簡易檢索 / 詳目顯示

回結果列表
研究生： 陳瑞強
Chen, Jui-Chiang
論文名稱： 生成式AI填補井測資料缺失段
Complementing Logging Data with Generative AI
指導教授： 吳泓昱
Wu, Hung-Yu
學位類別： 碩士
Master
系所名稱： 工學院 - 資源工程學系
Department of Resources Engineering
論文出版年： 2024
畢業學年度： 112
語文別： 英文
論文頁數： 101
中文關鍵詞： 機器學習 、特徵工程 、測井資料 、資料缺失
外文關鍵詞： Machine learning, Feature engineering, Well-logging data, Data missing
相關次數： 點閱：156  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 本研究旨在解決測井過程中受環境幹擾、地質複雜性和儀器限制等因素影響所造成的資料缺失問題。為了降低重新鑽井或測量的高昂成本，我們計劃利用機器學習技術開發一個通用性很強的模型。透過特徵工程提高模型性能，我們的目標是產生缺失數據，創建連續數據集，全面評估和分析地下變化和特徵，包括孔隙度、岩性和油藏分佈的連續性變化。
    機器學習的三大要素：資料相關性、無相依數學公式或經驗公式、大量資料，在處理多維度、非線性資料方面呈現優越性。先前的研究中，我們成功使用支援向量機(SVM)將測井資料分為不同類別，並可準確辨識原始DST資料中的薄氣層或油層。
    特徵工程是機器學習和數據挖掘中至關重要的步驟，涉及對數據進行轉換、選擇和新特徵創建，以提高模型泛化能力。特徵工程在處理缺失資料時較傳統方法更靈活，不僅填充缺失值，還強調基於數據特性創建新特徵，全面理解數據，提高模型性能。
    這項研究面臨地質和機器學習領域的挑戰，包括不同地質條件的適應性、來自惡劣環境和儀器的數據不確定性和噪聲等問題。在多源數據整合方面，需要集成學習或深度學習模型確保一致的數據標準和元數據信息。實時數據生成挑戰在於需要高效算法和實時計算能力，可利用輕量級機器學習模型或增量式學習方法，並使用硬體加速技術提高實時計算性能。
    最後，跨學科合作方面，需建立跨學科團隊，促進合作和知識共享，透過跨學科培訓增進不同領域專家之間的理解和合作，建立共同目標。

    This study aims to address the challenge of data missing in the well-logging process, influenced by factors such as environmental interference, geological complexity, and instrument limitations. To mitigate the high cost of re-drilling or measurements, we plan to develop a highly versatile model using machine learning. Through feature engineering to enhance model performance, we aim to generate missing data, create a continuous data set, and comprehensively assess and analyze underground variations and features, including the continuity changes in porosity, lithology, and reservoir distribution.
    The three key elements of machine learning—data correlation, independence from mathematical or empirical formulas, and a large volume of data—demonstrate superiority in handling multidimensional, nonlinear data. In previous research, we successfully used Support Vector Machines (SVM) to classify well-log data into different categories and accurately identify thin gas layers or oil layers in the original DST data.
    Feature engineering is a crucial step in machine learning and data mining, involving the transformation, selection, and creation of new features to enhance model generalization. Feature engineering is more flexible in dealing with missing data than traditional methods, focusing not only on filling missing values but also on creating new features based on data characteristics for a comprehensive understanding of the data and improved model performance.
    This study faces challenges in the fields of geology and machine learning, including adaptability to different geological conditions, data uncertainty, noise from adverse environments, and instrument limitations. In terms of integrating data from multiple sources, the study requires the use of ensemble learning or deep learning models to ensure consistent data standards and metadata information. Challenges in real-time data generation involve the need for efficient algorithms and real-time computing capabilities, which can be addressed using lightweight machine learning models or incremental learning methods, along with hardware acceleration techniques to enhance real-time computing performance.
    Finally, interdisciplinary collaboration requires the establishment of cross-disciplinary teams to facilitate cooperation and knowledge sharing. Cross-disciplinary training is essential to enhance understanding and collaboration among experts in different fields, working towards common goals.

