边缘特征和相关成像联合约束自适应网格剖分的重力三维反演方法  

3D gravity inversion based on adaptive meshing constrained by edge features and correlation imaging

作　　者：安国强 鲁宝亮[1,3,4] 李柏森 高新宇 AN GuoQiang;LU BaoLiang;LI BoSen;GAO XinYu(School of Geological Engineering and Geomatics,Chang’an University,Xi′an 710054,China;China Coal Aerial Survey and Remote Sensing Group Co.,Ltd.,Xi’an 710100,China;National Engineering Research Center of Offshore Oil and Gas Exploration,Beijing 100028,China;Key Laboratory of Western Mineral Resources and Geological Engineering,Ministry of Education,Chang'an University,Xi′an 710054,China)

机构地区：[1]长安大学地质工程与测绘学院,西安710054 [2]中煤航测遥感集团有限公司,西安710100 [3]海洋油气勘探国家工程研究中心,北京100028 [4]长安大学西部矿产资源与地质工程教育部重点实验室,西安710054

出　　处：《地球物理学报》2025年第3期1102-1115,共14页Chinese Journal of Geophysics

基　　金：国家自然科学基金(42374168,41904106);中央高校基本科研业务费(300102260202,300102262902)联合资助。

摘　　要：三维密度反演在地球物理反演中具有重要的作用.传统的重力三维密度反演是将反演区域均匀剖分为直立六面体网格,通过增加网格数量来提高对于复杂地质体的拟合程度,然而随着网格数量的增多导致反演效率降低以及多解性增加.针对此问题,本文提出了基于边缘特征和相关成像联合约束自适应网格剖分的重力三维反演方法.首先对反演区域进行粗糙网格剖分并进行重力相关成像计算,并且使用边缘特征信息对重力相关成像进行约束,得到更高分辨率的成像结果;然后对相关系数较大的重点区域使用八叉树结构进行加密,实现反演区域的自适应细致剖分;最后使用正则化方法对整个研究区域进行三维密度反演.通过引入最小长度约束、深度加权约束、体积约束、物性范围约束提高反演的精度,并使用共轭梯度法对目标函数进行快速求解.模型实验表明本文方法能够清晰地反映异常体的空间位置和密度分布,在提高反演结果分辨率的同时,相较于全局细致网格反演的计算效率大大提高.将该方法应用于澳大利亚Olympic Dam的实际重力数据,反演结果与已知的地质结构和钻井岩心密度曲线基本一致,进一步证实了该方法的正确性和有效性.The inversion of three-dimensional density is a crucial factor in geophysical inversions.The traditional method of gravity 3D density inversion divides the inversion regions into upright hexahedral grids,which increases the number of grids to improve the fitting degree for complex geological bodies.However,increasing the number of grids decreases inversion efficiency and increases non-uniqueness.This paper proposes a 3D gravity inversion method that uses adaptive meshing jointly constrained by edge features and correlation imaging.The inversion region is coarsely meshed,and gravity correlation imaging calculations are performed.The edge feature information is used to constrain gravity correlation imaging to obtain higher-resolution imaging results.The key regions with high correlation coefficients are encrypted using an octree structure to achieve adaptive and detailed subdivision of the inversion regions.The study area's 3D density is inverted using a regularization method.To improve the accuracy of the inversion,minimum length constraint,depth weighting constraint,volume constraint,and physical property range constraint are introduced.The conjugate gradient method is utilized to efficiently solve the objective function.Model experiments demonstrate that the proposed method accurately reflects the spatial position and density distribution of anomalous bodies,enhances the resolution of inversion results,and improves computational efficiency compared to global fine meshing.Applying this method to the field gravity data of Olympic Dam in Australia,the inversion results are consistent with known geological structures and drilling core density curves.This confirms the correctness and effectiveness of the proposed method.

关 键 词：重力异常 三维密度反演 自适应剖分 相关成像 边缘特征 

分 类 号：P631[天文地球—地质矿产勘探]