Algorithm for Extraction of Reflection Waves in Single-Well Imaging Based on MC-ConvTasNet
Source abstract
Abstract
Single-well imaging makes use of reflected waves to image geological structures outside a borehole, with a detection distance expected to reach tens of meters. However, in the received full wave signal, reflected waves have much smaller amplitudes than borehole-guided waves, which travel directly through the borehole. To obtain clear reflected waves, we use a deep neural network, the multi-channel convolutional time-domain audio separation network (MC-ConvTasNet), to extract reflected waves. In the signal channels of the common-source gather, there exists a notable arrival time difference between direct waves and reflected waves. Leveraging this characteristic, we train MC-ConvTasNet on the common-source gathers, ultimately achieving satisfactory results in wave separation. For the hard-to-hard single-interface, soft-to-hard single-interface and double-interface models, the reflected waves extracted by MC-ConvTasNet are closer to the theoretical reflected waves in both phase and shape (the average scale-invariant signal-to-distortion ratio exceeds 32 dB) than those extracted by parameter estimation, a median filter and an F-K filter. Meanwhile, MC-ConvTasNet naturally fits in the scenarios of various inclined interfaces and interfaces parallel to the borehole axis. As an application, our method is employed on field logging data and its ability to separate waves is verified.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Seismic Imaging and Inversion Techniques · Drilling and Well Engineering
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Mechanics, Harbin Institute of Technology, Harbin 150001, China
Funding
National Natural Science Foundation of China
References (43)
Show source-provided references
1. Imaging near-borehole structure using directional acoustic-wave measurement10.1190/1.1836812 ↗
2. Acoustic imaging of reservoir structure from a horizontal well10.1190/1.1438075 ↗
3. Hu, H., Xu, J., and Wang, Z. (2019, January 1–4). Analytical Simulation of the Waves Reflected From a Distant Slanted Interface Outside a Borehole. Proceedings of the 2019 13th Symposium on Piezoelectrcity, Acoustic Waves and Device Applications (SPAWDA), Shijiazhuang, China.10.1109/SPAWDA.2019.8681820 ↗
4. Combination of FDTD with Analytical Methods for Simulating Elastic Scattering of 3-D Objects Outside a Fluid-Filled Borehole10.1109/TGRS.2020.3011124 ↗
5. Elastic Reverse Time Migration Method for Single-Well Imaging and Elimination of Azimuthal Ambiguity With a Combined Receiver System10.1109/TGRS.2023.3328032 ↗
6. Borland, W., Holderson, J., Rougha, H., Sultan, A., Chakravorty, S., and Al-Raisi, M. (2005, January 11). Integration of Microelectrical and Sonic Reflection Imaging Around the Borehole-Offshore UAE. Proceedings of the IPTC International Petroleum Technology Conference, Doha, Qatar.10.2523/IPTC-11021-MS ↗
7. Wavefield separation for borehole acoustic reflection survey using parametric decomposition and waveform inversion10.1190/geo2018-0330.1 ↗
8. Processing array acoustic-logging data to image near-borehole geologic structures10.1190/1.2435083 ↗
9. Imaging of near-borehole structure using full-waveform sonic data10.1190/1.1442702 ↗
10. High-resolution Radon transforms for improved dipole acoustic imaging10.1111/1365-2478.12434 ↗
11. Extraction of reflected events from sonic-log waveforms using the Karhunen-Loève transform10.1190/geo2017-0031.1 ↗
12. Extracting Reflected Waves From Acoustic Logging Data Based on the Shearlet Transform10.1109/LGRS.2019.2908286 ↗
13. Wang, Q., Wang, H., Fang, Z., Xu, D., Qi, X., and Yu, Y. (2024, January 18–22). Weak Reflection Extraction in Borehole Acoustic Reflection Imaging Using an Unsupervised Machine Learning Method. Proceedings of the SPWLA Annual Logging Symposium, Rio de Janeiro, Brazil.
14. Li, Y., Zhou, R., Tang, X.M., Jackson, J., and Patterson, D. (2002, January 27–30). Single-Well Imaging with Acoustic Reflection Survey at Mounds, Oklahoma, USA. Proceedings of the 64th EAGE Conference & Exhibition, Florence, Italy.10.3997/2214-4609-pdb.5.P141 ↗
15. A Consolidated Perspective on Multimicrophone Speech Enhancement and Source Separation10.1109/TASLP.2016.2647702 ↗
16. Hershey, J.R., Chen, Z., Le Roux, J., and Watanabe, S. (2016, January 20–25). Deep clustering: Discriminative embeddings for segmentation and separation. Proceedings of the 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Shanghai, China.10.1109/ICASSP.2016.7471631 ↗
17. Jansson, A., Humphrey, E.J., Montecchio, N., Bittner, R.M., Kumar, A., and Weyde, T. (2017, January 23–27). Singing Voice Separation with Deep U-Net Convolutional Networks. Proceedings of the International Society for Music Information Retrieval Conference, Suzhou, China.
