Tengfei Lin

Tengfei Lin

Auto-generated profile, based on publicly available information

Tengfei Lin

CNPC 2 , Department of Middle East E&P, RIPED, Beijing, .

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
18 August 2019 in Geophysics
Verified
Journal Article
Closed access
HWHao Wu
Hao Wu
BZBo Zhang
Bo Zhang
TLTengfei Lin
Tengfei Lin
DCDanping Cao
Danping Cao
YLYihuai Lou
Yihuai Lou
The seismic horizon is a critical input for the structure and stratigraphy modeling of reservoirs. It is extremely hard to automatically obtain an accurate horizon interpretation for seismic data in which the lateral continuity of reflections is interrupted by faults and unconformities. The process of seismic horizon interpretation can be viewed as segmenting the seismic traces into different parts and each part is a unique object. Thus, we have considered the horizon interpretation as an object detection problem. We use the encoder-decoder convolutional neural network (CNN) to detect the “objects” contained in the seismic traces. The boundary of the objects is regarded as the horizons. The training data are the seismic traces located on a user-defined coarse grid. We give a unique training label to the time window of seismic traces bounded by two manually picked horizons. To efficiently learn the waveform pattern that is bounded by two adjacent horizons, we use variable sizes for the convolution filters, which is different than current CNN-based image segmentation methods. Two field data examples demonstrate that our method is capable of producing accurate horizons across the fault surface and near the unconformity which is beyond the current capability of horizon picking method.

1 September 2019 in Geophysics
Verified
Journal Article
Closed access
HWHao Wu
Hao Wu
BZBo Zhang
Bo Zhang
TLTengfei Lin
Tengfei Lin
FLFangyu Li
Fangyu Li
NLNaihao Liu
Naihao Liu
Seismic noise attenuation is an important step in seismic data processing. Most noise attenuation algorithms are based on the analysis of time-frequency characteristics of the seismic data and noise. We have aimed to attenuate white noise of seismic data using the convolutional neural network (CNN). Traditional CNN-based noise attenuation algorithms need prior information (the “clean” seismic data or the noise contained in the seismic) in the training process. However, it is difficult to obtain such prior information in practice. We assume that the white noise contained in the seismic data can be simulated by a sufficient number of user-generated white noise realizations. We then attenuate the seismic white noise using the modified denoising CNN (MDnCNN). The MDnCNN does not need prior clean seismic data nor pure noise in the training procedure. To accurately and efficiently learn the features of seismic data and band-limited noise at different frequency bandwidths, we first decomposed the seismic data into several intrinsic mode functions (IMFs) using variational mode decomposition and then apply our denoising process to the IMFs. We use synthetic and field data examples to illustrate the robustness and superiority of our method over the traditional methods. The experiments demonstrate that our method can not only attenuate most of the white noise but it also rejects the migration artifacts.

1 July 2019 in Geophysics
Verified
Journal Article
Closed access
YLYihuai Lou
Yihuai Lou
BZBo Zhang
Bo Zhang
RWRuiqi Wang
Ruiqi Wang
TLTengfei Lin
Tengfei Lin
DCDanping Cao
Danping Cao
Faults in the subsurface can be an avenue of, or a barrier to, hydrocarbon flow and pressure communication. Manual interpretation of discontinuities on 3D seismic amplitude volume is the most common way to define faults within a reservoir. Unfortunately, 3D seismic fault interpretation can be a time-consuming and tedious task. Seismic attributes such as coherence help define faults, but suffer from “staircase” artifacts and nonfault-related stratigraphic discontinuities. We assume that each sample of the seismic data is located at a potential fault plane. The hypothesized fault divides the seismic data centered at the analysis sample into two subwindows. We then compute the coherence for the two subwindows and full analysis window. We repeat the process by rotating the hypothesized fault plane along a set of user-defined discrete fault dip and azimuth. We obtain almost the same coherence values for the subwindows and the full window if the analysis point is not located at a fault plane. The “best” fault plane results in maximum coherence for the subwindows and minimum coherence for the full window if the analysis point is located at a fault plane. To improve the continuity of the fault attributes, we finally smooth the fault probability attribute along the estimated fault plane. We illustrate the effectiveness of our workflow by applying it to a synthetic and two real seismic data. The results indicate that our workflow successfully produces a continuous fault attribute without staircase artifacts and stratigraphic discontinuities.

