An Open-Access Benchmark of Statistical and Machine-Learning Anomaly Detection Methods for Battery Applications

Mei-Chin Pang, Suraj Adhikari, Takuma Kasahara, Nagihiro Haba, Saneyuki Ohno · 2025

Read An Open-Access Benchmark of Statistical and Machine-Learning Anomaly Detection Methods for Battery Applications by Mei-Chin Pang, Suraj Adhikari, Takuma Kasahara, Nagihiro Haba, Saneyuki Ohno (2025) 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-2511-01745.txt.

Description

Mei-Chin Pang and four colleagues release OSBAD, an open benchmark for finding faults in battery data. Missed faults can mean fires, failures, or costly downtime, in goods from phones to electric aircraft. Yet rival methods are most often compared on private data with patchy labels. That makes progress hard to judge.

OSBAD runs 15 methods head to head. The set spans statistical, distance-based, and unsupervised machine learning methods, across varied datasets. A physics- and statistics-guided change of features makes faults stand out more clearly. It breaks group faults into single-point ones. A Bayesian tuning pipeline sets each method's knobs on its own, working around the lack of labels. Tests span liquid and solid-state chemistries. The methods carry across both, a base for safer battery monitoring.

Catalog record

CREATORMei-Chin Pang, Suraj Adhikari, Takuma Kasahara, Nagihiro Haba, Saneyuki Ohno
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE108 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2511-01745

Browse the full library · About the Tower of Babel