Processing parameters optimization in hot forging of AISI 4340 steel using instability map and reinforcement learning

Jeong, Hee Jeong; Park, Joonhee; Kim, Yosep; Shin, Sumin; Kim, Naksoo

doi:10.1016/j.jmrt.2023.01.106

Cited by 17 publications

(3 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the present work, it was shown that the ordinary ANN can successfully be used to model the flow stress and inferring useful information regarding the hot deformation behaviour. Accordingly, more sophisticated machine learning techniques can be used in this regard, including evolutionary deep neural net (EvoDN2) [37,38], adaptive neuro-fuzzy inference system (ANFIS) [39], and deep and reinforcement learning of artificial neural network model [10,16], as well as multi-objective optimisation techniques such as genetic algorithm [23].…”

Section: Resultsmentioning

confidence: 99%

“…The characterisation of hot deformation behaviour during hot working is vital for developing metalforming processes [8][9][10][11], for which the appropriate constitutive equations are utilised to predict the hot deformation behaviour of the materials under the prevailing loading conditions [12,13]. For this purpose, the empirical, semi-empirical, phenomenological, and physically-based models [14,15], as well as machinelearning approaches such as artificial neural networks (ANN) models [16][17][18][19][20][21][22][23] have been proposed so far.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

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

Zamani

Mirzadeh

Malekan

2023

Materials Science and Technology

View full text Add to dashboard Cite

The potentials of artificial neural network (ANN) modelling as a potent machine learning approach for investigating the hot deformation behaviour of high-entropy alloys (HEAs) and multi-principal element alloys during thermomechanical processing are assessed and reviewed. Flow stress of CoCrFeNiMn (FCC Cantor alloy), HfNbTaTiZr (BCC refractory alloy), AlCoCuFeNi, and Al xCoCrFeNi alloys is accurately predicted based on the deformation temperature, strain rate, and strain. Moreover, in comparison with the limited experimental dataset, a significantly larger output dataset can be generated by ANN to gain valuable insights such as prediction of flow stress (and whole dynamic recovery/recrystallisation flow curves), elucidating the microstructural mechanisms such as dynamic precipitation reactions, and obtaining hot working parameters (e.g. deformation activation energy) for different ranges of deformation conditions.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

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

Zamani

Mirzadeh

Malekan

2023

Materials Science and Technology

View full text Add to dashboard Cite

show abstract

“…Recently, Park et al developed FEM simulation data that can plot on the 3D processing map to control process parameters, avoid flow instability and include high power dissipation efficiency [32]. Jeong et al employed a 3D processing map and utilized a learning environment founded on a Q-learning algorithm in order to optimize processing parameters, encompassing both the temperature and stroke speed of the workpiece [33]. However, microstructural defects inside the material during plastic deformation cannot be visually detected, and few studies have been conducted on predicting and observing them.…”

Section: Introductionmentioning

confidence: 99%

Numerical Simulation of Crack Condition in Forging Products of M50 Bearing Steel Based on Processing Map Theory

Park

Han

Kwon

et al. 2023

Metals

Self Cite

View full text Add to dashboard Cite

The microstructure of forged products significantly impacts their properties, and defects or carbide distribution are not visible to the naked eye. Isothermal compression tests on M50 steel with a Gleeble 3500 tester were conducted to study microstructure behavior during forging. Tests examined the hot deformation behavior within a temperature range of 900–1200 °C and a strain rate range of 0.01–10 s−1. Power dissipation efficiency (η) and flow instability (ξ), which are crucial processing map parameters, were employed to analyze the high-temperature deformation behavior of M50 steel. The 3D processing map determined the optimum forging conditions, indicating that hot working should start at an initial temperature of 1050 °C or higher and a strain rate of 1 s−1, decreasing the strain rate and temperature as the strain increases. The 3D power dissipation efficiency map displayed an average value of 0.43 or higher at a strain rate of 0.1 s−1 and a temperature of 1150 °C before reaching a strain rate of 0.8. The Finite Element Method (FEM) simulated results, revealing ξ and η distributions, and confirmed that microstructure observation during deformation matched the hot forging parameters. This approach can effectively predict microstructure changes during hot forging.

show abstract

Hot Deformation Behavior and Numerical Simulation of 40CrNiMo Steel for Wind Turbine Pulley Shafts

Wei,

Liu,

Zhang

et al. 2024

steel research int.

View full text Add to dashboard Cite

The hot deformation behavior (T = 800–1100 °C, = 0.01–10 s−1) of 40CrNiMo steel for wind turbine pulley shafts was studied by Gleeble‐3800 thermomechanical simulator. A constitutive equation and hot processing map are established based on the friction and temperature correction curves. The most available hot processing parameters are determined by combining microstructure analysis. The temperature fields and effective strain fields under different deformation conditions are simulated by Deform‐3D software. The results indicate that the effect of friction on the flow curves is greater than that of temperature rise, the activation energy Q of hot deformation for 40CrNiMo steel calculated based on the theoretical calculation is 368.292 kJ mol−1. The constitutive model based on strain compensation has high accuracy, with an average relative error of 6.65% and a correlation coefficient of 0.987. The optimum hot processing interval is at a deformation temperature of 950–1050 °C and a strain rate of 0.03–0.25 s−1, which has a high‐power dissipation value and avoids the instability region. Additionally, numerical simulation results show that the temperature field distribution is uniform in this deformation range, and the standard deviation of the effective strain is low, making it suitable for hot processing.

show abstract

Processing parameters optimization in hot forging of AISI 4340 steel using instability map and reinforcement learning

Cited by 17 publications

References 49 publications

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

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

Numerical Simulation of Crack Condition in Forging Products of M50 Bearing Steel Based on Processing Map Theory

Hot Deformation Behavior and Numerical Simulation of 40CrNiMo Steel for Wind Turbine Pulley Shafts

Contact Info

Product

Resources

About