节点文献
基于反射系数的物探数据融合成像方法及应用
Geophysical Data Fusion Method in Imaging and Application Based on Reflection Coefficient
【作者】 赵莹;
【导师】 苏茂鑫;
【作者基本信息】 山东大学 , 建筑与土木工程(专业学位), 2020, 硕士
【摘要】 借助不同地球物理手段在地层物性响应、探测深度和分辨率等方面的优势,开展综合物探勘察成为降低物探多解性,提高解释精度的主要手段。物探方法根据观测物理场的不同可分为电阻率法、反射波法、重力法和磁法等种类。每种方法各有其优缺点,如地震波法频率低,探测深度大,但精度较低;地质雷达分辨率高但探测深度浅。同样在电阻率法中,不同方法对高低阻的响应敏感程度不同。所以,无论是在地铁灾害勘察、隧道超前预报、滑坡体界面识别等方面,物探方法的综合使用体现出比单一方法更为丰富的地质信息。但是,目前对综合物探结果的呈现还是以先单独成像,再进行综合分析为主,对不同类型数据之间的相关程度和统一成像研究相对较少。本文基于反射系数的概念,试图将反射波数据和电阻率数据均转换为反射系数后进行融合分析和解释;在融合前,需要对不同反射系数进行重采样和特征重构以实现不同数据在空间和分辨率上的统一;最后利用主成分变换对重构的反射系数进行相关性计算和数据融合,获得多源物探数据基于反射系数的统一成像。从而初步建立起一套较为有效的多种物探数据融合成像方法。论文根据上述思路,首先对数值模拟获得的理论响应数据展开处理和分析。将地震波法、地质雷达法、高密度电法和瞬变电磁数据两两组合,形成三种数据融合模式:反射波法+反射波法,电阻率法+反射波法,电阻率法+电阻率法。针对每种模式,分别建立层状和复杂模型获取响应数据,按常规方法进行成像分析,发现单一成像分析的优势和不足;然后从不同理论响应数据中分别提取反射系数,对比反射系数成像与常规成像的异同;最后,借助主成分变换算法对不同源反射系数进行融合计算,实现基于反射系数的地质界面刻画和多源物探数据融合成像。将本文方法应用到江西某隧道工程的多源物探实测数据的处理中。针对现场探测某里程范围内拱底、拱肩两个部位的多源物探数据,利用反射系数对地质结构的界面特征进行刻画成像,然后分别融合实现统一成像,并与常规成像解释进行对比,证明该方法在提高地质结构界面识别效果和解决多源物探数据统一成像中的应用价值。
【Abstract】 Integrated geophysical investigation has become an important method to reduce multi-solution of geophysical exploration and improve the interpretation accuracy,which means the investigation was carried out with the help of various geophysical methods that have different advantages in formation physical response,investigation depth and resolution.According to the difference of physical parameters observed,geophysical methods can be divided into gravity method,magnetic method,resistivity method and reflection wave method.Each method has its advantages and disadvantages.For example,the seismic wave method characterized by greater detection depth but low accuracy because of the low detection frequency.Ground Penetrating radar(GPR)characterized by the high resolution but shallow detection depth.Also,different resistivity methods have different sensitive to the response of high and low resistivity abnormal body.Therefore,comprehensive use of geophysical exploration method could reflect more abundant geological information than the single method in terms of disaster investigation of subway,advance prediction of tunnel and interface recognition of landslide.At present,however,the expression of integrated geophysical results is still dominated by separate imaging and then comprehensive analysis.There are relatively few studies on the correlation degree and unified imaging between different types of observed data.Based on the concept of reflection coefficient,this paper attempts to convert the reflected wave data and resistivity data into reflection coefficient for fusion analysis and interpretation.Before fusion,resampling and feature reconstruction of different reflection coefficients are needed to unify the spatial and resolution.Finally,the correlation calculation and data fusion of the reconstructed reflection coefficients are carried out by means of Principal Component Analysis(PCA)transformation to obtain unified imaging of the multi-source geophysical data.Consequently,a set of more effective methods of geophysical data fusion imaging is preliminarily established.According to the above research ideas,this paper firstly processes and analyzes the theoretical response data obtained from numerical simulation.Three data fusion modes are formed by the seismic wave method,GPR,Electric Resistivity Tomography and Transient Electromagnetic:reflection wave data+reflection wave data;resistivity data+reflection wave data;resistivity data+resistivity data.For each mode,a layered and complex model are established to obtain response data,and conventional imaging analysis was conducted to find the advantages and disadvantages of single imaging analysis.Then the reflection coefficients were extracted from different theoretical response data and the imaging were compared with the conventional imaging.Finally,the PCA transformation algorithm is used to complete different reflection coefficients fusion,so as to realize the expression of geological interface based on the reflection coefficients and the fusion imaging of multi-source geophysical data.The method is also applied to the processing of measured data from multi-source geophysical exploration in a tunnel project in Jiangxi province.The reflection coefficient was extracted to image aiming at the multi-source geophysical data observed from the arch bottom and shoulder of a certain mileage range in tunnel.The features of geological structures were presented by the reflection coefficient of different geophysical methods.Then,the multi-source reflection coefficients of the two sites are fused to achieve unified imaging.The result is compared with the conventional imaging interpretation to prove the application value of this method in improving the recognition effect of geological structure interface and solving the unified imaging of multi-source geophysical data.