Natural Fracture Network Model Using Machine Learning Approach

Abstract A fracture network model is a powerful tool for characterizing fractured rock systems. In this paper, we present the by integrating machine learning algorithm in two-dimensional setting to predict natural topology porous media. We also use azimuth angle fault data from Kazakhstan. The results indicate that with LightGBM performs better designing parameter hidden areas based on known area. addition, numerical result of shows good randomly selected azimuth.