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

2025-01-23

AI Summary

Computational fluid dynamics (CFD) simulations are computationally expensive, while existing machine learning surrogate models suffer from low accuracy, poor generalization, and strong dependence on downsampled, aligned grids. Method: This paper proposes a decomposable multi-scale iterative neural operator architecture that operates directly on raw point clouds—without requiring global grid alignment—by integrating local geometric features, multi-scale feature decomposition, iterative residual updates, and geometry-aware attention. Implemented within NVIDIA Modulus, it preserves spatial fidelity and avoids mesh-based constraints. Contribution/Results: On the DrivAerML dataset, the model achieves CFD-level prediction accuracy with >100× faster inference. It demonstrates exceptional robustness on out-of-distribution (OOD) geometries and critical engineering metrics—including drag coefficient and pressure distribution. This work establishes the first point-cloud-native, mesh-agnostic neural operator paradigm, significantly enhancing cross-geometry generalization capability.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

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.

Problem

• Computational Fluid Dynamics
• Machine Learning
• Complex Geometry

Innovation

• DoMINO Model
• NVIDIA Modulus
• Engineering Simulation