A Survey of Machine Learning in Friction Stir Welding, including Unresolved Issues and Future Research Directions

Table 1

Explanation of variation in microstructures for different metals.

S.no.	FSW material composition	The characteristic feature of the microstructure	Ref.
1	Carbon steels	A gradual transformation occurs in the microstructure from austenite to martensite, increasing carbon content above 0.2%. The joint hardness is improved	[23]
2	Stainless steel (304 austenitic stainless steel)	Ferrite form and sigma form were observed. The ferrite form was mainly formed at the grain borders of austenite, which is recrystallized due to the high cooling rates, whereas the sigma phase was seen due to a decrease in cooling rate	[24]
3	AZ61 magnesium alloy	Stir zone: grains are well cultivated as a result of recrystallization. Compared to the base metal, the grain dimensions decrease as it goes onto the other weld zones, i.e., TMAZ and SZ; TMAZ region: distorted grain pattern is observed	[25]
4	Dissimilar aluminum alloy joints	Finer grains are formed near the interfaces of the two dissimilar alloys. The stir zone of dissimilar Al alloys consists of various grain sizes depending on the intermixing of two metals during FSW	[26]
5	Dissimilar magnesium alloys (A5052P-O and AZ31B-O alloys)	The differing microstructure is introduced near the bonded interface. It is visible by the zigzag pattern near the interface. With increasing tool speed, the surface structure turns smoother	[27]