Research On Surface Microseismic Noise Suppression And Source Position Inversion

With the developments of unconventional oil and gas fields in China,microseismic monitoring,as an important exploration technology,has received much attention from scholars.It is worth noting that the surface microseismic monitoring has gained more recognition due to its low cost,wide azimuth coverage,flexible geophone placement,and the increasing performance of surface microseismic technology,computer technology,and geophone.In this paper,we focus on two important steps of the surface microseismic data processing including microseismic data noise suppression and microseismic source location.First,the microseismic monitoring technology is briefly described,and the characteristics and noise types of the surface microseismic data are analyzed.Furthermore,the method of synthesizing microseismic data by the convolution model is introduced,and the surface microseismic noise suppression method and microseismic source location method are discussed.In particular,the microseismic source location methods are divided into two types including the one based on walking time and the other one based on offset and imaging.Moreover,the advantages and disadvantages of the two methods are analyzed,and it is concluded that the method based on the offset and imaging is more suitable for solving the microseismic source location problem with low signal-to-noise ratio data.Then,the denoising convolutional neural network(Dn CNN)is proposed to apply for microseismic noise suppression,and the problem of deeper structure of this network with high training difficulty is improved.Moreover,based on the convolution model,the simulation experiments are carried out to synthesize microseismic data using the Ricker wavelets as microseismic subwaves,and the effectiveness of the improved Dn CNN in noise suppression is verified by practical data.Finally,the whale optimization algorithm(WOA)is put forward to accelerate the cross-correlation interference imaging for the problem of slow computation speed.In addition,as for the problem that WOA is easy to fall into local extremes optimization,three aspects of adaptive weighting,boundary processing and Levy-based flight behavior are improved and the efficiency and accuracy of location are promoted.Furthermore,the improved WOA(LWOA)is accelerated for the cross-correlation interference imaging.At last,a comparison experiment is conducted using the microseismic data before and after noise suppression in the simulation experiment of the improved Dn CNN to accelerate the improved WOA for cross-correlation interference imaging.The effectiveness of the improved Dn CNN network for noise suppression of microseismic data and the advantage of the improved WOA for acceleration of cross-correlation interference imaging are also verified.