混合相位反褶积串联形态滤波技术提高地震资料分辨率

Seismic data resolution improvement based on mixed phase deconvolution with morphological filtering

针对中深层岩性油气藏地震资料分辨率低、有效信号弱的问题,提出了一种混合相位反褶积串联形态滤波技术,旨在提高地震资料分辨率的同时使资料具备较高的信噪比。该方法首先对地震资料进行多尺度分解;然后分析各尺度剖面信噪比情况,确定加权重构系数并进行剖面重构;最后对重构剖面进行混合相位反褶积处理。通过原始数据处理、叠加剖面对比分析等测试了串联方法的可行性。在吉林油田西部前缘带的应用结果表明,该串联处理方法不仅有效地提高了研究区深、浅层地震资料分辨率,而且改善了深层弱信号信噪比,增强了弱同相轴可识别性,利于后续构造解释和薄互层预测。

To solve the problems such as seismic low resolution and weak signals in middle-deep stratigraphic reservoirs, the article proposes a joint-application approach, a mixed phase deconvolution in connection with morphological filtering to improve data resolution and S/N. The article firstly discusses algorithm principle of the mixed phase deconvolution in connection with morphological filtering. Then multi-scale morphological filtering decomposes seismic data, S/N of various scales is analyzed to determine reconstruction coefficients, and multi-scale seismic data is reconstructed, and so seismic data noise is suppressed. Finally, the mixed phase deconvolution is used to process reconstructed seismic data. Some model data tests verified the feasibility of the proposed approach. Applications in the western front belt of Jilin Oilfield show that the proposed approach can not only improve data resolution in the shallow and deep layers, but also data S/N in deep parts, which may be beneficial to interbed prediction. © 2016, Editorial Department OIL GEOPHYSICAL PROSPECTING. All right reserved.