The recently emerging field of electrified aviation requires design automation which is able to quickly and accurately create light-weight, efficient, and reliable electric propulsion drives for potential mass production. This paper proposes a systematic multi-objective design methodology intended for aviation applications, specifically focusing on power electronic drives. Due to the high priority of reliability in such safety-critical applications, the lifetime estimations due to three major contributing factors are considered during the design optimization, namely, semiconductor power cycling lifetime, semiconductor cosmic radiation susceptibility, and DC-link capacitor power cycling lifetime. When used to design a back-to-back voltage source converter, the design optimization algorithm produced 114,817,329 solutions within approximately 2 hours. The resulting optimized performance with respect to power density, efficiency, and reliability was 11.66 kW/kg, 98.81%, and 223.48 Failures-in-Time (FIT), respectively.
|Title of host publication||2022 IEEE Energy Conversion Congress and Exposition, ECCE 2022|
|State||Published - 2022|
|Event||2022 IEEE Energy Conversion Congress and Exposition, ECCE 2022 - Detroit, United States|
Duration: Oct 9 2022 → Oct 13 2022
|Name||2022 IEEE Energy Conversion Congress and Exposition, ECCE 2022|
|Conference||2022 IEEE Energy Conversion Congress and Exposition, ECCE 2022|
|Period||10/9/22 → 10/13/22|
Bibliographical noteFunding Information:
This work was supported by the U.S. Department of Education’s GAANN Fellowship Program through the University of Kentucky Electrical and Computer Engineering Department.
© 2022 IEEE.
- Electric aircraft propulsion
- high efficiency
- high power density
- high reliability
- multi-objective design optimization
- power converters
ASJC Scopus subject areas
- Electrical and Electronic Engineering
- Mechanical Engineering
- Safety, Risk, Reliability and Quality
- Energy Engineering and Power Technology
- Renewable Energy, Sustainability and the Environment
- Control and Optimization