Date of Award


Publication Type

Master Thesis

Degree Name



Electrical and Computer Engineering

First Advisor

Narayan C. Kar


Electric vehicle, Heat transfer, LPTN, Thermal modelling, Traction motor



Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.


The electric motor is at the center focus as an alternative to the internal combustion engine for automotive applications since it does not produce greenhouse gas emissions and can contribute significantly to the reduction of fossil fuel consumption globally. As extensive research works are being done on electric vehicles at present, thermal analysis of traction motor is increasingly becoming the key design factor to produce electric motors with high power and torque capabilities in order to satisfy electric vehicle driving requirements. Motor losses cause active heat generation in the motor components and excessive temperature rise affects the electromagnetic performance of the traction motor. High torque and power requirements based on the driving conditions under urban and highway drive conditions demand high capacity motor cooling system in order to keep the temperature within the safe limit. Hence, it is critical to develop and design a temperature prediction tool to dynamically estimate the winding and magnet temperature and regulate cooling to remove excessive heat from the motor. Conventional thermal modeling of motors includes analytical and numerical modeling. Analytical modeling is done by using Lumped Parameter Thermal Network (LPTN) which is analogous to electric circuit and a fast method for predicting temperature. It uses heat transfer equations involving thermal resistances and thermal capacitances to analytically determine temperature at different nodes. Numerical modeling is done in two ways–Finite Element Analysis and Computational Fluid Dynamics. Numerical modeling can produce more accurate results, but it requires more computational time. Since the temperature of motor components has to be predicted very quickly, i.e. during driving, LPTN is more effective because LPTN can quickly predict temperature based on the heat transfer equations. This thesis proposes an LPTN model that predicts motor temperature and regulates the required coolant flow rate simultaneously. Thus, it is able to dynamically predict the temperature. MATLAB Simulink has been used for simulation of the LPTN model for a laboratory PMSM prototype. The thermal resistances in the thermal network model have been obtained from the motor geometrical parameters. The electromagnetic loss data with respect to torque and speed were taken as input, and thus the temperature results of motor components have been found. The future work will be to implement this model into full scale prototype of the motor.

Available for download on Thursday, July 29, 2021