ANN Assisted Multi Sensor Information Fusion for BLDC Motor Fault Diagnosis

Multiple sensor data fusion is necessary for effective condition monitoring as the electric machines operate in a wide range of diverse operations. This study investigates acquired vibration and current signals to establish reliable multi-fault diagnosis framework brushless DC (BLDC) motor. Faults stator rotor were created deliberately by shorting two adjacent windings creating hole on surface, respectively. The threshold different health states was obtained third harmonic analysis motor current. Later, key features from are selected based monotonicity reduced using principal component (PCA). For future predictions, an artificial neural network (ANN) used classify fault its performance evaluated several metrics. Analysis harmonics impulsive response at same time provides thorough estimation BLDC presence both electrical mechanical faults. information fused obtain better understanding characteristics mitigate randomness diagnosis. proposed model able detect multiple with higher accuracy compared other similar methods.