Prediction of Weld Quality in Dissimilar Metal Sheets using Machine Learning Algorithms in R

Mechanical Engineering

Dissimilar metals with different thicknesses are frequently required due to the increasing demand for lightweight vehicle body constructions. It can be difficult to weld dissimilar metals together because their characteristics change. The purpose of this research is to investigate how welding factors affect the quality and strength of the weld. The Central Campsite Design Matrix was used to perform spot-welding trials on mild steel and 304L SS sheets of varying thickness. The purpose of the study was to determine how the strength of dissimilar metal welds was affected by the weld current, weld time, and weld force. R-language is used to create an empirical relationship for weld strength prediction. To determine which process parameters had a substantial impact on weld strength, ANOVA was used. Analysis is also done on the differences in hardness between the various weld samples. Through confirmatory testing, the ideal weld strength parameters for different sheet thicknesses were confirmed. The study's conclusions offer important new information about how modifications to the weld's characteristics improve the quality of the weld joint.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.