Selective Analysis of a High-Strength Steel Shielded Metal Arc Weld Metal Database

The current effort performed a novel, two-stage selective analysis of a large, Shielded Metal Arc (SMA) Weld Metal (WM) database posted online at ResearchGate by Dr. Glyn M Evans [1] together with the previously published analysis [2] using Constraints-Based Modeling (CBM), an Artificial Neural Network (ANN) offered by Japan Welding Engineering Society (JWES) at its web-site [3] and a machine learning algorithm that provided a New A r3 equation [4] as appropriate and effective tools to organize and analyze Evans's database in gaining valuable insights on welding electrode development and/or evaluation: a.The first-stage selective analysis of TiBAlN series of WMs with N content below 85ppm (0.0085wt-%) used two metallurgical criteria involving CBM, New A r3 equation and JWES-ANN tool; and b.The second-stage selective analysis used the results of the first-stage selective analysis and identified 61 WMs belonging to different Fe-C-Mn systems with N content below 100ppm (0.0010wt-%) to reaffirm that control of selected characteristics is essential to obtain high-performance WM microstructures.