Friction Stir and Cold Metal Transfer Welding Of AA7075: Experimental Investigations and Machine Learning Approaches

Abstract

Frictions stir welding (FSW) and cold metal transfer (CMT) welding processes are ideal for joining aluminium and its alloys. FSW and CMT are clearly distinguished by their working principle, one is solid state type and other is fusion type welding process. This paper attempted to study the effectiveness of both processes on aluminium alloy. AA7075 is a heat treatable alloy and most suitable for dynamic applications owing to its exceptional characteristics. Structure and mechanical properties of the joints produced by FSW and CMT are investigated. Experimental results revealed that the performance of FSW is better than the CMT in metallurgical and mechanical aspects. The fine structure and homogenous distribution of particles in FSW joints results in strengthening mechanism unlike CMT, where the particles are uneven and coarse-grained. However, CMT can produce welds with higher speeds and can achieve performance at par with FSW. Further, machine learning techniques, polynomial regression and support vector machine are applied to derive the relationships between process parameters and strength parameters.