利用谱分解技术进行薄储层预测

Using spectrum decomposition technique for prediction of thin reservoir.

采用短时窗离散傅里叶变换的频谱分解技术,实现了在频率域内通过调谐振幅的成像特征来研究储层横向变化规律的目标,最大限度地挖掘了地震资料主频至高频端的地震分辨能力。本文介绍了谱分解技术的基本原理、谱分解数据体的类型及其解释方法,充分利用谱分解得到的一系列单一频率的调谐振幅数据体研究辽河盆地东部凹陷西斜坡铁匠炉地区近源沉积砂砾岩储层,查清了该套储层沉积相带的平面分布和有效储层横向展布特征,预测结果已被钻井所证实。

Using the spectrum decomposition technique by discrete Fourier transform in short window a-chieves the object using imaging feature of tuning amplitude in frequency domain to study the regularity of lateral variation in reservoir and fully dig up seismic resolution capability in dominant-high frequencies of seismic data. The paper introduced the basic principle of spectrum decomposition technique, categories of spectrum-decomposed data volumes and interpolation methods, fully used a series of amplitude-tuning data volume with single frequency from spectrum decomposition to study near-source sedimentary sand-gravel rock reservoir in Tiejianglu area in west slope of east sag, Liaohe basin,and clarified the planar distribution of sedimentary facies zone of the reservoirs and lateral distribution feature of effective reservoirs. The predicted results have been proved by drilling results.