Machine learning for flux regression in discrete fracture networks

Abstract In several applications concerning underground flow simulations in fractured media, the rock matrix is modeled by means of Discrete Fracture Network (DFN) model. The fractures are typically described through stochastic parameters sampled from known distributions. this framework, it worth considering application suitable complexity reduction techniques, also view possible uncertainty quantification analyses or other requiring a fast approximation network. Herein, we propose Neural Networks to flux regression problems DFN characterized trasmissivities as an approach predict fluxes.