To detect a stator winding fault caused by a short-circuited turn in a permanent magnet synchronous motor, a simple online fault detecting scheme is presented. The proposed scheme is based on the monitoring of the second-order harmonic components in the q-axis current through harmonic analysis. To verify the effectiveness of the proposed scheme, a test motor which allows an interturn short in the stator winding has been built, and the entire control system has been implemented using the DSP TMS320F28335.

A simple parameter-independent online compensation scheme for the dead time and inverter nonlinearity in an interior-permanent-magnet-synchronous-motor drive through waveform analysis is presented. The effectiveness of the proposed scheme is verified through the simulations and experiments using the DSP TMS320F28335. Without requiring any additional hardware, the proposed scheme can effectively compensate the dead time and inverter nonlinearity even in the presence of the parameter uncertainty.

To enhance the reliability in an inverter-fed linear induction motor (LIM) drive under simultaneous open-switch faults, practical online basis fault detection and localization schemes are presented. The open-switch faults in the inverter incur overcurrent stress to other electric components, resulting in a severe secondary fault in the entire system. As compared with the conventional schemes that the fault detection and faulty switch identification rely on the averaging or current error from the reference value, the proposed algorithms are achieved through signal processing schemes such as harmonic analysis, averaging, and waveform analysis by using only the measured current information. In addition, the proposed scheme classifies the open-switch faults as four faulty groups according to their unique fault signature, which enables the proposed scheme to identify faulty switches effectively. Since the fault detection is accomplished by using the relative magnitude of the second-order harmonic component in phase currents, the influence on steady-state error or dc offset of current is significantly reduced. Once the fault is detected, the operating mode is changed to identify the faulty group which the fault belongs to. By identifying the faulty group, the faulty switches can be found in a much more straightforward way. The entire LIM drive system is implemented using DSP TMS320F28335 to verify the feasibility of the proposed fault detection and faulty switch identification schemes. Without requiring any additional hardware or measuring apparatus, the proposed diagnostic algorithm can effectively detect a fault even under simultaneous open-switch faults by online basis.

A current control technique for a permanent magnet synchronous motor (PMSM) with a simple disturbance estimation scheme is proposed. Among the various current control schemes for a PMSM drive, predictive control is known to give a superior performance. This scheme, however, requires the full knowledge of motor parameters. To overcome such a limitation, the disturbances caused by the parameter variations will be estimated by using disturbance observer theory and used for the calculation of the reference voltages by feedforward control. Thus, the control performance can be significantly improved with a relatively simple control algorithm. The proposed control scheme is implemented on a PMSM using DSP TMS320C30 and the effectiveness is verified through the comparative experiments.

To analyse influences under various fault conditions, a fault model of an inverter-fed permanent magnet (PM) synchronous motor drive that can be used for a performance assessment of a diagnostic algorithm is presented. For fault conditions of a drive system, a short-circuited turn in the stator winding, switching device open in pulse width modulated inverter and the isolation between the inverter and motor are considered. Even though the conventional dq model is widely used to control an AC motor, it cannot be used directly for the analysis of a motor fault, since three-phase balanced condition does not hold under such conditions, and thus, it is not easy to obtain phase voltage inputs of motor from the pole voltage. To overcome this limitation, a fault model of an inverter-fed PM synchronous motor is derived using the line voltage, which can be effectively used to evaluate the performance of a fault detection algorithm. A test motor producing short-circuited turn in the stator winding is built. The validity of the proposed fault model is verified through the comparative simulations and experiments using digital signal processor TMS320F28335 under various fault conditions.