基于共形辛Euler算法的非金属地下管道精细化高效探地雷达正演模型

GPR forward model of underground non-metallic pipeline based on parallel conformal symplectic Euler algorithm

我国城市地下管线种类繁多、规模巨大.近年来,因城市管网家底不清、位置不明导致的管道挖断事故频发.探地雷达(GPR)是目前最为有效的地下设施探测定位设备,通过对探地雷达回波信号进行反演分析,可获取管道埋深、位置等信息.然而,地下管道等地下圆形目标,传统正演模型中阶梯近似方法会产生一定误差.本文结合辛Euler算法与曲面共形技术,建立探地雷达电磁波在含非金属管道地下结构中传播的精细化数值模型,并通过图形处理器(GPU)加速技术进一步提升模型计算效率.数值模拟结果显示,共形网格技术有效降低了由于阶梯近似造成的虚假绕射波,并通过结合GPU并行计算,在不增加网格密度的前提下,实现了对地下管道电磁响应的高效精细化计算.

There are many types and large scales of urban underground pipelines in China.In recent years,pipeline excavation accidents occur frequently due to unclear urban pipeline network location.Ground Penetrating Radar(GPR)is the most effective tool for detecting and locating underground facilities.The information of pipeline depth and location can be obtained by inverse analysis of GPR echo signal.However,the ladder approximation method in the traditional forward model can produce some errors in the circular target of underground pipelines.Based on the symplectic Euler algorithm and surface conformal technique,a fined numerical model of GPR electromagnetic wave propagation in underground structures containing non-metallic pipeline is established in this paper.The computational efficiency of the model is further improved by Graphics Processing Unit(GPU)acceleration technology.Numerical simulation results show that the surface conformal technique can greatly reduce the pseudo-waves that are generated by the ladder approximation.By combining GPU parallel calculation,the efficient and precise calculation of the electromagnetic response of the underground pipeline is realized without increasing the grid density.