through orbit parameter correction and pre-filtering, generating and processing image fine registration, interferogram and coherence map generation, post-filtering and phase unwrapping, calculating coherent phase to geographic information, mainly comprising phase to elevation conversion, geographic coding and DEM construction, and finally obtaining ground surface deformation information.
2) The tower-pole foundation geological destruction degree distribution map is obtained by analyzing related geological data in a geological region collected in the early stage, or by drawing an earth surface deformation map obtained in the early stage manually according to geographical coordinates, and the geological deformation field map and the tower-pole foundation geological destruction degree distribution map are input into a network according to the geographical position relationship for feature extraction.
3) The fault surface position and trend change data can be obtained from a local geological department by obtaining relevant vector or grid data, and can also be obtained by vectorizing a local geological map set according to actual needs, and a geographic coordinate system used in vectorizing the map layer is the same as a geological deformation field distribution map in the previous stage so as to ensure that superposition analysis processing can be carried out.
4) The deviation value needs to be calculated according to the surface deformation map acquired in the previous stage. The earth surface deformation value obtained by direct D-InSAR settlement is the deformation observed from the distance direction, the deformation quantity in a certain direction of the vertical deformation and the plane, namely the deviation value can be calculated according to the relation with the satellite incident angle, and the obtained deviation direction is vertical to the satellite flight direction; and different orbit data can be used, three-dimensional deformation calculation is carried out by utilizing two observation angles, and offset in the north-south direction and the east-west direction can be obtained. Because the monitoring precision can reach millimeter level, the deviant scope can be according to the offset that the magnitude arouses and carry out the interval partition, with deviant scope and magnitude correlation in order to quantitative evaluation tower pole damage degree.
5) The model structure comprises a convolution layer, a pooling layer, an activation layer and a full connection layer, which are described in the specification. Different initial weight values can be respectively given to the offset, the offset direction, the tower pole foundation geological damage degree and the fault trend when training is started, the initial weight values can be set to be 0.4, 0.1, 0.4 and 0.1, and the specific values are set according to the geological condition of the earthquake launching range.
In summary, the invention provides a power transmission line geological disaster analysis method and system based on deformation detection, wherein pole tower foundation geological damage degree distribution is obtained based on comparison of a plurality of groups of remote sensing images before and after an earthquake, a deformation field diagram is established and marked by adopting differential interference, segmentation is carried out according to an offset value range and an offset direction to form a sample library, and a plurality of artificial neural network models are trained; the method comprises the steps of obtaining remote sensing images before and after an earthquake, obtaining a deformation field through differential interference, carrying out segmentation based on an offset value and an offset direction, selecting a corresponding artificial neural network model to obtain the foundation geological damage degree of the pole breaking tower, and forming distribution of foundation geological damage conditions of the pole breaking tower. According to the method, the degree of geological damage of the tower foundation is obtained by establishing the deformation field diagram through differential interference, and the rapid delineation of the disaster range or the accurate evaluation of the earthquake damage degree are realized.
The foregoing shows and describes the general principles, essential features, and advantages of the invention. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (9)
1. A power transmission line geological disaster analysis method based on deformation detection is characterized by comprising the following steps:
obtaining tower foundation geological damage degree distribution based on the comparison of a plurality of groups of remote sensing images before and after the earthquake, establishing a deformation field diagram by adopting differential interference, marking the tower foundation geological damage degree distribution, and segmenting according to the range of an offset value and the offset direction to form a sample library;
constructing a plurality of artificial neural network models, and respectively adopting samples with different offset value ranges and offset directions for training and packaging;
the method comprises the steps of obtaining remote sensing images before and after an earthquake to be analyzed, establishing a deformation field through differential interference, carrying out segmentation based on an offset value and an offset direction, selecting an artificial neural network model corresponding to the offset value range and the offset direction to identify the pole tower foundation geological damage degree, and forming pole tower foundation geological damage condition distribution.
