Modeling of friction stir welding of aviation grade aluminium alloy using machine learning approaches. (2nd January 2022)

• Record Type: Journal Article
• Authors: Verma, Shubham; Misra, Joy Prakash; Popli, Dipesh
• Is Part Of: International journal of modelling & simulation. Volume 42:Number 1(2022)
• Page Start: 1
• Page End: 8
• Publication Date: 2022-01-02
• Subjects: AA6082 -- FSW -- UTS -- random forest regression -- M5P model tree -- ANN; Mathematical models -- Periodicals; Simulation methods -- Periodicals; Mathematical models; Simulation methods; Periodicals; 003.3
• DOI: 10.1080/02286203.2020.1803605
• Languages: English
• ISSNs: 0228-6203

Abstract

ABSTRACT: The machine learning methodology is gaining immense exposure as a potential methodology for solving and modeling manufacturing problems. The present study deals with the application of machine learning approaches in analyzing and predicting the tensile behavior of friction stir welded AA6082. Rotational speed and feed rate are used as input variables; ultimate tensile strength (UTS) is observed as a response parameter. Full factorial designed is used to perform the experiment. Random forest regression, M5P tree regression, and artificial neural network (ANN) are employed to validate the experimental results. These machine learning-based models are adopted to analyzing the absurdity in actual and predicted data. Random forest regression is observed best performing a machine-learning approach in predicting the tensile behavior of FSW joints. In addition, sensitivity analysis is also carried out to determine the most sensitive factor for UTS. It is observed that rotational speed is the most influencing factor for UTS.