| Asphalt pavement as the main form of road surface in China,in its routine maintenance or major repairs,there are often invisible disease can not be found in the problem.This can lead to a lack of timely road maintenance or incomplete treatment of major repairs,which ultimately leads to a series of problems such as reduced quality of traffic and increased maintenance costs.As a non-destructive testing tool,3D-Radar can avoid road excavation and conduct a comprehensive screening of the internal structure of the road and its defects.It makes the daily maintenance plan and the design of road expansion more scientific,reasonable and accurate.However,the current identification methods for radar images rely on manual experience and lack detailed protocols for data processing and image interpretation.Therefore,a systematic analysis of the characteristics of invisible asphalt pavement defects in the 3D-Radar response,to provide technical support for the engineering application of 3D-Radar inspection,has become an urgent issue in the current stage of asphalt pavement inspection work.This thesis firstly analyzes the composition of 3D-Radar system and electromagnetic wave theory,elaborates on the detection mode,parameter setting and operation procedure of 3D-Radar asphalt pavement,and forms a complete asphalt pavement detection method;carries out 3D-Radar forward simulation by using finite difference time domain method,and quantitatively analyzes the radar response of devoid,loose and cracks under different sizes,different locations and whether the water content is present or not.The ground-penetrating radar response of invisible diseases was determined by numerical analysis of the radar images of typical diseases;through the on-site detection results of the preset "typical invisible diseases",combined with the radar image results of the forward simulation,a 3D-Radar mapping interpretation method of typical invisible diseases was finally proposed;the 3D-Radar was used to inspect the road section of the dependent project.The proposed radar interpretation method was used to identify the abnormal areas in the data,and the results were verified by means of core drilling and field milling.The main research results of this thesis include: a proposed method for setting parameters for 3D-Radar inspection of asphalt pavements;a feature extraction method for3D-Radar images is proposed,and the radar images after processing have clear features,which are more effective in identifying the type and volume of the disease;a calculation method for the depth of the disease in 3D-Radar targets is proposed.The radar signal wave speed formula and the depth calculation formula are combined to correct the vertical dimension of the disease marked in the 3D ground-penetrating radar detection image;the A-Scan,B-Scan and B-Scan image binarisation results,the radar response characteristics of cross-section and horizontal section for different sizes,different locations and whether they contain water,etc.are derived;the radar image characteristics of the devoid disease are determined by the ratio of the devoid size to the ground-penetrating radar step.The ratio determines whether the feature is "up-convex","down-concave" or "straight";a3D-Radar based invisible asphalt pavement mapping library was established. |
