CN115423180A - Surface deformation monitoring and predicting method - Google Patents
Surface deformation monitoring and predicting method Download PDFInfo
- Publication number
- CN115423180A CN115423180A CN202211056420.2A CN202211056420A CN115423180A CN 115423180 A CN115423180 A CN 115423180A CN 202211056420 A CN202211056420 A CN 202211056420A CN 115423180 A CN115423180 A CN 115423180A
- Authority
- CN
- China
- Prior art keywords
- data set
- surface deformation
- model
- environment factor
- deformation
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B15/00—Measuring arrangements characterised by the use of electromagnetic waves or particle radiation, e.g. by the use of microwaves, X-rays, gamma rays or electrons
- G01B15/06—Measuring arrangements characterised by the use of electromagnetic waves or particle radiation, e.g. by the use of microwaves, X-rays, gamma rays or electrons for measuring the deformation in a solid
-
- 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
- G01S13/885—Radar or analogous systems specially adapted for specific applications for ground probing
-
- 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
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/05—Geographic models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
- G06T17/205—Re-meshing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Remote Sensing (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Electromagnetism (AREA)
- Geometry (AREA)
- Data Mining & Analysis (AREA)
- Business, Economics & Management (AREA)
- Operations Research (AREA)
- Computational Mathematics (AREA)
- Computer Graphics (AREA)
- Economics (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Computer Networks & Wireless Communication (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Computational Biology (AREA)
- Algebra (AREA)
- Marketing (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- General Engineering & Computer Science (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Geophysics And Detection Of Objects (AREA)
Abstract
The invention provides a method for monitoring and predicting surface deformation, which comprises the following steps: the method comprises the steps of obtaining an InSAR time sequence earth surface deformation data set of a research area, processing the InSAR time sequence earth surface deformation data set, obtaining an environment factor data set of the research area, processing the environment factor data set, merging the processed environment factor data set and the processed InSAR time sequence earth surface deformation data set to obtain a model data set, building an auto-former model, training the auto-former model through the model data set, adjusting model hyper-parameters, obtaining an optimal model, selecting a research point set with the largest surface deformation change in the research area, and inputting the optimal model to perform prediction and early warning. The method for monitoring and predicting the surface deformation provided by the invention does not need ground and manual cooperation, can fully play the advantages of satellite monitoring, can monitor the surface deformation in the whole range of a research area, can realize prediction, and can reduce the accident occurrence rate and the property loss rate.
Description
Technical Field
The invention relates to the technical field of surface deformation monitoring, in particular to a surface deformation monitoring and predicting method.
Background
In the monitoring means of surface deformation, traditional geodetic techniques, such as GPS and manual level gauge, are easily limited by low spatial resolution and high cost, and in addition, in the inspection of ground objects, the traditional techniques often need a large amount of professionals and heavy equipment, and the monitoring time is long and the monitoring cost is high.
In the prior art, a multi-time phase interference synthetic aperture radar (MT-InSAR) technology is used for realizing large-scale, high-precision and low-cost monitoring, wherein one of typical MT-InSAR technologies is a permanent scattering radar interference synthetic aperture radar (PS-InSAR), in the method, time stable points such as buildings and bare rocks are selected and have high stability and reflectivity in a plurality of SAR images, but the PS-InSAR is used in the prior art and is not matched with a long-term time sequence prediction processing analysis frame, the prediction of intensive ground object surface subsidence cannot be realized, and environmental factors influencing surface deformation are not considered. Therefore, it is necessary to design a surface deformation monitoring and predicting method based on an auto former and a PS-Insar.
Disclosure of Invention
The invention aims to provide a method for monitoring and predicting the surface deformation, which does not need the ground and manual cooperation, can fully exert the advantages of satellite monitoring, can monitor the surface deformation in the whole range of a research area, can realize prediction and can reduce the accident rate and the property loss rate.
