Acoustic impedance inversion in coal strata using the priori constraint-based TCN-BiGRU method

Abstract

Acoustic impedance inversion is a key technique for the seismic exploration of coalfield, which can determine subsurface lithological changes and coal seam distribution. The traditional method is highly subjective, has poor generalizability, and interpretation can be time and labor consuming. Due to the powerful nonlinear interpretation and feature extraction capabilities of neural networks, deep learning technology has demonstrated potential for geophysical exploration. To predict acoustic impedance accurately and efficiently, this study proposes the use of the initial geological model as the priori constraint for training. The low-frequency feature extraction capability of a bidirectional gated recurrent unit network and the high-frequency feature extraction capability of a temporal convolutional network are used to establish a new acoustic impedance inversion method in coal strata with a priori constraint data. The temporal convolutional network-bidirectional gated recurrent unit method was applied to data from the Xinjing Mining Area in Shanxi province, northern China. The results displayed good precision by accurately predicting the distribution and thickness variation of local coal seams. Compared with the traditional model-based method and the method using temporal convolutional network-bidirectional gated recurrent unit network, the proposed priori constraint-based temporal convolutional network-bidirectional gated recurrent unit network has better feature expression capability and provides more detailed coal seam information. In conclusion, the new method can improve the accuracy of acoustic impedance inversion, which is of great significance for coalfield seismic exploration.

Document Type: Original article

Cited as: Shi, S., Qi, Y., Chang, W., Li, L., Yao, X., Shi, J. Acoustic impedance inversion in coal strata using the priori constraint-based TCN-BiGRU method. Advances in Geo-Energy Research, 2023, 9(1): 13-24. https://doi.org/10.46690/ager.2023.07.03

References

Abdi, H., Williams, L. J. Normalizing data, in Encyclopedia of Research Design, edited by N. J. Salkind, SAGE Publications, State of California, 2010.

Adler, A., Araya-Polo, M., Poggio, T. Deep learning for seismic inverse problems: Toward the acceleration of geophysical analysis workflows. IEEE Signal Processing Magazine, 2021, 38(2): 89-119.

Alfarraj, M., AlRegib, G. Semi-supervised learning for acoustic impedance inversion. Paper Presented at the SEG International Exposition and Annual Meeting, San Antonio, USA, 15-20 September, 2019.

Biswas, R., Sen, M. K., Das, V., et al. Prestack and poststack inversion using a physics-guided convolutional neural network. Interpretation, 2019, 7(3): SE161-SE174.

Chen, H., Gao, J., Zhang, W., et al. Seismic acoustic impedance inversion via optimization-inspired semisupervised deep learning. IEEE Transactions on Geoscience and Remote Sensing, 2021a, 60: 111.

Chen, T., Zhang, Z., Zhou, X., et al. Prediction model of coalbed methane content based on well logging parameter optimization. Coal Geology & Exploration, 2021b, 49(3): 227-235. (in Chinese)

Das, V., Pollack, A., Wollner, U., et al. Convolutional neural network for seismic impedance inversion. Geophysics, 2019, 84(6): R869-R880.

Jais, I. K. M., Ismail, A. R., Nisa, S. Q. Adam optimization algorithm for wide and deep neural network. Knowledge Engineering and Data Science, 2019, 2(1): 41-46.

Janiesch, C., Zschech, P., Heinrich, K. Machine learning and deep learning. Electronic Markets, 2021, 31(3): 685-695.

Li, Y. E., O’Malley, D., Beroza, G., et al. Machine learning developments and applications in solid-earth geosciences: Fad or future? Journal of Geophysical Research: Solid Earth, 2023, 128(1): e2022JB026310.

Marques, C. R., dos Santos, V. G., Lunelli, R., et al. Analysis of deep learning neural networks for seismic impedance inversion: A benchmark study. Energies, 2022, 15(20): 7452.

Martin, G. S., Wiley, R., Marfurt, K. J. Marmousi2: An elastic upgrade for Marmousi. The Leading Edge, 2006, 25(2): 156-166.

