Cost-effective search for lower-error region in material parameter space using multifidelity Gaussian process modeling

Information regarding precipitate shapes is critical for estimating material parameters. Hence, we considered estimating a region of material parameter space in which a computational model produces precipitates having shapes similar to those observed in the experimental images. This region, called the lower-error region (LER), reflects intrinsic information of the material contained in the precipitate shapes. However, the computational cost of LER estimation can be high because the accurate computation of the model is required many times to better explore parameters. To overcome this difficulty, we used a Gaussian-process-based multifidelity modeling, in which training data can be sampled from multiple computations with different accuracy levels (fidelity). Lower-fidelity samples may have lower accuracy, but the computational cost is lower than that for higher-fidelity samples. Our proposed sampling procedure iteratively determines the most cost-effective pair of a point and a fidelity level for enhancing the accuracy of LER estimation. We demonstrated the efficiency of our method through estimation of the interface energy and lattice mismatch between MgZn2 and {\alpha}-Mg phases in an Mg-based alloy. The results showed that the sampling cost required to obtain accurate LER estimation could be drastically reduced.

[1]  Yunzhi Wang,et al.  Simulation study of precipitation in an Mg–Y–Nd alloy , 2012 .

[2]  K. Hono,et al.  Atom probe characterization of plate-like precipitates in a Mg–RE–Zn–Zr casting alloy , 2003 .

[3]  James E. Gubernatis,et al.  Multi-fidelity machine learning models for accurate bandgap predictions of solids , 2017 .

[4]  Kazuhiro Hono,et al.  Effect of double aging and microalloying on the age hardening behavior of a Mg–Sn–Zn alloy , 2011 .

[5]  Akinori Yamanaka,et al.  Data assimilation for massive autonomous systems based on a second-order adjoint method. , 2016, Physical review. E.

[6]  M. Nishijima,et al.  Characterization of β′ Phase Precipitates in an Mg-5 at%Gd Alloy Aged in a Peak Hardness Condition, Studied by High-Angle Annular Detector Dark-Field Scanning Transmission Electron Microscopy , 2006 .

[7]  A. G. Khachaturi︠a︡n Theory of structural transformations in solids , 1983 .

[8]  Yuhki Tsukada,et al.  Equilibrium shape of isolated precipitates in the α-Mg phase , 2014 .

[9]  Kazuhiro Hono,et al.  Age-hardening response of Mg-0.3 at.%Ca alloys with different Zn contents , 2009 .

[10]  Masayuki Karasuyama,et al.  Estimation of material parameters based on precipitate shape: efficient identification of low-error region with Gaussian process modeling , 2019, Scientific Reports.

[11]  K. Easterling,et al.  Phase Transformations in Metals and Alloys , 2021 .

[12]  Kenji Hiraga,et al.  Structural Changes of Precipitates in an Mg-5 at%Gd Alloy Studied by Transmission Electron Microscopy , 2007 .

[13]  Yunzhi Wang,et al.  A simulation study of the shape of β′ precipitates in Mg–Y and Mg–Gd alloys , 2013 .

[14]  Peter W Voorhees,et al.  THE EQUILIBRIUM SHAPE OF A MISFITTING PRECIPITATE , 1994 .

[15]  David J. C. MacKay,et al.  Information Theory, Inference, and Learning Algorithms , 2004, IEEE Transactions on Information Theory.

[16]  Toshio Mura,et al.  Micromechanics of defects in solids , 1982 .

[17]  Akinori Yamanaka,et al.  Data assimilation for phase-field models based on the ensemble Kalman filter , 2018 .

[18]  Kazuhiro Hono,et al.  The effect of Ag and Ca additions on the age hardening response of Mg–Zn alloys , 2013 .

[19]  Barry L. Mordike,et al.  Structural aspects of high performance Mg alloys design , 2002 .

[20]  J. G. Byrne,et al.  Precipitate strengthening mechanisms in magnesium zinc alloy single crystals , 1969 .

