This paper presents a workflow for near-surface velocity automatic estimation using the early arrivals of seismic data. This workflow comprises two methods, source-domain full traveltime inversion (FTI) and early-arrival waveform inversion. Source-domain FTI is capable of automatically generating a background velocity that can kinematically match the reconstructed plane-wave sources of early arrivals with true plane-wave sources. This method does not require picking first arrivals for inversion, which is one of the most challenging aspects of ray-based first-arrival tomographic inversion. Moreover, compared with conventional Born-based methods, source-domain FTI can distinguish between slower or faster initial model errors via providing the correct sign of the model gradient. In addition, this method does not need estimation of the source wavelet, which is a requirement for receiver-domain wave-equation velocity inversion. The model derived from source-domain FTI is then used as input to early-arrival waveform inversion to obtain the short-wavelength velocity components. We have tested the workflow on synthetic and field seismic data sets. The results show source-domain FTI can generate reasonable background velocities for early-arrival waveform inversion even when subsurface velocity reversals are present and the workflow can produce a high-resolution near-surface velocity model.
Presentation Date: Wednesday, September 27, 2017
Start Time: 9:20 AM
Presentation Type: ORAL
Lanjie, Jiang (China University of Petroleum) | Guangzhi, Zhang (China University of Petroleum) | Liu, Lu (China University of Petroleum) | Jiajie, Song (China University of Petroleum) | Zhonglin, Pei (China University of Petroleum)
Density can predict fluid saturation of reservoir and plays an important role in hydrocarbon interpretation. Due to the cross-talk effects between velocity and density, density is difficult to reconstruct in multi-parameter full waveform inversion. Based on the model test the analysis of the influence of density and velocity on wave propagation, we discover that density inversion result grows desirable while misfit of the initial velocity goes down slightly. Considering, multi-parameter FWI is capable of reconstructing a satisfactory velocity model. Therefore, the stepped multi-parameter FWI is performed which decouples the velocity and density during the reconstruction procedure. In the meantime, the L-BFGS method is carried out to update parameters iteratively cause it can search direction without storing the Hessian matrix approximation which save computation time and storage. Nonetheless, in the pre n times iterations of L-BFGS method, the energy of gradient is imbalance. Consequently, the pretreatment L-BFGS method is carried out based on the illumination analysis that improves the convergence property and the accuracy of inversion. The numerical examples testify the feasibility of our method that it can improve accuracy of result.
Presentation Date: Wednesday, October 19, 2016
Start Time: 2:20:00 PM
Location: Lobby D/C
Presentation Type: POSTER
In this paper, we present a new optimal second order traveltime formula based on Chebyshev polynomial and simulated annealing method in homogenous and VTI media. Compare with the high order equations based on Taylor expansion, this method not only reduces the calculation amount of traveltime through lowering the order of traveling-time formula, but also achieve almost the same accuracy with the high-order ones within the limit of mid-long offset. Then, we apply this formula in bend-ray Kirchhoff pre-stack time migration (PSTM) to test its correctness and accuracy. Finally, through the method verification of homogenous media, VTI media and field data, it is proved that this modification can cost less calculated amount of traveltime to acquire a comparatively same migration profile with the high order formulas.
Waveform inversion is a kind of method to reveal the underground structure and lithology information through minimizing the residual error between predicted wavefield and true seismic record using full-wavefield information. In this paper, we briefly state the principle of conventional Quasi-Newton algorithm, and then exploit a new modified Quasi-Newton equation to modify the conventional Davidon-Fletcher-Powell (DFP) and Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm (Zhang et al, 2001). Furthermore, we take BFGS for instance to implement a comparison between modified method and conventional one. Different from past Quasi-Newton methods, this modified one considers gradient value, model information and objective function value together to approximate the inverse matrix of Hessian matrix, which leads a fast convergence for inversion; moreover, it almost does not increase the calculation amount for each iteration. Finally, numerical experiment shows that compared with conventional Quasi-Newton method, modified BFGS algorithm can not only speed up convergence and decrease consuming time, but also preserve the inverse accuracy well.