This article proposes an improved thermal modeling approach for fractional-slot concentrated winding (FSCW) permanent magnet (PM) machines, based on the thermal network method (TNM). In the traditional TNM models, the core loss is considered uniformly distributes in the stator core, which actually leads to an unavoidable error in FSCW PM machine thermal prediction. In order to achieve a high accuracy, the discrete loss distribution model (DLDM) is built for machine stator, where the stator core is divided into amounts of blocks. Based on the Bertotti model, the accurate core loss distributions are calculated under different load rates. Then, the calculated core losses are allocated to corresponding blocks, and the temperature rise is predicted by TNM. The predicted results by the proposed model reveals that there exists the local hot spot in the stator tip caused by the local loss concentration in FSCW PM machines, which is not ever reported. In order to validate the predicted results, the experimental platform with a 12-slot 10-pole FSCW PM machine is built and tested. The experimental results reveal that the proposed model features a higher temperature prediction accuracy, while the local hot spot really exists in stator tip. It is believed that the proposed model can give a more reliable support in FSCW PM machine design considering magnetic-thermal coupling.
Thermal Modeling and Loss Analysis of Fractional-Slot Concentrated Winding Permanent Magnet Motors for Improved Performance
IEEE Transactions on Transportation Electrification ; 10 , 3 ; 7151-7159
2024-09-01
3598576 byte
Article (Journal)
Electronic Resource
English