Refined Deformation Measurement Of Urban Infrastructures Based On The Time-series D-InSAR Technique

With the rapid development of the social economy,a large number of infrastructures such as roads,buildings,high-speed railways,and bridges have been built in China.However,the increasing operation time and environmental loads would destroy the stability of their structures,leading to slow structural deformation.Such deformation,if accumulate or even exceed the specified limits,may cause safety hazards or even threaten their normal operations.Therefore,the operational safety of urban infrastructures has become an important practical issue,which has attracted increasing attention in the fields of public safety,earth observation and civil engineering.It is urgent to develop a scientific and efficient technical method to monitor their deformation conveniently,providing technical support for timely detecting the safety hazards and ensuring their safe operation.The time-series D-InSAR technology has become an effective method for measuring small deformation of the earth surface due to the unique advantage of acquiring ground information throughout all-day and all-weather.By analyzing stable point targets(such as buildings and artificial corner reflectors,etc.)that are not affected by the temporal/spatial de-correlation and atmospheric effects,it obtains accurate and reliable time-series deformation with the theoretical precision of millimeters,making it possible to realize the refined deformation monitoring of urban infrastructures.However,the detailed deformation analysis of such complex urban infrastructures puts forward higher accuracy requirements of InSAR deformation monitoring,so that the conventional time-series D-InSAR method is insufficient to meet the new monitoring requirements.Firstly,in the deformation monitoring of road networks,the more serious overlapping phenomenon of the signals in the high-resolution data would lead to inaccurate identification of the structural point targets.Therefore,how to accurately identifying high-density point targets upon road networks is one of the difficulties that need to be overcome.Secondly,the coverage of a single high-resolution SAR image is small,which is difficult to meet the overall monitoring requirement of large-scale road networks.Thus,integrating the road networks deformation results of multi-track SAR data sets is a problem to be solved.Moreover,the metal infrastructures are prone to be affected by thermal dilation,resulting in inaccurate deformation phase estimation.Therefore,approaching the appropriate temperature deformation estimation method is also an urgent problem to be solved.In addition,due to the more complicated structure of bridges,the conventional point targets identification method and simply deformation analysis method are insufficient to meet the practical application requirements.As a result,how to improve the bridge deformation analysis method so as to investigate more useful information is a problem that must be solved in practical applications.Finally,limited by the side-viewing geometry of SAR satellites,the twodimensional deformation maps of complex structure bridges are difficult for users to understand in multi-view.Therefore,improving the visualization of bridge deformation results and assisting the users in performing the refined deformation mechanism analysis is also an issue to be considered.In view of the above problems,this paper combines the specific structural and material characteristics of the research object,improving the key techniques including the point targets identification,temperature deformation estimation,deformation results interpretation and structural 3D visualization in the time-series D-InSAR method.Based on the improved method,we realized the detailed deformation monitoring(point targets refinement identification,temperature deformation estimation,3D deformation calculation,etc.)and refined deformation interpretation(temporal-spatial deformation characteristics analysis,deformation mechanism investigation,structural 3D visualization,etc.)of different types of urban infrastructures including the large-scale road networks,metal structures and complexly structured bridges.Regarding the deformation monitoring of different types of infrastructures,the main research results of this paper are as follows:(1)In the subsidence monitoring of large-scale road networks,the semantic information based structural point targets refinement identification method and the registration and fusion analysis method of multi-track D-InSAR points are proposed to maximize the number and accuracy of detectable structural point targets and to realize the integration of adjacent-orbit Terra SAR-X road networks results in Shanghai.We detected the subsidence along the large-scale elevated road,highway,and railway networks in Shanghai,and validated the results with the leveling data in terms of both spatial distribution and time-series.Based on the results,we explored the influence of different construction periods and structural styles on the temporal and spatial evolution of subsidence along the traffic networks.(2)In the deformation detection of metal structures,a structural temperature deformation modeling and analysis method based on coherent weighted least squares was proposed.It consists of the qualitative analysis and quantitative calculation of the temperature deformation,and was applied to the deformation monitoring of buildings and high-speed railways,realizing the separation of thermal dilation and trend deformation and obtaining more accurate deformation measurements.Based on the temperature deformation estimates,the disaster risk levels along the high-speed railways are evaluated.The reliability of the temperature deformation model is verified by comparing the derived thermal expansion coefficients with the actual property of the materials.The results investigated the general regularity of structural temperature deformation,that is the propagation direction,spatial distribution,and amplitude of the thermal dilation are associated with the geometry shape,static structural characteristics,and material properties of the targets.(3)In the case of complexly structured bridge deformation monitoring,the backscattering characteristics analysis of bridges are analyzed firstly,and the point target extraction strategy is improved according to the different levels of coherence of the bridges.As for the bridge deformation analysis,the potential relationship between the deformation characteristics of arch bridges and cable-stayed bridges and their respective structural and material properties is detected through the identification and analysis of deformation key points.Regarding the deformation mechanism interpretation,the 3D visualization effects of the deformation results are improved,assisting the spatial-temporal classification of points based on the time-series deformation models under the 3D perspective,and spatially linking the points to the specific bridge structures.The empirical deformation mechanisms of different components on arch bridges and cable-stayed bridges are then investigated based on the deformation analysis of different types of points.