Unsupervised Machine Learning for Anomaly Detection in Beam Position Monitors Data

Turn-by-turn beam position monitors (TbT-BPMs) are the primary diagnostic tool for optics measurement and correction in high-luminosity colliders. Reliable identification of faulty sensors is essential for any optics reconstruction pipeline, yet manual inspection does not scale to the hundreds of monitors in current rings or the thousands expected in future machines, such as the FCC. We apply two unsupervised machine-learning methods to TbT-BPM data from both rings of SuperKEKB: density-based spatial clustering of applications with noise (DBSCAN) with automatic feature selection via the Time2Feat library, and a one-dimensional convolutional autoencoder (1D-CNN AE) that operates directly on raw data without labelled training examples. Both methods are evaluated on 2024 data against a reference list of hardware-damaged BPMs provided by the SuperKEKB operations team. On the High Energy Ring (HER), DBSCAN detects faulty BPMs with a recall (TPR) of 83%, and the 1D-CNN AE reaches the same value. On the Low Energy Ring (LER), both methods recover all known-faulty BPMs (100% TPR). Read at a matched reporting depth, the two methods return the same recall and precision on both rings, and differ only in which monitors outside the reference list they flag. The models further identify strong candidates for unreported hardware faults, ranking them among the top outliers for the HER and LER measurement sessions. Nine such monitors are flagged independently by both methods and are absent from the reference list, so the precision reported here is a lower limit.