An Adaptive Kalman Filter-Based Condition-Monitoring Technique for Induction Motors

Induction motors are typical rotating machines that are widely used in various industrial processes. The condition of induction motors has to be monitored to avoid serious losses, which can be caused by various reasons. Over the last decades, although many studies have been performed on the condition monitoring (CM), there is still an increasing need for cost-effective and reliable CM techniques for induction motor. This paper presents an adaptive Kalman filter (AKF)-based CM technique for an induction motor driving a scrubber fan. In this work, AKFs are used to extract useful information about the induction motor's condition based on measured vibration signals. The main novelty of the proposed method is the use of multiple AKFs for the detection of outliers and anomalies. The output of the AKFs plays as the basis of severity assessment on the vibration signals. A set of AKFs are employed to deal with various anomaly conditions caused by different severity levels of vibration as the IM is deteriorated. Moreover, the effectiveness of the proposed method is demonstrated through experiments involving a real scrubber fan driven by an induction motor.