| Abstract Scope |
As the automotive industry shifts towards electrification, autonomous driving, and mobility to improve efficiency and meet environmental regulations on CO2 emissions, the parts industry is also transitioning from internal combustion engine components to those related to electric power and autonomous vehicles. As electric vehicles become more prevalent, the high-voltage systems used to power them lead to increased modularization and output. However, this also leads to higher temperatures, which can negatively impact the performance of electrical parts. To address this issue, it is essential to improve the durability and thermal management of electric vehicle parts, especially power semiconductor devices. We propose the use of machine learning to predict the tensile strength of electric vehicle parts produced through dissimilar material friction stir welding, and validate our model's reliability under new welding process conditions. |