A review of using computational fluid dynamic in simulating of friction stir welding and parametric studies

Friction Stir Welding (FSW) is still gradually evolving where it is newer than most thermomechanical processes and due to its ability to avoid many of the common defects in other welding techniques it has become largely used, particularly for those materials that are soft.[1]
Since the invention of friction stir welding by The Welding Institute (TWI), Cambridge, UK, there have been many attempts to comprehend the physical phenomena that take place during this process. These phenomena can be affected by the welding variables and determine the quality of the welded joint as well. Numerous studies have been carried out based on the experimental approach. However, the experiments were often time-consuming and costly. As an alternative and to overcome these problems that accompany the experimental works, other approaches have been used such as numerical and analytical modeling. Numerical modeling was represented in two main methods, which are fluid dynamic (simulation of material flow and temperature distribution) and solid mechanics (simulation of temperature distribution, stress and strain).[2]
This paper focuses on applying Computational Fluid Dynamic (CFD) in modelling of FSW as its one of the most powerful numerical analysis technique. Whereas FSW is a complex process, the so important coupled physical phenomena such as heat generation, heat transfer, and metal flow are discussed. In addition, the latest developments in analysis of FSW using CFD are reviewed. Although, this technique is promising, there are still important issues need to be addressed. Thus, some factors that affect the reliability and applicability of CFD modelling are highlighted, and some future opportunities are identified.