In order to achieve the purpose, the invention provides the following scheme:
a method for monitoring and predicting surface deformation comprises the following steps:
step 1: acquiring an InSAR time sequence earth surface deformation data set of a research area;
step 2: processing the obtained InSAR time sequence earth surface deformation data set;
and step 3: acquiring an environmental factor data set of a research area;
and 4, step 4: processing the environmental factor data set, and merging the processed environmental factor data set and the processed InSAR time sequence earth surface deformation data set to obtain a model data set;
and 5: building an auto-former model, training the auto-former model through a model data set, adjusting the hyper-parameters of the model, and obtaining an optimal model;
step 6: and selecting a research point set with the largest surface deformation change in the research area, and inputting the research point set into the optimal model for prediction and early warning.
Optionally, in step 1, the acquiring an InSAR time sequence earth surface deformation data set of the research area specifically includes:
acquiring N Sentinel-1A SAR images covering a research area, selecting one SAR image as a main image and the rest as auxiliary images according to space-time baseline parameters and Doppler center frequency, carrying out accurate registration and interference treatment on the main image and the auxiliary images to generate N-1 differential interferograms, acquiring external DEM data and satellite orbit data, registering the external DEM data and an image coverage range, carrying out orbit offset vector correction through the satellite orbit data to remove a plane effect, sampling the DEM data into a coordinate system of the SAR images, identifying PS points through an amplitude deviation method, setting a reasonable amplitude deviation threshold, extracting PS candidate points of the research area, and constructing a pixel point network, wherein the amplitude deviation method determines the PS points through evaluating the signal-to-noise ratio of image pixels, and the phase can use the amplitude deviation index to determine the signal-to-noise ratio, namely:
in the formula, σ A And mu A D is set for the standard deviation and average value of the amplitude of the SAR image respectively A When the amplitude deviation index of the position is smaller than x, the position is judged to be a PS point, otherwise, the position is judged not to be the PS point, and a deformation inversion model is established, wherein the threshold is as follows:
ψ x,i ＝W{φ D,x,i +φ A,x,i +Δφ S,x,i +Δφ θ,x,i +Δφ N,x,i }
wherein W is a phase winding operator, phi D,x,i For the phase induced by the ground phase, phi A,x,i For atmospheric delay error phase, Δ φ S,x,i For satellite orbital error phase, Δ φ θ,x,i For residual topographic phase, Δ φ N,x,i Establishing a deformation inversion model for registration errors, namely noise phases caused by thermal noise decorrelation, performing adjacent PS point phase parameter estimation, resolving residual elevation, linear deformation and atmospheric phase parameters, testing the fitting condition of interference phases of each parameter and pixel points resolved by the model, proving the accuracy of model solution values, removing the atmospheric phase, reselecting PS candidate points, performing deformation analysis again, finally decomposing residual errors to obtain nonlinear deformation information, obtaining a time sequence earth surface deformation data set, and obtaining an InSAR time sequence earth surface deformation data set after geocoding.
Optionally, in step 2, the obtained InSAR time sequence earth surface deformation data set is processed, specifically:
and cutting and smoothing the obtained InSAR time sequence earth surface deformation data set according to a research area, obtaining the length of the time sequence data set, carrying out file format conversion, confirming key points of the research area, namely a point set with the maximum deformation degree, and obtaining the processed InSAR time sequence earth surface deformation data set.
Optionally, in step 3, the obtaining of the environmental factor data set of the research area specifically includes:
and acquiring an atmospheric driving data set, an earth surface temperature analysis data set, a soil humidity data set, a soil temperature analysis data set and a soil relative humidity data set of the NetCDF format data to form an environment factor data set.
