Study On Deformation Disaster Monitoring In Huainan Mining Area Based On InSAR And Sentinel-1A

The development of coal industry has made a great contribution to the urbanization process and the coordinated development of regional economy in China.However,along with the large-scale mining of coal resources,the surface structure has been continuously destroyed,and a series of mining disasters,such as ground subsidence,collapse,building deformation and collapse,and damage to public facilities,have seriously threatened the safety of people’s lives and properties in the mining area.The sustainable development of China faces severe challenges.Therefore,it is of great theoretical and practical significance to carry out long-term and efficient monitoring of ground subsidence in mining areas,and to conduct research on dynamic damage assessment and prediction and early warning of buildings,for disaster prevention and reduction of mining areas and harmonious and steady development of the region.Therefore,it is of great theoretical and practical significance to carry out long-term and efficient monitoring of ground subsidence in mining areas,dynamic damage assessment and prediction and early warning of buildings,for disaster prevention and reduction of mining areas and harmonious and steady development of the region.In recent years,the emerging synthetic aperture radar interferometry technology has made up for the shortcomings of traditional optical remote sensing imaging difficulties in special environments with its advantages such as all-weather,all-weather,high-precision,and continuous coverage,and overcomes the difficulties of low sampling point density and poor continuity of traditional observation techniques.It has brought a new and efficient observation method for the deformation monitoring of the mining area..However,In SAR is only used as a means to obtain deformation information in mining areas,and its practical application value in disaster prevention and reduction in mining areas has not been fully exerted.How to better combine In SAR technology with mine deformation research,and truly and efficiently apply it to long-term ground subsidence monitoring,building dynamic damage assessment,and prediction and early warning for disaster prevention and mitigation work is still in the exploratory stage,which is also an urgent problem to be solved.Therefore,this paper takes Huainan mining area in east China as the research area,and integrates relevant theoretical knowledge such as ground observation technology,geology,structure and computer machine learning,etc.,to carry out spatial-temporal evolution analysis of mining area surface based on In SAR technology,as well as dynamic damage assessment and prediction and early warning research of construction structures affected by mining.It provides efficient method and reliable basis for surface subsidence monitoring,disaster assessment and early warning in mining area.The main research content and results of this paper are as follows:(1)Research on the identification and feature extraction of deformation area in Huainan mining area based on D-In SAR technology and Sentinel-1A,and verifies the ability of the D-In SAR technology and Sentinel-1A data in the mining subsidence monitoring of the Huainan mining area.Based on sentinel-1 radar image data and using D-In SAR technology,the results of surface deformation in the Huainan mining area were inverted,all deformation areas in the study area were identified and delineated,and the spatial distribution of the deformation areas in the study area was explored.It also analyzes the reasons affecting the regional deformation characteristics,reveals the spatial deformation characteristics and laws of the deformed area.The results show that there are 25 deformation areas in the study area,which are mainly distributed in the west and north of Huainan mining area.The deformation interference fringes caused by coal mining in Huainan mining area all present concentric circular or elliptical spatial features with small area and regular shape in the interference pattern.The deformation area presents a typical funnel-shaped ground subsidence in space.The maximum area of the subsidence is located in the center of the deformation area,and the subsidence decreases gradually from the center to the boundary of the deformation area.(2)Research on the temporal coherence of different feature types based on Sentinel-1A data,and reveal the temporal coherence change law of different ground feature types.The factors that affect the coherence of In SAR technology are discussed in depth,and Sentinel-1A data is used to screen out two different important influencing indicators: time and ground feature types.Based on this,the research on the decoherence effect of different feature types in the mining area on the D-In SAR technology in different time periods was carried out,and the temporal coherence change characteristics of different feature types in the mining area were quantitatively analyzed.Four types of typical features in Huainan mining area: farmland,woodland,bare land,and residential land are selected as representatives.Through the calculation of coherence coefficients,the results of the temporal coherence changes of the four feature types in each month of the year are obtained.The results show that the four types of features have a similar trend of coherence changes throughout the year.Among them,the coherence is the worst from June to September,and the coherence is the best from December to March;Residential land and bare land showed a relatively stable and high correlation throughout the year,while agricultural land and forest land showed strong seasonal change correlation;the average of the four types of ground features in the study area in a year The coherence coefficient values are all higher than 0.32,which meets the coherence requirements of long time sequence deformation monitoring.By comparing with the meteorological element map of the mining area,the result shows that the change trend of coherence is negatively correlated with the change trend of local precipitation and temperature.Field visits and surveys show that the strong seasonal changes in the regional coherence of agricultural land and woodland are closely related to the planting cycle of local crops and the growth cycle of vegetation.(3)Research on safety monitoring and damage assessment of buildings in mining area based on SBAS-In SAR technology,and verify the feasibility of the method.Through in-depth analysis and comparison of the factors and characteristics of the impact of the deformation of the mining area on the structural damage of the building,the damage indicators are screened.And according to the actual structural characteristics of the buildings in the mining area,the damage assessment model is selected.By using the characteristics of obtain the regional area deformation information quickly of SBAS-In SAR technology,extracting damage indicators and combining the damage evaluation model,and then realize the rapid dynamic safety monitoring and damage evaluation of the buildings in the mining area.Taking the Yangjuzhuang residential area in Huainan mining area as an example,based on Sentinel-1A satellite image data from July 2015 to August 20,2016,the deformation information of the study area during the monitoring period was retrieved by using SBAS-In SAR technology,and the time series deformation map of the study area was obtained.Based on the deformation results of time series,the dynamic damage assessment results of buildings in the study area are extracted,and the field investigation results of the damage of buildings in the study area are verified.The results show that the damage assessment results are highly consistent with the actual damage of the buildings in the study area.This method provides a new method for the rapid dynamic safety monitoring and damage assessment of buildings in the mining area using In SAR technology.(4)Research on mine deformation prediction based on time series In SAR and machine learning theory,and put forward the GM-SVR combination forecasting model.Through the use of time series In SAR technology to obtain the time series deformation results of the mining area,the temporal and spatial evolution law and development trend of the deformation are obtained.Combining the deformation law of the mining area and the characteristics of the time series In SAR data set,based on the idea of combined modeling,and Integrating the advantages of the powerful linear processing of the gray model and the excellent learning,generalization and high-dimensional nonlinear processing capabilities of the support vector machine,the GM-SVR combined prediction model is established.Through the prediction of linear and non-linear deformation components respectively,the prediction research of mining area deformation is finally realized.Based on the results of the Yangjuzhuang time series In SAR monitoring from December 27,2016 to May 20,2017,a case study of the prediction model was conducted to verify the global prediction capability of the GMSVR combined model.Through comparison with a single prediction model,the results show that GM-SVR combined model has the highest prediction accuracy and stable performance,which is an effective deformation prediction method.