Mustafa, A., Alfarraj, M., AlRegib, G. Estimation of acoustic impedance from seismic data using temporal convolutional network. Paper Presented at the SEG International Exposition and Annual Meeting, San Antonio, USA, 15-20 September, 2019.

Mustafa, A., Alfarraj, M., AlRegib, G. Joint learning for spatial context-based seismic inversion of multiple data sets for improved generalizability and robustness. Geophysics, 2021, 86(4): O37-O48.

Shen, Y., Ma, Y., Deng, S., et al. An ensemble model based on deep learning and data preprocessing for short-term electrical load forecasting. Sustainability, 2021, 13(4): 1694.

Smith, R., Nivlet, P., Alfayez, H., et al. Robust deep learningbased seismic inversion workflow using temporal convolutional networks. Interpretation, 2022, 10(2): SC41-SC55.

Song, L., Yin, X., Zong, Z. Deep learning seismic impedance inversion based on prior constraints. Oil Geophysical Prospecting, 2021, 56(4): 716-727. (in Chinese)

Tao, L., Ren, H., Gu, Z. Acoustic impedance inversion from seismic imaging profiles using self attention U-Net. Remote Sensing, 2023, 15(4): 891.

Wang, D., Chen, G. Seismic wave impedance inversion based on temporal convolutional network. Earth Science, 2022, 47(4): 1492-1506. (in Chinese)

Wang, H., Yan, J., Fu, G., et al. Current status and application prospect of deep learning in geophysics. Progress in Geophysics, 2020, 35(2): 642-655. (in Chinese)

Wang, B., Yin, X., Zhang, F. Elastic impedance inversion and its application. Progress in Geophysics, 2005, 20(1): 89-92. (in Chinese)

Wu, B., Meng, D., Wang, L., et al. Seismic impedance inversion using fully convolutional residual network and transfer learning. IEEE Geoscience and Remote Sensing Letters, 2020, 17(12): 2140-2144.

Wu, B., Meng, D., Zhao, H. Semi-supervised learning for seismic impedance inversion using generative adversarial networks. Remote Sensing, 2021a, 13(5): 909.

Wu, X., Yan, S., Bi, Z., et al. Deep learning for multidimensional seismic impedance inversion. Geophysics, 2021b, 86(5): R735-R745.

Xie, C., Du, S., Wang, J., et al. Intelligent modeling with physics-informed machine learning for petroleum engineering problems. Advances in Geo-Energy Research, 2023, 8(2): 71-75.

Yang, W., Xia, K., Fan, S. Oil logging reservoir recognition based on TCN and SA-BiLSTM deep learning method. Engineering Applications of Artificial Intelligence, 2023, 121: 105950.

Yoo, J., Kim, D., Choi, J., et al. Impedance inversion based on domain adaptation technique with reconstruction. IEEE Geoscience and Remote Sensing Letters, 2022, 19: 1-5.

Yoon, D., Yeeh, Z., Byun, J. Seismic data reconstruction using deep bidirectional long short-term memory with skip connections. IEEE Geoscience and Remote Sensing Letters, 2020, 18(7): 1298-1302.

Yuan, S., Wang, S., Luo, Y., et al. Impedance inversion by using the low-frequency full-waveform inversion result as an a priori model. Geophysics, 2019, 84(2): R149-R164.

Zhang, T., Sun, L., Yang, Y., et al. A method of collapse column recognition based on improved residual network. Coal Geology & Exploration, 2023, 51(5): 171-179. (in Chinese)

Zhang, Y., Wen, X., Wu, H., et al. Seismic acoustic impedance inversion using Lp quasi-norm sparse constraint and alternating direction multiplier algorithm. Geophysical Prospecting for Petroleum, 2022, 61(5): 856-864.(in Chinese)

Zhao, D., Yang, N., Xiong, J., et al. Model-constrained and data-driven double-supervision acoustic impedance inversion. Petroleum Science, 2023, in press, https://doi.org/10.1016/j.petsci.2023.03.019.