Surrogate-based Pareto optimization of annealing parameters for severely deformed steel

Ghiabakloo, H.,Lee, K.,Kazeminezhad, M.,Kang, B.S. Elsevier Ltd 2016 Materials & Design Vol.92 No.-

<P>Severe plastic deformation (SPD) is a metalworking technique that is used for the enhancement of the strength and hardness of metallic materials. As SPD causes ductility deterioration, materials typically necessitate annealing for ductility increase; however, annealing may conversely affect strength and hardness. Thus, to optimally balance strength, hardness, and ductility, this study determined annealing conditions with a severely deformed low carbon steel sheet by adjusting annealing time and temperature. For the facilitation of the annealing process optimization, measurements of strength, hardness, and ductility under various annealing conditions were represented by regression Kriging. Then, because of the conflicting nature of the desired metal properties, a set of optimal annealing conditions was identified by Pareto multi-objective optimization. Finally, the best combination on the Pareto front was selected with TOPSIS. The results of Pareto optimization with regression Kriging showed that the best candidates for annealing conditions can be determined at a significantly reduced experimental cost. (C) 2015 Published by Elsevier Ltd.</P>

Specialized finite elements for numerical simulation of the flexibly-reconfigurable roll forming process

Ghiabakloo, Hadi,Kim, Jeong,Kang, Beom-Soo Elsevier 2019 International journal of mechanical sciences Vol.151 No.-

<P><B>Abstract</B></P> <P>Flexibly-reconfigurable roll forming (FRRF) process is suggested in recent years as a response to the needs of modern manufacturing industries for small-lot or even single-lot production of doubly-curved sheet metal surfaces with reduced costs. In FRRF, a sheet metal is deformed by the use of two bent rollers with non-constant roll gap where the rolling and roll-bending mechanisms are imposed together. Thanks to FRRF process, the costs due to the manufacturing of dies are reduced; however, the computational cost for process design becomes significant as each design may be used for a few or even a single production. Hence, the net production rate is controlled by computational time that can be several times longer than the forming time. Considering the deformation mechanism in FRRF, this study tries to specialize an analysis method to FRRF that is accurate and efficient. This is accomplished by an in-house <SMALL>MATLAB</SMALL> implementation of an efficient finite element approach and testing various combinations of elements and variational formulations (mixed and irreducible) to find the best combination of them for numerical simulation of FRRF. For this purpose, first a set of preliminary comparisons are made to select: (1) the best integration method for the elements with irreducible formulation, (2) the best stress components to be included in the nodal parameters for the elements with mixed formulation, and (3) the best elemental class (Serendipity or Lagrangian) for the elements with nonlinear interpolation in at least two directions. Then, the best variational formulation is decided for each elemental family by comparing the simulation results in a range of numerical model parameters. Finally, the FE model parameters are optimized for each candidate by a surrogate-based Pareto optimization method, and the best element/formulation couple is selected among all the candidates by the technique for order of preference by similarity to ideal solution (TOPSIS). Moreover, in order to identify the most decisive parameters in the analysis and design of FRRF process, the response of FRRF to variation of several process parameters including thickness, width, rollers' curvature radius, roll gap distribution, and material properties is investigated and the results are compared with those reported by the other studies.</P> <P><B>Highlights</B></P> <P> <UL> <LI> The most efficient element/FE formulation couple for numerical simulation of the flexibly-reconfigurable roll forming process is identified. </LI> <LI> Several shell and solid elements from Serendipity and Lagrangian families with various interpolation degrees in rolling, transverse, and thickness directions are studied. </LI> <LI> Irreducible and mixed FE approaches are coupled with the elements and their performances are compared. </LI> <LI> The element/formulation couples are compared after a surrogate-based Pareto optimization of their performance with respect to the FE input parameters. </LI> <LI> The sensitivity of the flexibly-reconfigurable roll forming process to various parameters including thickness, width, initial transverse radius, roll gap distribution, and material properties is investigated. </LI> </UL> </P> <P><B>Graphical abstract</B></P> <P>[DISPLAY OMISSION]</P>

An efficient finite element approach for shape prediction in flexibly-reconfigurable roll forming process

Ghiabakloo, Hadi,Kim, Jeong,Kang, Beom-Soo Elsevier 2018 International journal of mechanical sciences Vol.142 No.-

<P><B>Abstract</B></P> <P>Flexibly-reconfigurable roll forming (FRRF) process is recently introduced for the production of doubly-curved sheet metal surfaces with convenient change of the curvatures for each part. In FRRF, the sheet metal is deformed by a pair of small-diameter bent rollers with non-constant roll gap distribution from center to edge locations; the resulted part has two curvatures in transverse and in longitudinal directions. In this process, the sheet experiences a bending deformation before rolling, a plane strain compression during rolling, and another bending deformation after rolling. The previous finite element (FE) simulations of FRRF have been performed using 3D solid elements. In the present study, in order to increase the efficiency and accuracy of shape prediction process, the domain is decomposed into three subdomains of pre-rolling, rolling, and post-rolling deformation steps; each subdomain is solved separately by hiring an appropriate method for it. The pre-rolling and rolling deformation steps are solved only one time as the deformation is steady-state in these subdomains. For the post-rolling deformation step in which the sheet undergoes unknown curvature changes, an elastoplastic FEM with curvilinear shell elements is used with an initial stress state and accumulative plastic strain fields recorded from the other subdomains. The mathematical foundation of the method, the numerical computation procedure, and the sensitivity analysis of the method to different parameters are presented. In order to validate the numerical method with experimental data, several specimens with various dimensions are processed by FRRF and the results are compared to the numerical data.</P> <P><B>Highlights</B></P> <P> <UL> <LI> An efficient FE approach based on domain decomposition is suggested for shape prediction in FRRF process. </LI> <LI> Quadratic and cubic curvilinear Lagrangian shell elements are utilized for elastoplastic FEA of the process. </LI> <LI> The sensitivity of the numerical results to different simulation parameters is analyzed. </LI> <LI> Various specimens with different dimensions are processed by FRRF and the results are compared to numerical values. </LI> </UL> </P> <P><B>Graphical abstract</B></P> <P>[DISPLAY OMISSION]</P>

Design of the flexibly-reconfigurable roll forming process by a progressively-improving goal seeking approach

Ghiabakloo, Hadi,Park, Ji-Woo,Kil, Min-Gyu,Lee, Kyunghoon,Kang, Beom-Soo Elsevier 2019 International journal of mechanical sciences Vol.157 No.-

<P><B>Abstract</B></P> <P>In order to design the flexibly-reconfigurable roll forming (FRRF) process, i.e., to find the set of process parameters that lead to the target 3D surface, an efficient approach is suggested in the present study that can significantly improve the production rate by FRRF where each process design is usually applied for a few or a single forming. According to this approach, the results of the already performed simulations and experiments are used as the sample data for constructing a surrogate model that receives a set of process parameters and returns the longitudinal radius. Then, the genetic algorithm (GA) is hired to explore the surrogate model to predict a set of process parameters that would lead to the target longitudinal radius. This is accomplished by a goal seeking approach and maximization of a conditional likelihood function. The predicted set of the process parameters is given to an efficient finite element method (FEM) to evaluate the corresponding longitudinal radius, and the new input-output pair is added to the sample data to improve the subsequent modeling and predictions. Then, the sequential modeling-prediction-evaluation (MPE) steps are repeated until the output is close enough to the target value. Since the results of all the simulations and experiments are saved in a database to be used for the next designs, the efficiency of the procedure is progressively improved because the surrogate model becomes more informative that in turn enables more exact predictions. However, constructing the surrogate model becomes slower as the number of sample points increases. For preventing this phenomenon, a modified version of the Morris-Mitchel criterion is utilized for subset sampling that selects a desired number of sample points with maximum space-filling property. In order to examine the suggested procedure, an initial database containing 54 sample points collected from experimental study is constructed. Then, the suggested approach is hired to reproduce 3 of the samples which are randomly selected from the database. The results show that this method can be used for design of FRRF process with a few or even one MPE that can significantly improve the production rate in this process.</P> <P><B>Highlights</B></P> <P> <UL> <LI> The material shape as a response of FRRF to the set of process parameters is modeled by Kriging regression method. </LI> <LI> The genetic algorithm is utilized to maximize a conditional likelihood function for predicting the process parameters which lead to the target shape. </LI> <LI> Each prediction during the goal seeking procedure is evaluated by an efficient finite element approach based on a domain decomposition method, and the result is added to the samples to improve the surrogate model. </LI> <LI> All the numerical and experimental results achieved during a design are added to a database for improving the subsequent modeling and design steps. </LI> <LI> A modified Morris-Mitchell criterion is utilized for selecting a subset of desired size from the set of all the previously saved sample points in the database for maintaining the efficiency of modeling and design procedures. </LI> <LI> The FRRF apparatus is used to produce 54 experimental sample points for constructing the initial database, and the suggested approach is hired to reproduce 3 of them as examples. </LI> </UL> </P> <P><B>Graphical abstract</B></P> <P>[DISPLAY OMISSION]</P>

Rapid Annealing of Severely Deformed Low Carbon Steel in Subcritical Temperature Range

H. Ghiabakloo,M. Kazeminezhad 대한금속·재료학회 2017 METALS AND MATERIALS International Vol.23 No.5

A low-carbon steel sheet containing 0.05 C, 0.203 Mn, and 0.0229 Si (all in wt%) was rapidly annealed in a temperaturerange of 300 °C to 600 °C after severe plastic deformation by using constrained groove pressing (CGP)technique. Microstructure evolution was investigated by scanning electron and optical microscopes. Mechanicalproperties were evaluated by hardness measurements and shear punch test. The results showed a thermal stabilityup to 400 °C where recrystallization did not occur in the specimens even after 7200 s. This thermal stability is inagreement with previously reported results of conventional annealing of the same steel after CGP. However,annealing at 500 °C and 600 °C led to recrystallization which started after holding times of 600 s and 20 s, respectively. Longer holding times resulted to grain growth and deterioration of strength and hardness, but the finalstrength and hardness were still higher than those of conventionally annealed specimens. The reason has beenattributed to no abnormal grain growth in the present study, in contrast to that occurs after conventional annealingof CGPed low carbon steel. The kinetics of recrystallization at 600 °C was studied using the celebrated Johnson-Mehl-Avrami-Kolmogorov (JMAK) model; the results showed a bi-linear JMAK plot indicating two differentstages of recrystallization rate before and after 70% recrystallization.

Numerical Simulation of the Flexibly-Reconfigurable Roll Forming Process

H. Ghiabakloo,M.G. Kil,J.W. Park,J. Kim,B.S. Kang 한국소성가공학회 2018 한국소성가공학회 학술대회 논문집 Vol.2018 No.5

An Elastoplastic Mixed Finite Element Method for Numerical Analysis of 3D Sheet Metal Forming in Flexibly-Reconfigurable Roll Forming Process

H. Ghiabakloo,J. Kim,B. S. Kang 한국소성가공학회 2016 한국소성가공학회 학술대회 논문집 Vol.2016 No.10