Optionally, in step 4, the environmental factor data set is processed, and the processed environmental factor data set and the processed inssar time series earth surface deformation data set are merged to obtain a model data set, which specifically includes:
converting an environment factor data set of NetCDF format data into an environment factor data set in a csv format through a Pandas library and a Numpy library based on Python language, extracting environment factor parameters in the environment factor data set, dividing the environment factor data set according to latitude and longitude grids of a research area, standardizing the data format of the extracted environment factor parameters, examining the data format of the environment factor data set after standardization is completed, removing data noise which does not conform to a conventional format, supplementing abnormal values to the environment factor data set after the removal is completed, supplementing null values of the environment factor data set by using average values of the environment factor parameters to obtain a processed environment factor data set, calculating the distance between each research point in the research area and an environment factor meteorological station for collecting the environment factor parameters, selecting the environment factor parameter collected from the nearest environment factor meteorological station as a reference by each research point, clustering each research point, combining the environment factor data set and an InEarth's surface deformation time sequence data set to obtain a model data set.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects: the invention provides a surface deformation monitoring and predicting method, which comprises the steps of obtaining an InSAR time sequence surface deformation data set of a research area, processing the obtained InSAR time sequence surface deformation data set, obtaining an environmental factor data set of the research area, processing the environmental factor data set, combining the processed environmental factor data set and the processed InSAR time sequence surface deformation data set to obtain a model data set, building an auto-former model, training the auto-former model through the model data set, adjusting the super-parameters of the model, obtaining an optimal model, selecting a research point set with the largest surface deformation change in the research area, and inputting the optimal model to perform prediction and early warning; according to the method, the Autoformer and the PS-Insar are combined, environmental factors influencing the deformation of the earth surface are fully considered, the deformation of the earth surface can be monitored in a large range and with high precision, the economic investment of monitoring the deformation of the earth surface is effectively reduced, a new idea is provided for preventing ground surface settlement disasters, the deformation of the earth surface is monitored by using the PS-Insar, millimeter-level wide-area large-range monitoring can be achieved, the precision is high, and the precision of a prediction model of the Autoformer is far higher than that of a same-class prediction model.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments 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 it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic flow chart of a method for monitoring and predicting surface deformation according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an InSAR time series earth surface deformation data set acquisition process.
FIG. 3 is a comparison graph of evaluation indexes of respective models;
FIG. 4 is a comparison chart of assessment indexes of the Autoformer experimental model at each research point.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a method for monitoring and predicting the surface deformation, which does not need the ground and manual cooperation, can fully play the advantages of satellite monitoring, can monitor the surface deformation in the whole range of a research area, can realize prediction and can reduce the accident rate and the property loss rate.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
As shown in fig. 1, the method for monitoring and predicting surface deformation provided by the embodiment of the present invention includes the following steps:
step 1: acquiring an InSAR time sequence earth surface deformation data set of a research area;
step 2: processing the obtained InSAR time sequence earth surface deformation data set;
and step 3: acquiring an environmental factor data set of a research area;
and 4, step 4: processing the environmental factor data set, and merging the processed environmental factor data set and the processed InSAR time sequence earth surface deformation data set to obtain a model data set;
and 5: building an auto-former model, training the auto-former model through a model data set, adjusting model hyper-parameters, and obtaining an optimal model;
step 6: selecting a research point set with the largest surface deformation change in a research area, and inputting an optimal model for prediction and early warning;
the earth surface deformation can be monitored according to the model data set, and can be predicted through the optimal model.
In step 1, an InSAR time sequence earth surface deformation data set of a research area is obtained, which specifically comprises the following steps:
acquiring N Sentinel-1A SAR images covering a research area, selecting one SAR image as a main image and the rest as auxiliary images according to a space-time baseline parameter and Doppler center frequency, carrying out accurate registration and interference treatment on the main image and the auxiliary images to generate N-1 differential interferograms, wherein the phase difference between adjacent PS is acquired through the differential interferograms, establishing a phase difference model solution of the adjacent PS, obtaining surface subsidence information by solving model parameters, acquiring external DEM data and satellite orbit data, carrying out registration by using the external DEM data and an image coverage range, carrying out orbit offset vector correction through the satellite orbit data to remove a plane effect, sampling the DEM data into a coordinate system of the SAR images, identifying PS points through an amplitude dispersion method, setting a reasonable amplitude dispersion threshold, extracting PS candidate points of the research area, and constructing a pixel point network, wherein the amplitude dispersion method confirms the PS points through evaluating the signal-to-noise ratio of pixel images, and the signal-to-noise ratio of the phase can be measured by a dispersion index, namely:
in the formula, σ A And mu A D is set for the standard deviation and average value of the amplitude of the SAR image respectively A The threshold value of (2) is x, when the amplitude deviation index of the position is less than x, the position is judged to be a PS point, otherwise, the position is not the PS point, a deformation inversion model is established through a phase difference, an atmospheric delay error, a satellite orbit error phase, a residual terrain phase, a registration error and a phase unwrapping operator caused by a ground phase, and parameters and model parameters of adjacent PS point phases are estimated through the deformation inversion model, wherein the deformation inversion model is as follows:
ψ x,i ＝W{φ D,x,i +φ A,x,i +Δφ S,x,i +Δφ θ,x,i +Δφ N,x,i }
wherein W is the phase winding operator, phi D,x,i For the phase induced by the ground phase, phi A,x,i Is the atmospheric delay error phase, Δ φ S,x,i For satellite orbital error phase, Δ φ θ,x,i For residual topographic phase, Δ φ N,x,i Establishing a deformation inversion model for registration errors, namely noise phases caused by thermal noise decorrelation, estimating adjacent PS point phase parameters, resolving residual elevation, linear deformation and atmospheric phase parameters, testing the fitting condition of interference phases of each parameter and pixel points resolved by the model, proving the accuracy of model solution values, after removing the atmospheric phase, re-selecting PS candidate points, performing deformation analysis again, and finally decomposing residual errors to obtain nonlinear deformation informationAnd acquiring a time series earth surface deformation data set, and obtaining the InSAR time series earth surface deformation data set after geocoding, wherein a specific flow chart is shown in FIG. 2.
In step 2, the obtained InSAR time sequence earth surface deformation data set is processed, specifically:
and cutting and smoothing the obtained InSAR time sequence earth surface deformation data set according to a research area, obtaining the length of the time sequence data set, carrying out file format conversion, confirming key points of the research area, namely a point set with the maximum deformation degree, and obtaining the processed InSAR time sequence earth surface deformation data set.
In step 3, an environmental factor data set of the research area is obtained, specifically:
acquiring a NetCDF format data product Atmospheric Drive Data Set (ADDS), an earth Surface Temperature Analysis Data Set (STADS), a soil humidity data set (SMDS), a Soil Temperature Analysis Data Set (STADS) and a Soil Relative Humidity Data Set (SRHDS) provided by a Chinese meteorological network to form an environmental factor data set. Through comparing the root mean square error, deviation and correlation coefficient of each parameter of the data set, the environmental factor data set is confirmed to have high space-time resolution and quality superior to international data of the same kind, and the specific data quality parameters are shown in table 1:
table 1 environmental factor data quality description
In step 4, the environmental factor data set is processed, and the processed environmental factor data set and the processed InSAR time series earth surface deformation data set are merged to obtain a model data set, which specifically comprises the following steps:
in order to construct an accurate and effective environment factor data set, the method carries out data preprocessing on an original environment factor data set, and specifically comprises data format conversion, data rationality processing, data denoising and data abnormal value supplement;
wherein, the data format is converted into: converting the environmental factor data set of NetCDF format data into an environmental factor data set of a csv format through a Python language-based Pandas library and a Numpy library, and extracting environmental factor parameters in the environmental factor data set;
the data rationality treatment is: the environmental factor data set is divided according to longitude and latitude grids of a research area, and data format standardization is carried out on the extracted environmental factor parameters, the environmental factor data used by the method is 1-day resolution monitoring data, in order to solve the contradiction between slow change of surface deformation and too high environmental factor monitoring frequency, the environmental factor data are subjected to averaging processing, daily average environmental factor parameters are obtained, data redundancy is removed, and data reasonability is improved;
the data denoising is as follows: after the standardization is finished, carrying out data format examination on the environment factor data set, and removing data noise which does not conform to the conventional format;
data outliers were supplemented as: after the data noise is removed, supplementing abnormal values to the environmental factor data set, and supplementing null values of the environmental factor data set by using average values of all environmental factor parameters to obtain a processed environmental factor data set;
and calculating the distance between each research point in the research area and the environmental factor meteorological station for acquiring the environmental factor parameters, selecting the environmental factor parameters acquired by the environmental factor meteorological station closest to each research point as reference for each research point, clustering each research point, and combining the environmental factor data set and the InSAR surface deformation time sequence data set to obtain a model data set.
The method also carries out precision comparative analysis on the Autoformer, the Transformer (an encoder-decoder model based on an attention mechanism), the Informer (a long-term sequence prediction model) and the Reformer (an efficient encoder-decoder model), and finally confirms the effectiveness and the accuracy of the Autoformer on the research problem;
the method uses five precision indexes of a square root error (MSE), a Mean Absolute Error (MAE), a Root Mean Square Error (RMSE), a Mean Absolute Percentage Error (MAPE) and a Mean Square Percentage Error (MSPE) to evaluate the precision of the model, and the specific formula is as follows:
wherein, y t A monitored value representing the deformation of the earth's surface at time t,
and the predicted value of the surface deformation at the time t is shown, N is the size of the test set, a comparison graph of evaluation indexes of each model is shown in figure 3, and a comparison graph of evaluation indexes of the auto-former experimental model of each research point is shown in figure 4.
The auto-former is a deep learning network model based on a deep decomposition framework and an autocorrelation mechanism, aiming at the problems that a complex time mode in long-term sequence prediction is difficult to process and high in operation efficiency, the long-term sequence prediction efficiency is greatly improved through progressive decomposition and sequence level connection, the auto-former network structure comprises an internal series decomposition block, the autocorrelation mechanism and a corresponding decoder and an encoder, wherein the encoder eliminates a long-term trend period part through the sequence decomposition block, and focuses on seasonal mode modeling, the decoder gradually accumulates trend parts extracted from hidden variables, and the encoder and the decoder utilize past seasonal information in the encoder to perform autocorrelation.
According to the method, the PS-InSAR technology is adopted to monitor the deformation of the earth surface, millimeter-scale monitoring and prediction are carried out on the deformation of the earth surface, the deformation of the earth surface in a research area can be accurately monitored, and the property loss rate is reduced;
the method realizes the method flow of data processing, deformation monitoring, key point early warning and key point prediction, can be applied to various ranges with dense earth surface and ground objects, such as monitoring, routing inspection and prediction of urban infrastructure, and when in use, a user can determine the prediction duration according to the own requirements, considers the influence of environmental factors on the deformation of the earth surface, and has more accurate and convincing prediction results.
The invention provides a surface deformation monitoring and predicting method, which comprises the steps of obtaining an InSAR time sequence surface deformation data set of a research area, processing the obtained InSAR time sequence surface deformation data set, obtaining an environmental factor data set of the research area, processing the environmental factor data set, combining the processed environmental factor data set and the processed InSAR time sequence surface deformation data set to obtain a model data set, building an auto-former model, training the auto-former model through the model data set, adjusting the super-parameters of the model, obtaining an optimal model, selecting a research point set with the largest surface deformation change in the research area, and inputting the optimal model to perform prediction and early warning; according to the method, the Autoformer and the PS-Insar are combined, environmental factors influencing the deformation of the earth surface are fully considered, the deformation of the earth surface can be monitored in a large range and with high precision, the economic investment of monitoring the deformation of the earth surface is effectively reduced, a new idea is provided for preventing ground surface settlement disasters, the deformation of the earth surface is monitored by using the PS-Insar, millimeter-level wide-area large-range monitoring can be achieved, the precision is high, and the precision of a prediction model of the Autoformer is far higher than that of a same-class prediction model.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (5)
1. A method for monitoring and predicting surface deformation is characterized by comprising the following steps:
step 1: acquiring an InSAR time sequence earth surface deformation data set of a research area;
step 2: processing the obtained InSAR time sequence earth surface deformation data set;
and step 3: acquiring an environmental factor data set of a research area;
and 4, step 4: processing the environmental factor data set, and merging the processed environmental factor data set and the processed InSAR time sequence earth surface deformation data set to obtain a model data set;
and 5: building an auto-former model, training the auto-former model through a model data set, adjusting model hyper-parameters, and obtaining an optimal model;
step 6: and selecting a research point set with the largest surface deformation change in the research area, and inputting the optimal model for prediction and early warning.
