XGBoost-based Fault Classification System for Induction Motors with Real-Time Dashboard Integration

In industrial applications, unplanned failures of induction motors can lead to significant operational delays and maintenance costs. This work proposes a predictive maintenance solution using the XGBoost algorithm for early fault classification in three-phase squirrel cage induction motors. The system focuses on detecting faults such as bearing failures, broken rotor bars, and stator short circuits by analysing key sensor inputs including current, vibration, and axial flux data. The machine learning model is trained on labelled datasets collected from both healthy and faulty conditions, enabling accurate and timely identification of motor anomalies. To complement the fault detection system, a real-time monitoring dashboard is developed using FastAPI for the backend and Chart.js for the frontend. This dashboard provides live graphical visualization of sensor data and fault indicators, offering operators an interactive interface to track motor health status. The proposed system achieves high classification performance and offers scalable deployment potential in industrial settings. Its integration of real-time monitoring with intelligent fault analysis supports proactive maintenance, improves.