Conference Agenda

Electric conductors are key components of electromagnetic actuators. While being responsible for the controllable magnetic field, and subject to numerous losses, their design is often overlooked relying on simple geometries. Efforts have been made by the community in the design of ferromagnetic parts using numerical methods such as topology optimization, but the application of such an approach to the design of electric conductors seems to be left out. This work establishes the foundation of a topology optimization formulation designing conductor tracks in an environment made of air and ferromagnetic regions.

In the present work, we apply an extension of the space-time finite element method to the problem of a rotating synchronous reluctance machine in two space dimensions. Further on, we consider the shape optimization of the machine’s rotor core so as to maximize the average torque of the machine, with the eddy current equation as PDE constraint and thus taking into account the time-dependent effects. Finally, we add to our model the coupling of the electromagnetic and thermal effects, considering the dependency of the parameters on the temperature.

This paper proposes a novel optimization method to design transmitting and receiving coils of a wireless power transfer (WPT) device considering magnetic and circuit properties. In the proposed optimization, the coil shapes are determined by parameter optimization for manufacturability, while the magnetic core shapes introduced in the vicinity of them are determined by topology optimization. The power transfer efficiencies of the optimized WPT device are evaluated with the SPICE simulation, after the circuit parameters of the optimized coils are computed with the homogenization-based finite element method, which can effectively evaluate the eddy current loss in coil windings. It is shown that the optimized WPT device has over 90 % power transfer efficiencies regardless of misalignment patterns.

Electromagnetic fields and eddy currents in thin electrical steel laminations are governed by the laws of magnetodynamics with hysteresis. Conventional homogenization techniques are however complex and very time-consuming. In consequence, hysteresis and eddy currents in ferromagnetic laminated cores are usually outright disregarded in finite element simulations, considering only saturation, and magnetic losses are only evaluated a posteriori, by means of a Steinmetz-Bertotti like empirical formula. This model simplification yields however potentially inaccurate results in the presence of non-sinusoidal B-fields, common in modern electrical devices. Assuming a time-periodic excitation of the system, a more accurate and fast approach, based on homogenization and neural networks (NN), is presented. A parametric homogenized material law is used in the macroscopic model, whose parameters are given element-wise by a NN according to the actual local waveform of the magnetic field. It is shown that, with an appropriately trained NN, this NN-based material law allows computing fields and losses inside ferromagnetic laminated stacks efficiently and accurately.

In this paper, the topology optimization (TO) technique is proposed to design a 12/8 switch reluctance machine (SRM) to reduce the resonance-induced stator vibration. To this end, the electromagnetic-structure coupled finite element modeling is first established, and the rotation speeds that may evoke stator resonance are elaborated. Then, the normalized Gaussian network method (NGnet) method is employed to form the rotor topology for reducing the resonance-induced vibration. As the optimization results, adopting the anchor-shaped rotor pole and opening a window inner the rotor pole are discovered to effectively reduce the stator vibration, the vibration of the SRM has been reduced by 42% with only a 5% average torque loss.