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

Research has shown that for every 10% weight reduction, fuel economy can be increased by 6 to 8%. Aluminum has a positive strength to mass ratio, making it an excellent choice for vehicle mass savings however it is also notorious for forming poor bonds with other metals. Spot welding (also known as resistance spot welding) is a resistance welding technique used to join resistive metals such as aluminum and low carbon steel by applying pressure and heat from an electric current to the weld area.

Joint performance is well known to be dependent on the spot weld attributes developed during processing, and several factors influence the weld attributes during processing. Unfortunately, traditional statistical techniques do not provide a comprehensive understanding of the various process development parameters during the welding process due to the data structure and the number of variables involved. In an attempt to address this gap, a supervised machine learning approach has been applied to identify the key factors that influence aluminum and steel joint properties and performance, as well as to predict the process parameters needed to manufacture joints with predetermined performance. A number of theoretical methods are used to generate uniform designs to investigate the parametric space of the factors most responsible for the best performance joints and recommendations on the parametric space beneficial for exploration.