Diagnosis of BLDC motor faults at different speed conditions using LSSVM

A BLDC motor is a type of DC motor used in electric vehicles. The availability of this component determines the operation of the vehicles. Early detection of motor faults can help to avoid sudden motor failure. This paper aims to diagnose the damage to the BLDC motor that operates at different speeds using the least square support vector machine (LSSVM) method. The motor in healthy condition and the motor with a bearing fault are examined. The vibration signals are captured from the BLDC motor operated at 450 rpm, 500 rpm, and 550 rpm. The vibration signals are acquired at a 20 kHz sampling rate. The rms and peak-to-rms ratio are extracted from the vibration signals and used as fault indicators. The diagnosis process is performed by classifying the motor condition using the LSSVM model. The model is built from the training data. Then the model is tested using random data. The results show that the LSSVM model performs excellently in diagnosing the BLDC motor fault. The accuracies are 100%, 99.6%, and 100% for the motor speeds of 450, 500, and 550 rpm, respectively. It indicates that the method is stable when used at various operation speeds. Thus, the LSSVM models have the potential as a tool for the diagnosis BLDC motor damage.