论文详情
山前带复杂构造区蕴藏了丰富的油气资源。地震资料信噪比低和速度建模困难是目前山前带地震数据精确成像面临的主要问题。以龙门山山前带海棠铺地区地震数据为例, 开展了叠前深度域地震成像处理方法攻关研究, 重点在叠前噪声压制和叠前深度偏移速度建模等两方面开展了研究, 具体如下。①加强低频弱信号保护, 采用二次信噪分离技术, 对噪声记录中有效信号进行二次信噪分离, 最大限度保护低频弱信号不受损伤。②优化叠前深度偏移速度建模流程: 首先优化传统深度域速度场更新输入道集与偏移输入道集共用一个道集的方式, 对驱动速度场更新的输入道集进行五维数据规则化处理以提高其信噪比; 然后以地质导向为基础, 精细刻画强反射速度缺失界面, 联合层析反演并融合深度域近地表速度模型, 建立具有地质模型约束的深度域速度场; 得到了精确的速度模型后, 再修改偏移成像输入道集及偏移方法, 将五维数据规则化前的地震道集作为输入道集, 将全方位角度域叠前深度偏移作为最终偏移成像方法。将该方法应用于川西北海棠铺地区的地震数据处理, 结果表明所提方法大幅提高了研究区的地震成像精度, 最终得到的叠前深度偏移成像结果信噪比高, 偏移归位合理, 为进一步推动该区的深入勘探开发, 提供了可供借鉴的深度域地震成像处理经验。
The complex structural area in the piedmont zone has abundant oil and gas resources, but the low signal-to-noise ratio(SNR) of the data and the difficulty of velocity modelling bring challenges to accurate seismic imaging for the piedmont zone.Based on the data from the Haitangpu area in the piedmont zone of Longmen Mountain, this paper presents an applied research-based work on prestack depth-domain seismic imaging processing, especially with respect to prestack noise suppression and prestack depth migration velocity modeling.This work includes two procedures: ①protection of low-frequency weak signals, and ② optimization of prestack depth migration velocity modeling process.The secondary signal-to-noise separation technique is adopted to perform a secondary signal-to-noise separation on effective signals in noise recording, so as to protect the low-frequency weak signals to the maximum extent.To optimize the prestack depth migration velocity modeling process, the situation that input gather and offset input gather share a common gather in traditional depth-domain velocity field update is optimized by five-dimensional data regularization of input gather driving the velocity field update to improve its SNR.Then, based on geosteering, the interface with strong reflection but missing velocity information is finely characterized, and the depth-domain near-surface velocity model is combined with joint tomographic inversion to establish a depth-domain velocity field with geological model constraints.After an accurate velocity model is obtained, the migration input gather and migration method are modified, that is, the gather before five-dimensional data regularization is taken as the input gather, and the full-azimuth angle-domain prestack depth migration as the final migration imaging technique.Application of this method to the Haitangpu area demonstrated a significantly improved seismic imaging accuracy, a high SNR in the final prestack depth migration imaging results, and a rational migration positioning.This provides a valuable reference for depth-domain seismic imaging processing to further promote the exploration and development in the area.