| About this Abstract |
| Meeting |
2023 TMS Annual Meeting & Exhibition
|
| Symposium
|
Late News Poster Session
|
| Presentation Title |
N-26: Machine Learning to Model the Relationship between Heat Affected Zone and Weld Join Quality Performance of Aluminum-Steel Resistance Spot Welds |
| Author(s) |
Narmadha Meenu Mohankumar, Moses Obiri, Deb Fagan, Alejandro Ojeda, Luke Durell, Shoieb Ahmed Chowdhury, Hassan Ghassemi-Armaki, Keerti Kappagantula |
| On-Site Speaker (Planned) |
Narmadha Meenu Mohankumar |
| Abstract Scope |
Resistance spot welding (RSW) of Aluminum-steel dissimilar materials is a prominent line of development in the automotive industry to reduce vehicle weight and improve fuel economy. However, properties of Aluminum such as the low melting point, low electrical resistivity, and high thermal conductivity pose unique challenges for the RSW process affecting the Aluminum-steel joint performance. The properties of the heat-affected zone (HAZ) are assumed to influence the Aluminum-steel joint properties and performance but have not been studied extensively. In our work, we use principal component analysis to reduce the dimension of weld performance variables (e.g., peak load, total energy, extension at the break, etc.) and apply machine learning techniques to identify the relationship between the weld performance and properties of HAZ. Moreover, we investigate and account for the uncertainty in the measurements of HAZ properties to improve model accuracy. |
| Proceedings Inclusion? |
Undecided |
| Keywords |
Machine Learning, Aluminum, Mechanical Properties |