Application of Machine Learning to Mechanical Equipment Diagnostics
Machines of all varieties have failure modes which if predicted can reduce business costs, reduce downtime and increase safety. Failure prediction can be model- or data-based. Both methods require that patterns be recognized which provide an indication of the failure mode. Machine learning methods such as convolutional neural networks are being used to increase the speed and accuracy of failure predictions. This course is an introduction to the application of machine learning to equipment diagnostics. Data requirements and sources, types of applicable machine learning, implementation approaches to machine learning and use cases demonstrating the application of machine learning for failure prediction will be discussed.
Benefits and Learning Outcomes
- Learn the steps necessary to apply machine learning to equipment diagnostics
- Understand requirements for equipment operating data and sources of data to support the implementation of equipment diagnostics
- Examine examples of machine learning applied to diagnostics which can be used as references for action toward its application