Modeling of steady state hot flow behavior of API-X70 microalloyed steel using genetic algorithm and design of experiments

Autor: Bahman Mirzakhani, H. Abarghooei, Seyed Hossein Seyedein, Hossein Arabi
Rok vydání: 2017
Předmět:
Zdroj: Applied Soft Computing. 52:471-477
ISSN: 1568-4946
DOI: 10.1016/j.asoc.2016.10.021
Popis: Comparison between measured and predicted stress in various temperatures, grain sizes and strain rates.Display Omitted Hot torsion test was performed to study hot flow behavior of API-X70 steel.Genetic algorithm was used for the first time to model steady state hot flow behavior of API-X70 steel.Taguchi Design of Experiments method was used to reach an optimal value for Genetic Algorithm parameters.The model extracted from GA has higher accuracy with respect to the conventional methods.The GA models take the effect of metallurgical phenomena and predict hot flow behavior with good accuracy. API-X70 microalloyed steel is one of the most conventional materials that has been used to produce the pipelines used in oil and gas industry. This steel is produced by thermo mechanical processing (TMP). Prediction of steady state hot flow behavior of metals during TMP, for design of its forming process is of great importance. In this research, flow curves of API-X70 were obtained using hot torsion test at temperature range of 9501150C and strain rates of 0.0013s1. Genetic algorithm (GA) was used to find parameters of steady state stress semi-empirical model in the way that minimizing the difference between experimental data and model output. The optimal combination of GA parameters were chosen by Taguchi design of experiments(DOE) method in order to increase efficiency of GA. Accuracy of developed model to predict flow stress in steady state region was evaluated through statistical methods. Results showed a good agreement between developed model and experimental data with R2=0.99 and this model can predict steady state flow stress well.
Databáze: OpenAIRE
Externí odkaz: