Comprehensive modeling and optimization of high-performance vernier machines

• Main Author: Zhu, Jingwei
• Other Authors: Christopher H. T. Lee
• Format: Thesis-Doctor of Philosophy
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Electrical and electronic engineering::Electric power
• Online Access: https://hdl.handle.net/10356/170974
• Institution: Nanyang Technological University

Electric motors occupy approximately 45% of the electricity consumption and over 99% of the total electricity originates from electric generators. As a result, electric machines act as an essential role in the energy conversion, such as pumping, drilling, cutting, etc. and enable the potential for sustainable energy technologies, such as electric vehicles (EVs) in transportation and wind turbines in clean energy generation. However, the torque performance of conventional electric machines is restricted by the amplitude of air gap flux density due to the magnetic saturation in the electric steel. Therefore, a high-speed electric machine with a reduction gear box provides a solution for high-torque applications. However, this introduces additional transmission loss & maintenance issue, increases the total volume, and reduces the reliability of the motor system. Permanent magnet vernier machines (PMVMs) break through the limitation of air gap flux density by utilizing multiple air gap flux density components and achieve significant improvement in torque performance, indicating a potential for direct drive. However, the low power factor challenge makes it unacceptable for commercial direct-drive applications. As for further improvement of the performance of this type of motor, the difficulties lie in the following four aspects: a lack of the accurate analytical model to identify the impact of geometric parameters on high torque density and power factor designs, an absence of motor optimization framework to take the performance metrics over the entire speed range into consideration, fixed control method (id = 0 control) at the rated point not appropriate for all motor geometries and how to develop a new topology with enhanced performance. Hence, the main objective of the thesis is to present a deep investigation into the PMVM through accurate modeling, motor optimization, optimal control method at the rated operating point and new motor topology, to achieve significant performance improvement. Firstly, a deep-investigated new analytical model for PMVMs with maximum torque control is proposed. The details include an exact air gap permeability model based on the conformal mapping approach, a new leakage flux evaluation method, the introduction of the concept of slot opening factor and its impact on high order harmonics, and the motor performance metrics calculation approach, including the average torque and power factor. The proposed analytical model can provide motor designers with tools to select appropriate motor geometric parameters for high torque density and power factor designs. Secondly, a new comprehensive optimization framework developed for PMVMs to achieve the performance improvement over the entire speed range is proposed. A new armature reaction modeling for the open-slot structure is developed for the armature flux linkage and winding inductance evaluations. The harmonic analysis, air gap flux density component computation, slot leakage evaluation and flux linkage, inductance calculation approach are implemented. Then, the analytical model is combined with finite element analysis (FEA) to obtain the constant power speed range (CPSR) in the flux weakening region accurately. This computation approach is applied to a semi-analytical optimization framework to ensure high torque & power density, power factor, efficiency and wide CPSR in the whole operating speed range. Thirdly, a new magnetic saturation model in the stator core to identify the optimal current control angle for the rated operating point of PMVMs is proposed. The model is able to evaluate the flux density values in different parts of the stator core accurately, thus providing the accurate magnetomotive force (MMF) drop in the stator core to identify the optimal current control angle. This modeling technique shows its potential to be used at the initial stage of motor optimization since id = 0 control might not be the optimal one for all candidates with various geometric parameters. Suitable selection of current control angle is able to achieve higher torque density, higher power factor and wider CPSR than id = 0 control at the rated point. Finally, with regard to the issue of large back EMF and considerable amount of PM usage for conventional PMVMs, a new hybrid excitation vernier motor topology is proposed to enhance the PM utilization and reduce the no-load back EMF while maintaining similar torque density. The new design possesses high quality flux interaction between PMs, field winding and armature winding. When there is no field or armature current, large amount of PM flux flow inside the rotor core and not into the air gap, thus significantly decreasing the no-load back EMF. Under load condition when id = 0 control is used, the flux interaction between PMs, field winding and armature winding results in more PM flux flowing into the air gap, which significantly increases the torque density. Thereby, the air gap flux density is improved significantly under load condition and the torque performance is comparable to conventional PMVM, but with significantly reduced PM amount.