Research On Seismic High Resolution Processing Based On Multi-band Fusion Reconstruction

With the transformation of domestic oil and gas exploration target from tectonic reservoir to complex lithological gas reservoir,this presents new requirements for exploration technology.Due to the small scale and thickness of rock oil and gas reservoir,the seismic reflection characteristics are relatively weak same-phase axial amplitude,easily submerged in the background noise,and becomes an effective signal that is not easy to identify.In order to better obtain the high-resolution data,this paper studies the frequency domain information compensation method based on the harmonic extrapolation method and the compressed sensing method to compensate for the missing high and low-frequency components of the original seismic data to obtain the high-resolution seismic data.In order to ensure the smooth development of seismic data frequency band compensation research,this paper first introduces the basic knowledge of seismic data resolution,analyzes the factors affecting the data resolution,compares the effects of missing high and low frequency components on the resolution,and then analyzes the absorption effect of the formation,the limitation of seismic signal reception,and the signal to noise ratio.The high frequency composition of the seismic data affects the overall resolution of the data.The lack of high frequency composition will lead to the significantly reduced refinement of the data,and the thin layers and other geological structures are difficult to be identified.Through harmonic extrapolation,the missing components in the frequency domain response characteristics of the data itself are compensated to obtain higher accurate seismic data.Restricted by the field acquisition equipment,seismic information below 8Hz is often missing.Low frequency compensation for seismic data can reduce the side lobe of the wavelet,thus improving the resolution of seismic data,which also plays an important role in the formation division.The compression sensing method can reconstruct the wide band data in the sparse domain to obtain the missing low frequency.The research analysis believes that the frequency band of seismic data is affected and limited by various factors,which leads to reduced resolution and brings difficulties to the subsequent inversion interpretation.According to the loss of different frequency band signals,two seismic data processing methods of high frequency compensation based on harmonic extrapolation and low frequency compensation based on compression sensing are formed for different frequency bands.The results of model test and actual data show that the band compensation method can effectively compensate the missing high and low band data through the fusion of different band data,so as to improve data resolution and reliability.