改进的EMD时频分析地震资料高分辨率处理方法及其应用研究

Research on the improved approach and application of EMD time-frequency for high-resolution seismic data processing

针对地震波地下传播过程中深层的高频成分损耗严重,导致分辨率下降的现象,而常规高分辨率处理受到局限性较高、成本高、处理周期长问题,基于EMD(经验模态分解)自适应分频的思想,提出了一种改进常规高分辨率处理的方法—基于EMD时频分析地震高分辨率处理。该方法的核心在于依据地震数据本身的时间尺度特征,使用EMD算法进行模态分解,能够在保留该数据本身的特性情况下,将地震信号自适应分解为有限个IMF(本征模态函数),分别对得到的每个IMF进行高分辨率处理,最后叠加得到高分辨率的地震数据。本方法适用于非平稳信号,无假设条件,且无需知道Q值,为解决高频补偿问题提供了一条全新的思路,理论模型测试和实际地震资料的运用表明,该方法能有效提高地震资料分辨率,在薄互层识别方面也有较好的应用前景。

As the development and increased difficulty of global hydrocarbon exploration, obtaining high fidelity and high resolution seismic data is of great significance to the exploration and development of complex oil and gas reservoirs. In order to solve the problems, such as the loss of high frequency components in the seismic pulse propagation process trough deep stra turn, the resolution is degraded, and the conventional high resolution processing is limited and high cost. According to the i dea of EMD adaptive frequency division, a method is proposed to improve the conventional high resolution processing high resolution processing based on EMD in this paper. According to the time scale feature of the seismic data, the core of the meth od is to decompose used by the EMD algorithm. The characteristics of the data are preserved during this process, and the seis mic signal is decomposed into a finite number of intrinsic modal functions IMF. High resolution processing is conducted on each IMF. And the high resolution seismic data is obtained by accumulating all the processing results mentioned above. The method in this paper adapt non stable signal, non assumed condition, and non Q value, and provides a new idea for the resolution of high frequency compensation issue. By applying it to the test results based on theoretical model data and actual seismic data show that the proposed method can effectively improve the seismic resolution effectively, and it has a good prospect in thin interbed seismic recognition.