Detection of bearing faults in asynchronous motors using Luenberger speed observer

This paper deals with the application of Luenberger speed observer for the detection of damaged rolling bearings in asynchronous machines. Vibration monitoring of mechanical characteristic frequencies related to the bearings is widely used to detect faulty operations. However, vibration measurement is expensive and can not always be performed. An alternative is to base the monitoring on the available electrical quantities e.g. the machine stator current which is often already measured for control and protection purposes. This work investigates another detection approach using estimated variables such as mechanical rotating speed. Considering that bearing faults induce load torque oscillations, a theoretical stator current model in case of load torque oscillations demonstrates the presence of phase modulations. Then, a theoretical estimated rotor flux and speed model in case of load torque oscillations demonstrates the presence of harmonics related to load torque oscillations. These components appear in the estimated speed spectrum and can be used for detection purposes. Measurements show that speed spectral energy is modified in specific frequency ranges corresponding to bearing faults. A fault detector is then proposed. The efficiency of the indicator is studied on long data records of measurements for different operating conditions. The investigated indicator proves that the detection of bearing faults is ensured by using estimated mechanical speed.