19 August 2025 in IEEE Sensors Journal
Electrical erosion caused by bearing currents is one of the main reasons for bearing failure in inverter-fed motors. The research of non-intrusive measurement of bearing currents contributes to risk assessment and early warning of bearing electrical erosion in electric motors, thereby enhancing the reliability of motor drive systems. This paper proposes a novel non-intrusive measurement method of bearing currents by measuring external electrical signals (bearing voltage and common-mode currents). Firstly, the relationship between bearing currents, bearing voltage, and common-mode (CM) currents is analyzed based on the CM model of the motor. Subsequently, the bearing voltage and CM currents are measured to detect and extract the characteristic signals induced by the discharge of the bearing oil film. The bearing voltage characteristic signals are detected using the instantaneous rate of change, while the CM current characteristic signals are extracted through a time-frequency features clustering method. The time-domain waveform of bearing currents is estimated by using the relationship between bearing currents and measured signals. Finally, the accuracy and effectiveness of the proposed method are verified by comparison with the intrusive measurement method. The proposed method demonstrates higher accuracy and reliability than traditional non-intrusive measurement methods.
12 March 2026 in IEEE Transactions on Industrial Electronics
The article proposes a model predictive suspension force control (MPSFC) to suppress the rotor’s radial displacement of bearingless switched reluctance motor (BSRM) during suspension. It can handle the complex coupling between the torque and suspension systems and between suspension forces in multiple directions which exhibit time-varying characteristics. Within each control step, model predictive control (MPC) will be first implemented for the torque system, where the predicted torque winding current can be obtained to achieve decoupling between the torque winding and the suspension winding with the mutual inductance model. Then, the suspension winding current and suspension force will be predicted sequentially through the prediction model considering one-step delay compensation. The concept of suspension force vector (SFV) is also proposed in this article to present the combined effect of mutually coupled suspension forces. The optimal control voltage vector will be selected by cost function to minimize the deviation between the predictive SFV and the desired one. The experiment results on a three-phase 12/8 BSRM demonstrate the proposed method can significantly reduce the energy consumption and radial displacement of the rotor and achieve robust suspension under low speed and external interference.
1 January 2026 in IEEE Transactions on Transportation Electrification