Scholarlink

<p>The current research investigates the usage of ML (Machine learning) algorithms in practical scenarios to predict the mechanical properties, specifically Ultimate tensile strength (UTS), elongation and microhardness of friction stir welded (FSW) aluminium Al6061 alloys. The welding process parameters like tool rotational speed, weld speed and axial forces act as input parameters and the UTS, elongation and micro hardness as the response variables. In this study 6&#xa0;mm thickness similar Al6061 plates used for joints. Supervised ML regression model algorithms like Extreme Gradient Boosting, Decision Tree (DT), Random Forest (RF), Recurrent Neural Networks (RNN) and Artificial Neural Networks (ANN) were implemented to envision the mechanical characteristics of FSW Al60601. Different algorithms showed different performances; the ANN method performed the best, as demonstrated by the findings. The experimental values of UTS, elongation, and microhardness and prediction values of ANN are closer. Among the ML techniques, a high coefficient of determination (0.96) is achieved by using ANN. This indicates that ANN can be used as an alternative way to calculate mechanical properties.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction of Mechanical Properties in Friction Stir Welded Al6061 Alloys Using Machine Learning Algorithms

  • 期刊论文

  • 发表日期：

    2025年5月9日

查看全文
  • Priyadarsini Morampudi,
  • Gadde Raghu Babu,
  • R. Koushik,
  • K. Durga Prasad,
  • G. Aravind,
  • T. Teja Sri Vaishawehwara

摘要

The current research investigates the usage of ML (Machine learning) algorithms in practical scenarios to predict the mechanical properties, specifically Ultimate tensile strength (UTS), elongation and microhardness of friction stir welded (FSW) aluminium Al6061 alloys. The welding process parameters like tool rotational speed, weld speed and axial forces act as input parameters and the UTS, elongation and micro hardness as the response variables. In this study 6 mm thickness similar Al6061 plates used for joints. Supervised ML regression model algorithms like Extreme Gradient Boosting, Decision Tree (DT), Random Forest (RF), Recurrent Neural Networks (RNN) and Artificial Neural Networks (ANN) were implemented to envision the mechanical characteristics of FSW Al60601. Different algorithms showed different performances; the ANN method performed the best, as demonstrated by the findings. The experimental values of UTS, elongation, and microhardness and prediction values of ANN are closer. Among the ML techniques, a high coefficient of determination (0.96) is achieved by using ANN. This indicates that ANN can be used as an alternative way to calculate mechanical properties.