Band 57, Heft 7

In this study, dissimilar friction stir welding (FSW) processes were performed using Taguchi L9 orthogonal layout on AA 5083 H111 and AA 2024 T351 plates. The ultimate tensile strength (UTS) and percentage of elongation (∊ t ) were evaluated as the performance characteristics of the FS welded joints. A Taguchi based grey relational analysis was conducted using a grey relational grade (GRG) to obtain a welded joint with an optimal combination of high strength and ductility. Taking the largest mean value of GRG into consideration, the optimal welding combination was obtained using a tool with straight-threaded pin shape with the process parameters of 500 min −1 tool rotational speed (TRS) and 63 mm/min welding speed (WS). This combination was obtained from the experimental combinations in the Taguchi L9 orthogonal layout. Additionally, analysis of variance was performed to investigate the effect of parameters on GRG value and significance of the parameters. It was revealed that the effectiveness of each parameter was ranked: TRS (48.02%) > tool type (30.60%) > WS (21.12%). Tunnel and cavity-type defects were found to occur in the welded joints produced with the TRS of 630 and 800 min −1 . Notably, the weakest welded joints were produced with the WS of 63 mm/min and 80 mm/min.

Material characterization is one of the major challenges faced in the field of materials research. The general approach is the assessment of quantitative properties, which are dependent on the utilization of destructive/non-destructive techniques. Conventional methods require the user to manually assess the obtained micrographs to identify the microstructural patterns followed by physical tests to quantify properties and characterization. A recent development in this area is the use of the concept of machine learning (ML) in image segmentation and analysis. Over the years, research in this area has resulted in the development of stable, robust and reliable systems, which yield consistently good results. This paper is aimed at introducing the use of one such machine learning approach based on Artificial Neural Networks (ANN) for image segmentation and quantification of material properties and discussion of some use cases. The results of the ML based method are compared with the results obtained from the traditional threshold based segmentation method.

This short article presents a simple case of failure, from a materials engineering point of view. One may therefore ask what makes it worth publishing. The inclined reader, though, will notice the somewhat surprising nature of the macroscopic damage pattern. The type of defect reported in the context of the first examination of the component, which had to be performed non-destructively, was therefore wrong. This is an example of the risk of misinterpretation in cases where the customer dictates how the examination is carried out that all those working in the field of damage analysis are aware of. It also demonstrates that using the scanning electron microscope is not a luxury, not even in the manufacturing control, as some damage patterns cannot be elucidated without using this device that has revolutionized the component failure analysis.