Search

The design of power electronic converters for the integration of wind generated power into the\ud grid is more and more important due to a new class of Superconducting Generators (SG) with power\ud ratings of up to 20 MW. High efficiency of power converters for high power applications is mandatory in\ud order to reduce the overall cost of the system. This paper proposes a design method to minimise the cost of\ud the system by finding the optimal number of power devices and capacitors for different high power\ud converter topologies. The investigation focuses on determining the optimal number of voltage levels for a\ud Back-To-Back (BTB) Neutral Point Clamped (NPC) converter. The design method is demonstrated by\ud estimating the cost of different BTB NPC power converter topologies for the integration of a 10 MW SG\ud to the grid.

The fault tolerant capabilities are compared in this\ud paper for the conventional double layer switched flux permanent magnet machine and its single layer counterparts, i.e. C-core, Ecore and modular. The comparison includes the inter-turn shortcircuit and irreversible demagnetization faults. A combination of Simulink and finite element models is used in the study. Based on the predictions, it is found that the modular topology produces the lowest short-circuit current and also has the best demagnetization withstand capability while the conventional one produces the highest short-circuit current and has the worst demagnetization withstand capability. The frozen permeability method is employed to separate the flux produced by armature current and magnets, and the results showed that, besides the influence of short-circuit current, the available magnet volume and magnetic circuit configuration play an important role in the demagnetization process. It is also found that removing half of the magnets, such as using C-core, E-core and modular topologies, generally improves the demagnetization withstand capability and also increases the torque per magnet volume. Measured results are also presented to validate the short-circuit current predictions\ud and magnet demagnetization.

This paper investigates the influence of mutual fluxes (inductances) on the resultant torque in three-phase conventional switched-reluctance machine (CSRM) and mutually coupled SRM (MCSRM) using the frozen permeability (FP) method. Under saturation conditions, the FP method allows accurately separating the torques due to self-flux and mutual flux, hence quantifying their contributions to torque generation. Then, appropriate current waveforms (unipolar or bipolar, square wave or sinewave) can be established to maximize the output torques. It is well known that the mutual torque of CSRM can be negligible. However, this paper has shown that when sinewave current is employed and under full or overload conditions, the torque will be significantly reduced due to non-negligible negative mutual torques. Different from CSRM, the self-torque and the mutual torque of MCSRM can be added if current waveform is properly chosen, e.g., sinewave currents. This can significantly boost the resultant torque. The predictions have been validated by experiments.

This paper presents simple analytical modelling for 2 types of 3-phase surface-mounted permanent magnet (SPM) machines such as modular and unequal tooth width (UNET) machines with different slot/pole number combinations. It is based on the slotless open-circuit air-gap flux density and the slotted air-gap relative permeance calculations. This model allows calculating the open-circuit air-gap flux density, phase flux linkage and back electromotive force (EMF), average torque of both the modular and UNET machines. Its accuracy has been validated by 2D finite element (FE) method. A flux focusing/defocusing factor has also been introduced to analyse the influence of flux gaps in alternate stator teeth of modular machines on the machine electromagnetic performances. It has been found that by properly choosing the slot/pole number combination and flux gap width, the open-circuit air-gap flux density, the winding factor and the flux focusing effect can all be improved. As a result, the performance of modular machine can be significantly boosted. The analytical and FE models have been validated by experiments.