Prognosis Of BLDC Drive Faults Using Autoregressive Integrated Moving Average Algorithm

Abstract

Abstract Generally, Brushless DC (BLDC) machines attract many industrialists due to their unique characteristics like better output, stabilized performance, and high torque to current ratio. BLDC drive has a long life, and they do not need maintenance; however, the drive has low starting torque and high cost. Thus, Non-stop monitoring and future prediction methods can reduce fault occurrence and improve system performance. In this paper, we have proposed the prognosis of BLDC drive faults using the Autoregressive Integrated Moving Average (ARIMA) Algorithm. Here, we consider the open circuit (OC) and short circuit (SC) faults in BLDC drive to prognosis by ARIMA technique. The ARIMA has a fixed structure, and it is particularly built for time series data. By data acquisition system, the drive parameters such as current, torque, and speed will be continuously obtained. Filtering out the high-frequency noise present in the data is the main principle of the ARIMA model. Matlab/Simulink platform is used to implement the process and analyze the results using prediction efficiency, the fault analysis in speed, flux, torque, current, and voltage.