Study On Marine Controlled-Source Electromagnetic Data Preprocessing Technology

With the decreasing oil and gas resources on land,countries in the world have gradually shifted the focus of oil and gas exploration to deep sea areas.The Marine Controlled-Source Electromagnetic Method(MCSEM)is a method which uses the difference in electrical resistivity between a hydrocarbon reservoir and its surrounding deposits to determine whether or not the seabed contains a high-resistivity reservoir,and is widely used to evaluate the value of reservoirs prior to drilling.At present,the domestic research on MCSEM is still in its infancy,and the data preprocessing technology is not perfect.This paper summarized MCSEM data preprocessing processes such as data conversion,unit conversion,normalization,azimuth correction,noise suppression,merge channel data,and plot MVO curves.At the same time we focused on the data seawater disturbance noise suppression algorithm.Firstly,on the basis of detailed analysis of the characteristics of emission signals,received signals,and data noise during the exploration process,combined with the one-dimensional forward model Occam1 DCSEM,by setting typical exploration simulation parameters,a variety of subsea formation models are constructed to calculate MCSEM theoretical signals.Analyze the influence of different emission frequencies,field components and high resistance layers on the theoretical signal.At the same time,the simulated seawater disturbance noise is superimposed on the theoretical signal to form a noisy MCSEM data model.Then,the similarity between the MCSEM signal and the wavelet filter of the construction filter bank is proposed as the constraint condition.The method of solving the key parameters of the filter bank is designed based on the Particle Swarm Optimization(PSO)algorithm.And then we get the low-pass decomposition and reconstruction filter coefficients.Finally,we analyzed the effects of two types of seawater disturbance noise suppression on the simulated noise-containing data and the measured data.The results show that the method has a good suppression effect on the seawater disturbance noise,both for the simulated data and the measured data.The signal-to-noise ratio and mean-square error of the simulated noise-containing data after denoising are significantly improved,the data interpretation accuracy is improved,the noise level after the measured data denoising is reduced,and the offset range of the data interpretation is increased.And the filter bank constructed in this paper is superior to the traditional mother wavelets in terms of data edge processing and noise suppression.After verification,the preprocessing process method and the wavelet filter bank constructed in this paper can significantly improve the signal-to-noise ratio of MCSEM data and the offset range of data interpretation to ensure the accuracy of data retrieval in later period,and has important reference value for improving MCSEM data preprocessing technology.