论文部分内容阅读
为了进一步保留大地电磁低频段的有用信息、提高矿集区复杂噪声环境下大地电磁测深深部探测能力,在形态滤波的基础上结合信号子空间增强和端点检测做二次信噪分离处理.首先,针对形态滤波预提取的噪声轮廓运用信号子空间增强分离出信号子空间和噪声子空间.然后,将信号子空间和重构信号相结合并将噪声子空间置零.最后,借鉴端点检测做后处理,以识别波形突变的起止点.仿真结果表明,卡尼亚电阻率曲线在低频段的数据质量得到了明显改善、视电阻率值相对稳定;有效地补偿了形态滤波处理过程中损失的低频有用信号,其结果更加真实地反映了测点本身所固有的大地电磁深部构造信息.
In order to further preserve the useful information of the electromagnetic band in the earth and improve the deep detection ability of the magnetotelluric sounding in the complex noise environment in the mining area, based on the morphological filtering, the second signal-noise separation processing is performed by combining the signal subspace enhancement and the endpoint detection. , The signal subspace and the noise subspace are separated by using the signal subspace enhancement for the noise contour pre-extracted by morphological filtering. Then, the signal subspace is combined with the reconstructed signal and the noise subspace is set to zero.Finally, Post-processing to identify the starting and ending point of waveform mutation.The simulation results show that the data quality of the Karnia resistivity curve has been significantly improved in the low frequency range, the apparent resistivity value is relatively stable, the effective compensation for the loss of the morphological filtering process Low-frequency signal, the result more truly reflects the local electromagnetic inherent deep structure of the earth’s tectonic information.