چکیده مقاله
In this research, the optimization of friction stir welding FSW of Al5083 alloy wasinvestigated using artificial neural network ANN For this purpose, parameters such asrotation speed and tool advance speed were selected as neural network input and tensilestrength as output of neural network The microstructural and mechanical studies of thecreated joint showed that the high traverse speed of the tool leads to the undesiredshaking of the central area of the joint and pitting of the joint due to the lack of propermaterial flow to the back ofthe tool A very low traverse tool advance speed also leads todifferences in the structure and properties of the central area of the joint and the adjacentareas of joint loss of the joint properties Similarly, the maximum strength was obtainedat the optimal rotation speed of the tool The investigations showed that the neuralnetwork has a high capability in predicting the strength of the established connection andthe multi layered perceptron neural network with three intermediate layers with a valueof MSE equal to 0 033 and correlation coefficient of 0 94 has a favorable ability of andhigh accuracy for estimation oftensile strength ofFSW bonded Al5083
کلیدواژهها
نویسندگان
شیوه ارجاع
Mosallaee Pour, M and Azizi, G and Hossein-morshedy, A,1401,Prediction of Tensile Strength of Friction Stir Welded Al-5083 by UsingArtificial Neural Network,5th International Conference on Welding and Non Destructive Testing & 23rd National Conference on Welding & Inspection & 12th National Conference on NDT,Isfahan
ارائهشده در
مجموعه مقالات پنجمین کنفرانس بین المللی جوشکاری و آزمایش های غیرمخرب،بیست و سومین کنفرانس ملی جوش و بازرسی و دوازدهمین کنفرانس ملی آزمایش های غیرمخرب و اولین کنفرانس ملی ساخت افزایشی10 اسفند 1401 · اصفهان