Metraipolis › archive › dc_ca4ca5465449

FreqCondNorm: Towards Cross-domain Predictive Maintenance through a Frequency-Conditioned Transformer Foundation Model — arXiv:2609.20535

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.20535
License (as archived)
CC0-1.0
Archived
2026-10-10T03:22:11.030Z
SHA-256 of the archived text
e8d17430fd321daa8ba39f32f463bd34819170086df8dde06dcf68325333bbcd
Stable URL of this page
https://metraipolis.com/documento/dc_ca4ca5465449

How to cite

FreqCondNorm: Towards Cross-domain Predictive Maintenance through a Frequency-Conditioned Transformer Foundation Model — arXiv:2609.20535 — arXiv (en), https://arxiv.org/abs/2609.20535 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ca4ca5465449 (Metraipolis archive dc_ca4ca5465449, sha256 e8d17430fd321daa8ba39f32f463bd34819170086df8dde06dcf68325333bbcd)
BibTeX
@misc{dcca4ca5465449,
  title = {FreqCondNorm: Towards Cross-domain Predictive Maintenance through a Frequency-Conditioned Transformer Foundation Model — arXiv:2609.20535},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.20535},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ca4ca5465449; sha256 e8d17430fd321daa8ba39f32f463bd34819170086df8dde06dcf68325333bbcd},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.20535. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Deep learning predictive maintenance models suffer from poor transferability across machines and operating conditions, especially when labelled data are scarce and signals span five orders of magnitude in sampling frequency (1 Hz to ~100 kHz). We propose FreqCondNorm, a Transformer-based architecture that introduces a FiLM-style frequency-conditioned normalization layer to unify heterogeneous time-series within a single model. The architecture is pretrained on five public predictive maintenance datasets (CWRU, MFPT, UOC18, PRONOSTIA, CMAPSS) using masked auto-encoding and contrastive learning with balanced domain sampling. On fault diagnosis, the model achieves 99.2% accuracy on CWRU (+6.4 pp over CNN) and 82.1% zero-shot accuracy on MFPT, demonstrating strong transfer across sampling frequencies. However, the approach does not improve remaining useful life prediction, suggesting a mismatch between pretraining and RUL objectives that warrants future investigation.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.20535 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-10T03:04:13.749Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).