Machine learning and artificial neural networks have shown to be applicable in modeling and simulation of complex physical phenomena as well as creating surrogate models trained with physics-based simulation data for numerous applications that require good computational performance. In this article, we review widely the surrogate modeling concept and its applications in the electrical machine context. We present comprehensively a workflow for developing data-driven surrogate models including data generation with physics-based simulation and design of experiments, preprocessing of training data, and training and testing of the surrogates. We compare neural networks and gradient boosting decision trees in modeling and simulation of torque behavior of a permanent magnet synchronous machine together with selected design of experiments approaches with respect to surrogate accuracy and computational efficiency. In addition, an approach to utilizing domain knowledge to create a hybrid surrogate model in order to improve the surrogate accuracy is shown. The accuracy of the selected hybrid neural network was better than with the gradient boosting approach and was close to the finite element simulation, whereas its run-time efficiency was significantly better compared to the finite element simulation with a speed-up factor of over 2,000. In addition, combining the sampling methods provided better results than the selected methods alone.
Purpose: The purpose of this paper is to report on the developments in manufacturing soft magnetic materials using laser powder bed fusion (L-PBF). Design/methodology/approach: Ternary soft magnetic Fe-49Co-2V powder was produced by gas atomization and used in an L-PBF machine to produce samples for material characterization. The L-PBF process parameters were optimized for the material, using a design of experiments approach. The printed samples were exposed to different heat treatment cycles to improve the magnetic properties. The magnetic properties were measured with quasi-static direct current and alternating current measurements at different frequencies and magnetic flux densities. The mechanical properties were characterized with tensile tests. Electrical resistivity of the material was measured. Findings: The optimized L-PBF process parameters resulted in very low porosity. The magnetic properties improved greatly after the heat treatments because of changes in microstructure. Based on the quasi-static DC measurement results, one of the heat treatment cycles led to magnetic saturation, permeability and coercivity values comparable to a commercial Fe-Co-V alloy. The other heat treatments resulted in abnormal grain growth and poor magnetic performance. The AC measurement results showed that the magnetic losses were relatively high in the samples owing to formation of eddy currents. Research limitations/implications: The influence of L-PBF process parameters on the microstructure was not investigated; hence, understanding the relationship between process parameters, heat treatments and magnetic properties would require more research. Originality/value: The relationship between microstructure, chemical composition, heat treatments, resistivity and magnetic/mechanical properties of L-PBF processed Fe-Co-V alloy has not been reported previously.
After its recent improvements described here, open source finite element software Elmer is shown to be a highly efficient option for electrical machine modeling. The parallelization of computational burden is shown to be a necessity. The methods implemented enable applying fully parallel computation to electrical machine models, including rotation and electrical circuits. Computational experiments performed demonstrate that Elmer can effectively utilize several hundreds of computational cores in parallel, making it an attractive alternative when computational speed is of high importance.
Torque density of radial, axial and transverse flux machine topologies is investigated. A 10-kW, 200-rpm motor is chosen as a test case. Radial and axial flux motors that fulfill the key specifications of the selected test case are designed employing in-house analytical dimensioning tools and MATLAB genetic multi-objective optimization. Rather simple numerical approach is taken to study the transverse flux motor. A 20-pole-pair radial flux motor is found to outperform its axial and transverse flux counterparts in terms of torque density.
In electric vehicle (EV) traction, energy conversion by electric motors must be material and energy efficient to reduce environmental burden caused by the manufacture and operation of the systems and to help in efficient transition towards net zero future. The scarcity of key materials like rare earths and to some extent copper must be addressed in the design. Novel traction motors need a different approach compared to traditional industrial motor designs. Here, we focus on innovative design and optimization strategies to enhance material and energy efficiency and meet the escalating demand for sustainable transportation solutions. We are developing a prototype motor that achieves a continuous specific power of 7 kW/kg, significantly exceeding current automotive standards. This will be achieved by elevated operational speed of the traction motor, integrating advanced materials, and innovative cooling techniques such as direct liquid cooling (DLC) using hollow hairpin conductors, and insulating them with polyether-ether-ketone (PEEK) extrusion and expandable mainwall insulation material. The approach reduces reliance on rare earth permanent magnet (PM) materials by 60% compared to existing motors, aligning with global sustainability objectives. Through simulation, modeling, and practical case studies, the research demonstrates the feasibility of these innovations in real-world applications, highlighting potential advancements in EV propulsion systems. The article not only represents the insights into the latest developments in the design of electric motors for EV but also delineates the journey towards the creation of such a motor with considering accompanying electromagnetic and mechanical challenges.