This paper reviews the recent developments of design optimization methods for electromagnetic devices, with a focus on machine learning methods. First, the recent advances in multi-objective, multidisciplinary, multilevel, topology, fuzzy, and robust design optimization of electromagnetic devices are overviewed. Second, a review is presented to the performance prediction and design optimization of electromagnetic devices based on the machine learning algorithms, including artificial neural network, support vector machine, extreme learning machine, random forest, and deep learning. Last, to meet modern requirements of high manufacturing/production quality and lifetime reliability, several promising topics, including the application of cloud services and digital twin, are discussed as future directions for design optimization of electromagnetic devices.

Machine Learning for Design Optimization of Electromagnetic Devices: Recent Developments and Future Directions

The magnetic coupling phenomenon widely exists in the type of double rotor motor, which is difficult to be eliminated completely due to the interactions between two layers of air-gap magnetic fields in a compact geometry structure. This article studies magnetic coupling in the perspective of air-gap harmonic groups, where the novelty is to propose the concept of positive coupling harmonics and the negative one. And, a double rotor flux-switching PM machine is chosen as the research subject. Then, the article begins with the analytical investigation of the modulation principle considering magnetic coupling, where the coupling harmonic group factor is defined. According to this factor, the coupling harmonics are analyzed and selected, and the positive and negative coupling harmonic groups are further determined. Next, the influence of coupling harmonic group on the machine performance is investigated in detail, including the back-EMF, torque performances, overload capability, and so on. The results show that the raise of positive coupling harmonic group factor brings the improvement of machine performances. In addition, a prototype machine is manufactured for experimental testing. Overall, theoretical analysis, simulation and test results verify the validation of the whole investigation to a large extent.

Research on Magnetic Coupling Characteristic of a Double Rotor Flux-Switching PM Machine From the Perspective of Air-Gap Harmonic Groups

This article proposes a new partitioned stator hybrid excited (PS-HE) machine with dc-biased sinusoidal current. On the inner stator, the consequent pole permanent magnets (PMs) are employed as the PM excitation. On the outer stator, a set of integrated dual three-phase winding is adopted to produce the wound and armature fields at the same time by injecting the sinusoidal current with dc bias. So, the flexible flux regulation can be realized by controlling the dc bias component. The rotor composed of segmented iron cores is sandwiched between two stators. By employing the air-gap field modulation theory, the air-gap magnetic field harmonics are analyzed to explain the operating and flux regulation principle. Furthermore, a design optimization and the electromagnetic performances of the proposed machine are investigated by finite element analysis. When compared with the existing series and parallel PS-HE machines, the proposed machine can avoid the disadvantage of low torque density of PARALLEL HE machine, and it can solve the problem of the poor flux regulation ability of series HE machine as well, which is mainly contributed by the reasonable space allocation of PMs and field winding excitations. Finally, a prototype is built and tested to verify the theoretical analyses. 

Partitioned Stator Hybrid Excited Machine With DC-Biased Sinusoidal Current

In this article, a parallel-series magnetic circuit multi-permanent magnet (M-PM) motor is designed and optimized considering magnet characteristic variation effects, where two kinds of permanent magnets (PMs) are utilized as co-magnetic. Based on the equivalent magnetic circuit method, the detailed design method for the M-PM motor is illustrated. Considering the various operation conditions in potential electric vehicle applications, the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">B–H</i> curves of two PMs of different typical operation conditions are obtained. Then, a multiobjective optimization method is newly proposed which not only focuses on the magnet characteristics variation effect but also considers output torque, torque ripple as well as antidemagnetization capability under three typical operation conditions. The performances of the initial motor and optimal motor are compared under typical operation conditions. Finally, a prototype machine is built and tested. Both theoretical analysis and experimental results verify the validity of the motor and the proposed design optimization method. 

Multi-Objective Optimization Design of a Multi-Permanent-Magnet Motor Considering Magnet Characteristic Variation Effects

A comparative study of new partitioned stator hybrid excitation axial flux switching permanent magnet (HEAFSPM) motors with different partitioned excitation stator topologies for in-wheel traction applications is presented in this paper. The realization approaches of the partitioned stator HEAFSPM motor are discussed, and the motor topology with windings and permanent magnets (PMs) separately setting on different stators is selected in view of the advantageous structure for thermal management. On this basis, two partitioned excitation stator topologies including C-type tooth core and T-type tooth core stator configuration are proposed, and the air-gap flux modulation and flux regulation principle of the motors are analyzed. In addition, the design method of the partitioned stator HEAFSPM motor considering the effect of hybrid excitation sources is provided, and the two motors with different tooth cores are designed respectively. Meanwhile, for a fair comparison, the multiple objectives optimization method coupled with the sensitivity analysis is employed to optimize the two motors. The electromagnetic performances, anti-demagnetization capability, and flux regulation performance of the two motors are all investigated in detail. Finally, two prototype motors are fabricated and tested. Both the theoretical analysis and experimental results verify the effectiveness and reasonability of the motor.

Comparative Analysis and Design of Partitioned Stator Hybrid Excitation Axial Flux Switching PM Motors for In-Wheel Traction Applications

In this article, a variable-saliency-ratio permanent magnet motor is designed and optimized considering driving cycles. To satisfy the demands of multi-operation conditions of EVs, a variable resistance is adopted and a new variable-saliency-ratio design concept is proposed to enhance output torque, widen speed range and lower irreversible demagnetization risk effectively. A multiobjective optimization considering driving cycles is proposed, where the design objective of saliency ratio is firstly selected and optimized with multiobjective genetic algorithm method. Then, the operation performances of the motor in different operation conditions are investigated in detail. Finally, a prototype motor is built and tested. Both the simulation and experimental results verify the validity of the proposed method.

Multi-objective Optimization Design of Variable-Saliency-Ratio PM Motor Considering Driving Cycles

In this article, an optimization design method of power factor is proposed for vernier permanent magnet (VPM) machine, in which the key is to analyze and investigate the power factor from the perspective of air-gap flux harmonic modulation. To improve the design efficiency, air-gap harmonics are analyzed and serve as the crucial bridge between power factor and machine geometry. It is noted that, during the operation of the VPM machine, more attention is required for the power factor of low-speed operation area. Based on the popular new European driving cycle of the electric vehicles, three low-speed driving conditions are chosen for the optimization design. For the systematic investigation, a V-shape VPM machine is selected as an optimization example. In order to evaluate conveniently, the comprehensive power factor is defined and sensitive air-gap flux harmonics of different conditions are selected as the optimization objectives, for improving the power factor characteristics in low-speed conditions. In addition, power factor and machine performances are evaluated to validate the proposed method. Finally, the prototype machine is built and tested. Both the simulation and experimental results reveal the feasibility and effectiveness of the proposed method and investigated VPM machine.

Optimization Design of Power Factor for an In-Wheel Vernier PM Machine From the Perspective of Air-Gap Harmonic Modulation

In this article, a new controllable leakage flux multilayer barrier permanent magnet (PM) motor is proposed by adopting a new design concept of controllable leakage flux, where the PMs are uniquely placed in the d -axis magnetic circuit of the synchronous reluctance motor (SynRM). Compared to the conventional q -axis PM placement SynRM, the higher PM utilization and better flux-weakening capability can be obtained. For further comparison and verification, both of the controllable leakage flux multilayer barrier PM motor and the motor with traditional q -axis PM setting are designed. In addition, a multiobjective optimization method based on the correlation analysis method is proposed to obtain the optimized output torque and flux-weakening performances. Then, the performances of the two optimal motors are analyzed and compared in detail. Finally, two prototype motors are manufactured and tested. Both the theoretical and experimental analyses verify the effectiveness of the flux-weakening enhancement design and the proposed motor.

Flux-Weakening Capability Enhancement Design and Optimization of a Controllable Leakage Flux Multilayer Barrier PM Motor

https://doi.org/10.1016/j.ultsonch.2023.106289

Yongfeng Liu

Papers

Multi-objective Optimization Design of Variable-Saliency-Ratio PM Motor Considering Driving Cycles

Optimization Design of Power Factor for an In-Wheel Vernier PM Machine From the Perspective of Air-Gap Harmonic Modulation

Flux-Weakening Capability Enhancement Design and Optimization of a Controllable Leakage Flux Multilayer Barrier PM Motor

High-efficiency degradation catalytic performance of a novel Angelica sinensis polysaccharide-silver nanomaterial for dyes by ultrasonic cavitation

Synthesis of Functionalized Pamam Dendrimer/Apple Residue Bioadsorbents for Removal of Aqueous Hg(Ii)