Masoud Safdari | Searcharxiv

arXiv2021

Smart Adaptive Mesh Refinement with NEMoSys

Adaptive mesh refinement (AMR) offers a practical solution to reduce the computational cost and memory requirement of numerical simulations that use computational meshes. In this work, we introduce a novel smart methodology for adaptive mesh refinement. Smart adaptive refinement blends classical AMR with machine learning to address some of the known issues of the conventional approaches. We provide an algorithm for adaptive refinement. Subsequently, we introduce a modular object-oriented structure for our smart AMR algorithm. Then we present procedures used for the training of a smart AMR model. The study follows with a demonstration of preliminary numerical studies indicating the feasibility of performing adaptive mesh refinement on a few demonstrative problems selected from the CFD domain. Finally, we conclude with a few comments about future work.

physics.flu-dyn↗

arXiv2020

A Supervised Machine Learning Approach for Accelerating the Design of Particulate Composites: Application to Thermal Conductivity

A supervised machine learning (ML) based computational methodology for the design of particulate multifunctional composite materials with desired thermal conductivity (TC) is presented. The design variables are physical descriptors of the material microstructure that directly link microstructure to the material's properties. A sufficiently large and uniformly sampled database was generated based on the Sobol sequence. Microstructures were realized using an efficient dense packing algorithm, and the TCs were obtained using our previously developed Fast Fourier Transform (FFT) homogenization method. Our optimized ML method is trained over the generated database and establishes the complex relationship between the structure and properties. Finally, the application of the trained ML model in the inverse design of a new class of composite materials, liquid metal (LM) elastomer, with desired TC is discussed. The results show that the surrogate model is accurate in predicting the microstructure behavior with respect to high-fidelity FFT simulations, and inverse design is robust in finding microstructure parameters according to case studies.

physics.comp-ph↗