| Supervised fault diagnosis of inverter-driven permanent magnet synchronous motors (PMSMs) typically requires large volumes of expert-annotated data, which are expensive to obtain in industrial settings. This paper presents a systematic benchmark of six pool-based active learning (AL) strategies, random sampling, uncertainty (entropy), Bayesian Active Learning by Disagreement (BALD), Query-by-Committee (vote entropy), CoreSet (k-center), and a hybrid uncertainty–diversity method, combined with four tree-based classifiers on a nine-class PMSM fault dataset comprising 10 857 samples and 25 engineered features. Experiments are conducted under stratified five-fold cross-validation across ten labelling budgets (1–100%). Results show that committee-based strategies paired with Random Forest reach a Macro-F1 of 0.9998 ± 0.0004 using only 10% of the available labels, matching full-budget performance. Across all 24 model–strategy combinations the median annotation saving is 97%, with several configurations requiring as few as 2% of labels to retain 95% of peak performance (defined relative to the 100% budget). These findings demonstrate that active learning can drastically reduce labelling effort for PMSM fault diagnosis without sacrificing classification quality. |
*** Title, author list and abstract as submitted during Camera-Ready version delivery. Small changes that may have occurred during processing by Springer may not appear in this window.