Wave Impedance Inversion Based On Frequency Division And Topology Technology

Improving the resolution of seismic data has always been one of the urgent tasks of the oil and gas exploration industry,and it is also an important direction for geophysical workers to study.Based on the time-frequency domain analysis method,this paper focuses on the frequency-reconstructed wave impedance inversion method of well-controlled fast sparse S-transform and the principle of throttling technology based on compressed sensing,and the earthquake in a certain industrial area in southeastern China.The data processing improves the inversion accuracy and the resolution of seismic data,and achieves certain practical processing effects,which lays a foundation for subsequent reservoir prediction and has certain theoretical and practical significance.The frequency division inversion method of seismic data is often used for data with low signal-to-noise ratio.By making full use of the effective information of each frequency band,the resolution and inversion precision of seismic data are improved.Conventional wave impedance inversion is generally performed using full-band poststack data.Most of the inversion results are controlled by the dominant frequency of the seismic wave.The potential of the high-frequency and low-frequency parts of the effective band in the data is not fully utilized and excavated.The result is difficult to achieve the desired results.Compared with other time-frequency analysis methods,the fast sparse S-transform analysis method optimizes the sparse window parameters to adaptively adjust the window functions of different frequency components,reducing the time cost required to adjust the parameters.The calculated spectrum has the characteristics of high resolution and more energy aggregation.To this end,this paper uses a frequency-reconstructed wave impedance inversion method based on fast sparse S transform to segment the seismic data into data bodies of several different frequency bands and use the frequency-dividing curve constraint to invert to improve the pair.The recognition accuracy of the thin layer.Both the model trial and the actual data application results show that the method has good stability and can effectively improve the seismic inversion resolution.The topology method is often used for seismic data with narrow frequency bands and low resolution,which is difficult to effectively identify and describe reservoir information.Such methods can expand the frequency bandwidth of earthquakes and improve seismic resolution,which is of great significance for reservoir prediction and identification.Relative to high-frequency components,the absence of low-frequency information in the data can cause false high-resolution phenomena on the seismic section.The traditional topology method only deals with seismic wavelets without considering the changes in the wavelet structure.In this paper,combined with the emerging compressed sensing algorithm in recent years,the full-band fitting of seismic data is carried out,and the low-frequency part of the result is compensated into the seismic data,and the post-tuned seismic data is inversed by wave impedance.The theoretical model and actual data processing results show that the inversion wave impedance profile after the extension is more consistent with the well data and higher resolution,indicating that the method has better development potential in improving seismic data resolution and improving inversion accuracy.