2. The surface deformation monitoring and predicting method according to claim 1, wherein in step 1, the InSAR time series surface deformation data set of the research area is obtained, specifically:
acquiring N Sentinel-1A SAR images covering a research area, selecting one SAR image as a main image and the rest as auxiliary images according to space-time baseline parameters and Doppler center frequency, carrying out accurate registration and interference treatment on the main image and the auxiliary images to generate N-1 differential interferograms, acquiring external DEM data and satellite orbit data, registering the external DEM data and an image coverage range, carrying out orbit offset vector correction through the satellite orbit data to remove a plane effect, sampling the DEM data into a coordinate system of the SAR images, identifying PS points through an amplitude deviation method, setting a reasonable amplitude deviation threshold, extracting PS candidate points of the research area, and constructing a pixel point network, wherein the amplitude deviation method determines the PS points through evaluating the signal-to-noise ratio of image pixels, and the phase can use the amplitude deviation index to determine the signal-to-noise ratio, namely:
in the formula, σ A And mu A D is respectively set for the standard deviation and the average value of the amplitude of the SAR image A When the amplitude deviation index of the position is smaller than x, the position is judged to be a PS point, otherwise, the position is judged not to be the PS point, and a deformation inversion model is established, wherein the threshold is as follows:
ψ x,i ＝W{φ D,x,i +φ A,x,i +Δφ S,x,i +Δφ θ,x,i +Δφ N,x,i }
wherein W is the phase winding operator, phi D,x,i For the phase induced by the ground phase, phi A,x,i For atmospheric delay error phase, Δ φ S,x,i For satellite orbital error phase, Δ φ θ,x,i For residual topographic phase, Δ φ N,x,i Establishing a deformation inversion model for registration errors, namely noise phases caused by thermal noise decorrelation, performing phase parameter estimation on adjacent PS points, resolving residual elevation, linear deformation and atmospheric phase parameters, testing the fitting condition of interference phases of all parameters resolved by the model and pixel points, proving the accuracy of model resolving values, removing the atmospheric phases, reselecting PS candidate points, performing deformation analysis again, finally decomposing residual errors to obtain nonlinear deformation information, obtaining a time sequence earth surface deformation data set, and obtaining an InSAR time sequence earth surface deformation data set after geocoding.
3. The surface deformation monitoring and predicting method according to claim 2, wherein in step 2, the acquired InSAR time series surface deformation data set is processed, specifically:
and cutting and smoothing the obtained InSAR time sequence earth surface deformation data set according to a research area, obtaining the length of the time sequence data set, carrying out file format conversion, confirming key points of the research area, namely a point set with the maximum deformation degree, and obtaining the processed InSAR time sequence earth surface deformation data set.
4. A method for monitoring and predicting surface deformation according to claim 3, wherein in step 3, the data set of environmental factors of the study area is obtained, specifically:
and acquiring an atmospheric driving data set, an earth surface temperature analysis data set, a soil humidity data set, a soil temperature analysis data set and a soil relative humidity data set of the NetCDF format data to form an environment factor data set.
