Process optimization in friction stir welding of γ-TiAl alloys — A machine learning approach

[Formula: see text]-TiAl alloys with high strength-to-weight ratio, and excellent high-temperature properties are used in the low-pressure turbine sections of advanced aero-gas turbine and automotive combustion engines. Their joining processes are highly in demand as it is necessary for wider application of the alloys. Friction stir welding (FSW) has emerged as a promising solid-state welding technique, especially in the context of [Formula: see text]-TiAl alloys, owing to its ability to mitigate thermal distortion and preserve material properties. This research focuses on the process optimization of FSW in joining similar [Formula: see text]-TiAl alloys using artificial intelligence (AI)-based machine learning (ML) algorithms and enhancing the tensile strength of the welded joints. The optimization process involves the identification and refinement of welding parameters such as rotational speed, friction time, and traverse speed. By leveraging ML algorithms, a predictive model is developed to estimate tensile strength based on the selected welding parameters.