Mitigating Hardness Degradation in the Heat Affected Zone of AA7075-T6 Aluminum Alloy: Machine Learning Enabled Friction Stir Welding

Friction stir welding (FSW) of precipitation-strengthened aluminum alloys remains to be a challenging problem due to a significant reduction in strength or hardness within the heat- affected zone (HAZ). The hardness in the HAZ is affected by various parameters, including tool rotational rate, tool traverse speed, and tool shoulder diameter among others. In this study, we utilized machine learning (ML) tools to identify the key processing parameters that determine the hardness in the HAZ of AA7075-T6 alloy. Subsequently, we employed an adaptive design strategy to iteratively identify the combination of processing parameters required to achieve a targeted hardness in the HAZ. The goal is to maximize the improvement in targeted hardness with each subsequent experiment until the desired hardness is achieved. Initial data was collected from our random experiments as well as data within reported literature. The initial dataset was then used to train adaptive design tools, and the ML predicted processing parameters were validated via subsequent experiments. It was demonstrated that the ML-guided approach required the generation of only six experimental data points to attain a hardness difference of 41.00 HV 0.3 in the HAZ and the base metal, compared to trial-and-error based conventional method in which 32 data points were needed to obtain a HAZ hardness difference of 42.00 HV 0.3, thereby significantly increasing the efficiency of process optimization for a targeted mechanical property. Our machine learning guided approach revealed that only a small set of data needs to be generated experimentally to attain the targeted hardness in the HAZ. Finally, high fidelity mathematical equations were extracted from the data to predict hardness difference in the HAZ as a function of processing parameters using explainable artificial intelligence.