论文部分内容阅读
地震、测井资料联合反演技术，是利用地震、测井数据的互补特性，从井间地震数据中反演地层的物性参数。共分三步进行：①查清井间地层构造形态，应用Born近似反散射线性反演技术，从叠前或叠后地震记录中得到反映层速度界面的反射系数剖面，效果相当于叠前或叠后的偏移剖面，并用作非线性反演的井间几何形态约束信息；②求取井间地震波参数，应用最优拟合牛顿法非线性反演技术，分别得到地层反射系数、波阻抗和介质速度参数；③求取地层物性参数，先应用神经网络技术从测井资料中求取井中物性参数，并以此作为非线性反演的井参数约束信息，在非线性反演地震波参数基础上，应用神经网络定量分析，从井间地震资料中分别得到地层的孔隙率、渗透率和含水饱和度参数的空间分布信息。该技术为实用的系列软件包，有I860微机版本和高档工作站版本。现已处理实际地震剖面近1000km。本技术特色为2．5维线性波动方程反演、二维非线性波动方程反演和神经网络分析的联合应用。地质效果特点是分层分辨率可以与井参数逐层对比，现已在多个油田使用。
The joint inversion technique of seismic data and well logging data utilizes the complementary features of seismic data and well log data to invert the physical property parameters of the formation from the cross-well seismic data. It is divided into three steps: (1) to ascertain the formation structure between wells and apply the Born approximate backscatter linear inversion technique to obtain the reflection coefficient profile reflecting the layer velocity interface from prestack or post-stack seismic records, the effect is equivalent to prestack or Post-stack offset profile, and used as non-linear inversion of cross-well geometrical constraint information; ② Obtaining crosswell seismic wave parameters, using the best fit Newton’s nonlinear inversion technique, respectively, the formation reflection coefficient, wave impedance And velocity parameters of the medium; (3) Obtaining the physical property parameters of the formation, the neural network technique is firstly used to obtain the well property parameters from the well logging data, and then the well parameter information is used as the non-linear inversion well parameter information. , The neural network quantitative analysis is used to obtain the spatial distribution information of the porosity, permeability and water saturation parameters of the formation from the crosswell seismic data. The technology is a practical series of software packages, I860 computer version and upscale workstation version. Now the actual seismic section has been processed nearly 1000km. The technical characteristics of 2.5D linear wave equation inversion, two-dimensional nonlinear wave equation inversion and neural network analysis of the joint application. The geologic effect is characterized by a layered resolution that can be compared with the well parameters layer by layer and is now used on multiple fields.