Artificial Neural Network Modelling of the Effect of Vanadium Addition on the Tensile Properties and Microstructure of High-Strength Tempcore Rebars

Abstract

In high-strength rebar, the various microstructures obtained by the Tempcore process and the addition of V have a complex effect on the strength improvement of rebar. This study investigated the mechanism of strengthening of high-strength Tempcore rebars upon the addition of vanadium through artificial neural network (ANN) modelling. Various V contents (0.005, 0.072 and 0.14 wt.%) were investigated, and a large amount of bainite and V(C, N) were precipitated in the core of the Tempcore rebar in the high-V specimens. In addition, as the V content increased, the number of these fine precipitates (10–30 nm) increased. The precipitation strengthening proposed by the Ashby–Orowan model is a major contributing factor to the yield-strength increase (35 MPa) of the Tempcore rebar containing 0.140 wt.% V. The ANN model was developed to predict the yield and tensile strengths of Tempcore rebar after the addition of various amounts of V and self-tempering at various temperatures, and it showed high reproducibility compared to the experimental values (R-square was 93% and the average relative error was 2.6%). ANN modelling revealed that the yield strength of the Tempcore rebar increased more significantly with increasing V content (0.01–0.2 wt.%.) at relatively high self-tempering temperatures (≥530 °C). These results provide guidelines for selecting the optimal V content and process conditions for manufacturing high-strength Tempcore rebars.