DoMINO: A Decomposable Multi-scale Iterative Neural Operator for Modeling Large Scale Engineering Simulations

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Ranade, Rishikesh, Nabian, Mohammad Amin, Tangsali, Kaustubh, Kamenev, Alexey, Hennigh, Oliver, Cherukuri, Ram, Choudhry, Sanjay
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909464262606848
author Ranade, Rishikesh
Nabian, Mohammad Amin
Tangsali, Kaustubh
Kamenev, Alexey
Hennigh, Oliver
Cherukuri, Ram
Choudhry, Sanjay
author_facet Ranade, Rishikesh
Nabian, Mohammad Amin
Tangsali, Kaustubh
Kamenev, Alexey
Hennigh, Oliver
Cherukuri, Ram
Choudhry, Sanjay
contents Numerical simulations play a critical role in design and development of engineering products and processes. Traditional computational methods, such as CFD, can provide accurate predictions but are computationally expensive, particularly for complex geometries. Several machine learning (ML) models have been proposed in the literature to significantly reduce computation time while maintaining acceptable accuracy. However, ML models often face limitations in terms of accuracy and scalability and depend on significant mesh downsampling, which can negatively affect prediction accuracy and generalization. In this work, we propose a novel ML model architecture, DoMINO (Decomposable Multi-scale Iterative Neural Operator) developed in NVIDIA Modulus to address the various challenges of machine learning based surrogate modeling of engineering simulations. DoMINO is a point cloudbased ML model that uses local geometric information to predict flow fields on discrete points. The DoMINO model is validated for the automotive aerodynamics use case using the DrivAerML dataset. Through our experiments we demonstrate the scalability, performance, accuracy and generalization of our model to both in-distribution and out-of-distribution testing samples. Moreover, the results are analyzed using a range of engineering specific metrics important for validating numerical simulations.
format Preprint
id arxiv_https___arxiv_org_abs_2501_13350
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DoMINO: A Decomposable Multi-scale Iterative Neural Operator for Modeling Large Scale Engineering Simulations
Ranade, Rishikesh
Nabian, Mohammad Amin
Tangsali, Kaustubh
Kamenev, Alexey
Hennigh, Oliver
Cherukuri, Ram
Choudhry, Sanjay
Machine Learning
Computational Physics
Numerical simulations play a critical role in design and development of engineering products and processes. Traditional computational methods, such as CFD, can provide accurate predictions but are computationally expensive, particularly for complex geometries. Several machine learning (ML) models have been proposed in the literature to significantly reduce computation time while maintaining acceptable accuracy. However, ML models often face limitations in terms of accuracy and scalability and depend on significant mesh downsampling, which can negatively affect prediction accuracy and generalization. In this work, we propose a novel ML model architecture, DoMINO (Decomposable Multi-scale Iterative Neural Operator) developed in NVIDIA Modulus to address the various challenges of machine learning based surrogate modeling of engineering simulations. DoMINO is a point cloudbased ML model that uses local geometric information to predict flow fields on discrete points. The DoMINO model is validated for the automotive aerodynamics use case using the DrivAerML dataset. Through our experiments we demonstrate the scalability, performance, accuracy and generalization of our model to both in-distribution and out-of-distribution testing samples. Moreover, the results are analyzed using a range of engineering specific metrics important for validating numerical simulations.
title DoMINO: A Decomposable Multi-scale Iterative Neural Operator for Modeling Large Scale Engineering Simulations
topic Machine Learning
Computational Physics
url https://arxiv.org/abs/2501.13350