A Tensor-Network Chebyshev--Galerkin Method for Boundary-Value Problems in Computational Mechanics
Abstract
Quantum-inspired numerical methods for partial differential equations represent discretized fields and operators as quantics tensor trains (QTT), so that their cost depends on how compressible the problem is rather than on the number of unknowns. Global spectral formulations in this setting have so far relied on Fourier bases, restricted to periodic problems, or on collocation, which does not preserve the variational structure exploited by tensor-network solvers. We introduce a QTT Chebyshev--Galerkin method that assembles and solves global spectral systems directly in compressed form, without forming the underlying arrays. Starting from the weak formulation, the method handles coupled elliptic systems with variable coefficients, smoothly mapped geometries, and mixed Dirichlet--Robin boundary conditions, and yields symmetric positive-definite systems suited to variational tensor-network solvers. Parameters and random inputs enter as additional coordinates, so the same construction returns the solution over an entire parameter space in a single solve. In linear elasticity, the method retains spectral convergence and reaches errors below 10−8 on a three-dimensional benchmark, beyond those attained by QTT finite differences and Q2/Q4 FEniCSx finite elements over the tested range. A six-parameter cantilever design study covering 1.7×107 designs is solved in a single 92~s run, and a stochastic problem with ten random coordinates is represented with 2.4×106 times less memory than its full coefficient array.
Related
- Stochastic Galerkin Method for Fractional Boundary Value Problems: Convergence Analysis and Numerical Treatment
- A nonoverlapping spectral additive Schwarz method for interior penalty discontinuous Galerkin discretizations of Anisotropic Elliptic Problems
- A Discontinuous Galerkin Method for Biharmonic Dirichlet Control Problems With Patch Reconstruction
- A super-convergent quasi-discontinuous Galerkin method for advection-diffusion problems
- Deep learning methods for stochastic Galerkin approximations of random domain problems
BibTeX
@misc{rodriguezaldavero2026tensornetwork,
title = {{A Tensor-Network Chebyshev--Galerkin Method for Boundary-Value Problems in Computational Mechanics}},
author = {Juan José Rodríguez-Aldavero and Ángel de la Rosa Velasco and Juan José García-Ripoll},
howpublished = {arXiv},
year = {2026},
month = nov,
doi = {10.48550/arxiv.2611.06552},
eprint = {2611.06552},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2611.06552},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - A Tensor-Network Chebyshev--Galerkin Method for Boundary-Value Problems in Computational Mechanics
AU - Juan José Rodríguez-Aldavero
AU - Ángel de la Rosa Velasco
AU - Juan José García-Ripoll
PY - 2026
DA - 2026/11/04
PB - arXiv
DO - 10.48550/arxiv.2611.06552
UR - https://arxiv.org/abs/2611.06552
AB - Quantum-inspired numerical methods for partial differential equations represent discretized fields and operators as quantics tensor trains (QTT), so that their cost depends on how compressible the problem is rather than on the number of unknowns. Global spectral formulations in this setting have so far relied on Fourier bases, restricted to periodic problems, or on collocation, which does not preserve the variational structure exploited by tensor-network solvers. We introduce a QTT Chebyshev--Galerkin method that assembles and solves global spectral systems directly in compressed form, without forming the underlying arrays. Starting from the weak formulation, the method handles coupled elliptic systems with variable coefficients, smoothly mapped geometries, and mixed Dirichlet--Robin boundary conditions, and yields symmetric positive-definite systems suited to variational tensor-network solvers. Parameters and random inputs enter as additional coordinates, so the same construction returns the solution over an entire parameter space in a single solve. In linear elasticity, the method retains spectral convergence and reaches errors below 10^−8 on a three-dimensional benchmark, beyond those attained by QTT finite differences and Q2/Q4 FEniCSx finite elements over the tested range. A six-parameter cantilever design study covering 1.7×10^7 designs is solved in a single 92~s run, and a stochastic problem with ten random coordinates is represented with 2.4×10^6 times less memory than its full coefficient array.
ER -