    Abstract I 中文摘要 III Acknowledgments V Table of Contents VII List of Tables IX List of Figures X Symbol Description XII 1. Introduction 1 1.1 Preface 1 1.2 Research Motivation and Objectives 2 2. Literature Review 5 3. Methodology 17 3.1 Research Background and Framework 17 3.2 Preliminary Exploration and Research Methodology 21 3.2.1 Preprocessing 21 3.2.2 Experimental Design and Model Adjustment 23 3.2.3 Post-processing 25 3.3 Supervised Learning Models 29 3.3.1 Linear Regression 29 3.3.2 Random Forest 30 3.3.3 XGBoost 31 3.3.4 LightGBM 32 3.4 Feature Engineering 33 3.4.1 Correlation 33 3.4.2 Recursive Feature Elimination (RFE) 33 3.4.3 Random Forest Feature Importance 34 3.4.4 LassoCV 35 3.5 Ensemble Models 36 3.5.1 Bagged Decision Trees 36 3.5.2 AdaBoost (Adaptive Boosting) 37 3.5.3 Gradient Boosting 37 3.5.4 Stacking 37 4. Research Process and Findings 39 4.1 Early Exploration 39 4.2 Supervised Learning Models 52 4.3 Feature Engineering 55 4.4 Integrated Models 58 5. Results and Discussion 67 5.1 Conclusion 67 5.2 Discussion 67 References 72 Appendices 77 Appendix A: Generated Results of GLCM Under Various Statistical Features 77 Appendix B: The Results of Generating Data for RM_HRLT with Different Input Data 78 Appendix C: The Results of Generating Data for RXOZ with Different Input Data 80 Appendix D: Generation Results of Top 7 Training Logs Based on Correlation Analysis |r| 82 Appendix E: Generation Results Using All Well Logs as Training Logs 85

    1. Adeniran, A. A., Adebayo, A. R., Salami, H. O., Yahaya, M. O., & Abdulraheem, A. (2019). A competitive ensemble model for permeability prediction in heterogeneous oil and gas reservoirs. Applied Computing Geosciences, 1, 100004. https://doi.org/10.1016/j.acags.2019.100004
    2. Aleryani, A., Wang, W., & De La Iglesia, B. (2020). Multiple imputation ensembles (MIE) for dealing with missing data. SN Computer Science, 1, 1–20. https://doi.org/10.1007/s42979-020-00131-0
    3. Batista, G. E., & Monard, M. C. (2003). An analysis of four missing data treatment methods for supervised learning. Applied Artificial Intelligence, 17(5–6), 519–533. https://doi.org/10.1080/713827181
    4. Bauer, E., & Kohavi, R. (1999). An empirical comparison of voting classification algorithms: Bagging, boosting, and variants. Machine Learning, 36(1–2), 105–139. https://doi.org/10.1023/A:1007515423169
    5. Bühlmann, P., & Hothorn, T. (2007). Boosting algorithms: Regularization, prediction and model fitting. Statistical Science, 22(4). https://doi.org/10.1214/07-sts242
    6. Chen, Y., Wong, M.-L., & Li, H. (2014). Applying ant colony optimization to configuring stacking ensembles for data mining. Expert Systems with Applications, 41(6), 2688–2702. https://doi.org/10.1016/j.eswa.2013.10.063
    7. De Silva, H., & Perera, A. S. (2016). Missing data imputation using evolutionary k-nearest neighbor algorithm for gene expression data. In 2016 Sixteenth International Conference on Advances in ICT for Emerging Regions (ICTer) (pp. 141–146). IEEE. https://doi.org/10.1109/ICTER.2016.7829911
    8. Dietterich, T. G. (2002). Ensemble learning. In M. A. Arbib (Ed.), The Handbook of Brain Theory and Neural Networks (pp. 110–125). MIT Press. https://doi.org/10.7551/mitpress/3413.001.0001
    9. Donders, A. R. T., Van Der Heijden, G. J., Stijnen, T., & Moons, K. G. (2006). A gentle introduction to imputation of missing values. Journal of Clinical Epidemiology, 59(10), 1087–1091. https://doi.org/10.1016/j.jclinepi.2006.01.014
    10. Dzeroski, S., & Zenko, B. (2002). Is combining classifiers better than selecting the best one? In Proceedings of the Nineteenth International Conference on Machine Learning (ICML) (pp. 123–130). Citeseer. https://doi.org/10.1023/b:mach.0000015881.36452.6e
    11. García-Laencina, P. J., Sancho-Gómez, J.-L., Figueiras-Vidal, A. R., & Verleysen, M. (2009). K nearest neighbours with mutual information for simultaneous classification and missing data imputation. Neurocomputing, 72(7–9), 1483–1493. https://doi.org/10.1016/j.neucom.2008.11.026
    12. Ghazanfar, M. A., & Prugel, A. (2013). The advantage of careful imputation sources in sparse data-environment of recommender systems: Generating improved SVD-based recommendations. Informatica, 37(1).
