井測資料是一個隨深度變化的連續性資料，在石油工程中需要完整的井測資料建立三維地質模型以訂定油氣田的開發計畫，然而，要取得完整的井測資料並不是一件容易的事情。測量出來的井測資料常常會遇到部分區間井測資料缺失的問題，其可能的原因有三個：測量儀器損壞、井眼條件不佳、或是因預算考量而無法施測。
　　然而，若有部分區間的井測資料缺失則會無法得知該區間段落的地層特性與資訊，讓本來可測量地層連續性資料優勢的井下測量無法發揮其最大效益。因此，重建、還原缺失區間的井測資料是一件非常重要的議題，在遇到井測資料缺失的狀況時，最直接的解決方法為重新鑽井或再次進行井下量測。但重新鑽井需要花費的成本太高，部分井測資料更需要在裸孔的環境下量測，在成本與經濟效益的考量下，上述的方法接無法實施。
　　為了解決井測資料區段缺失的問題，本研究將應用ADMM-ADAM架構來重建缺失段的井測資料組，在深度學習網絡架構下，包含Deep Image Prior以及Residual Encoder Decoder Networks來初步還原缺失段的井測資料，接著再透過凸優化算法求解井測還原問題，期能藉此提高井測資料的完整度，並模擬出缺失段的井測資料組，以對地層及儲集層能有更準確的解釋與判讀。
　　本研究成功的以影像領域中常用的二維卷積神經網絡來生成井測資料，彌補井測資料缺失之問題。而以歸一化互相關係數(Normalized Cross Correlation, NCC)來評估模型的生成結果時，歸一化互相關係數值皆達九成以上，代表透過ADMM-ADAM架構生成出的井測資料能與原始資料趨勢相符。本研究期在深度學習的演算法的增益下，補遺井測資料的缺失，並提高其在資源探勘及解釋上的應用。

Well Log data are geophysical tools used to understand the geophysical properties of subsurface formation. In petroleum engineering, complete well log data are needed to establish a reliable 3D reservoir model to formulate oilfield development and production planning. However, in some intervals, well log data might have the problem of missing due to sensor issues. To solve this problem, the algorithm used in the image field, namely ADMM-ADAM Framework is applied to generate and reconstruct missing well log data. In this study, Deep Image Prior is chosen as the deep learning architecture and RED-Net is chosen as the convolutional neural network architecture. The missing well log data are initially reconstructed by training deep learning through ADAM optimization, and then the missing well log data problem can be described as a convex function solved by ADMM optimization.

According to the result, the problem of missing well log data was successfully generated by using the 2D convolutional neural network commonly used in the image field. According to the results, the missing well log data generated by the ADMM-ADAM Framework fitted the trend of the original well log data. The normalized cross-correlation (NCC) of each case is above 0.9, which means the trend of generated well log data is consistent with the original data. However, the Pearson correlation coefficient (r) is between 0.5 and 0.8, indicating that the confidence in the absolute value of generated well log data is still insufficient.

