Research On Noise Suppression Technology Of Marine Controlled-source Electromagnetic Data Based On VMD

There are abundant minerals,oil and gas and other strategic resources under the ocean.The exploration and development of these strategic resources has become the focus of attention of all countries in the world.Marine Controlled-Source Electromagnetic(MCSEM)method is a new type of geophysical exploration method,which plays an important role in oil and gas exploration,marine geological structure survey and marine environment survey.As we all know,the ocean exploration process is seriously affected by the weather,the construction is complicated,the operation cycle is long and the risk is high,the exploration cost is high,and the data is hard to come by.However,the data is only the "root",and it must be processed and inverted before the "results" become valuable data reflecting the characteristics of underground geological structures or revealing the mysteries of deep underground.Therefore,the processing of MCSEM data,especially noise suppression,is of great significance to improve the data quality and the accuracy of inversion interpretation in the later stage.Firstly,this article is based on the MCSEM exploration principle,combined with the analysis of measured data,to study the statistical characteristics of MCSEM data and noise;Based on the one-dimensional isotropic seafloor stratum model,the MCSEM theoretical data is forward calculated,and the MCSEM random noise model is constructed according to the characteristics of random noise.Secondly,the MCSEM data random noise suppression method based on variational modest decomposition(VMD)is designed.The method is: using the hybridized particle swap algorithm to optimize the VMD parameters,the VMD decomposition MCSEM data optimized by the parameter is obtained by VMD components of the center frequency of data base frequency or the primary harmonic;based on soft threshold function and Stein unbiased risk estimate.The estimated threshold is compressed with respective components,and the MCSEM data that is suppressed by compression random noise is reconstructed by each component.Finally,the method is compared with the traditional db8 wavelet method and EMD method,and the MCSEM simulation data and the measured data are used,and the noise suppression experiment is performed,and the effectiveness of this method is verified.The results show that the method in this paper can effectively suppress the random noise of MCSEM data,and is better than the noise suppression effect of the traditional db8 wavelet method and EMD method.It can increase the offset range of data interpretation and improve the signal-to-noise ratio of the data,which has research significance and practical value for improving the accuracy of data inversion in the later stage.