PINNA: Physics‐Informed Neural Networks Architecture for Implicit Governing Physics With Interpretability.
Saved in:
| Title: | PINNA: Physics‐Informed Neural Networks Architecture for Implicit Governing Physics With Interpretability. |
|---|---|
| Authors: | Yao, Zhengtao1,2 (AUTHOR) zyao9248@alumni.usc.edu, Hawi, Philippe1 (AUTHOR), Aitharaju, Venkat2 (AUTHOR), Mahishi, Jay2 (AUTHOR), Ghanem, Roger1 (AUTHOR) |
| Source: | International Journal for Numerical Methods in Engineering. Sep2026, Vol. 127 Issue 17, p1-30. 30p. |
| Subjects: | Engineering simulations, Deep learning, Composite materials, Combustion engineering, Multiscale modeling |
| Abstract: | Modern engineering simulations frequently operate in regimes where explicit governing equations are incomplete or experiments are prohibitively expensive. To remain accurate and trustworthy under such constraints, we propose the Physics‐Informed Neural Network Architecture (PINNA), a deep‐learning surrogate that fuses data with domain knowledge inside the network. Concretely, a fully connected encodergϕ$$ {g}_{\phi } $$ embeds raw inputs (e.g., material descriptors, loads, or operating conditions) into a latent space, while an intermediate‐physics head is explicitly supervised to predict expert‐selected quantities with clear physical meaning, such as strain‐energy densities or chemically relevant indicators. These intermediate predictions are residual‐concatenated with the original inputs and passed to a task decoderhθ$$ {h}_{\theta } $$, enabling the network to learn and correct mismatches between approximate physics and observed responses. This architecture achieves (i) higher accuracy and sample efficiency than purely data‐driven baselines, (ii) negligible computational overhead at inference time, and (iii) intrinsic interpretability, as intermediate predictions expose physically meaningful internal representations. We further introduce a scalable extension, Generalized PINNA (G‐PINNA), which stacks multiple physics heads to accommodate multiscale and multi‐physics constraints within the same residual‐concatenation framework. PINNA is validated across three fundamentally different benchmarks: two composite‐material problems involving nonlinear stress‐strain behavior and multistage failure, and a large‐scale 1‐D laminar combustion problem governed by stiff chemical kinetics, thermal transport, and reduced fluid mechanics. In the composite benchmarks, PINNA reduces test errors by up to an order of magnitude relative to Fourier Neural Operators and DeepONets while using fewer parameters. In the combustion benchmark, PINNA accurately predicts both interpretable intermediate chemical indicators and high‐dimensional flame quantities, including scalar metrics and full spatial profiles, demonstrating that intermediate supervision enables robust generalization beyond solid mechanics. These results confirm that embedding expert knowledge directly into neural architectures provides a practical, interpretable, and general framework for learning implicit physics across diverse engineering domains. [ABSTRACT FROM AUTHOR] |
| Copyright of International Journal for Numerical Methods in Engineering is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
| FullText | Text: Availability: 0 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 196946482 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: PINNA: Physics‐Informed Neural Networks Architecture for Implicit Governing Physics With Interpretability. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Yao%2C+Zhengtao%22">Yao, Zhengtao</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> zyao9248@alumni.usc.edu</i><br /><searchLink fieldCode="AR" term="%22Hawi%2C+Philippe%22">Hawi, Philippe</searchLink><relatesTo>1</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Aitharaju%2C+Venkat%22">Aitharaju, Venkat</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Mahishi%2C+Jay%22">Mahishi, Jay</searchLink><relatesTo>2</relatesTo> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Ghanem%2C+Roger%22">Ghanem, Roger</searchLink><relatesTo>1</relatesTo> (AUTHOR) – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22International+Journal+for+Numerical+Methods+in+Engineering%22">International Journal for Numerical Methods in Engineering</searchLink>. Sep2026, Vol. 127 Issue 17, p1-30. 30p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Engineering+simulations%22">Engineering simulations</searchLink><br /><searchLink fieldCode="DE" term="%22Deep+learning%22">Deep learning</searchLink><br /><searchLink fieldCode="DE" term="%22Composite+materials%22">Composite materials</searchLink><br /><searchLink fieldCode="DE" term="%22Combustion+engineering%22">Combustion engineering</searchLink><br /><searchLink fieldCode="DE" term="%22Multiscale+modeling%22">Multiscale modeling</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Modern engineering simulations frequently operate in regimes where explicit governing equations are incomplete or experiments are prohibitively expensive. To remain accurate and trustworthy under such constraints, we propose the Physics‐Informed Neural Network Architecture (PINNA), a deep‐learning surrogate that fuses data with domain knowledge inside the network. Concretely, a fully connected encodergϕ$$ {g}_{\phi } $$ embeds raw inputs (e.g., material descriptors, loads, or operating conditions) into a latent space, while an intermediate‐physics head is explicitly supervised to predict expert‐selected quantities with clear physical meaning, such as strain‐energy densities or chemically relevant indicators. These intermediate predictions are residual‐concatenated with the original inputs and passed to a task decoderhθ$$ {h}_{\theta } $$, enabling the network to learn and correct mismatches between approximate physics and observed responses. This architecture achieves (i) higher accuracy and sample efficiency than purely data‐driven baselines, (ii) negligible computational overhead at inference time, and (iii) intrinsic interpretability, as intermediate predictions expose physically meaningful internal representations. We further introduce a scalable extension, Generalized PINNA (G‐PINNA), which stacks multiple physics heads to accommodate multiscale and multi‐physics constraints within the same residual‐concatenation framework. PINNA is validated across three fundamentally different benchmarks: two composite‐material problems involving nonlinear stress‐strain behavior and multistage failure, and a large‐scale 1‐D laminar combustion problem governed by stiff chemical kinetics, thermal transport, and reduced fluid mechanics. In the composite benchmarks, PINNA reduces test errors by up to an order of magnitude relative to Fourier Neural Operators and DeepONets while using fewer parameters. In the combustion benchmark, PINNA accurately predicts both interpretable intermediate chemical indicators and high‐dimensional flame quantities, including scalar metrics and full spatial profiles, demonstrating that intermediate supervision enables robust generalization beyond solid mechanics. These results confirm that embedding expert knowledge directly into neural architectures provides a practical, interpretable, and general framework for learning implicit physics across diverse engineering domains. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of International Journal for Numerical Methods in Engineering is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=196946482 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1002/nme.70399 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 30 StartPage: 1 Subjects: – SubjectFull: Engineering simulations Type: general – SubjectFull: Deep learning Type: general – SubjectFull: Composite materials Type: general – SubjectFull: Combustion engineering Type: general – SubjectFull: Multiscale modeling Type: general Titles: – TitleFull: PINNA: Physics‐Informed Neural Networks Architecture for Implicit Governing Physics With Interpretability. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Yao, Zhengtao – PersonEntity: Name: NameFull: Hawi, Philippe – PersonEntity: Name: NameFull: Aitharaju, Venkat – PersonEntity: Name: NameFull: Mahishi, Jay – PersonEntity: Name: NameFull: Ghanem, Roger IsPartOfRelationships: – BibEntity: Dates: – D: 15 M: 09 Text: Sep2026 Type: published Y: 2026 Identifiers: – Type: issn-print Value: 00295981 Numbering: – Type: volume Value: 127 – Type: issue Value: 17 Titles: – TitleFull: International Journal for Numerical Methods in Engineering Type: main |
| ResultId | 1 |
