| The effective bandwidth of seismic data plays an important role in enhancing resolution.In the face of increasing exploration accuracy,if only focusing on the high-frequency extension we can’t meet current demand,so we must pay more attention to low-frequency extension while enhancing resolution.Low-frequency signal not only affects resolution of seismic data,but also plays an important role in improving quality of inversion.Full waveform inversion can make full use of the kinematic and dynamic information of seismic waves and has the potential to reveal the structure and reservoir properties in complex geological background.However,the lack of ultra-low frequency data will lead to the dependence of full waveform inversion on the initial model.Full waveform inversion based on low-frequency extended data is one of methods to solve this problem.This paper includes two main contents: full-band extension of seismic data and full waveform inversion based on low-frequency extended data.In the first part of this paper,the full-band extension filter is used to extend bandwidth of post-stack seismic data.We achieve low-frequency extension and resolution enhancement through three steps: wavelet extraction,filter construction and data filtering.In this part we use Fourier transform theorem to extend the bandwidth of constant phase wavelet and construct the full-bandwidth extended filter in frequency domain by using extended wavelet and extraction wavelet.Finally,we demonstrate the advantages of the filter in extending low frequency and enhancing resolution through profile comparison and spectrum analysis.When using the full-band extension filter on prestack data,we need to maintain AVO relation of shot gather in frequency extension process.In the second part of this paper,we use generalized wavelet to construct filter to extend low frequency of prestack seismic data,and then the data after filtering will be as input data of full waveform inversion,which solves the problem of inversion instability caused by the lack of low-frequency data in traditional full waveform inversion.Finally,we use Foothill model to compare and analyze the inversion results.The result show that full waveform inversion with extension of low frequency is more stable and reliable than inversion without low-frequency extension. |
