地震数据POCS插值重建中的一种新型质量控制准则

2019年 58卷 第No. 1期
阅读:84
查看详情
A novel quality control criterion in the POCS interpolation of seismic data
(1.同济大学海洋地质国家重点实验室,海洋与地球科学学院,高等研究院,上海200092;2.中国石油大学(北京)油气资源与探测国家重点实验室,北京102249)
(1.State Key Laboratory of Marine Geology,School of Ocean and Earth Science,Institute for Advanced Study,Tongji University,Shanghai 200092,China; 2.State Key Laboratory of Petroleum Resources and Prospecting,China University of Petroleum,Beijing 102249,China)

大多数地震数据插值重建方法利用的是迭代策略,因此,有必要建立合理的质量控制准则来适时终止迭代,在保证插值精度的同时,提高计算效率。利用迭代硬阈值方法及投影算子推导出凸集投影方法(POCS),基于观测数据置入前、后的数据体建立了新的质量控制准则,该控制准则仅利用了观测数据的信息,具有较好的适应性,可以进一步提高POCS插值重建方法的计算效率。利用两套模拟数据详细比较了3种质量控制准则(J1,J2和J3)的应用效果,验证了新提出的质量控制准则的有效性。实际资料应用效果表明该准则可以在保证插值精度的同时提高计算效率。

Most seismic data interpolation methods use iterative strategies; thus,it is important to reasonably define a quality control criterion (QCC) to duly terminate the iterations for efficient and highly accurate seismic data interpolation.In this article,a projection onto convex sets (POCS) method is derived using the iterative hard thresholding (IHT) algorithm and a projection operator.Then a novel POCS-based QCC is established before and after the insertion of observed seismic data.The QCC has better flexibility and can further reasonably improve the POCS efficiency.Testing on the 2D numerical examples compared the performances of different QCCs,namely,J1,J2,and J3,validating the proposed QCC.The proposed QCC can be easily extended into the cases with higher dimensions.Field data interpolation performance demonstrates that the proposed QCC can improve and ensure the interpolation efficiency and accuracy respectively.

迭代硬阈值方法; 凸集投影方法; 质量控制准则; 稀疏变换; 计算效率; Curvelet变换; 压缩感知;
iterative hard thresholding (IHT),; projection onto convex sets (POCS),; quality control criterion,; sparse transform,computational efficiency,; Curvelet transform,; compressed sensing;

国家自然科学基金项目(41604096,41874128)、国家科技重大专项课题(2016ZX05024001-004)及同济大学启动资金(180072)共同资助。

10.3969/j.issn.1000-1441.2019.01.009