Reproduce “Data-Driven Physics-Informed Neural Networks: A Digital Twin Perspective” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Sunwoong Yang, Hojin Kim, Yoonpyo Hong, Kwanjung Yee, Romit Maulik, Namwoo Kang

The authors develop a parametric model for lid-driven cavity flow, combining governing fluid equations with guide measurements and using Reynolds number as an input. At Reynolds number 4000, multi-fidelity DD-PINN case 4 has a root-mean-square difference (RMSE) of 0.0214 between predicted y-velocity and a 160 × 160-grid DNS reference, versus 0.0565 for single-fidelity DD-PINN, about 62% lower. This numerical demonstration is limited to cavity flow, not a deployed real-time digital twin.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.