3 July 2019 in Geophysics
Verified
Journal Article
Closed access
YLYihuai Lou
Yihuai Lou
BZBo Zhang
Bo Zhang
TLTengfei Lin
Tengfei Lin
NLNaihao Liu
Naihao Liu
HWHao Wu
Hao Wu
DCDanping Cao
Danping Cao
Seismic volumetric dip and azimuth are widely used in assisting seismic interpretation to depict geologic structures such as chaotic slumps, fans, faults, and unconformities. Current popular dip and azimuth estimation methods include the semblance-based multiple window scanning (MWS) method and gradient structure tensor (GST) analysis. However, the dip estimation accuracy using the semblance scanning method is affected by the dip of seismic reflectors. The dip estimation accuracy using the GST analysis is affected by the analysis window centered at the analysis point. We have developed a new algorithm to overcome the disadvantages of dip estimation using MWS and GST analysis by combining and improving the two methods. The algorithm first obtains an estimated “rough” dip and azimuth for reflectors using the semblance scanning method. Then, the algorithm defines a window that is “roughly” parallel to the local reflectors using the estimated rough dip and azimuth. The algorithm next estimates the dip and azimuth of the reflectors within the analysis window using GST analysis. To improve the robustness of GST analysis to noise, we used analytic seismic traces to compute the GST matrix. The algorithm finally uses the Kuwahara window strategy to determine the dip and azimuth of local reflectors. To illustrate the superiority of this algorithm, we applied it to the F3 block poststack seismic data acquired in the North Sea, Netherlands. The comparison indicates that the seismic volumetric dips estimated using our method more accurately follow the local seismic reflectors than the dips computed using GST analysis and the semblance-based MWS method.

26 November 2019 in Geophysics
Verified
Journal Article
Closed access
YLYihuai Lou
Yihuai Lou
BZBo Zhang
Bo Zhang
TLTengfei Lin
Tengfei Lin
DCDanping Cao
Danping Cao
Seismic horizons are the compulsory inputs for seismic stratigraphy analysis and 3D reservoir modeling. Manually interpreting horizons on thousands of vertical seismic slices of 3D seismic survey is a time-consuming task. Automatic horizon interpreting algorithms are usually based on the seismic reflector dip. However, the estimated seismic reflector dip is usually inaccurate near and across geologic features such as unconformities. We are determined to improve the quality of picked horizons using multiple seismic attributes. We assume that seismic horizons follow the reflector dip and that the same horizons should have similar instantaneous phase values. We first generate horizon patches using a reflector dip attribute, which is similar to current methods. We use seismic coherence attribute as the stop criteria for tracking the horizon within each patch. Considering the inaccuracy of reflector dip estimates at and near the discontinuous structures such as fault and unconformities, we use the seismic instantaneous phase attribute to improve the quality of the generated horizon patches. We generate horizons by merging the residual horizon patches and only outputting the best horizon in each iteration. Our method is capable of generating a horizon for each reflection within the 3D seismic survey, and the generated horizons strictly follow the seismic reflections over the whole seismic survey. Finally, each time sample of seismic traces is assigned a chronostratigraphic relative geologic time value according to the tracked horizons.

24 March 2021 in Seismological Research Letters
Verified
Journal Article
Closed access
PDF file
YYYuefeng Yan
Yuefeng Yan
CSChengyu Sun
Chengyu Sun
TLTengfei Lin
Tengfei Lin
JWJiao Wang
Jiao Wang
JYJidong Yang
Jidong Yang
DWDunshi Wu
Dunshi Wu
In exploration and earthquake seismology, most sources used in subsurface structure imaging and rock property estimation are fixed in certain positions. Continuously moving seismic sources, such as vehicles and the metro, are one kind of important passive sources in ambient noise research. Commonly, seismic data acquisition and processing for moving sources are based on the assumption of simple point passive sources, and the dispersion curve inversion is applied to constrain near-surface velocity. This workflow neglects the Doppler effects. Considering the continuously moving properties of the sources, we first derive the analytical solution for the Rayleigh waves excited by heavy vehicles and then analyze their Doppler effects and dispersion curves. We observe that the moving source data have the Doppler effect when compared with the changes in the frequency of the source intensity, but this effect does not affect the frequency dispersion of Rayleigh waves. The dispersion curves computed for moving source records are consistent with the analytical dispersion solutions, which provide a theoretical foundation for velocity estimation using moving source data.