Condition Monitoring of Ball Bearing Using MEMS-based Accelerometer

Abstract

Industry 4.0 has attracted a lot of attention in recent times. Intelligent machines are at the heart of Industry 4.0. Condition monitoring and predictive maintenance of critical components in rotating machines are crucial for intelligent machines. This paper studies the use of a commercially available Micro Electro Mechanical System (MEMS) accelerometer for the condition monitoring of ball bearings. Embedding the sensor in the housing reduces the transmission path between the sensor and the fault providing efficient condition monitoring for low-speed applications. A comparison between MEMS and piezoelectric accelerometers has been made. Defects are induced artificially into the bearing and fault classification of the bearing has been done using a machine learning algorithm.