Intelligent lithology mapping method based on multi-source geoscience information database

Abstract

The invention discloses an intelligent lithology mapping method based on a multi-source geoscience information database, and belongs to the technical field of basic geological survey, and the method comprises the steps: S1, obtaining original lithology imbalance modeling data; s2, converting the original lithologic imbalance modeling data in the step S1 into a lithologic balance data set through a hybrid sampling algorithm; s3, obtaining classification categories of the lithology balance data set through a supervised machine learning algorithm considering uncertainty quantization; and S4, on the basis of obtaining the classification categories, sequentially correcting classification results through a post-regularization algorithm until a final result has no isolated lithologic grid, and outputting three-dimensional lithologic and uncertainty modeling results. According to the intelligent lithologic mapping method based on the multi-source geoscience information database, the supervised lithologic mapping method based on the multi-source geoscience information database is input through computer software, the lithologic mapping accuracy and efficiency can be improved, and the uncertainty of the mapping process can be quantified.

Images