OPTIMIZATION AND PREDICTION OF LIQUIDUS TEMPERATURE OF MILD STEEL WELD METAL USING RSM AND ANN

Abstract

This study is focused on the optimization and prediction of Liquidus Temperature of Mild steel weld metal using Response Surface Methodology (RSM) and Artificial Neural Network (ANN) from Tungsten Inert Gas (TIG) welding process. Welding Current, Welding Voltage and Gas Flow Rate are the process input parameters and the response variable is Liquidus Temperature. The final solution of the optimization process is to determine the most appropriate percentage combination of the Liquidus Temperature with the optimum values of Current, Voltage and the Gas Flow Rate that will adequately optimize (minimize) the Liquidus Temperature of the Mild Steel weld metal. Optimizing this process is one sure way of producing a quality weld.

The RSM model produced the numerical optimal solution for the weldment of Mid Steel (MS). The model Coefficient of Determination (R²) and Adjusted R² for Liquidus Temperature are 94.69% and 89.92% respectively. The Optimal Solutions for the input parameters are; Welding Current, 180.00Amps, Welding Voltage, 21.672Volts and Gas Flow Rate, 15.504L/min. The Optimal Solution for the response variable, Liquidus Temperature is 1484.783⁰C. From the analysis of variance (ANOVA), it was observed that welding current (WC) input parameter has more significant effect on the Liquidus temperature response variable.

The ANN analysis predicted an optimal solution for the Liquidus temperature response variable to be 1464.49, with an overall strong correlation (R) between the input factors and the response variable to be 99.89%. Therefore, it is advised that the models be used to navigate the design space.

KEYWORDS:  TIG, Mild Steel, Liquidus Temperature, RSM, ANN