ABSTRACT
This chapter discusses a fault diagnosis system was shaped based on the results and a well-known pattern recognition technique. Bayesian decision theory was introduced first and used as a core to detect and classify possible faults. The problem of fault diagnosis can be viewed as a classification problem if a proper selection of features and classifier is chosen. Pattern-based methods generally consist of templates or patterns distinguishing acceptable and unacceptable operations that are then compared to the system observations to determine whether a fault has occurred. The volume of training data required may be extensive, and only faults represented in the training data can be diagnosed. The faults being investigated and looked at were mainly eccentricity and broken rotor bar faults. However, recognition technique is quite useful for detecting other types of fault as long as proper features are gathered and utilized.
