Detection of a motor bearing shield fault using neural networks

Condition-based maintenance (CBM) has attracted more attention and interest due to its advantages over the conventional breakdown-based or time-based maintenance. CBM of electrical machines such as motors is based on using data obtained by real-time condition monitoring, and fault detection and diagnosis to recommend an optimized maintenance. This paper presents an application of an Artificial Neural Network (ANN) for detecting a very small fault in a bearing shield of an induction motor. The experimental results show that the incipient fault can be efficiently detected. An alarm may be activated so that corrective actions are promptly taken before the detected fault manifests itself to be further serious failures.