基于地震振幅谱分维数的油气预测软件系统研制

Development of Oil Development and Gas Prediction Software System Based on Fractal Dimension of Amplitude Spectrum

【目的】地震油气预测技术在油气勘探开发中发挥着越来越重要的作用,利用吸收系数地震属性进行油气预测技术在国内外已取得较大进展。【方法】使用非平稳地震信号谱分析的时频分析技术计算地震信号的时频剖面,并从中提取与吸收衰减有关的特征参数,是吸收衰减研究的一个重要方向。与吸收衰减关系紧密的时频谱属性有衰减梯度、振幅谱分维数、高频能量及平均频率等。【结果】我们把振幅谱分维数与广义S变换时频分析技术结合起来,提出了振幅谱曲线的Mandelbrot分维数,使得它们具有了瞬时特性,可以很好地进行油气预测。【结论】经实际工区验证,利用振幅谱曲线的Mandelbrot分维数法开展油气检测具有较高的预测精度。

[Objectives]Seismic hydrocarbon prediction technology plays an increasingly important role in oil and gas exploration and development,Oil and gas prediction technology using the seismic attribute of absorption coefficient has made great progress both at home and abroad.[Methodology]Using the time-frequency analysis of non-stationary seismic signal spectrum to calculate the time-frequency profile of seismic signals and extracting the characteristic parameters related to absorption and attenuation is an important direction of absorption and attenuation research.Time-frequency spectrum properties closely related to absorption and attenuation include attenuation gradient, fractal dimension of amplitude spectrum,high-frequency energy and average frequency.[Results]We combine the fractal dimension of amplitude spectrum with the time-frequency analysis technique of generalized S transform, and propose the Mandelbrot fractal dimension of the amplitude spectrum curve, which makes them have instantaneous properties and can well predict oil and gas.[Conclusions]The actual work area verification, the use of amplitude spectrum curve Mandelbrot fractal dimension method to carry out oil and gas detection with high prediction accuracy.