This study focuses on the application of AdaBoost for stator fault diagnosis in three-phase permanent magnet synchronous motors (PMSMs) using vibration–current data fusion analysis. PMSMs are widely used in industrial applications due to their precise control capabilities, but they are prone to operational faults that can affect safety and performance. The study introduces a qualification-based methodology to detect early failures through health state monitoring. Stator faults were induced using bypassing resistances, and experimental datasets were acquired from a test rig, including current and vibration time-domain signals. These signals were transformed into statistical features, which were then analyzed using AdaBoost, a machine learning model. The results showed that vibration statistical features alone achieved an accuracy of 83.0%, while vibration–current data fusion achieved 90.7%, the highest accuracy. The precision, F1 score, and recall values were all 0.907, validating the effectiveness of the data fusion methodology. This study highlights the potential of data fusion analysis in early fault diagnosis, enabling proactive maintenance strategies and enhancing the reliability of PMSMs in various industrial and renewable energy applications.This study focuses on the application of AdaBoost for stator fault diagnosis in three-phase permanent magnet synchronous motors (PMSMs) using vibration–current data fusion analysis. PMSMs are widely used in industrial applications due to their precise control capabilities, but they are prone to operational faults that can affect safety and performance. The study introduces a qualification-based methodology to detect early failures through health state monitoring. Stator faults were induced using bypassing resistances, and experimental datasets were acquired from a test rig, including current and vibration time-domain signals. These signals were transformed into statistical features, which were then analyzed using AdaBoost, a machine learning model. The results showed that vibration statistical features alone achieved an accuracy of 83.0%, while vibration–current data fusion achieved 90.7%, the highest accuracy. The precision, F1 score, and recall values were all 0.907, validating the effectiveness of the data fusion methodology. This study highlights the potential of data fusion analysis in early fault diagnosis, enabling proactive maintenance strategies and enhancing the reliability of PMSMs in various industrial and renewable energy applications.