一种基于稀疏窗S变换的分频-重构波阻抗反演方法

A frequency division-reconstruction wave impedance inversion method based on sparse window S transform

常规波阻抗反演一般使用全频带叠后数据进行反演,大多数反演结果受地震波主频控制,数据中有效频带的高频与低频部分的潜力没有得到充分利用,很难达到理想效果。本文提出了一种基于稀疏窗S变换的分频-重构波阻抗反演方法,首先对窗参数进行基于振幅谱的稀疏优化,根据不同信号的振幅谱作不同适应性窗参数优化,然后利用稀疏窗S变换对井数据进行处理,最后将分频井曲线反演与分频地震反演方法相结合,建立新的分频-重构波阻抗反演流程。模型试算、方法验证及实例应用表明,基于稀疏窗S变换的时频分析方法具有高分辨率和能量聚集的特性,分频-重构波阻抗反演结果与井吻合度更高,能够有效提高对断层及薄层的识别精度,具有良好的应用潜力。

Conventional wave impedance inversion generally adopts the full-band post-stack data for inversion. For most of the inversion results are controlled by the dominant frequency of seismic wave, the potentials of the high-frequency and low-frequency parts of the effective band in the data are not fully utilized, and therefore, the inversion results can hardly achieve the desired results. In this paper, a method of sparse window S transform- based frequency division-reconstruction wave impedance inversion is proposed. Firstly, the window parameters are sparsely optimized based on amplitude spectrum, and different adaptive window parameters are optimized according to the amplitude spectrum of different signals;and then, the sparse window S transform is utilized to process the well data;finally, both the frequency division well curve inversion method and the frequency division seismic inversion method are used to establish a new frequency division -reconstruction wave impedance inversion process. The results of model trial calculation and method verification match well with practical application results;this fact proves that the sparse window S transform-based time-frequency analysis method presents the characteristics of high-resolution and energy concentration, and the results of frequency division-reconstruction wave impedance inversion are more consistent with well data. This method can effectively improve the identification accuracy of faults and thin layers, and exhibits a good application prospect.