[21]  Kazuhiro Hono,et al.  Precipitation of prismatic plates in Mg–0.3Ca alloys with In additions , 2011 .

[22]  T. C. T. Ting,et al.  Micromechanics of Defects in Solids (2nd rev. ed.) , 1989 .

[23]  Kazuhiro Hono,et al.  Precipitation hardenable Mg―Bi―Zn alloys with prismatic plate precipitates , 2009 .

[24]  Shigeharu Kamado,et al.  Effect of microalloyed Zr on the extruded microstructure of Mg–6.2Zn-based alloys , 2014 .

[25]  Dietmar Gross,et al.  The effect of elastic inhomogeneity on equilibrium and stability of a two particle morphology , 1998 .

[26]  Kazuhiro Hono,et al.  Microalloying Effect on the Precipitation Processes of Mg-Ca Alloys , 2012, Metallurgical and Materials Transactions A.

[27]  Kazuhiro Hono,et al.  TEM and 3DAP characterization of an age-hardened Mg–Ca–Zn alloy , 2005 .

[28]  Shigeharu Kamado,et al.  Strong and ductile age-hardening Mg-Al-Ca-Mn alloy that can be extruded as fast as aluminum alloys , 2017 .

[29]  J. B. Clark Transmission electron microscopy study of age hardening in a Mg-5 wt.% Zn alloy , 1965 .

[30]  S. Celotto TEM study of continuous precipitation in Mg–9 wt%Al–1 wt%Zn alloy , 2000 .

[31]  M. Gibson,et al.  An enhanced age hardening response in Mg–Sn based alloys containing Zn , 2006 .

[32]  Xiang Gao,et al.  Enhanced age hardening response and creep resistance of Mg-Gd alloys containing Zn , 2005 .

[33]  Dietmar Gross,et al.  The equilibrium shape of an elastically inhomogeneous inclusion , 1997 .

[34]  Kenji Hiraga,et al.  Characterization of β' Precipitate Phase in Mg-2 at%Y Alloy Aged to Peak Hardness Condition by High-Angle Annular Detector Dark-Field Scanning Transmission Electron Microscopy (HAADF-STEM) , 2007 .

[35]  A. O'Hagan,et al.  Predicting the output from a complex computer code when fast approximations are available , 2000 .

[36]  Shigeharu Kamado,et al.  Strong and ductile heat-treatable Mg–Sn–Zn–Al wrought alloys , 2015 .

[37]  Robert B. Ash,et al.  Information Theory , 2020, The SAGE International Encyclopedia of Mass Media and Society.

[38]  Kazuhiro Hono,et al.  Enhanced age hardening response by the addition of Zn in Mg–Sn alloys , 2006 .

[39]  M. Kahlweit,et al.  Ostwald ripening of precipitates , 1975 .

[40]  Christopher M Wolverton,et al.  Predicting β′ precipitate morphology and evolution in Mg–RE alloys using a combination of first-principles calculations and phase-field modeling , 2014 .

[41]  Shin-ichi Ito,et al.  Grain growth prediction based on data assimilation by implementing 4DVar on multi-phase-field model , 2017, Science and technology of advanced materials.

[42]  Yunzhi Wang,et al.  Shape evolution of a precipitate during strain-induced coarsening , 1991 .

[43]  L. Goddard Information Theory , 1962, Nature.

[44]  Henning Friis Poulsen,et al.  Determining material parameters using phase-field simulations and experiments , 2017 .

[45]  Christopher K. I. Williams,et al.  Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning) , 2005 .

[46]  T. Sasaki,et al.  Compositional optimization of Mg–Sn–Al alloys for higher age hardening response , 2013 .

[47]  J. B. Vander Sande,et al.  Precipitation hardening in Mg-Ca-Zn alloys , 2006 .

[48]  Kazuhiro Hono,et al.  Enhanced age hardening in a Mg–2.4 at.% Zn alloy by trace additions of Ag and Ca , 2007 .