PT-PINNs: A Parametric Engineering Turbulence Solver based on Physics-Informed Neural Networks

2025-03-22

AI Summary

This work addresses the critical bottlenecks of data dependency and low accuracy in Physics-Informed Neural Networks (PINNs) for parametric engineering turbulent flow optimization. We propose a purely physics-driven PINN solver—requiring no experimental or CFD data—for modeling three-dimensional turbulent backward-facing step flows across continuous variations in Reynolds number (Re = 3,000–200,000) and expansion ratio (ER = 1.1–1.5). Our method innovatively incorporates a turbulence viscosity soft-constraint mechanism and a mass-flow-conservation-guided pretraining strategy, while embedding parametric boundary conditions and source terms. The resulting framework significantly enhances generalization capability and numerical robustness under zero-data conditions: prediction accuracy matches that of high-fidelity experiments and CFD simulations; training time is reduced to 39 hours—just 1/16 of conventional methods—and single-inference latency is only 40 seconds, i.e., 0.5% of typical CFD runtime.

Technology Category

• Search and Optimization: Learning to Search
• Natural Language Processing: Learning & Optimization for NLP
• Machine Learning: Hardware-aware ML

Application Category

• 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
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Physics-informed neural networks (PINNs) demonstrate promising potential in parameterized engineering turbulence optimization problems but face challenges, such as high data requirements and low computational accuracy when applied to engineering turbulence problems. This study proposes a framework that enhances the ability of PINNs to solve parametric turbulence problems without training datasets from experiments or CFD-Parametric Turbulence PINNs (PT-PINNs)). Two key methods are introduced to improve the accuracy and robustness of this framework. The first is a soft constraint method for turbulent viscosity calculation. The second is a pre-training method based on the conservation of flow rate in the flow field. The effectiveness of PT-PINNs is validated using a three-dimensional backward-facing step (BFS) turbulence problem with two varying parameters (Re = 3000-200000, ER = 1.1-1.5). PT-PINNs produce predictions that closely match experimental data and computational fluid dynamics (CFD) results across various conditions. Moreover, PT-PINNs offer a computational efficiency advantage over traditional CFD methods. The total time required to construct the parametric BFS turbulence model is 39 hours, one-sixteenth of the time required by traditional numerical methods. The inference time for a single-condition prediction is just 40 seconds-only 0.5% of a single CFD computation. These findings highlight the potential of PT-PINNs for future applications in engineering turbulence optimization problems.

Problem

• Enhancing PINNs for parametric turbulence without training data
• Improving accuracy via soft constraint and pre-training methods
• Validating efficiency and accuracy in 3D turbulence problems

Innovation

• Soft constraint method for turbulent viscosity
• Pre-training based on flow rate conservation
• Efficient parametric turbulence solver PT-PINNs