Akinnikawe, O., Lyne, S., & Roberts, J. (2018). Synthetic Well Log Generation Using Machine Learning Techniques. Paper presented at the SPE/AAPG/SEG Unconventional Resources Technology Conference.
Al Alfy, I. M. (2018). Mathematical derivation of density log from total gamma ray and neutron logs in clastic rocks, a case study, Egypt. Applied Radiation and Isotopes, 142, 42-45.
Azadpour, M., Manaman, N. S., Kadkhodaie-Ilkhchi, A., & Sedghipour, M.-R. (2015). Pore pressure prediction and modeling using well-logging data in one of the gas fields in south of Iran. Journal of Petroleum Science and Engineering, 128, 15-23.
Bader, S., Wu, X., & Fomel, S. (2019). Missing log data interpolation and semiautomatic seismic well ties using data matching techniques. Interpretation, 7(2), T347-T361.
Boyd, S., Boyd, S. P., & Vandenberghe, L. (2004). Convex optimization: Cambridge university press.
Breiman, L. (2001). Random Forests. Machine Learning, 45(1), 5-32.
Chen, Y., & Zhang, D. (2020). Well Log Generation via Ensemble Long Short-Term Memory (EnLSTM) Network. Geophysical Research Letters, 47(23).
Cohen, J. (1992). A power primer. Psychological bulletin, 112(1), 155.
Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine Learning, 20(3), 273-297.
Darling, T. (2005). Well logging and formation evaluation: Elsevier.
Delfiner, P., Delhomme, J. P., & Pelissier-Combescure, J. (1983). Application Of Geostatistical Analysis To The Evaluation Of Petroleum Reservoirs With Well Logs. Paper presented at the SPWLA 24th Annual Logging Symposium.
Feng, R., Grana, D., & Balling, N. (2021). Imputation of missing well log data by random forest and its uncertainty analysis. Computers & Geosciences, 152, 104763.
Gabay, D., & Mercier, B. (1976). A dual algorithm for the solution of nonlinear variational problems via finite element approximation. Computers & mathematics with applications, 2(1), 17-40.
Gardner, G., Gardner, L., & Gregory, A. (1974). Formation velocity and density—The diagnostic basics for stratigraphic traps. Geophysics, 39(6), 770-780.
Glowinski, R., & Marroco, A. (1975). Sur l'approximation, par éléments finis d'ordre un, et la résolution, par pénalisation-dualité d'une classe de problèmes de Dirichlet non linéaires. ESAIM: Mathematical Modelling and Numerical Analysis-Modélisation Mathématique et Analyse Numérique, 9(R2), 41-76.
Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural computation, 9(8), 1735-1780.
Jensen, J. L. W. V. (1906). Sur les fonctions convexes et les inégalités entre les valeurs moyennes. Acta mathematica, 30(1), 175-193.
Kingma, D. P., & Ba, J. (2014). Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980.
Kwon, M., Yoo, J., Kang, C., Hong, Y., Jeong, H., & Yoon, S. (2020). Accurate and Convenient Missing Well Log Synthesis using Generative Model. Paper presented at the Abu Dhabi International Petroleum Exhibition & Conference.
Lin, C. H., Lin, Y. C., & Tang, P. W. (2022). ADMM-ADAM: A New Inverse Imaging Framework Blending the Advantages of Convex Optimization and Deep Learning. IEEE Transactions on Geoscience and Remote Sensing, 60, 1-16.
Luthi, S. (2001). Geological well logs: Their use in reservoir modeling: Springer Science & Business Media.
Mao, X.-J., Shen, C., & Yang, Y.-B. (2016). Image restoration using convolutional auto-encoders with symmetric skip connections. arXiv preprint arXiv:1606.08921.
Mirhashemi, M., Khojasteh, E. R., Manaman, N. S., & Makarian, E. (2022). Efficient sonic log estimations by geostatistics, empirical petrophysical relations, and their combination: Two case studies from Iranian hydrocarbon reservoirs. Journal of Petroleum Science and Engineering, 213, 110384.
Moore II, D. H. (1987). Classification and regression trees. Cytometry, 8(5), 534-535.
Newell, A., Yang, K., & Deng, J. (2016). Stacked hourglass networks for human pose estimation. Paper presented at the European conference on computer vision.
Ronneberger, O., Fischer, P., & Brox, T. (2015). U-net: Convolutional networks for biomedical image segmentation. Paper presented at the International Conference on Medical image computing and computer-assisted intervention.
Roth, S., & Black, M. J. (2009). Fields of experts. International Journal of Computer Vision, 82(2), 205-229.
Senosy, A. H., Ewida, H. F., Soliman, H. A., & Ebraheem, M. O. (2020). Petrophysical analysis of well logs data for identification and characterization of the main reservoir of Al Baraka Oil Field, Komombo Basin, Upper Egypt. SN Applied Sciences, 2(7), 1293.
Sidorov, O., & Hardeberg, J. Y. (2019, 27-28 Oct. 2019). Deep Hyperspectral Prior: Single-Image Denoising, Inpainting, Super-Resolution. Paper presented at the 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW).
Smith, J. H. (2007). A Method for Calculating Pseudo Sonics from E-Logs in a Clastic Geologic Setting.
Tatsipie, N. R. K., & Sheng, J. J. (2021). Generating pseudo well logs for a part of the upper Bakken using recurrent neural networks. Journal of Petroleum Science and Engineering, 200, 108253.
Ulyanov, D., Vedaldi, A., & Lempitsky, V. (2017). Deep Image Prior. arXiv:1711.10925.
Wang, J., Cao, J., You, J., Cheng, M., & Zhou, P. (2021). A method for well log data generation based on a spatio-temporal neural network. Journal of Geophysics and Engineering, 18(5), 700-711.
Wu, L., Dong, Z., Li, W., Jing, C., & Qu, B. (2021). Well-Logging Prediction Based on Hybrid Neural Network Model. Energies, 14(24), 8583.
Xia, Y., Zhang, L., Ravikumar, N., Attar, R., Piechnik, S. K., Neubauer, S., Petersen, S. E., & Frangi, A. F. (2021). Recovering from missing data in population imaging – Cardiac MR image imputation via conditional generative adversarial nets. Medical Image Analysis, 67, 101812.
Yu, J., Lin, Z., Yang, J., Shen, X., Lu, X., & Huang, T. S. (2018). Generative Image Inpainting with Contextual Attention. arXiv:1801.07892.
Zare, A., Bagheri, M., & Ebadi, M. (2020). Reservoir facies and porosity modeling using seismic data and well logs by geostatistical simulation in an oil field. Carbonates and Evaporites, 35(3), 65.
謝秉志, 林. (2006). 井測法—地層岩性及石油蘊藏探勘. 科學發展, 407, 44-51.