Conference Agenda

In this paper, a Finite Element (FE) analysis for structure optimization of an aluminum screen in a Wireless Power Transfer System (WPTS) for electric vehicles has been performed. The aluminum sheet, located above the receiver coil, plays a significant role in the magnetic field distribution. The geometrical characteristics of the screen affect the power transferred to the load, the screen effectiveness and the weight of the screen itself. In this study, the aluminum plate has 16 equal rectangular holes, whose base and height are geometrical parameters that can vary. An optimization problem is solved to achieve multiple objectives: minimizing the magnetic field 10 cm above the plate, maximizing the power transferred to the receiver, and minimizing the aluminum volume.

The Virtual Element Method (VEM) is a recent and successful method for the numerical solution of partial differential equations. The methodological background of the VEM is presented and its application in magnetic field analysis is shown; accordingly, the

numerical solution of non-linear magnetostatic problems is worked out in 2D domains. In this paper, we are interested in the analysis and synthesis of an internal permanent magnet motor. In this framework, the VEM is particularly suitable thanks to its flexibility in handling generally shaped polygons, specifically the possibility of treating in a smart way both hanging nodes and mesh gluing.

Reconfigurable devices are of paramount importance in many applications and a method inspired to Pareto-like optimality can generate a sequence of configurations, each of which is the best trade-off between design criteria. Accordingly, an original concept of Pareto meta-front is presented: given a family of fronts depending on a variable parameter, the meta-front is defined as the curve which joins all the closest-to-utopia points. It describes the variation of the best combinations of two conflicting design criteria against a cost parameter like e.g. a material property. The case study considered in order to prove the validity of the proposed method is the optimal shape design of an interior permanent-magnet motor (IPM) which is assumed to be magnetically reconfigurable, meaning that the variable parameter indexes a set of several permanent-magnet materials, each of which is featured by its own curve of demagnetization.

Model-based systems engineering (MBSE) is a design methodology for complex systems via models and their interactions, shifting the focus from documents and reports to model implementations and their queries.

We present a software framework applying MBSE methodology in the context of designing and optimizing accelerator magnets. We cover model implementation, interaction, dependency resolution, storage, and exchange. In this context, we encapsulate the necessary software to run each model in containers. This provides flexibility in combining different simulation codes while keeping track of dependencies on specific snapshots of the software used.

By combining these components, we can fully utilize the MBSE design methodology, react to design changes, and guarantee the future execution of models in systems-generation engineering.

High voltage direct current gas insulated transmission lines (HVDC GIL) are becoming more relevant due to the transmission to green energy, since GIL enables the transmission of electrical energy over long distances in a compact structure and with a high voltage capacity. However, due to the DC condition, slow charge accumulation in the GIL insulators occur, which may lead to high electric field stress. GIL consists of a high voltage conductor, an insulating gas, an enclosure and a solid spacer to sustain the system. Especially the interface between the insulating gas and the solid spacer is sensitive to space charge accumulation, hence high electric fields, where surface flashover may occur. Therefore the design of the spacer with regard to high electric fields and space charge accumulation is crucial for the safety of the system. Numerical simulations are a convenient tool to investigate the design of the system and its influence on the electric field distribution. In this work deep neural networks are utilized to optimize the geometrical design of the spacer. Numerical simulations results are used to train the deep neural network. The optimization of the geometry of the spacer results in a significant reduction of the electric field distribution and the maximum electric field value.

This paper presents a characterization and design optimization environment suitable for electric vehicle Dynamic Wireless Power Transfer charging systems. It utilizes Particle Swarm Optimization search algorithm and Artificial Neural Networks. The system identifier employs electromagnetic 3D Finite Element and State Space models, and it is used in a case study of a Dynamic Wireless Power Transfer charging system with multi-ground transmitter coils and a receiver coil.

Synchronous reluctance machines are receiving attention as they combine high efficiency with low material cost. The design of a high-performance synchronous reluctance rotors is complex, and benefits from using computationally intensive topology optimization. Significant discrepancy can occur between measured prototype performance and such FEA based simulation, attributed in part to the manufacturing processes. These factors can be large when stator and rotor core packs are fabricated by laser cutting. This paper presents a level set optimization approach with degradation modelling to adapt designs for manufacture during the optimization process. The results show topological changes and a reduction in torque deficit when manufacturing degradation is included. Further work will explore the effect of design parameters using a high-performance computing platform to accelerate the optimization process.

Permanent magnet synchronous motors (PMSMs) have attracted increased research interest recently within electrification initiative, as they meet the challenges of high-speed operation in electric vehicle applications. The paper proposes an optimization procedure for interior permanent magnet (IPM) motor with the magnets arranged in V-type configuration in the rotor implementing advanced iron laminations; the methodology is based on genetic algorithms (GAs) in conjunction with finite element analysis (FEA) enabling to meet the desired performance while minimizing the involved losses. The proposed method has been applied in a 31 kW motor example and the simulated results have been validated by measurements in a prototype.

We present a topology optimization framework for electric motors that maximizes the torque under the constraint of the torque ripple and the rotor weight. By projecting the rotor design onto a static background mesh at various orientations, the proposed method eliminates the need for remeshing, simplifying the handling of the relative motion between the stator and rotor. Furthermore, we explore interpolation factors that improve design exploration during optimization. A synchronous reluctance motor (SynRM) is optimized to achieve substantial torque improvement and ripple reduction, demonstrating strong potential for advancing high-efficiency SynRM design.

This research proposes a parametric-design methodology for rotors of Internal Permanent Magnet Synchronous Motors (IPMSMs) to optimize performance while integrating geometric constraints. The method is implemented to systematically refine flux barrier shape and magnet area and location through a Genetic Algorithm (GA). The study demonstrates the ability to efficiently explore high-dimensional design spaces, resulting in improved rotor Total Harmonic Distortion (THD) metrics. This approach provides a robust and scalable framework for the development of electric motors.

Preliminary Breakdown Pulses (PBP) are a series of electromagnetic signals observed tens of milliseconds prior to the first Return

Stroke (RS) in a lightning flash. Due to the complexity of the phenomenon and since the risk associated with PBPs is typically lower

than that produced by RSs, few studies in the literature exist about PBP modeling. However, observation of PBPs and accurate models

may be useful to develop RS prediction tools, allowing to include preventive actions in the actual lightning protection strategies.

This contribution presents a preliminary study for the optimization of the parameters of a PBP discharge current numerical model

for Cloud-to-Ground (CG) flashes simulations. This is to be intended as the first step to obtain statistical distributions of the model

parameters from a large dataset of experimental measurements.

Magnets are essential in High Energy Physics (HEP) experiments, serving critical functions in particle accelerators and detectors. In detectors, like those in many experiments at CERN’s Large Hadron Collider (LHC), large magnets are essential for particle identification by analyzing the curvature of particle tracks within a magnetic field. A consolidated design process follows a sequential process (i.e., identification of specification, analytical modeling followed by numerical validation), to balance performance and cost, meeting the physical constraints. Nowadays, any up-to-date design approach includes, within the fundamental goals, minimizing the use of energy/power and more in general material resources for sustainability, which is reflected in new guidelines at CERN. The present work aims at presenting some reformulation of resistive large magnet design procedures supporting this novel paradigm. This is obtained by developing an integrated optimization framework, that prioritizes energy efficiency, alongside performance and technical constraints, leading to a new perspective in the design exploration of the parameters space. A case study is reported for the design of spectrometer magnets for the SND@LHC upgrade project at CERN, showing the possibilities offered by such a new approach in balancing sustainability with experimental requirements and demand.

In this paper, a novel hybrid multi-objective optimization (MOO) algorithm is proposed by combining a beluga whale optimization (BWO) with an improved NSGA-II. The original BWO is transformed to a MOO version and NSGA-II is enhanced by using opposite-learning initialization and revised crowding-distance strategy. Besides, material field series expansion method is adopted to form a multi-objective topology optimization methodology. To validate the proposed methodology, topology of an electromagnetic actuator prototype is optimized.