Reproduce “Physical-Variable-Informed Deep Learning for Robust Diagnosis in Industrial Fluid Transport Systems” 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
Yu Sha, Junqi Tao, Dixin Zhou, Horst Stoecker, Domagoj Vnucec, Andreas Widl, Kai Zhou
PhyNet combines routine physical measurements with local response signals for fault diagnosis without point-by-point temporal alignment. It encodes physical sequences as tokens and response spectrograms as patches, using a global response token to pass operating context to local evidence. On Cavitation-Long, accuracy is 96.59%, versus 94.03% for PerceiverIO; across four fluid-transport benchmarks, it ranks first on four reported metrics.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.