基于改进的Morlet小波变换在伊犁盆地南缘层序地层划分中的应用

Application of improved morlet wavelet transform in sequence stratigraphic division in the southern margin of Yili basin

阐明了小波变换的基本原理及其包含的砂/泥比、砂泥岩互层频率等地质信息,处理手段选择适用性较好的Morlet小波。由于传统的Morlet小波不满足小波的容许条件,引入带宽参数对其进行了改进,用频率参数替换尺度参数,从而获得更高的分辨率,使其能够更加明确地表达出所包含的地球物理意义。利用改进后的Morlet小波基函数,对测井信号进行多频段分解,将一维信号转换为二维深度-频率信息,并进行逆变换恢复为深度-尺度信息。以新疆伊犁盆地南缘苏阿苏地区A4560钻孔为例,选择测井数据中最能体现泥质含量变化的自然伽马测井信号,进行连续小波分析和复小波连续变换,联合应用振幅谱和相位谱进行层序划分,获得了旋回叠加模式、层序界面。与地质资料反复对比确认后,确定了适用于该钻孔的划分标准,推广应用到该地区其他钻孔,经统计与人为划分的结果基本一致。基于改进的Morlet小波分析,能够在排除干扰因素的情况下,有效地利用自然伽马测井曲线准确地识别层序界面,为层序地层划分提供了借鉴。

The basic principles of wavelet transform and the included geological information such as sand/mud ratio,frequency of sand and mudstone interlayers were introduced to select Morlet Wavelet with Better Suitability.Since the traditional Morlet wavelet does not satisfy the permissible condition of the wavelet,the bandwidth parameters are introduced to improve it.Replacing scale parameters with frequency parameters was used to achieve higher resolution,and allow it to more clearly express the geophysical implications involved.Using the improved Morlet wavelet basis function,the logging signal is decomposed by multi-band.The one-dimensional signal is converted into two-dimensional depth-frequency information,and the inverse transform is used to restore depth-scale information.Take the A4560 hole in the Suasu region in the southern margin of the Yili basin in Xinjiang as an example,the natural gamma logging signal that best reflects the change of shale content in the logging data was selected for perform continuous wavelet analysis and continuous wavelet transform,and combine the amplitude spectrum and phase spectrum to sequence the sequence.And then obtain the cyclic superposition mode and sequence interface.After repeated comparison and confirmation with geological data,the classification criteria applicable to the drilling were determined and applied to other boreholes in the area.The results were basically consistent with the results of artificial and statistical classification.In summary,based on the improved Morlet wavelet analysis,it is possible to effectively use the natural gamma log curves to accurately identify the sequence boundaries and eliminate the disturbing factors,providing a certain basis for the sequence stratigraphic division.