Application of Machine Learning to Predict the Mechanical Properties of High Strength Steel at Elevated Temperature

Abstract

for conventional strength steels. For HSS, EN 1993-1-12 currently states that the rules in EN 1993-1-2 are applicable to high strength steels up to S700.The production of HSS is different from that of other steel types in two main aspects, the steel chemical composition and the heat treatment process that are adopted to achieve their high strengths. There are four main processing routes for the production of high strength steel plates, namely: Thermo-Mechanical (TM) rolling, Thermo-Mechanical rolling with Accelerated Cooling (TM-ACC), Thermo-Mechanical rolling with Direct Quenching (TM-DQ) and Conventional route for QT plates [4]. The different chemical composition and heat treatment processes employed in the production of the HSS by these routes lead to material stress-strain behaviours that are different from those of conventional mild steels at both room and elevated temperatures. At room temperature, the ductility and the ratio of the ultimate-to-yield strength of HSS reduce with increasing strength. At elevated temperatures, high strength steels, like all steels, lose strength and stiffness with increasing temperatures, but the rate of degradation is influenced by the chemical composition and the heat treatment process. The microstructure of mild steel can be either pearlite, bainite, or combination of both depending on the rate of cooling of austenite steel [5]. Bainite and pearlite transformations are stable phases, and do not transform to other phases without reheating to form austenite (at about 900°C). On the other hand, the microstructure of

AISI 1040 Çeliğinin Mikroyapı Resimlerinden Mekanik Özelliklerinin Derin Öğrenme ile Tahmini

“…Sıcaklık, nem ve diğer çevresel faktörler, numunenin malzemeyi en iyi temsil edecek yerden alınması, mekanik testlerde kullanılan cihazların hassas ve kalibrasyonlu olması gibi durumlar mekanik özellik tespitini zorlaştırmaktadır. Bu sorunların üstesinden gelmek için son yıllardaki çalışmalarda Makine Öğrenmesi ve Derin Öğrenme algoritmaları kullanılmıştır [1][2][3][4][5][6][7][8][9][10].…”

The Initial Assessment of the Possibility of Simulation Fire Standard Curve in the Electric Furnace with the Assessment of Chemical Composition Changes in Steel after Cooling Phase

“…Furthermore, these codes do not consider a cooling phase. This exciting approach is presented in [19], where authors presented the results obtained using machine learning (ML) algorithms to predict the mechanical properties, including ultimate tensile strength, yield strength, 0.2% proof strength, and elastic modulus, of high-strength steel plate material at elevated temperatures. Some works [20,21] investigated the mechanical properties of steel during the entire fire process, i.e., the heating and cooling phases.…”