    13. Gimpy, D., & Rajan Vohra, M. (2014). Estimation of missing values using decision tree approach. International Journal of Computer Science and Information Technology, 5(4), 5216–5220.
    14. Gosiewska, A., Kozak, A., & Biecek, P. (2021). Simpler is better: Lifting interpretability-performance trade-off via automated feature engineering. Decision Support Systems, 150, 113556. https://doi.org/10.1016/j.dss.2021.113556
    15. Honghai, F., Guoshun, C., Cheng, Y., Bingru, Y., & Yumei, C. (2005). A SVM regression based approach to filling in missing values. In International Conference on Knowledge-Based and Intelligent Information and Engineering Systems (pp. 581–587). Springer. https://doi.org/10.1007/11553939_83
    16. Horton, N. J., Lipsitz, S. R., & Parzen, M. (2003). A potential for bias when rounding in multiple imputation. The American Statistician, 57(4), 229–232. https://doi.org/10.1198/0003130032314
    17. Huang, J., & Sun, H. (2016). Grey relational analysis based k nearest neighbor missing data imputation for software quality datasets. In 2016 IEEE International Conference on Software Quality, Reliability and Security (QRS) (pp. 86–91). IEEE. https://doi.org/10.1109/qrs.2016.20
    18. Huang, J., Keung, J. W., Sarro, F., Li, Y.-F., Yu, Y.-T., Chan, W., & Sun, H. (2017). Cross-validation based k nearest neighbor imputation for software quality datasets: An empirical study. Journal of Systems and Software, 132, 226–252. https://doi.org/10.1016/j.jss.2017.07.012
    19. Jerez, J. M., Molina, I., García-Laencina, P. J., Alba, E., Ribelles, N., Martín, M., & Franco, L. (2010). Missing data imputation using statistical and machine learning methods in a real breast cancer problem. Artificial Intelligence in Medicine, 50(2), 105–115. https://doi.org/10.1016/j.artmed.2010.05.002
    20. Khan, S. S., Ahmad, A., & Mihailidis, A. (2019). Bootstrapping and multiple imputation ensemble approaches for classification problems. Journal of Intelligent & Fuzzy Systems, 37(6), 7769–7783. https://doi.org/10.3233/JIFS-182656
    21. Khan, S. I., & Hoque, A. S. M. L. (2020). SICE: An improved missing data imputation technique. Journal of Big Data, 7(1), 1–21. https://doi.org/10.1186/s40537-020-00313-w
    22. Khalid, S., Khalil, T., & Nasreen, S. (2014). A survey of feature selection and feature extraction techniques in machine learning. In 2014 Science and Information Conference (pp. 372-378). IEEE. https://doi.org/10.1109/SAI.2014.6918213
    23. Kuhn, M., & Johnson, K. (2019). Feature engineering and selection: A practical approach for predictive models. Chapman and Hall/CRC. https://doi.org/10.1080/00031305.2020.1790217
    24. Little, R. J., & Rubin, D. B. (2019). Statistical analysis with missing data. Wiley. https://doi.org/10.1002/9781119482260
    25. Lu, C.-B., & Mei, Y. (2018). An imputation method for missing data based on an extreme learning machine auto-encoder. IEEE Access, 6, 52930–52935. https://doi.org/10.1109/ACCESS.2018.2868729
    26. Maillo, J., Ramírez, S., Triguero, I., & Herrera, F. (2017). kNN-IS: An iterative Spark-based design of the k-nearest neighbors classifier for big data. Knowledge-Based Systems, 117, 3–15. https://doi.org/10.1016/j.knosys.2016.06.012
    27. Nguyen, C. D., Carlin, J. B., & Lee, K. J. (2017). Model checking in multiple imputation: An overview and case study. Emerging Themes in Epidemiology, 14(1), 8. https://doi.org/10.1186/s12982-017-0062-6
    28. Oehmcke, S., Zielinski, O., & Kramer, O. (2016). kNN ensembles with penalized DTW for multivariate time series imputation. In 2016 International Joint Conference on Neural Networks (IJCNN) (pp. 2774–2781). IEEE. https://doi.org/10.1109/ijcnn.2016.7727549
    29. Pelckmans, K., De Brabanter, J., Suykens, J. A., & De Moor, B. (2005). Handling missing values in support vector machine classifiers. Neural Networks, 18(5–6), 684–692. https://doi.org/10.1016/j.neunet.2005.06.025
    30. Ponti Jr, M. P. (2011). Combining classifiers: From the creation of ensembles to the decision fusion. In 2011 24th SIBGRAPI Conference on Graphics, Patterns, and Images Tutorials (pp. 1–10). IEEE. https://doi.org/10.1109/sibgrapi-t.2011.9
    31. Rahman, G., & Islam, Z. (2011). A decision tree-based missing value imputation technique for data pre-processing. In Proceedings of the Ninth Australasian Data Mining Conference, vol. 121 (pp. 41–50).
