Improvements in time domain FWI and its applications

Yoon, Kang Sup; Suh, Sang; Cai, James; Wang, Bin

doi:10.1190/segam2012-1535.1

Cited by 12 publications

(7 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The coarsely sampled active-source data (Mirandola example) and the recorded ambient-noise data (SW HUB example and USArray example) are inputs for the interferometric processing. We also convolve the interferometric Green's functions with a known source wavelet (Yoon et al, 2012) and do a 3D-to-2D phase correction (Pica et al, 1990) to avoid the source wavelet estimation and adopt a 2D rather than 3D inversion. After data preprocessing, we first transform the common-shot-gathers (t-x domain) to the frequency domain (f-x domain) and calculate their f-v spectra using the linear Radon transform (equation A-4).…”

Section: Methodsmentioning

confidence: 99%

Rayleigh Wave Dispersion Spectrum Inversion Across Scales

Zhang

Saygin

et al. 2021

Surv Geophys

View full text Add to dashboard Cite

show abstract

Section: Methodsmentioning

confidence: 99%

Rayleigh Wave Dispersion Spectrum Inversion Across Scales

Zhang

Saygin

et al. 2021

Surv Geophys

View full text Add to dashboard Cite

show abstract

“…A modified free-surface boundary condition, which can suppress strong surface waves, is used in the simulation (He et al, 2016). We convolve the observed data with the half-order differentiation of the known wavelet, and thus, we avoid source estimation, while correcting the phase discrepancy between the 3D acquisition and the 2D simulation (Pica et al, 1990;Yoon et al, 2012). The initial model is a 1D…”

Section: Numerical Examplementioning

confidence: 99%

High-resolution reservoir characterization using deep learning-aided elastic full-waveform inversion: The North Sea field data example

Zhang

Alkhalifah

2020

GEOPHYSICS

View full text Add to dashboard Cite

Reservoir characterization is an essential component of oil and gas production, as well as exploration. Classic reservoir characterization algorithms, deterministic and stochastic, are typically based on stacked images and rely on simplifications and approximations to the subsurface (e.g., assuming linearized reflection coefficients). Elastic full-waveform inversion (FWI), which aims to match the waveforms of prestack seismic data, potentially provides more accurate high-resolution reservoir characterization from seismic data. However, FWI can easily fail to characterize deep-buried reservoirs due to illumination limitations. We have developed a deep learning-aided elastic FWI strategy using observed seismic data and available well logs in the target area. Five facies are extracted from the well and then connected to the inverted P- and S-wave velocities using trained neural networks, which correspond to the subsurface facies distribution. Such a distribution is further converted to the desired reservoir-related parameters such as velocities and anisotropy parameters using a weighted summation. Finally, we update these estimated parameters by matching the resulting simulated wavefields to the observed seismic data, which corresponds to another round of elastic FWI aided by the a priori knowledge gained from the predictions of machine learning. A North Sea field data example, the Volve Oil Field data set, indicates that the use of facies as prior knowledge helps resolve the deep-buried reservoir target better than the use of only seismic data.

show abstract

“…It is known that field data contain quite high‐frequency components, and the source wavelet is important for imaging and inversion. Therefore, instead of estimating the source wavelet from the field data and to avoid numerical dispersion, a 15 Hz Ricker wavelet is convolved with the original observed data (Yoon et al, ). In addition to this, it is then used for inversion.…”

Section: Numerical Examplesmentioning

confidence: 99%

Wave‐Equation Migration Velocity Analysis Using Radon‐Domain Common‐Image Gathers

Liu

2020

JGR Solid Earth

View full text Add to dashboard Cite

The accuracy of the background velocity analysis is critical to image the subsurface structure in seismology. Wave‐equation migration velocity analysis using source‐domain common‐image gathers requires picking the relative moveouts based on semblance analysis. However, the picking process is complex and time‐consuming. To avoid picking and maintain efficiency in generating CIGs, an automatic wave‐equation migration velocity analysis method is proposed utilizing the focusing properties of the Radon‐domain common‐image gathers. Seismic events in the Radon‐domain common‐image gathers appear focused if an accurate migration velocity model is used; otherwise, the resultant events will be unfocused. The objective function is defined in the Radon domain to measure focus of the common‐image gathers automatically. The proposed migration velocity analysis method links the defocusing information to the migration velocity update under the assumption that model perturbations only result in slope shifts in the common‐image gathers. This method provides a reasonably accurate background velocity model for imaging and full‐waveform inversion without requiring an accurate initial velocity model. Tests on synthetic and field data demonstrate the effectiveness of the method. This method is tolerable to both inaccurate initial velocity models and the lack of low‐frequency data.

show abstract

Improvements in time domain FWI and its applications

Cited by 12 publications

References 4 publications

Rayleigh Wave Dispersion Spectrum Inversion Across Scales

Rayleigh Wave Dispersion Spectrum Inversion Across Scales

High-resolution reservoir characterization using deep learning-aided elastic full-waveform inversion: The North Sea field data example

Wave‐Equation Migration Velocity Analysis Using Radon‐Domain Common‐Image Gathers

Contact Info

Product

Resources

About