This thesis addresses the challenge of reducing the use of Critical and Strategic Raw Ma- terials (CRMs), which are essential in key technologies such as e-mobility and renewable energy systems but whose supply chains are increasingly vulnerable in Europe. The first chapter introduces the broader context of the European Green Transition and circular economy, highlighting the urgent need to minimize material dependency through inno- vative design methods. In this framework, topology optimization (TO) is identified as a powerful computational tool capable of reducing material usage while maintaining high performance in electrical machines. The second chapter develops the numerical foundation by implementing a finite element analysis (FEA) framework for two-dimensional magnetostatic problems. Starting from Maxwell’s equations, the weak formulation is derived and discretized with triangular ele- ments. Both linear and nonlinear material models are considered, with validation against FEMM simulations confirming the accuracy of the in-house MATLAB solver. This step provides a reliable computational basis for the subsequent optimization tasks. The third chapter focuses on the formulation and implementation of a density-based topology optimization procedure for a simplified transformer benchmark. A Rational Approximation of Material Properties (RAMP) interpolation is adopted to link the design variable to magnetic material properties, while sensitivities are computed through the Adjoint Variable Method. Different optimization schemes are explored, including fixed- multiplierandAugmentedLagrangianformulations, thelatterofferingafullyautonomous way to satisfy the volume constraint. Additional features include strategies for gray- scale suppression and the construction of Pareto fronts through multi-start runs, which demonstrate the potential of the method to balance conflicting design objectives. Overall, the work demonstrates the feasibility and robustness of combining FEA with topology optimization for material-efficient design in electromagnetics. The results are encouraging and confirm the relevance of this approach as a contribution toward sustain- able engineering solutions in the context of Europe’s transition to renewable energy and e-mobility.
Questa tesi affronta la sfida della riduzione dell’utilizzo di Critical and Strategic Raw Materials (CRMs), materiali fondamentali per tecnologie chiave quali la mobilità elettrica e i sistemi di energia rinnovabile, ma la cui catena di approvvigionamento in Europa risulta sempre più vulnerabile. Il primo capitolo introduce il contesto della transizione verde europea e della circular economy, sottolineando la necessità di ridurre la dipendenza da tali materiali attraverso metodologie di progettazione innovative. In questo quadro, l’ottimizzazione topologica (TO) viene individuata come uno strumento computazionale avanzato, in grado di ridurre il consumo di materiale mantenendo elevate prestazioni nelle macchine elettriche. Il secondo capitolo sviluppa le basi numeriche implementando un framework di anal- isi agli elementi finiti (FEA) per problemi magnetostatici bidimensionali. A partire dalle equazioni di Maxwell viene derivata la formulazione debole, discretizzata tramite elementi triangolari. Sono stati considerati modelli sia lineari che non lineari, con una validazione condotta attraverso simulazioni FEMM che ha confermato l’accuratezza del solver MAT- LAB sviluppato. Questo passaggio fornisce una base computazionale affidabile per le successive fasi di ottimizzazione. Il terzo capitolo si concentra sulla formulazione e implementazione di una procedura di ottimizzazione topologica basata sulla densità, applicata a un trasformatore semplificato. È stato adottato lo schema di interpolazione RAMP (Rational Approximation of Mate- rial Properties) per collegare la variabile di progetto alle proprietà magnetiche, mentre le sensibilità sono state calcolate tramite l’Adjoint Variable Method. Sono stati esplorati diversi schemi di ottimizzazione, tra cui la formulazione a moltiplicatore fisso e quella a Lagrangiano Aumentato, quest’ultima capace di soddisfare in modo completamente autonomo il vincolo di volume. Ulteriori sviluppi hanno riguardato strategie di soppres- sione delle gray scales e la costruzione di fronti di Pareto tramite approcci multi-start, dimostrando la capacità del metodo di bilanciare obiettivi di progetto contrastanti. Nel complesso, il lavoro ha dimostrato la fattibilità e la robustezza dell’integrazione tra FEA e ottimizzazione topologica per la progettazione a ridotto consumo di materiale in ambito elettromagnetico. I risultati ottenuti sono incoraggianti e confermano la rilevanza di questo approccio come contributo verso soluzioni ingegneristiche sostenibili nel contesto della transizione europea verso le energie rinnovabili e la mobilità elettrica.
Raw material minimization in electrical machines manufacturing: a topology optimization approach
Albanese, Stefano
2024/2025
Abstract
This thesis addresses the challenge of reducing the use of Critical and Strategic Raw Ma- terials (CRMs), which are essential in key technologies such as e-mobility and renewable energy systems but whose supply chains are increasingly vulnerable in Europe. The first chapter introduces the broader context of the European Green Transition and circular economy, highlighting the urgent need to minimize material dependency through inno- vative design methods. In this framework, topology optimization (TO) is identified as a powerful computational tool capable of reducing material usage while maintaining high performance in electrical machines. The second chapter develops the numerical foundation by implementing a finite element analysis (FEA) framework for two-dimensional magnetostatic problems. Starting from Maxwell’s equations, the weak formulation is derived and discretized with triangular ele- ments. Both linear and nonlinear material models are considered, with validation against FEMM simulations confirming the accuracy of the in-house MATLAB solver. This step provides a reliable computational basis for the subsequent optimization tasks. The third chapter focuses on the formulation and implementation of a density-based topology optimization procedure for a simplified transformer benchmark. A Rational Approximation of Material Properties (RAMP) interpolation is adopted to link the design variable to magnetic material properties, while sensitivities are computed through the Adjoint Variable Method. Different optimization schemes are explored, including fixed- multiplierandAugmentedLagrangianformulations, thelatterofferingafullyautonomous way to satisfy the volume constraint. Additional features include strategies for gray- scale suppression and the construction of Pareto fronts through multi-start runs, which demonstrate the potential of the method to balance conflicting design objectives. Overall, the work demonstrates the feasibility and robustness of combining FEA with topology optimization for material-efficient design in electromagnetics. The results are encouraging and confirm the relevance of this approach as a contribution toward sustain- able engineering solutions in the context of Europe’s transition to renewable energy and e-mobility.| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_10_Albanese.pdf
accessibile in internet per tutti
Descrizione: Topology Optimization approach (master thesis)
Dimensione
79.1 MB
Formato
Adobe PDF
|
79.1 MB | Adobe PDF | Visualizza/Apri |
|
2025_10_Albanese_ExecutiveSummary.pdf
accessibile in internet per tutti
Descrizione: Executive Summary
Dimensione
1.48 MB
Formato
Adobe PDF
|
1.48 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/243216