Thermal Analysis of Direct Air-Cooled PM Machine for UAV Propulsion

Wrobel, R; Graham-Watson, A; Mason, C; Chakraborty, N; Zouein, R

doi:10.1049/icp.2022.1086

Thermal Analysis of Direct Air-Cooled PM Machine for UAV Propulsion

Lookup NU author(s): Dr Rafal Wrobel, Archie Graham-Watson, Professor Nilanjan Chakraborty ORCiD

Downloads

Accepted version [.pdf]

Licence

This is the authors' accepted manuscript of a conference proceedings (inc. abstract) that has been published in its final definitive form by IET, 2022.

For re-use rights please refer to the publisher's terms and conditions.

Abstract

This paper presents findings from a combined theoretical and computational thermal analysis of alternative heat removal concepts for an open-frame PM machine. The analysed machine design has been developed for propulsion of an unmanned aerial vehicle (UAV), which operates in a regime characterised by a high torque overload (x2.5) at take-off and a high efficiency (+95%) at high-altitude cruise operation. The thermal management of such a machine has a significant impact when balancing the design for a lightweight/compact solution. Here, direct air-cooling of the machine is analysed by focussing on the winding body that is the dominant heat source during the UAV’s take-off phase. As the airflow into the machine body is generated by the forward motion of the UAV, a careful consideration needs to be taken when guiding the air through the machine body. In this analysis, alternative airflow channelling via the stator-winding assembly is investigated. The results suggest that the proposed motor design with reduced conductor slot utilisation and stator bore airflow guiding feature allows for achieving the required convection heat transfer coefficient (75W/m2K at 1m/s), which is 150% improvement as compared with a more conventional design with a good conductor slot utilisation. However, the reduced conductor volume needs to be compensated for by appropriately adjusting the machine outer diameter and profile during the initial design sizing exercise.