研究目的
Investigating the relationship between B-scan wave outlines and the profiles of layered rough interfaces in ground-penetrating radar (GPR) responses, and analyzing the influence of upper rough surface profiles on the echoes from lower surfaces.
研究成果
The shape of the B-scan wave outline from the top rough surface can reflect its profile, but for lower surfaces, the outlines are affected by scattering from upper surfaces. When all layered surfaces have identical profiles, their wave outlines match the profiles. For surfaces with different profiles or roughness, the lower the surface position, the greater the deviation between wave outline and profile, especially with larger upper surface roughness. These findings aid in understanding GPR echoes for improved imaging of layered rough interfaces.
研究不足
The study is based on numerical simulations using the FDTD method, which may have limitations in accurately modeling real-world complexities such as varying material properties and environmental conditions. The use of Gaussian rough surfaces may not fully represent all natural rough interfaces. The model is 2-D, which might not capture 3-D scattering effects. The analysis focuses on specific roughness parameters and media, potentially limiting generalizability.
1：Experimental Design and Method Selection:
A 2-D finite-difference time-domain (FDTD) computational model with a uniaxial perfectly matched layer (UPML) boundary is constructed for GPR demining of layered rough interfaces. The FDTD method is chosen for its effectiveness in simulating subsurface layers with varying geometries and roughness.
2：Sample Selection and Data Sources:
Gaussian randomly rough interfaces are generated using the Monte Carlo method, with parameters such as root-mean-square height (δ) and correlation length (l) defining the roughness. The model includes two-layered and three-layered rough surfaces with different profiles and degrees of roughness.
3：List of Experimental Equipment and Materials:
The model uses a differential Gaussian pulse as the GPR source with a central frequency of
4：5 GHz. Media include air, concrete (εr=0, σ=0001 S/m), clay (εr=0, σ=05 S/m), and metal (σ=72×10^7 S/m). Experimental Procedures and Operational Workflow:
The GPR antenna moves along a B-scan line with a step of 2 cm, generating 500 traces. Simulations are performed with spatial steps Δx=Δy=1 cm, time step Δt=
5：0167 ns, and UPML absorbing boundary of 8 cm thickness. Model dimensions are set to L=10 m, h=5 m, h0=15 m, d=-1 m. Data Analysis Methods:
Numerical results of B-scan echoes are compared with surface profiles to analyze relationships and influences. Statistical comparisons are made for different scenarios of surface profiles and roughness.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