    32. Rahman, M. G., & Islam, M. Z. (2013). Missing value imputation using decision trees and decision forests by splitting and merging records: Two novel techniques. Knowledge-Based Systems, 53, 51–65. https://doi.org/10.1016/j.knosys.2013.08.023
    33. Re, M., & Valentini, G. (2012). Ensemble methods. In Advances in Machine Learning and Data Mining for Astronomy (pp. 563–593). CRC Press. https://doi.org/10.1201/b11822-34
    34. Rubin, D. B. (1976). Inference and missing data. Biometrika, 63(3), 581–592. https://doi.org/10.1093/biomet/63.3.581
    35. Smola, A. J., Vishwanathan, S., & Hofmann, T. (2005). Kernel methods for missing variables. In AISTATS. Citeseer.
    36. Song, Q., & Shepperd, M. (2007). Missing data imputation techniques. International Journal of Business Intelligence and Data Mining, 2(3), 261–291. https://doi.org/10.1504/IJBIDM.2007.015485
    37. Stewart, T. G., Zeng, D., & Wu, M. C. (2018). Constructing support vector machines with missing data. Wiley Interdisciplinary Reviews: Computational Statistics, 10(4), e1430. https://doi.org/10.1002/wics.1430
    38. Tran, C. T., Zhang, M., Andreae, P., Xue, B., & Bui, L. T. (2017). Multiple imputation and ensemble learning for classification with incomplete data. In Intelligent and Evolutionary Systems (pp. 401–415). Springer. https://doi.org/10.1007/978-3-319-49049-6_29
    39. Twala, B. (2009). An empirical comparison of techniques for handling incomplete data using decision trees. Applied Artificial Intelligence, 23(5), 373–405. https://doi.org/10.1080/08839510902872223
    40. Ulyanov, D., Vedaldi, A., & Lempitsky, V. (2017). Deep image prior. arXiv:1711.10925. https://doi.org/10.1007/s11263-020-01303-4
    41. Uusitalo, L., Lehikoinen, A., Helle, I., & Myrberg, K. (2015). An overview of methods to evaluate uncertainty of deterministic models in decision support. Environmental Modelling & Software, 63, 24–31. https://doi.org/10.1016/j.envsoft.2014.09.017
    42. Zhang, C., & Ma, Y. (2012). Ensemble machine learning: Methods and applications. Springer. https://doi.org/10.1007/978-1-4419-9326-7
    43. Zhang, X.-F., Ou-Yang, L., Yang, S., Zhao, X.-M., & Hu, X. (2019). Enimpute: Imputing dropout events in single-cell RNA-sequencing data via ensemble learning. Bioinformatics, 35(22), 4827–4829. https://doi.org/10.1093/bioinformatics/btz435
    44. Zhao, Y., & Long, Q. (2016). Multiple imputation in the presence of high-dimensional data. Statistical Methods in Medical Research, 25(5), 2021–2035. https://doi.org/10.1177/0962280213511027
    45. Zheng, A., & Casari, A. (2018). Feature engineering for machine learning: Principles and techniques for data scientists. O'Reilly Media.
    46. Zhu, M., & Cheng, X. (2015). Iterative KNN imputation based on GRA for missing values in TPLMS. In 2015 4th International Conference on Computer Science and Network Technology (ICCSNT) (vol. 1, pp. 94–99). IEEE. https://doi.org/10.1109/ICCSNT.2015.7490714
    47. Lin, Y.-T. (2022). Generating missing well log data by ADMM-ADAM framework.
    48. Huque, M. H., Carlin, J. B., Simpson, J. A., & Lee, K. J. (2018). A comparison of multiple imputation methods for missing data in longitudinal studies. https://doi.org/10.1186/s12874-018-0615-6
    49. Mao, X.-J., Shen, C., & Yang, Y.-B. (2016). Image restoration using convolutional auto-encoders with symmetric skip connections. arXiv preprint arXiv:1606.08921.
    50. Li, P., Pei, Y., & Li, J. (2023). A comprehensive survey on design and application of autoencoder in deep learning. https://doi.org/10.1016/j.asoc.2023.110176

    下載圖示
    2026-07-02公開
    QR CODE