This study proposes a method to detect and quantify broken rotor bar fault using zero-sequence voltage in a wye-connected squirrel-cage induction motor. The zero-sequence voltage is analysed and two fault severity factors are defined. The factors are almost independent of motor speed and load torque. Furthermore, it is proved that closed-loop control has little influence on the diagnosis results. Consequently, the method can detect broken rotor bar fault for induction motors in both open-and closed-loop drives. Moreover, the method can discriminate the broken bar fault from a low-frequency load oscillation, even when these two phenomena occur simultaneously. In addition, the influence of unbalanced voltage supply on the diagnosis results is negligible. The effectiveness and robustness of the proposed method have been validated by experimental results.

In an induction motor with broken rotor bars, stator currents contain fault-related components. The fault information propagates to voltages when the motor is in a closed-loop control. The control has an effect on the c u r r e n t sa n dv o l t a g e s .R e l y i n gs o l e l yo nt h ec u r r e n t so rv oltages is difficult to obtain an accurate detection and quantification of the fault. A new online diagnosis method, which takes both currents and voltages into consideration, is proposed in this study. A fault severity factor is proposed to evaluate the extension of the fault, which is almost independent of the control parameters. Furthermore, the me t h o dc a nb ea p p l i e dt od i f f e r e nt control strategies. The effectiveness and robustness of the proposed method have been validated by experimental results.

This study proposes a method for stator fault detection of inverter-fed induction machine in closed-loop control. The sideband response of inverter switching harmonics in common-mode voltage (CMV) is employed to detect the stator inter-turn fault, thus no additional injected signal is necessary. However, the PWM nature harmonics introduce undesired CMV signals, depending on the inherent imbalance or resistance asymmetry. For this reason, the CMV amplitude is not suitable to indicate the fault directly. In this study, the behaviour of switching harmonic is analysed first, and then a simple but efficient stator fault model is demonstrated. To suppress the adverse interference from inverter, an inter-turn fault indicator is proposed by utilising both of the switching sidebands. This method can obtain fault location and severity with low sensitivity to steady operating condition and stator imbalance. The experimental results validate the effectiveness of the proposed method.

In this paper, a seven-phase inset-type permanent magnetic bearingless motor is proposed with characters as the treble frequency suspension force compensation and elimination of the affection of the third plane to the suspension plane. The winding configuration and harmonic distribution are presented. Then, the analysis formulations of inductance and suspension force of the proposed motor are presented and verified by Finite Element Analysis (FEA). Furthermore, the control system is proposed based on the derived voltage model, torque model and suspension force model with the rotor eccentricity being considered. Finally, the Simulink simulation is provided on a prototype motor to verify the validity of the analysis.

For permanent magnet synchronous motors with parallel branches, circulating currents are induced by a partial demagnetisation fault. The induced currents increase the losses of the motor. Therefore, detecting such a fault becomes possible by analysing the motor losses. This study analyses the effects of the partial demagnetisation fault and defines an index, which is the ratio of increased losses and total losses when the motor is under no-load conditions, to monitor the condition of the rotor magnets. The method is easily implemented and sensitive to the fault. The experimental results validate the effectiveness of the proposed method.