摘要
由于地层的吸收衰减作用,地震波在传播过程中不断衰减,振幅谱有效频带变窄,降低了地震剖面的分辨率。经窗函数能量归一化后的广义S变换能在保持时间分辨率不降低的前提下,提高频率域的分辨率。将改进广义S变换与谱模拟反褶积方法相结合,克服了傅里叶变换中复杂的时窗大小问题,能更好地适应地震信号是平稳信号的假设条件。通过模型试算和实际地震资料处理,有效地提高了地震资料的分辨率,以及薄储层的识别和精细储层的描述能力,验证了该方法的可行性。
The energy of seismic wave attenuates continuously in spreading process, and its effective band of frequency amplitude spectrum is striation, which declines the resolution of seismic data. Improved generalized S transform of window function normalized energy holds time resolution constant,and improves the frequency resolution at the same time. Spectral modeling method is drawn in time-frequency domain, in order to avoid the complex window dimension program of the Fourier transform. Spectral modeling method is adapted to the assumption condition that seismic data are smooth and steady signals. Through model experiment and actual seismic data processing, it is proved that this method can improve the resolution of seismic data effectively, which is favorable for thin reservoir identification and fine reservoir description.
出处
《岩性油气藏》
CSCD
2013年第1期116-121,共6页
Lithologic Reservoirs
基金
国家高新技术研究发展计划"863"项目"陆上非一致性时延地震
微地震油藏监测方法研究"(编号:2011AA060303)资助
关键词
广义S变换
谱模拟反褶积
薄储层
子波振幅谱
地震分辨率
general S transform
spectral modeling deconvolution
thin reservoir
wavelet amplitude spectrum
seismic resolution
