Permanent magnet synchronous machines (PMSM) with neodymium iron boron (NdFeB) permanent magnets (PM) are typically used in applications, such as electric transportation, that require high power/torque density and efficiency. NdFeB is popular because of its exceptionally high remanent flux density ($B_{r}$) and high coercivity ($H_{c}$) at room temperature, but supply chain limitations of rare-earth elements (REE), such as neodymium (N d) and dysprosium (Dy), have recently caused significant REE price fluctuations. As a result, researchers are searching for ways to reduce or eliminate using NdFeB in PMSM designs. In the past, ferrite PMs were the next best option, but recently, manganese bismuth (MnBi) has emerged as another possible alternative to ferrites, with magnetic properties superior to ferrites but inferior to NdFeB at room temperature. MnBi is relatively new compared to commercial PMs and shows a unique trend of significantly increasing coercivity with increasing temperature. This makes rotor PMs significantly more susceptible to irreversible demagnetization at low temperatures. Reduced order models (ROMs), or meta-models, can estimate nonlinear machine phenomenon well enough for multi-objective optimization, but no literature has used this tool to characterize MnBi low temperature demagnetization. Furthermore, few sources explain how to work through the challenges of multi-objective optimization when using ROMs. Therefore, this paper presents a case study for limiting low temperature irreversible demagnetization in MnBi IPMSMs using ROMs and multi-objective optimization. The choices made and the tools available to produce a sufficiently accurate demagnetization ROM can be applied to a wide variety of nonlinear multiphysics phenomena.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Irreversible Demagnetization Reduced Order Modeling in MnBi Interior Permanent Magnet Synchronous Motor Multi-Objective Optimization


    Beteiligte:
    Brody, Ryan (Autor:in) / Ohodnicki, Paul (Autor:in) / Grainger, Brandon (Autor:in)


    Erscheinungsdatum :

    2023-06-21


    Format / Umfang :

    2824307 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch