A spectral–physical manifold for scalable knowledge transfer in partial differential equation families
Authors:
Preprint
- 6 February 2026
- preprint
- Published by Springer Nature in Research Square
Abstract
High-fidelity simulation across geometrically heterogeneous physical systems is fundamentally constrained by the cost of resolving multiscale features over large design spaces. Although governing partial differential equations remain invariant, topological variations induce structured distortions in the solution space that existing neural operators and transfer learning methods fail to generalize across. Multi-task PDE learning is reformulated as navigation on a spectral–physical manifold, in which invariant differential operators are explicitly disentangled from geometry-specific adaptation. This formulation enables physically consistent knowledge diffusion across tasks with distinct topologies. Validated on 970 three-dimensional multi-physics problems with diverse geometries and canonical benchmarks in elasticity and heat conduction, the framework achieves fine-grid accuracy (R2>0.95) from coarse-grid supervision, converges faster with lower physical residuals than state-of-the-art operator-learning baselines, and reduces computational cost by approximately one to two orders of magnitude compared with conventional numerical simulation workflows, establishing a scalable and data-efficient paradigm for physics-based digital twins.Keywords
All Related Versions
This publication has 33 references indexed in Scilit:
- A physics-informed meta-learning framework for the continuous solution of parametric PDEs on arbitrary geometriesComputers & Structures, 2026
- From PINNs to PIKANs: recent advances in physics-informed machine learningMachine Learning for Computational Science and Engineering, 2025
- iPINNs: incremental learning for Physics-informed neural networksEngineering with Computers, 2024
- Linel2D-Net: A deep learning approach to solving 2D linear elastic boundary value problems on image domainsiScience, 2024
- Transferable Neural Networks for Partial Differential EquationsJournal of Scientific Computing, 2024
- Deep learning operator network for plastic deformation with variable loads and material propertiesEngineering with Computers, 2023
- Data-driven and physics-informed deep learning operators for solution of heat conduction equation with parametric heat sourceInternational Journal of Heat and Mass Transfer, 2022
- Predicting high-fidelity multiphysics data from low-fidelity fluid flow and transport solvers using physics-informed neural networksInternational Journal of Heat and Fluid Flow, 2022
- Finding the Homology of Submanifolds with High Confidence from Random SamplesDiscrete & Computational Geometry, 2006
- From graph to manifold Laplacian: The convergence rateApplied and Computational Harmonic Analysis, 2006