2. The power transmission line geological disaster analysis method based on deformation detection as claimed in claim 1, further comprising evaluating seismic intensity based on the obtained damage condition and outputting seismic intensity distribution.
3. The power transmission line geological disaster analysis method based on deformation detection according to claim 1 or 2, characterized in that the tower foundation geological damage degree distribution comprises the percentage of each damaged type, the changes of SAR remote sensing images before and after earthquake and corresponding high-spectrum optical remote sensing images are judged according to the detection results, and qualitative labeling confirmation is carried out through field visual investigation;
the damage types comprise collapse damage, no collapse damage, serious damage, slight damage and basic integrity, and respectively correspond to the area change rate of the front image and the back image being more than or equal to 80 percent, the area change rate of 60 percent to less than 80 percent, the area change rate of 40 percent to less than 60 percent, the area change rate of 20 percent to less than 40 percent and the area change rate of less than 20 percent; the region change rate is the root mean square error of the inclination displacement of the sequence images of the two time phases in the same region.
4. The power transmission line geological disaster analysis method based on deformation detection according to claim 1 or 2, wherein the establishing of the deformation field by differential interference comprises: and obtaining a deformation point bitmap covering the earthquake affected area for quantitative marking, and calculating the oblique displacement information of the tower area.
5. The power transmission line geological disaster analysis method based on deformation detection as claimed in claim 4, wherein the artificial neural network model comprises a convolution layer, a pooling layer, an activation layer and a full-connection layer; the method comprises the steps of extracting the characteristic of the convolutional layer, extracting deformation value distribution according to a deformation contour line and extracting whether the convolutional layer is positioned in a fault zone or not according to a cross-fault section line; the pooling layer reduces the dimension of the extracted features; the activation layer adopts a Relu function; and outputting a tower foundation geological damage distribution result by the full connection layer.
6. The power transmission line geological disaster analysis method based on deformation detection as claimed in claim 1, wherein the segmentation based on the offset value and the offset direction comprises: the area is divided into a certain range of offset values and offset directions.
7. A power transmission line geological disaster analysis system based on deformation detection is characterized by comprising an input module, a segmentation module, a selection module, a plurality of damage degree identification modules and an output module;
the input module is used for acquiring a deformation field;
the segmentation module is used for segmenting the deformation field based on the deviation value and the deviation direction;
the selection module is used for selecting a corresponding damage degree identification module to identify the geological damage degree of the tower foundation based on the offset value and the offset direction;
the output module is used for quantitatively outputting the geological damage degree of the tower foundation and combining corresponding geographic information to form a tower foundation geological damage distribution result.
8. The power transmission line geological disaster analysis system based on deformation detection as claimed in claim 7, wherein the damage degree recognition module is constructed and trained by an artificial neural network model; obtaining tower foundation geological damage degree distribution based on comparison of a plurality of groups of remote sensing images before and after an earthquake, obtaining a deformation field graph by adopting differential interference, marking the tower foundation geological damage degree distribution, segmenting according to an offset value range and an offset direction to form a sample library, selecting a training sample from the sample library to train the artificial neural network model to meet error levels, and packaging.
