Fast Bayesian equipment condition monitoring via simulation based inference: applications to heat exchanger health
Peter Collett, Alexander Johannes Stasik,, Simone Casolo, Signe Riemer-Sørensen · 2026
Read Fast Bayesian equipment condition monitoring via simulation based inference: applications to heat exchanger health by Peter Collett, Alexander Johannes Stasik,, Simone Casolo, Signe Riemer-Sørensen (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-20735.txt.
Description
Collett, Stasik, Casolo, and Riemer-Sørensen propose an AI-driven SBI framework to diagnose failure modes in heat units. By training neural density models on sim data, their approach learns a direct mapping from thermal readings to the full posterior distribution of degradation values. They benchmark this framework against an MCMC standard across fouling and leakage cases, including rare faults. The results show that SBI gives comparable accuracy and reliable uncertainty, while speeding checks by 82 times versus sampling.
Catalog record
| CREATOR | Peter Collett, Alexander Johannes Stasik,, Simone Casolo, Signe Riemer-Sørensen |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 54778 |
| LICENSE | http://creativecommons.org/licenses/by-sa/4.0/ |
| RIGHTS | CC BY SA |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.20735