Progressive Geological Modeling and Uncertainty Analysis Using Machine Learning

Three-dimensional geological modeling is a process of interpreting features from limited sample data and making predictions, which can be converted into classification task for grid units in the space. In sedimentary settings, it difficult single to comprehensively express complex spatio-temporal relationships underground response this problem, we proposed progressive strategy reconstruct subsurface based on machine learning approach. The work consisted two-stage classifications. first stage, stratigraphic classifier was built by mapping spatial coordinates classes, reflected time information unit. Then, obtained class used as new feature training lithologic second allowed implicitly rule condition enabled us output with implications. Finally, joint Shannon entropy two classifications calculated evaluate uncertainty total steps. experiment fine-grained 3D model integrated expression validated effectiveness strategy. Moreover, compared conventionally trained classifier, misclassification between different strata results has been reduced, improvement F1-score 0.75 0.78.