5. The method for monitoring and predicting surface deformation according to claim 4, wherein in step 4, the environmental factor data set is processed, and the processed environmental factor data set and the processed InSAR time series surface deformation data set are merged to obtain a model data set, which specifically comprises:
converting an environment factor data set of NetCDF format data into an environment factor data set in a csv format through a Pandas library and a Numpy library based on Python language, extracting environment factor parameters in the environment factor data set, dividing the environment factor data set according to latitude and longitude grids of a research area, standardizing the data format of the extracted environment factor parameters, examining the data format of the environment factor data set after standardization is completed, removing data noise which does not conform to a conventional format, supplementing abnormal values to the environment factor data set after the removal is completed, supplementing null values of the environment factor data set by using average values of the environment factor parameters to obtain a processed environment factor data set, calculating the distance between each research point in the research area and an environment factor meteorological station for collecting the environment factor parameters, selecting the environment factor parameter collected from the nearest environment factor meteorological station as a reference by each research point, clustering each research point, combining the environment factor data set and an InEarth's surface deformation time sequence data set to obtain a model data set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211056420.2A CN115423180A (en) | 2022-08-31 | 2022-08-31 | Surface deformation monitoring and predicting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211056420.2A CN115423180A (en) | 2022-08-31 | 2022-08-31 | Surface deformation monitoring and predicting method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115423180A true CN115423180A (en) | 2022-12-02 |
Family
ID=84200120
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211056420.2A Pending CN115423180A (en) | 2022-08-31 | 2022-08-31 | Surface deformation monitoring and predicting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115423180A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116068511A (en) * | 2023-03-09 | 2023-05-05 | 成都理工大学 | Deep learning-based InSAR large-scale system error correction method |
-
2022
- 2022-08-31 CN CN202211056420.2A patent/CN115423180A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116068511A (en) * | 2023-03-09 | 2023-05-05 | 成都理工大学 | Deep learning-based InSAR large-scale system error correction method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220043182A1 (en) | Spatial autocorrelation machine learning-based downscaling method and system of satellite precipitation data | |
| CN104459696B (en) | SAR interference baseline precise estimating method based on flat-earth phase | |
| CN111059998B (en) | High-resolution-based time sequence InSAR deformation monitoring method and system | |
| CN104122553B (en) | Regional ground settlement monitoring method based on multiple track and long strip CTInSAR (coherent target synthetic aperture radar interferometry) | |
| Papasaika et al. | Fusion of digital elevation models using sparse representations | |
| CN111401602B (en) | Assimilation method for satellite and ground rainfall measurement values based on neural network | |
| CN111856459B (en) | Improved DEM maximum likelihood constraint multi-baseline InSAR phase unwrapping method | |
| CN115423180A (en) | Surface deformation monitoring and predicting method | |
| CN115357847B (en) | Solar scale satellite-ground precipitation fusion method based on error decomposition | |
| CN114812491B (en) | Transmission line earth surface deformation early warning method and device based on long-time sequence analysis | |
| CN113281749A (en) | Time sequence InSAR high-coherence point selection method considering homogeneity | |
| Abdalla et al. | Estimating random errors of scatterometer, altimeter, and model wind speed data | |
| CN112526515A (en) | Surface deformation detection method based on synthetic aperture radar interferometry | |
| CN115079172A (en) | MTInSAR landslide monitoring method, equipment and storage medium | |
| CN117592005B (en) | PM2.5 concentration satellite remote sensing estimation method, device, equipment and medium | |
| CN116910041A (en) | Daily correction method for remote sensing precipitation product based on scale analysis | |
| CN116381680A (en) | Urban earth surface deformation monitoring method based on time sequence radar interferometry technology | |
| CN111025294A (en) | InSAR phase unwrapping method based on mean square volume Kalman filter | |
| CN113408111B (en) | Atmospheric precipitation inversion method and system, electronic equipment and storage medium | |
| CN110596017B (en) | Hyperspectral image soil heavy metal concentration assessment method based on space weight constraint and variational self-coding feature extraction | |
| CN116068511B (en) | Deep learning-based InSAR large-scale system error correction method | |
| CN117054022A (en) | Method and system for extracting fine deformation of bridge | |
| CN113884054B (en) | Tower displacement settlement monitoring method and system | |
| Zhang et al. | Comparisons of OCO-2 satellite derived XCO2 with in situ and modeled data over global ocean | |
| CN115830476A (en) | Terrain factor space downscaling method |
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 |