Abstract
Progress in the field of power electronics within electric vehicles has generally been driven by conventional engineering design principles and experiential learning. Power electronics is inherently a multidomain field where semiconductor physics and electrical, thermal, and mechanical design knowledge converge to achieve a practical realization of desired targets in the form of conversion efficiency, power density, and reliability. Due to the promising nature of artificial intelligence in delivering rapid results, engineers are starting to explore the ways in which it can contribute to making power electronics more compact and reliable. Here, we conduct a brief review of the foray of artificial intelligence in three distinct subtechnologies within a power electronics system in the context of electric vehicles: semiconductor devices, power electronics module design and prognostics, and thermal management design. The intent is not to report an exhaustive literature review, but to identify the state of the art and opportunities for artificial intelligence to play a meaningful role in power electronics design from a mechanical and thermal standpoint, as well as to discuss a few promising future research directions.
Original language | American English |
---|---|
Article number | Article No. 034501 |
Number of pages | 7 |
Journal | Journal of Electronic Packaging, Transactions of the ASME |
Volume | 145 |
Issue number | 3 |
DOIs | |
State | Published - 1 Sep 2023 |
Bibliographical note
Publisher Copyright:© 2023 American Society of Mechanical Engineers (ASME). All rights reserved.
NREL Publication Number
- NREL/JA-5400-81720
Keywords
- artificial intelligence
- electric vehicles
- machine learning
- power electronics
- prognostics
- semiconductors
- thermal management