Date of Award

1-23-2020

Publication Type

Master Thesis

Degree Name

M.A.Sc.

Department

Electrical and Computer Engineering

Keywords

Computational Fluid Dynamics, Electric machine, Electric Vehicle, Permanent Magnet Synchronous Motor, Thermal Analysis, Thermal Modeling

Supervisor

Narayan Kar

Rights

info:eu-repo/semantics/embargoedAccess

Abstract

Permanent magnet synchronous motor (PMSM) is a better choice as a traction motor since it has high power density and high torque capability within compact structure. However, accommodating such high power within compact space is a great challenge, as it is responsible for significant rise of heat in PMSM. As a result, there is considerable increase in operating temperature which in turn negatively affects the electromagnetic performance of the motor. Further, if the temperature rise exceeds the permissible limit, it can cause demagnetization of magnets, damage of insulation, bearing faults, etc. which in turn affect the overall lifecycle of the motor. Therefore, thermal issues need to be dealt with carefully during the design phase of PMSM. Hence, the main focus of this thesis is to develop efficient ways for thermal modeling to address thermal issues properly. Firstly, a universal lumped parameter thermal network (LPTN) is proposed which can be used for all types of PMSMs regardless of any winding configuration and any position of magnets in the rotor. Further, a computationally efficient finite element analysis (FEA) thermal model is proposed with a novel hybrid technique utilizing LPTN strategy for addressing the air gap convection in an efficient way. Both proposed LPTN and FEA thermal models are simplified ways to predict motor temperature with a comparatively less calculation process. Finally, the proposed thermal models have been experimentally validated for the newly designed interior and surface mounted PMSM prototypes. Again, a procedure for effective cooling design process of PMSM has been suggested by developing an algorithm for cooling design optimization of the motor. Further, a computational fluid dynamics (CFD) model with a proposed two-way electro-thermal co-analysis strategy has been developed to predict both thermal and electromagnetic performance of PMSM more accurately considering the active cooling system. The developed step algorithm and CFD modeling approach will pave the way for future work on cooling design optimization of the newly designed interior and surface mounted PMSM prototypes.

Share

COinS