论文部分内容阅读
由于实际工作中测量的位场数据存在混叠效应,并且这些观测数据具有多源性、非线性和非稳态性的特点,如何从其中提取研究对象所产生的异常信息是位场数据处理中的关键环节.二维经验模态分解(Bidimensional Empirical Mode Decomposition,BEMD)在处理非线性、非平稳信号方面具有明显的优势,且具有分解的自适应性,实现了对信号的多尺度分解.文中采用BEMD方法首先对模型试验中带噪声的重力数据进行分解,得到各级固有模态函数和剩余分量,去掉一个或多个固有模态函数实现初步去噪,或结合小波阈值实现准确、圆滑去噪;再次,对三峡地区的布格重力异常进行了多尺度分解,并且利用径向对数功率谱分析方法估计了各级模态函数所反映的地下场源似深度,定性或半定量地解释了不同深度下的场源产生的异常,较好地揭示了场源赋存的地质信息,对上述分解得到的各分量进行组合叠加,重构出局部异常和区域异常,并且同常规分场方法进行了对比分析,结果表明BEMD方法能够准确、有效地实现位场的异常分离.
Because of the aliasing effect of the field data measured in practice and the characteristics of these observations being multi-sourced, non-linear and non-steady state, how to extract the anomalous information from the research object is the field data processing . Bidimensional Empirical Mode Decomposition (BEMD) has obvious advantages in dealing with non-linear and non-stationary signals and has the adaptability of decomposition and realizes the multi-scale decomposition of signals. Firstly, BEMD method is used to decompose the noisy gravity data in the model test to obtain the intrinsic mode functions and residual components at all levels. One or more intrinsic mode functions are removed to achieve initial denoising, or the wavelet threshold is used to achieve accurate and smooth Noise; Thirdly, the multi-scale decomposition of Bouguer gravity anomalies in the Three Gorges area is carried out, and the depth of underground field source reflected by the modal functions at various levels is estimated by radial logarithmic power spectrum analysis method, and qualitatively or semi-quantitatively explained The abnormalities caused by the field sources at different depths reveal the geological information of the field sources well, Combination of superimposed rows, and the reconstructed local abnormal regional abnormalities, and were compared with conventional methods breakout, BEMD results show that the method can accurately and effectively achieve the separation potential field anomalies.