TY  - EJOU
AU  - Shin, Sojin 
AU  - Kim, Guk Heon 
AU  - Kim, Seung Hwan 
AU  - Kim, Jaemin 

TI  - A Surrogate Deep-Learning Super-Resolution Framework for Accelerating Finite Element Method-Based Fluid Simulations
T2  - Computer Modeling in Engineering \& Sciences

PY  - 2026
VL  - 146
IS  - 3
SN  - 1526-1506

AB  - This study develops a surrogate super-resolution (SR) framework that accelerates finite element method (FEM)-based computational fluid dynamics (CFD) using deep learning. High-resolution (HR) FEM-based CFD remains computationally prohibitive for time-sensitive applications, including patient-specific aneurysm hemodynamics where rapid turnaround is valuable. The proposed pipeline learns to reconstruct HR velocity-magnitude fields from low-resolution (LR) FEM solutions generated under the same governing equations and boundary conditions. It consists of three modules: (i) offline pre-training of a residual network on representative vascular geometries; (ii) lightweight fine-tuning to adapt the pretrained model to geometric variability, including patient-specific aneurysm morphologies; and (iii) an unstructured-to-structured sampling strategy with region-of-interest upsampling that concentrates resolution in flow-critical zones (e.g., the aneurysm sac) rather than the full domain. This targeted reconstruction substantially reduces inference and post-processing cost while preserving key HR flow features. Experiments on cerebral aneurysm models show that HR velocity-magnitude fields can be recovered with accuracy comparable to direct HR simulations at less than 1% of the direct HR simulation cost per analysis (LR simulation and SR inference), while adaptation to new geometries requires only lightweight fine-tuning with limited target-specific HR data. While clinical endpoints and additional variables (e.g., pressure or wall-based metrics) are left for future work, the results indicate that the proposed surrogate SR approach can streamline FEM-based CFD workflows toward near real-time hemodynamic analysis across morphologically similar vascular models.
KW  - Surrogate modeling; deep learning; super-resolution; finite element method (FEM); fluid simulation

DO  - 10.32604/cmes.2026.079127