Research on a Rapid Three-Dimensional Compressor Flow Field Prediction Method Integrating U-Net and Physics-Informed Neural Networks
This paper presents a neural network model, PINN-AeroFlow-U, for reconstructing full-field aerodynamic quantities around three-dimensional compressor blades, including regions near the wall. This model is based on structured CFD training data and physics-informed loss functions and is proposed for d...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-07-01
|
| Series: | Mathematics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-7390/13/15/2396 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849405835686445056 |
|---|---|
| author | Chen Wang Hongbing Ma |
| author_facet | Chen Wang Hongbing Ma |
| author_sort | Chen Wang |
| collection | DOAJ |
| description | This paper presents a neural network model, PINN-AeroFlow-U, for reconstructing full-field aerodynamic quantities around three-dimensional compressor blades, including regions near the wall. This model is based on structured CFD training data and physics-informed loss functions and is proposed for direct 3D compressor flow prediction. It maps flow data from the physical domain to a uniform computational domain and employs a U-Net-based neural network capable of capturing the sharp local transitions induced by fluid acceleration near the blade leading edge, as well as learning flow features associated with internal boundaries (e.g., the wall boundary). The inputs to PINN-AeroFlow-U are the flow-field coordinate data from high-fidelity multi-geometry blade solutions, the 3D blade geometry, and the first-order metric coefficients obtained via mesh transformation. Its outputs include the pressure field, temperature field, and velocity vector field within the blade passage. To enhance physical interpretability, the network’s loss function incorporates both the Euler equations and gradient constraints. PINN-AeroFlow-U achieves prediction errors of 1.063% for the pressure field and 2.02% for the velocity field, demonstrating high accuracy. |
| format | Article |
| id | doaj-art-a34aa964e08848b6bde0bbd46373ab81 |
| institution | Kabale University |
| issn | 2227-7390 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Mathematics |
| spelling | doaj-art-a34aa964e08848b6bde0bbd46373ab812025-08-20T03:36:34ZengMDPI AGMathematics2227-73902025-07-011315239610.3390/math13152396Research on a Rapid Three-Dimensional Compressor Flow Field Prediction Method Integrating U-Net and Physics-Informed Neural NetworksChen Wang0Hongbing Ma1School of Intelligence Science and Technology, Xinjiang University, Urumqi 830046, ChinaSchool of Intelligence Science and Technology, Xinjiang University, Urumqi 830046, ChinaThis paper presents a neural network model, PINN-AeroFlow-U, for reconstructing full-field aerodynamic quantities around three-dimensional compressor blades, including regions near the wall. This model is based on structured CFD training data and physics-informed loss functions and is proposed for direct 3D compressor flow prediction. It maps flow data from the physical domain to a uniform computational domain and employs a U-Net-based neural network capable of capturing the sharp local transitions induced by fluid acceleration near the blade leading edge, as well as learning flow features associated with internal boundaries (e.g., the wall boundary). The inputs to PINN-AeroFlow-U are the flow-field coordinate data from high-fidelity multi-geometry blade solutions, the 3D blade geometry, and the first-order metric coefficients obtained via mesh transformation. Its outputs include the pressure field, temperature field, and velocity vector field within the blade passage. To enhance physical interpretability, the network’s loss function incorporates both the Euler equations and gradient constraints. PINN-AeroFlow-U achieves prediction errors of 1.063% for the pressure field and 2.02% for the velocity field, demonstrating high accuracy.https://www.mdpi.com/2227-7390/13/15/2396flow field predictionU-Netgradient constraintscoordinate transformationEuler equations |
| spellingShingle | Chen Wang Hongbing Ma Research on a Rapid Three-Dimensional Compressor Flow Field Prediction Method Integrating U-Net and Physics-Informed Neural Networks Mathematics flow field prediction U-Net gradient constraints coordinate transformation Euler equations |
| title | Research on a Rapid Three-Dimensional Compressor Flow Field Prediction Method Integrating U-Net and Physics-Informed Neural Networks |
| title_full | Research on a Rapid Three-Dimensional Compressor Flow Field Prediction Method Integrating U-Net and Physics-Informed Neural Networks |
| title_fullStr | Research on a Rapid Three-Dimensional Compressor Flow Field Prediction Method Integrating U-Net and Physics-Informed Neural Networks |
| title_full_unstemmed | Research on a Rapid Three-Dimensional Compressor Flow Field Prediction Method Integrating U-Net and Physics-Informed Neural Networks |
| title_short | Research on a Rapid Three-Dimensional Compressor Flow Field Prediction Method Integrating U-Net and Physics-Informed Neural Networks |
| title_sort | research on a rapid three dimensional compressor flow field prediction method integrating u net and physics informed neural networks |
| topic | flow field prediction U-Net gradient constraints coordinate transformation Euler equations |
| url | https://www.mdpi.com/2227-7390/13/15/2396 |
| work_keys_str_mv | AT chenwang researchonarapidthreedimensionalcompressorflowfieldpredictionmethodintegratingunetandphysicsinformedneuralnetworks AT hongbingma researchonarapidthreedimensionalcompressorflowfieldpredictionmethodintegratingunetandphysicsinformedneuralnetworks |