Cost-efficient real-time condition monitoring and fault diagnostics system for BLDC motor using IoT and Machine learning

Abstract

A cost-efficient condition monitoring and fault diagnostic system are presented in this paper using the Internet of Things and machine learning. Most condition monitoring systems nowadays are either costly or used to monitor current values without emphasizing the analysis part. On the other hand, predictive maintenance of different electrical machines, including BLDC motors, is becoming the need of the hour. It reduces the cost needed for maintenance and can also be used to evade more significant faults in the machine. The data is transmitted in real-time using a data acquisition system onto the cloud, which is further processed to determine if there is a chance of any fault occurring in the motor. A short comparison of the results of different machine learning algorithms is also discussed related to predictive maintenance.