An offline multi‐scale unsaturated poromechanics model enabled by self‐designed/self‐improved neural networks

Abstract

Supervised machine learning via artificial neural network (ANN) has gained significant popularity for many geomechanics applications that involves multi‐phase flow and poromechanics. For unsaturated poromechanics problems, the multi‐physics nature and the complexity of the hydraulic laws make it difficult to design the optimal setup, architecture, and hyper‐parameters of the deep neural networks. This paper presents a meta‐modeling approach that utilizes deep reinforcement learning (DRL) to automatically discover optimal neural network settings that maximize a pre‐defined performance metric for the machine learning constitutive laws. This meta‐modeling framework is cast as a Markov Decision Process (MDP) with well‐defined states (subsets of states representing the proposed neural network (NN) settings), actions, and rewards. Following the selection rules, the artificial intelligence (AI) agent, represented in DRL via NN, self‐learns from taking a sequence of actions and receiving feedback signals (rewards) within the selection environment. By utilizing the Monte Carlo Tree Search (MCTS) to update the policy/value networks, the AI agent replaces the human modeler to handle the otherwise time‐consuming trial‐and‐error process that leads to the optimized choices of setup from a high‐dimensional parametric space. This approach is applied to generate two key constitutive laws for the unsaturated poromechanics problems: (1) the path‐dependent retention curve with distinctive wetting and drying paths. (2) The flow in the micropores, governed by an anisotropic permeability tensor. Numerical experiments have shown that the resultant ML‐generated material models can be integrated into a finite element (FE) solver to solve initial‐boundary‐value problems as replacements of the hand‐craft constitutive laws.

Feed-Forward Neural Networks for Failure Mechanics Problems

“…DL has many architectures, especially in mechanics, and research is focused mainly on using the feed-forward neural network (FFNN), recurrent neural network (RNN), and convolutional neural network (CNN). So far, applications of these types of neural networks devoted to the field of mechanics can be seen in the recent works of [27][28][29][30][31][32][33][34][35][36][37][38][39][40][41][42][43]. A detailed description of those approaches is summarized as follows:…”

A kd-tree-accelerated hybrid data-driven/model-based approach for poroelasticity problems with multi-fidelity multi-physics data

“…However, owing to the difficulty to conduct highly precise experiments and the lack of parametric space to characterize the hydraulic responses, a typical prediction of permeability based on the porosity-permeability model is expected to have a much higher variance and, in many cases, is considered accurate even if predicted benchmark permeability is just within the same order of magnitude [9,[22][23][24][25][26][27][28]. In this case, a calibrated hydraulic model that minimizes the mean square error of the Darcy's velocity or pressure gradient does not yield a reliable forecast due to the much wider confidence intervals.…”

Residual stresses in gas tungsten arc welding: a novel phase-field thermo-elastoplasticity modeling and parameter treatment framework

“…Therefore, the underlying modeling framework together with the developed algorithms will be implemented in future works in the context of modeling direct metal laser melting (DMLM) in the additive manufacturing process. Additionally, future works will address the possible embedment of machine learning approaches, like in [35,49,50,60,99,100], to capture the microscopic thermo-mechanical processes in the continuum modeling.…”