基于多种地震反演方法的哈拉哈塘地区火成岩识别及速度建模

Igneous rock identification and velocity modeling based on several seismic inversion methods in Halahatang area

塔里木盆地北部哈拉哈塘地区油气成藏条件良好,普遍发育二叠系火成岩储层,地震震资料显示二叠系岩性、速度变化剧烈,影响其下伏奥陶系油藏"串珠"的叠前深度偏移成像及低幅构造圈闭的变速成图。在分析哈拉哈塘南部工区地质资料基础上,采用约束稀疏脉冲反演、人工神经网络反演、多参数反演方法对二叠系火成岩速度识别进行对比;采用db4小波对声波测井曲线进行基于小波变换的分频重构,将反演得到的速度模型应用在叠前深度偏移中。结果表明,约束稀疏脉冲反演方法更适用于工区巨厚的、岩相变化复杂的火成岩的快速建模;声波测井曲线重构后反演的数据体对岩性的识别能力明显提高,有助于火成岩速度建模。文中速度模型对"串珠"的刻画取得较好效果,表明该方法可为哈拉哈塘及类似地区火成岩研究提供初始速度模型。

There are good conditions for hydrocarbon accumulation in Halahatang area in north of Tarim basin,where the Permian igneous rock has universally developed and the lithology and velocity changed violently.These will affect its underburden Ordovician reservoir ' beads-shaped cave' pre-stack depth migration and the variable velocity mapping of low amplitude trap structures.On the basis of analyzing geological data of Halahatang southern work area,the paper uses constrained sparse spike inversion,artificial neural network inversion and log parameters inversion to compare the Permian igneous velocity recognition.Based on the wavelet transform,using the db4 wavelet for the acoustic logging curve to reconstruct the frequency division and get the velocity model which will be applied pre stack depth migration.The results showed that the constrained sparse spike inversion method is more applicable to velocity rapid modeling of the very thick igneous rock which has complex lithofacies.Using the reconstructed log data to participate in the inversion and get the inverted data which improves the recognition ability of lithology obviously and helps construct the velocity model of the igneous rock.The velocity model obtained by CSSI method gets favourable effects in the ' beads-shaped cave' depicting,which indicates that the CSSI method can provide an accurate original velocity model for the research of igneous rock in Halahatang and similar areas.