Current Data Fusion through Kalman Filtering for Fault Detection and Sensor Validation of an Electric Motor

In this paper, a data fusion method through Kalman filtering for condition monitoring (CM) and fault detection (FD) of electrical motors (EM) is proposed. Moreover, sensor validation (SV) and tracking the fault source (either the sensor or the process) are possible through this approach. A current signal, obtained from different sensors, is used for the case study. A fused current information is calculated through Kalman filtering. Afterwards, the effects of the measurement and process noises on the fused signal, are discussed, respectively. Then, it is noticed distinctive features by the comparison of the fused and original signal in terms of spectral and statistical properties. In addition, Kalman gain is monitored to investigate the impact of the process noise and measurement noise to perform SV. The proposed method is developed on the artificial data and then tested on the real data collected from an Induction Motor (IM).