18. Multitalker Speech Separation With Utterance-Level Permutation Invariant Training of Deep Recurrent Neural Networks10.1109/TASLP.2017.2726762 ↗
19. Supervised Speech Separation Based on Deep Learning: An Overview10.1109/TASLP.2018.2842159 ↗
20. Luo, Y., and Mesgarani, N. (2018, January 15–20). TasNet: Time-Domain Audio Separation Network for Real-Time, Single-Channel Speech Separation. Proceedings of the 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Calgary, AB, Canada.10.1109/ICASSP.2018.8462116 ↗
21. Stoller, D., Ewert, S., and Dixon, S. (2018). Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separation. arXiv.
22. Conv-TasNet: Surpassing Ideal Time–Frequency Magnitude Masking for Speech Separation10.1109/TASLP.2019.2915167 ↗
23. A deep learning workflow for weak reflection extraction in pitch-catch measurements in the cased hole10.1190/geo2022-0243.1 ↗
24. Gu, R., Zhang, S.X., Chen, L., Xu, Y., Yu, M., Su, D., Zou, Y., and Yu, D. (2020, January 4–8). Enhancing End-to-End Multi-Channel Speech Separation Via Spatial Feature Learning. Proceedings of the ICASSP 2020–2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Barcelona, Spain.10.1109/ICASSP40776.2020.9053092 ↗
25. A survey of sound source localization with deep learning methods10.1121/10.0011809 ↗
26. Acoustic multipole sources in fluid-filled boreholes10.1190/1.1442028 ↗
27. Numerical evaluation of the transient acoustic waveform due to a point source in a fluid-filled borehole10.1190/1.1440932 ↗
28. Solutions of P-SV and SV-P waves in single-well imaging through reciprocity relations10.1190/geo2019-0551.1 ↗
29. Luo, Y., Chen, Z., and Yoshioka, T. (2020, January 4–8). Dual-Path RNN: Efficient Long Sequence Modeling for Time-Domain Single-Channel Speech Separation. Proceedings of the ICASSP 2020–2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Barcelona, Spain.10.1109/ICASSP40776.2020.9054266 ↗
30. Shear-wave radiation, reception, and reciprocity of a borehole dipole source: With application to modeling of shear-wave reflection survey10.1190/geo2013-0096.1 ↗
31. Aki, K., and Richards, P.G. (2002). Quantitative Seismology: Theory and Methods, University Science Books. [2nd ed.].
32. Reciprocity relations for the elastodynamic fields generated by multipole sources in a fluid–solid configuration10.1093/gji/ggv308 ↗
33. Asymptotic solution to a 3D dipole single-well imaging system with combined monopole and dipole receivers with an application in elimination of azimuth ambiguity10.1190/geo2018-0658.1 ↗
34. Relationships between compressional-wave and shear-wave velocities in clastic silicate rocks10.1190/1.1441933 ↗
35. Velocity regularities of p and s-waves in formations
36. Relationship among conpressional wave, shear wave velocities and density in rocks
37. On Loss Functions for Supervised Monaural Time-Domain Speech Enhancement
38. Roux, J.L., Wisdom, S., Erdogan, H., and Hershey, J.R. (2019, January 12–17). SDR—Half-baked or Well Done?. Proceedings of the ICASSP 2019—2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brighton, UK.10.1109/ICASSP.2019.8683855 ↗
39. Kingma, D.P., and Ba, J. (2014). Adam: A Method for Stochastic Optimization. arXiv.
40. Migration of seismic data by phase shift plus interpolation10.1190/1.1441643 ↗
41. Sridhar, K., Cutler, R., Saabas, A., Parnamaa, T., Loide, M., Gamper, H., Braun, S., Aichner, R., and Srinivasan, S. (2021, January 6–11). ICASSP 2021 Acoustic Echo Cancellation Challenge: Datasets, Testing Framework, and Results. Proceedings of the ICASSP 2021—2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Toronto, ON, Canada.10.1109/ICASSP39728.2021.9413457 ↗
42. Chapter Four—Seismic signal augmentation to improve generalization of deep neural networks
43. Yan, Z., Zhang, Z., and Liu, S. (2021). Improving Performance of Seismic Fault Detection by Fine-Tuning the Convolutional Neural Network Pre-Trained with Synthetic Samples. Energies, 14.10.3390/en14123650 ↗