宽频数据在致密砂岩储层预测中的应用

Application of broadband seismic data in tight sand reservoir prediction

西湖凹陷深层致密砂岩储层具有良好的勘探开发前景,受埋深影响,目的层地震资料品质较差.A构造通过斜缆宽频采集和处理获取宽频地震数据,提升了资料品质,然而应用常规子波提取方法对宽频数据进行子波提取并反演计算纵波阻抗,结果与井上实测数值差异较大,影响储层的定量解释.针对这一问题,提出统计性子波和确定性子波相结合的长短子波合并宽频子波提取方法,提取的宽频子波比常规子波低频丰富、旁瓣小,能更真实地反映地震信息,约束稀疏脉冲反演的纵波阻抗结果与测井曲线吻合度更高.基于宽频数据和常规数据分别进行约束稀疏脉冲弹性波阻抗反演,预测A构造优质储层分布,经已钻井证实,宽频数据比常规数据储层预测精度高,预测的储层展布特征与研究区地质沉积认识一致.结果表明:这种基于宽频子波提取的宽频资料应用方法有效降低了致密砂岩储层预测的多解性.

The tight sand reservoir in deep layer of Xihu Sag has good exploration and development prospect according to previous research.Therefore,determining high-quality reservoir distribution area is of great importance for reservoir fracturing and development.However,under the influence of the buried depth,the frequency of seismic data attenuates,which reduces seismic data quality.The data quality could not satisfy the demand for reservoir prediction.To improve seismic data quality,variable depth streamer acquisition and broadband data processing techniques are applied in A structure of Xihu Sag.As a result,broadband data is acquired.However,when we use conventional wavelet extraction method to extract wavelet for broadband data and calculate P-impedance by inversion,the difference between the inverted data and the well logging data is big,which affect reservoir quantitative interpretation.To solve this problem,in this paper,long and short wavelet merging wavelet extraction method for broadband data with the combination of statistical wavelet and determinacy wavelet is proposed.Compared with conventional wavelet,the extracted broadband wavelet is rich in low frequency information and it has smaller sidelobe,which could reflect seismic data information more precisely.Additionally,the inverted P-impedance by constrained sparse impulse method is much more consistent with well logging curve,which proves the reliability of the wavelet extraction method.Then constrained sparse impulse elastic impedance inversion is conducted both for broadband data and conventional data to predict high-quality reservoir distribution in A structure.It is proved by the drilling data that the prediction precision of broadband data is higher than that of conventional data.The predicted reservoir distribution feature using broadband data fits with the geological knowledge in the research area.The results show that the broadband data application method based on broadband wavelet extraction reduces the multiplicity of tight sand reservoir prediction,which provides the guidance for reservoir fracturing and development.