aiqbench analysis
The paper adds a hard-constraint projection layer to a physics-informed neural network (PINN) so that predicted velocities satisfy a discretised 1D Burgers' equation exactly. The authors report that the approach keeps reducing loss after a standard PINN plateaus and reaches a data error about an order of magnitude lower on unseen grid points. These are preliminary results from a short workshop paper on a single 1D nonlinear PDE, so readers should treat the gains as unproven for higher-dimensional or more complex fluid problems.
Hard Constraint Projection in a Physics Informed Neural Network (Burgers)
arXiv:2610.42106v1 Announce Type: cross Abstract: Physics informed machine learning provides a framework to study and simulate fluid dynamics, potentially at a fraction of the computational cost of traditional numerical methods, with the goal of improving the explainability, accuracy and robustness of purely data driven machine learning methods. In this work, we extend the hard constraint projection (HCP) method developed by Chen et al. [Journal of Computational Physics, 445, 110624, 2021] from a linear partial differential equation (PDE) to a nonlinear PDE: the 1D Burgers' equation. To strictly impose a discrete approximation of the governing equation, the Burgers' equation is first locally discretised (using a finite difference stencil) then decomposed into a constraint matrix and a prediction matrix. Using tools from linear algebra, the constraint matrix is used to construct a projection matrix. A physics informed neural network (PINN) [Raissi et al. Journal of Computational Physics, 378, 686, 2019] is used to predict the velocity, and a final unlearnable layer projects the velocity prediction to the hyperplane of exact solutions to a discretised form of the Burgers' equation. Our initial results indicate that the HCP-PINN initially trains similarly to an equivalent PINN, and as the PINN loss plateaus, the HCP-PINN loss continues to decline. When tested on a previously unseen uniform grid of points covering the domain, the HCP-PINN exhibits higher accuracy, converging to a data error approximately an order of magnitude lower than the comparable PINN. We also observe that the inclusion of the HCP layer results in better observance of the governing PDE. Our findings suggest that the HCP-PINN can be more robust to the selection of physics collocation points, and less dependent on collocation point quantity.
Abstract:Physics informed machine learning provides a framework to study and simulate fluid dynamics, potentially at a fraction of the computational cost of traditional numerical methods, with the goal of improving the explainability, accuracy and robustness of purely data driven machine learning methods. In this work, we extend the hard constraint projection (HCP) method developed by Chen et al. [Journal of Computational Physics, 445, 110624, 2021] from a linear partial differential equation (PDE) to a nonlinear PDE: the 1D Burgers' equation. To strictly impose a discrete approximation of the governing equation, the Burgers' equation is first locally discretised (using a finite difference stencil) then decomposed into a constraint matrix and a prediction matrix. Using tools from linear algebra, the constraint matrix is used to construct a projection matrix. A physics informed neural network (PINN) [Raissi et al. Journal of Computational Physics, 378, 686, 2019] is used to predict the velocity, and a final unlearnable layer projects the velocity prediction to the hyperplane of exact solutions to a discretised form of the Burgers' equation. Our initial results indicate that the HCP-PINN initially trains similarly to an equivalent PINN, and as the PINN loss plateaus, the HCP-PINN loss continues to decline. When tested on a previously unseen uniform grid of points covering the domain, the HCP-PINN exhibits higher accuracy, converging to a data error approximately an order of magnitude lower than the comparable PINN. We also observe that the inclusion of the HCP layer results in better observance of the governing PDE. Our findings suggest that the HCP-PINN can be more robust to the selection of physics collocation points, and less dependent on collocation point quantity.
6 pages, 3 figures, In proceedings of 1st international Symposium on AI and Fluid Mechanics 2025 (Paper No. S11P2)
Fluid Dynamics (physics.flu-dyn); Machine Learning (cs.LG)
arXiv:2610.42106 [physics.flu-dyn]
(or arXiv:2610.42106v1 [physics.flu-dyn] for this version)
https://doi.org/10.48550/arXiv.2610.42106 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 389 words
Related coverage
SNAP-FM: Sparse Nonlinear Accelerated Projection for Physics-Constrained Generative Modeling
arXiv · Sep 5, 2026
EPC-RawDC: Equivariant Physics Constrained 3D Diffusion with Raw Data Consistency for CBCT to CT Image Synthesis
arXiv · Oct 31, 2026
A SPIDER-type stochastic subgradient method for expectation-constrained nonconvex nonsmooth optimization
arXiv · Nov 3, 2026
RMRRT: Riemannian Barrier Metric RRT for Inequality-Aware Steering on Equality Manifolds
arXiv · Oct 7, 2026