Coupled Magnetic Field and Thermal Analysis Model for an IPMSM With Modular Three-Phase Winding Topologies for Fault-Tolerant Applications

Multi three-phase machines are frequently recommended for high current and high power applications due to their ability to minimize phase current while maintaining rated power and phase voltage. Furthermore, multi three-phase machines are also proposed because of their fault-tolerant capability to operate in partial working conditions due to their redundant structure. A partial working condition may occur when one or more three-phase winding sets go into faulty condition. This article focuses on evaluating the thermal behavior and efficiency of a multi three-phase machine comprising different modular winding topologies to investigate its normal, partial, and partially overloaded operation. The machine prototype used in the analysis is an interior permanent magnet synchronous machine (IPMSM) having six three-phase winding sets configured in modular three-phase winding arrangements. Firstly, the prototype machine’s losses, efficiency, and output torque performances are evaluated through electromagnetic finite element analysis (FEA) for the different modular three-phase winding topologies applying similar operating conditions. Then the machine’s thermal behavior is evaluated for the modular (concentrated, distributed, and combined concentrated and distributed) three-phase winding topologies through a coupled magnetic field and thermal analyses under the machine’s healthy, partial, and partially overload operations. Finally, a comparative thermal analysis is presented for the machine prototype along with experimental validations for the analyzed modular three-phase windings topologies. This research study suggests the best modular three-phase winding topology for a multi three-phase machine considering efficiency performance and mutual heat exchange phenomenon between different three-phase independent winding sets, particularly for the machine’s partial and partially over-load working conditions.

Abstract licence: CC BY-NC-ND

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.