Machine Learning for Joint Quality Performance-Determining relationship between process parameters and weld microstructure of Al/steel resistance spot welds

To decrease vehicle weight, and thereby increase fuel efficiency, the properties of resistance spot welding (RSW) of aluminum to steel are being studied. The resistance spot welding of Al and steel results in a layer of brittle intermetallic compounds along the interface of the Al and steel sheets. Previous work has documented the properties of microstructure variables (fracture mode, hardness, thickness) in the intermetallic layer resulting from RSW of Al – steel welds. The role of weld process parameters in weld intermetallic layer, on the other hand, has yet to be completely investigated. We use supervised machine learning techniques to identify key welding parameters that result in optimal weld microstructure and intermetallic layer characteristic. The results are used to plan intelligent design of experiments in a subset of the parameter space leading to optimization for production.