论文详情
针对单通道盲源分离中由于缺少先验信息而导致的源信号数目无法确定的问题,提出了一种基于单通道盲源分离的大地电磁工频干扰噪声消除方法。该方法首先将含有50Hz工频干扰噪声的大地电磁观测信号进行经验模态分解,然后利用衰减率确定其源信号的数目和待处理的有效主成分的个数,解决了源信号个数无法确定的问题。由于大地电磁观测信号的客观规律符合源信号相互统计独立的假设,因此可以将盲源分离中的独立分量分析方法应用于大地电磁工频干扰噪声压制。实验结果表明,在大地电磁原始信号被50Hz工频干扰噪声完全淹没的情况下,该方法仍能较好地压制工频干扰噪声。对添加了50Hz工频干扰信号的野外实测大地电磁信号去噪处理后发现,该方法在一定程度上改善了视电阻率曲线和相位曲线的偏移。
With the continuous development of industry and economy,the power-line interference noise (PLIN) in magnetotelluric (MT) signals has been increasing,seriously hindering MT exploration.In this paper,a method is proposed to eliminate the PLIN in MT signals based on single-channel blind source separation (SCBSS).The inability to determine the number of source signals due to the lack of a priori information in SCBSS is a common issue.To overcome this issue,empirical mode decomposition was performed on MT signals affected by a PLIN of 50Hz,and a decay ratio was applied to determine the number of source signals and effective main components to be processed.The analysis of MT signals is based on the assumption that the source signals are statistically independent;therefore,the independent component analysis method was applied to achieve blind source separation for PLIN elimination.Experimental results showed that the proposed method can significantly reduce the PLIN,even if the original MT signal is completely submerged by a PLIN of 50Hz.After denoising the field-measured MT signal,significant improvements in the apparent resistivity curve and phase curve were observed.
国家自然科学基金项目(41674107)、湖北省教育厅科学研究计划指导性项目(B2019034)、油气资源与勘探技术教育部重点实验室(长江大学)开放基金资助项目(K2018-04)和湖北省高等教育综合奖补资金资助项目共同资助。