This article employs surrogate models for large-scale high-speed (HS) electrical machine optimization to reduce heavy computational burden caused by the finite-element method (FEM) based on nondominated sorting genetic algorithm II. Three artificial neural networks, namely, multilayer perceptron (MLP), support vector regression (SVR), and generalized regression neural network (GRNN), and the classical Kriging model are developed based on the ${K}$ -fold crossing validation method. To find out the most suitable surrogate model, a HS permanent magnet synchronous machine (PMSM) applied in electrically assisted turbocharger (EATC) considering its multiphysics characteristics is optimized by different models. A detailed comparison is provided in terms of prediction accuracy and modeling time consumption. The optimal design is verified to guarantee the accuracy of optimization results. The key contributions of this article include an automatic HSPMSM optimization process is proposed, where complete modeling and tuning method as well as a reasonable comprehensive evaluation metric of different surrogate models are conducted. It is found that invalid samples from the initial dataset contain useful information to improve prediction accuracy. Hence, an entire surrogate process should benefit from a pretrained classifier. Also, it reveals that MLP, SVR, and Kriging model can greatly improve the design optimization effectiveness with high accuracy, whereas GRNN is not suitable for this specific optimization scenario.

Brushless modular spoke type permanent magnet (MSTPM) machine has exhibited prior electromagnetic performance over conventional surface-mounted PM machines for in-wheel traction system in electric vehicles (EVs). However, analysis of thermal behavior of MSTPM machines is rather insufficient, even accurate thermal analysis is of significant importance due to poor heat dissipation condition inside the wheel hub. Conventionally, the losses produced in electrical machines by EM prediction is equivalent to the heat source in normal thermal analysis, and only the resultant thermal behavior is investigated. However, the reaction of temperature rising on the EM performance is neglected. In this paper, a bi-directional coupled electromagnetic-thermal analysis method is proposed and carried out by a combination of lumped parameter thermal network (LPTN) model and finite element method (FEM), where both steady-state temperature distribution and transient-state temperature rise are investigated for MSTPM machines. By finite number of iterations between magnetic and thermal fields, the electromagnetic performance and thermal behavior can be predicted more accurately due to the coupling effect considered. The coupled model-predicted results are verified by 3D-FEM and experimental measurement, which shows that the proposed method has advantages in both computational efficiency and accuracy as well as can be applied to other electrical machines.

This paper addresses the effect of rotor structure on the magnetic adjusting performance of an axial-radial permanent magnetic machine via the comparative performance investigation on the axial-radial flux type permanent magnet machines with three different structures. By using the three-dimensional time stepping finite-element method, the distributions of the air-gap flux density are obtained when the machine operates with different axial magnetomotive force, and the key factor that affects the magnetic adjusting ability is determined. The variation of the magnetic field is investigated, and the back electromotive forces of the three machines with different structures are also studied. Meanwhile, the eddy current distribution and the eddy current loss in machines with different structures during load operation are analyzed. The theoretical claims are verified by both numerical simulations and experimental tests.

An internal permanent magnet synchronous machine (IPMSM) was designed for heavy-load traction vehicles applied in port transportation. Based on finite element analysis (FEA), the rotor iron core topology was optimized with the most attention paid to cogging torque and torque ripple. The influences of the iron core on the air-gap magnetic flux density, the back electro-motive-force harmonic, the cogging torque and the torque ripple were investigated. The design scheme of minimizing cogging torque and output torque ripple was obtained. Focused on the relationship between the rotor parameters and the torque ripple, the relative sensitivity factor was proposed and analyzed. Finally, the torque ripple was reduced from 14.4% to 3.84%, after further optimization of the rotor design parameters. The reliability and stability of the IPMSM were also covered. Additionally, the experimental study of the prototype was carried out to verify the FEA results.

In the high-speed flux-switching permanent magnet (FSPM) machine, the AC copper loss issue in the armature winding conductor is very prominent, due to the high-frequency leakage magnetic field in stator slot, which seriously affects the machine performance. To model and calculate AC copper loss more accurately, a coupled electric circuit-magnetic field model is constructed based on 2D-finite element method (FEM), and consequently the influence of modeling methods and winding types on AC copper loss calculation of a 12-slot/10-pole (12/10) 54.7kW/10krpm FSPM machine is investigated. Firstly, 10 simplified AC copper loss calculation models are established to deal with the problem of large consumption of calculation time and computer resources caused by the excessive number of wires, small size of winding conductors, high precision modeling, and mesh refinement. Secondly, the influences of voltage-/current-source and steady/rotating rotor are considered, and the 10 models are evaluated in terms of computing time and accuracy. Thirdly, based on the preferred model from the 10 models, the effects of frequency, round/flat wires and wire sizes on AC copper loss are further studied. Fourthly, to investigate the influence of flux leakage at the end of winding on AC copper loss, a 3D AC copper loss calculation model with different axial lengths of winding end space is established. Fifthly, the change process of AC copper loss and transient temperature of winding under short time over-load condition are obtained combined with the bidirectional coupling model between electromagnetic and thermal field method through multiple iterations of electromagnetic loss and temperature. Finally, the stator modules of round and flat copper wires are manufactured respectively, and an experimental modular device is built to measure the AC copper loss, which confirms the improved accuracy of the predictions by the proposed model.

Thermal management of electrical machines is an important issue to improve electromagnetic performance, reduce system size and prolong lifetime. Based on the equivalent thermal circuit (ETC) and computational fluid dynamics (CFD) methods, the temperature rise characteristic of a high-speed stator-permanent magnet-type flux switching (HS-SPMFS) machine for fuel-cell electric vehicle compressor is predicted. Firstly, the structural features, key dimensions, material properties and heat generation rate of the HS-SPMFS machine and its cooling system are introduced. Secondly, the ETC model of the machine with a liquid cooling channel is proposed, and the size of the cooling system is preliminarily designed. The influence mechanism of the number, size and flow rate of the water channel on the convective heat transfer area, convective heat transfer coefficient, water head loss and winding temperature rise is revealed. On this basis, combined with the CFD method, the temperature distribution of the machine is studied, and the detailed temperature distributions of each component of the machine are obtained and verified by the ETC method considering the cooling system. Finally, a HS-SPMFS machine prototype is manufactured and the temperature test platform of the prototype is built. The temperature rise behaviors of the machine under different working conditions are tested, verifying the theoretical analysis and the prediction methods.