Prediction of the Ultimate Tensile Strength in API X70 Line Pipe Steel Using an Artificial Neural Network Model

Abstract

An artificial neural network (ANN) model has been developed for the analysis and simulation of the correlation between the chemical composition and mechanical properties of high strength low alloy (HSLA) steel X70. The input parameters of the model consist of the base metal chemical composition (C, Si, Mn, the sum of Cr+Cu+Ni+Mo, the sum of Nb+Ti+V, carbon equivalent CEpcm) and the yield strength (YS). The outputs of the ANN model include the ultimate tensile strength (UTS) of the test material. Scatter plots, correlation coefficient (R) and mean relative error (MRE) were used to assess the performance of the developed neural network. Interestingly, the model output is efficient to calculate the mechanical properties of high strength low alloy steels, especially the ultimate tensile strength as a function of chemical composition and yield strength of the used material. The obtained results are in a good agreement with experimental ones, with high correlation coefficient and low mean relative error. The predictions accuracy of the developed model also conforms to the results of mean paired T-test.