9. The power transmission line geological disaster analysis system based on deformation detection as claimed in claim 8, wherein the artificial neural network model comprises a convolutional layer, a pooling layer, an activation layer and a full-link layer; the method comprises the steps of extracting the characteristic of the convolutional layer, extracting deformation value distribution according to a deformation contour line and extracting whether the convolutional layer is positioned in a fault zone or not according to a cross-fault section line; the pooling layer reduces the dimension of the extracted features; the activation layer adopts Relu function to carry out nonlinear calculation; outputting a tower foundation geological damage distribution result by the full connection layer;
the distribution of the geological damage degree of the tower foundation comprises the proportion of each damage type, wherein the damage types comprise damage collapse, damage non-collapse, serious damage, slight damage and basic intact;
the deformation field includes: and covering a deformation point bitmap of the earthquake influence area, and carrying out quantitative labeling based on the offset of the deformation point bitmap.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111340434.2A CN114239379A (en) | 2021-11-12 | 2021-11-12 | Transmission line geological disaster analysis method and system based on deformation detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111340434.2A CN114239379A (en) | 2021-11-12 | 2021-11-12 | Transmission line geological disaster analysis method and system based on deformation detection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114239379A true CN114239379A (en) | 2022-03-25 |
Family
ID=80749424
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111340434.2A Pending CN114239379A (en) | 2021-11-12 | 2021-11-12 | Transmission line geological disaster analysis method and system based on deformation detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114239379A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117612339A (en) * | 2023-11-09 | 2024-02-27 | 应急管理部大数据中心 | Geological disaster monitoring method and system based on iron tower big data |
-
2021
- 2021-11-12 CN CN202111340434.2A patent/CN114239379A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117612339A (en) * | 2023-11-09 | 2024-02-27 | 应急管理部大数据中心 | Geological disaster monitoring method and system based on iron tower big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111142119B (en) | Mine geological disaster dynamic identification and monitoring method based on multi-source remote sensing data | |
| Jaboyedoff et al. | New insight techniques to analyze rock-slope relief using DEM and 3Dimaging cloud points: COLTOP-3D software | |
| Liu et al. | A complete high-resolution coastline of Antarctica extracted from orthorectified Radarsat SAR imagery | |
| CN109165622B (en) | InSAR technology-based method for determining effective region for early dynamic identification and monitoring of regional landslide | |
| CN110132237A (en) | A kind of method of urban ground deformation disaster EARLY RECOGNITION | |
| CN107479065B (en) | Forest gap three-dimensional structure measuring method based on laser radar | |
| CN109100719B (en) | Terrain map joint mapping method based on satellite-borne SAR (synthetic aperture radar) image and optical image | |
| Xu et al. | Remote sensing for landslide investigations: A progress report from China | |
| CN113866764A (en) | Landslide susceptibility improvement evaluation method based on InSAR and LR-IOE model | |
| Xie et al. | A comparison and review of surface detection methods using MBL, MABEL, and ICESat-2 photon-counting laser altimetry data | |
| CN112444188B (en) | Multi-view InSAR sea wall high-precision three-dimensional deformation measurement method | |
| Li et al. | An iterative-mode scan design of terrestrial laser scanning in forests for minimizing occlusion effects | |
| CN115877421A (en) | Deformation detection method and device for geological sensitive area of power transmission channel | |
| Scott et al. | Measuring change at Earth’s surface: On-demand vertical and three-dimensional topographic differencing implemented in OpenTopography | |
| CN114239379A (en) | Transmission line geological disaster analysis method and system based on deformation detection | |
| Hao et al. | Extraction and analysis of tree canopy height information in high-voltage transmission-line corridors by using integrated optical remote sensing and LiDAR | |
| Razi et al. | Multi-temporal land deformation monitoring in V shape area using quasi-persistent scatterer (Q-PS) interferometry technique | |
| CN111368716A (en) | Geological disaster catastrophe farmland extraction method based on multi-source time-space data | |
| Sui et al. | Processing of multitemporal data and change detection | |
| Ajayi et al. | Modelling 3D Topography by comparing airborne LiDAR data with Unmanned Aerial System (UAS) photogrammetry under multiple imaging conditions | |
| Jakopec et al. | A novel approach to landslide monitoring based on unmanned aerial system photogrammetry | |
| CN113625241A (en) | Differential settlement monitoring and early warning method | |
| CN113344866A (en) | Point cloud comprehensive precision evaluation method | |
| Wang et al. | Deformation monitoring and evaluation of mountain slope stability combined with ground-based radar and spaceborne InSAR methods | |
| CN117541679B (en) | Forest canopy height mapping method and system based on sample point individual representativeness |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |