What the study found
The study proposes an efficient and accurate framework for nonlinear space-time fractional diffusion equations. The authors report that the resulting spectral deferred correction method keeps arbitrary order accuracy and excellent stability.
Why the authors say this matters
The authors state that the method is efficient and accurate, and they present a dual accelerated algorithm for exact matrix-vector product computation because the fractional Laplacian approximation is dense. They also conclude that the approach maintains stability while handling the nonlinear term without Newtonian methods.
What the researchers tested
The researchers based their approach on spectral deferred correction, using a compact difference scheme as a preconditioner through the Picard integral collocation formulation. They incorporated the nonlinear term into the preconditioner in a way similar to linear systems and introduced a discrete sine transform to accelerate matrix-vector products.
What worked and what didn't
The preconditioner is proven to be a stable operator. Numerical results show that the proposed methods are highly efficient and precise. The abstract does not describe any specific failures or cases where the approach did not work.
What to keep in mind
The available summary does not provide detailed numerical settings, comparison baselines, or problem-specific limitations. It also does not state any caveats beyond noting the dense matrix property of the fractional Laplacian approximation.
Key points
- The paper proposes a framework for nonlinear space-time fractional diffusion equations.
- The method is based on spectral deferred correction with a compact difference preconditioner.
- A discrete sine transform is used to accelerate exact matrix-vector product computation.
- The preconditioner is proven stable, and the method preserves arbitrary order accuracy.
- Numerical results are reported as highly efficient and precise.
Disclosure
- Research title:
- Accelerated methods improve nonlinear fractional diffusion solving
- Authors:
- Yiyin Liang, Shichao Yi
- Institutions:
- Jiangsu University of Science and Technology, Shanghai Shipbuilding Technology Research Institute
- Publication date:
- 2026-04-24
- OpenAlex record:
- View
Get the weekly research newsletter
Stay current with peer-reviewed research without reading academic papers — one filtered digest, every Friday.