RSM-Based Predictive Modelling of Gravimetric Corrosion Characteristics of Welded and Tempered UNS G10400 Carbon Steel in a Seawater Environment

Abstract

In this study, gravimetric analysis was used to determine corrosion properties, i.e., WL and CR of welded and tempered UNS G10400 CS exposed to Sw. In order to determine optimal WL and CR, RSM approach was utilized to develop a prediction model. 20 runs of the experiment were used for statistical ANOVA. As response variables, the experimental design matrix included WL and CR. To optimize the results of GC method and develop a predictive regression model, CCD was utilized. By ANOVA, the generated models for WL and CR indicated a significant p-value (which is probability, under the null hypothesis assumption, of obtaining a result equal to or higher than the one observed) of < 0.0001. F-values and R2 showed good statistical correlation between experimental and predicted values. Optimal conditions for WL and CR occurred at 25 °C(as-welded), with an ET of 17 days, and their optimum values were 0.744 g and 0.769 mm/year, respectively. Validation of optimized predicted results showed an UE of 1.6 and 2.9%, for WL and CR, respectively, which was below 5%. This revealed that the generated model adequately predicted CP of welded and tempered UNS G10400 CS in a Sw environment.