Simulation-driven machine learning: Bearing fault classification

• Published in: Mechanical systems and signal processing Vol. 99; pp. 403 - 419
• Main Authors: Sobie, Cameron, Freitas, Carina, Nicolai, Mike
• Format: Journal Article
• Language: English
• Published: Berlin Elsevier Ltd 15.01.2018; Elsevier BV
• Subjects: Algorithms; Artificial intelligence; Artificial neural networks; Bearing races; Classification; Computer simulation; Condition monitoring; Failure detection; Fault detection; Faults; Machine learning; Machinery condition monitoring; Prognostic health monitoring; Roller bearing; Roller bearings; Statistical analysis; Statistical methods; Studies; Condition monitoring; Roller bearing; Fault detection; Machine learning; Prognostic health monitoring
• ISSN: 0888-3270; 1096-1216
• DOI: 10.1016/j.ymssp.2017.06.025

•Data driven condition monitoring is limited to cases where in-service data exists.•Simulation models are proposed as a data source for machine learning training data.•Simulation-trained classifiers achieve up to 94% accuracy on four exp. datasets. Increasing the accuracy of mechanical fault detection has the potential to improve system safety and economic performance by minimizing scheduled maintenance and the probability of unexpected system failure. Advances in computational performance have enabled the application of machine learning algorithms across numerous applications including condition monitoring and failure detection. Past applications of machine learning to physical failure have relied explicitly on historical data, which limits the feasibility of this approach to in-service components with extended service histories. Furthermore, recorded failure data is often only valid for the specific circumstances and components for which it was collected. This work directly addresses these challenges for roller bearings with race faults by generating training data using information gained from high resolution simulations of roller bearing dynamics, which is used to train machine learning algorithms that are then validated against four experimental datasets. Several different machine learning methodologies are compared starting from well-established statistical feature-based methods to convolutional neural networks, and a novel application of dynamic time warping (DTW) to bearing fault classification is proposed as a robust, parameter free method for race fault detection.