Supervised Machine Learning Approach for Modeling Hot Deformation Behavior of Medium Carbon Steel

Abstract

The hot metalworking process is a typical procedure in the metallurgy industries for manufacturing many daily-life products that can not be produced using the cold working process. [1,2] The crucial advantages of the hot forming process are the higher material formability. Besides, the produced parts have no oriented grain structure due to hot processing conditions, resulting in highly isotropic strength characteristics. [3,4] In the metal forming process, the material will experience various loading conditions through strain, stress, strain rate, and temperature. Therefore, the material can be tested at different temperatures to explain the deformation behavior and strain rates, and besides essential key parameters can also be obtained. [5,6] The constitutive model can be established and implemented into the finite element analysis (FEA) software using received stressstrain (SS) curves for performing forming simulations to save experimental time and cost. [7] Many researchers have been devising various constitutive models by establishing the relationship between the external loading conditions and the thermomechanical behavior of test materials. [8] Because precise flow stress models could be exploited to perform accurate forming simulations, and optimal forming conditions can eventually be received. Studies showed that Johnson-Cook (JC) and modified Johnson-Cook (MJC) models are widely used to predict material behavior under hot deformation conditions. Therefore, Song et al., [9] Wang et al., [10] Tan et al., [11] Liang et al., [12] and Li et al., [13] had investigated the selected material dynamical behavior at higher strain rates and elevated temperatures using JC and MJC models. They documented that the MJC model could accurately describe the studied material's flow behavior. For example, Liang et al. [12] examined Al-Si-Mg alloy material at strain rates of 10 À3 s À1 and deformation temperatures of to 500 °C. They reported that the prediction error from the MJC model was about 1.65% and held well-agreement with the actual data. Similarly, Li et al. [13] studied T24 steel material at hot deformations using the MJC model. They also confirmed that the proposed model wellcaptured the material's dynamic behavior and agreed with the experimental observations.Researchers have also adopted the strain compensated constitutive equation, repeatedly called, the Arrhenius-Type constitutive model, to describe the material hot deformation characteristics. For example, researchers Zhao et al., [14] Li et al., [15] Zhang et al., [16] Liu et al., [17] and Li et al. [18] had exploited the Zener-Hollomon exponent-type equation for the hard-toform materials such as 14Cr ODS steel, 2219 aluminum, TA15 titanium, titanium-aluminum, and magnesium alloys, respectively. Liu et al.

Warm Deformation Behavior and Flow Stress Modeling of AZ31B Magnesium Alloy under Tensile Deformation

“…where A, B, n, C, T, and m are the model coefficients [27][28][29]33]. Equation (1) [30] represents the elasto-plastic term, which shows the work hardening effect and viscosity term, which reveals strain-rate-strengthening effect and thermal softening term, which reveals the temperature effect that influences the material flow stress [34,35].…”

“…Additionally, the metallographic microstructure on the sample surface was observed via optical microscope (OM), as depicted in Figure 3 [ 26 ]. Furthermore, a field emission scanning electron microscopy (FESEM) (MIRA3 TESCAN, secondary electron detector, Seoul national university of science and technology, Seoul, South Korea) [ 27 , 28 ] was used to examine the tested samples for reviewing the fractured morphology during warm tensile deformation, as depicted in Figure 4 .…”

Flow Stress Characteristics and Constitutive Modeling of Typical Ultrahigh‐Strength Steel under High Temperature and Large Strain

“…where A 1 , A 2 , A, n 1, and β are material constants, α is equal to β/n 1 , R is gas constant, Z is the Zener-Hollomon parameter, Q is the deformation activation energy, and n is the stress index. Equation ( 14) can be obtained by taking logarithms of both sides of Equation (13).…”

“…The prediction accuracy of the constitutive model is 0.9995. Murugesan et al [13] explored the accuracy of different supervised machine learning methods in predicting flow stress and pointed out that the random forest regression model has the highest accuracy. The flow behaviors can be correctly predicted by artificial neural network models.…”

Artificial neural network applicability in studying hot deformation behaviour of high-entropy alloys

“…In 2004, Yeh et al [1] and Cantor et al [2] introduced high-entropy alloys (HEAs), which exhibit high mixing entropy that encourages the formation of random solid-solution phases [3]. Hot working, as a major processing step during forming/shaping of these alloys, is normally used for grain refinement via dynamic recrystallisation (DRX), reduction of casting defects, homogenisation of the microstructure, and improvement of the mechanical properties [4,5]. Similar to other alloys, the HEAs would be fabricated through elevated-temperature thermomechanically controlled processing, for which understanding the hot deformation behaviour is of utmost importance [6,7].…”