Risk-sensitive collaborative parameter tuning via calibrated deep surrogates for rare-earth electrolysis energy efficiency
Yu Liu, Jun Peng, Fan Yang, Haitao Guo · 2026
Read Risk-sensitive collaborative parameter tuning via calibrated deep surrogates for rare-earth electrolysis energy efficiency by Yu Liu, Jun Peng, Fan Yang, Haitao Guo (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/pmc-13473459.txt.
Description
Liu and co-authors propose a cautious way to tune rare-earth electrolysis plants, where chasing energy savings with aggressive settings can breach hard operating limits. Their 2026 framework pairs a deep ensemble surrogate model, several neural networks voting with stated uncertainty, with calibration that makes those uncertainty estimates honest before optimization trusts them.
Five process knobs are tuned jointly: current, voltage, temperature, feeding rate, and electrolyte concentration. In test runs under a fixed evaluation budget, the risk-sensitive method converged faster, reached a higher best-so-far energy-efficiency score, and produced a far lower rate of constraint violations than uncertainty-blind alternatives. The authors note validation so far sits at the aggregated interval level, with live closed-loop plant trials still ahead.
Catalog record
| CREATOR | Yu Liu, Jun Peng, Fan Yang, Haitao Guo |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 137 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |