End-to-end robust system discovery in electrical dynamical systems using scientific machine learning
Jhelum Chakravorty, Nicolò Ripamonti, Tor Laneryd · 2026
Read End-to-end robust system discovery in electrical dynamical systems using scientific machine learning by Jhelum Chakravorty, Nicolò Ripamonti, Tor Laneryd (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/PMC13645738.txt.
Description
Jhelum Chakravorty, Nicolò Ripamonti, and Tor Laneryd present a 2026 paper on finding hidden models of electrical machines from data, a task called system discovery. Power engineers need such models to manage equipment, yet real data are often sparse and noisy. The trio use a method that blends machine learning with known physics. Their test case is the heat behavior of electrical induction machines.
Their end-to-end approach beat nearly all rival methods, cutting validation error by wide margins in most comparisons. One rival Bayesian method scored slightly better but took far longer to run, while their method cut computing time sharply. A sampling trick made the predictions hold up when data were thin. The authors frame the work as a balance of accuracy and cost that could suit real asset management in power systems.
Catalog record
| CREATOR | Jhelum Chakravorty, Nicolò Ripamonti, Tor Laneryd |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 84519 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |