DEPARTMENT OF COMPUTATIONAL AND DATA SCIENCES
M.Tech Research Thesis Colloquium
Speaker : Ms. Diya Nag Chaudhury
S.R. Number : 06-18-01-10-22-23-1-23885
Title : “A physics-informed framework for super-resolution of fluid flows”
Research Supervisor : Prof. Sashikumaar Ganesan
Date & Time : January 15, 2026, 10.00 AM
Venue : # 102 CDS Seminar Hall
ABSTRACT
The reconstruction of high-resolution flow fields from low-resolution data remains a persistent challenge within fluid dynamics. Super-resolution using deep learning is a valuable tool for enhancing visualizations of flow fields to recover detailed features from low-resolution data. This work introduces the Bicubic Fourier Neural Operator (Bicubic FNO), which combines the benefits of both bicubic interpolation and FNO.
We compare our architecture against standard FNO and two convolutional neural network (CNN) models reported in the literature for super-resolution: Super-resolution Convolutional Neural Network (SRCNN), widely used for natural images, and the DSC/MS model, designed for reconstructing high-resolution turbulent flow fields. We tested our model on two-dimensional decaying homogeneous isotropic turbulence and achieved a 10.47% improvement over the standard FNO and a 19.59% improvement over the DSC/MS model in terms of Peak Signal-to-Noise Ratio (PSNR). Moreover, for a similar performance, the Bicubic FNO led to a significant reduction of 74% in parameter count over standard FNO. We extend our research to integrate physics constraints into the framework, ensuring that the model is able to reconstruct physically consistent flow fields even with extremely sparse data.
This research highlights the contributions of Bicubic FNO, its extension with physical constraints and demonstrates that we can extract meaningful turbulent features while significantly reducing the dependency on massive datasets.
ALL ARE WELCOME
