Abstract
With the advent of integrated computational materials engineering, there is a drive to exchange statistical confidence in a design obtained from repeated experimentation with one developed through modeling and simulation. Since these models are often missing physics or include incomplete knowledge or simplifying assumptions into their mathematical construct, they may not capture the physical system or process adequately over the entire domain. This can lead to a systematic discrepancy, otherwise known as model misfit, between the model output and the system it represents over at least part of the domain. However, by accounting for this discrepancy in uncertainty analyses, reliable inference, and prediction on the model parameters and material behavior can be made despite the missing physics. The previous statement is contingent on two conditions: (1) the discrepancy is systematic, and (2) the structure of the discrepancy is well understood, which is required to minimize issues with non-identifiability among unknown model components. We illustrate these insights via a case study of inference and prediction in a phenomenological meso-scale VPSC crystal plasticity model, which does not contain physics describing the elastic regime of deformation. Inference is performed via a Bayesian approach enabled by posterior simulation. Posterior uncertainty about unknown model parameters takes into account observation error, uncertainty stemming from aleatoric sample-to-sample variability, and model form error. Posterior uncertainty in the unknown Voce hardening parameters is propagated to generate a distribution of the stress–strain response in both the elastic and plastic regimes. Additionally, posterior predictive distributions are simulated to establish uncertainty bounds for future unobserved data.
Similar content being viewed by others
Notes
The observant reader might notice some observations indicate a negative position due to the noise; while this example is not perfect, it serves the purpose of emphasizing the importance of considering model misfit in the analysis.
The hardening rates are traditionally represented by \(\theta _{0}\) and \(\theta _{1}\) in the Voce hardening law. However, to avoid confusion with parameters in the statistical model, this notation is adopted as in [16].
The code and required data to reproduce these results is made available on GitHub at https://github.com/mesoOSU/UQ.
References
Honarmandi P, Paulson N, Arroyave R, Stan M (2019) Uncertainty quantification and propagation in calphad modelling. Model Simul Mater Sci Eng. https://doi.org/10.1088/1361-651X/ab08c3
Paulson NH, Bocklund BJ, Otis RA, Liu Z-K, Stan M (2019) Quantified uncertainty in thermodynamic modeling for materials design. Acta Mater 174:9–15
Olbricht W, Chatterjee ND, Miller K (1994) Bayes estimation: a novel approach to derivation of internally consistent thermodynamic data for minerals, their uncertainties, and correlations. Part I: theory. Phys Chem Miner 21(1–2):36–49
Königsberger E (1991) Improvement of excess parameters from thermodynamic and phase diagram data by a sequential bayes algorithm. Calphad 15(1):69–78
Malakhov DV (1997) Confidence intervals of calculated phase boundaries. Calphad 21(3):391–400
Otis RA, Liu Z-K (2017) High-throughput thermodynamic modeling and uncertainty quantification for ICME. JOM 69(5):886–892
Stan M, Reardon BJ (2003) A bayesian approach to evaluating the uncertainty of thermodynamic data and phase diagrams. Calphad 27(3):319–323
Kouchmeshky B, Zabaras N (2009) The effect of multiple sources of uncertainty on the convex hull of material properties of polycrystals. Comput Mater Sci 47(2):342–352
Yeratapally SR, Glavicic MG, Argyrakis C, Sangid MD (2017) Bayesian uncertainty quantification and propagation for validation of a microstructure sensitive model for prediction of fatigue crack initiation. Reliab Eng Syst Saf 164:110–123
Kouchmeshky B, Zabaras N (2010) Microstructure model reduction and uncertainty quantification in multiscale deformation processes. Comput Mater Sci 48(2):213–227
Koslowski M, Strachan A (2011) Uncertainty propagation in a multiscale model of nanocrystalline plasticity. Reliab Eng Syst Saf 96(9):1161–1170
Fezi K, Krane MJM (2016) Uncertainty quantification of modelling of equiaxed solidification. In: IOP conference series: materials science and engineering, vol 143, p 012028. IOP Publishing
Plotkowski A, Krane MJM (2017) Quantification of epistemic uncertainty in grain attachment models for equiaxed solidification. Metall Mater Trans B 48(3):1636–1651
Sun G, Li G, Zhou S, Wei X, Yang X, Li Q (2011) Multi-fidelity optimization for sheet metal forming process. Struct Multidiscipl Optim 44(1):111–124
Talapatra A, Boluki S, Duong T, Qian X, Dougherty E, Arróyave R (2018) Autonomous efficient experiment design for materials discovery with bayesian model averaging. Phys Rev Mater 2(11):113803
Ricciardi DE, Chkrebtii OA, Niezgoda SR (2019) Uncertainty quantification for parameter estimation and response prediction. Integr Mater Manuf Innov 8(3):273–293
Brynjarsdóttir J, O’Hagan A (2014) Learning about physical parameters: the importance of model discrepancy. Inverse Prob 30(11):114007
Gelfand AE, Smith AFM (1990) Sampling-based approaches to calculating marginal densities. J Am Stat Assoc 85(410):398–409
Gilks WR, Richardson S, Spiegelhalter D (1995) Markov chain Monte Carlo in practice. Chapman and Hall/CRC, London
Brooks S, Gelman A, Jones G, Meng X-L (2011) Handbook of markov chain monte carlo. CRC Press, London
Robert C, Casella G (2013) Monte Carlo statistical methods. Springer, Berlin
Metropolis N, Rosenbluth AW, Rosenbluth MN, Teller AH, Teller E (1953) Equation of state calculations by fast computing machines. J Chem Phys 21(6):1087–1092
Hastings WK (1970) Monte carlo sampling methods using markov chains and their applications. Biometrika. https://doi.org/10.2307/2334940
Geman S, Geman D (1984) Stochastic relaxation, gibbs distributions, and the Bayesian restoration of images. IEEE Trans Pattern Anal Mach Intell 6:721–741
Barnard J, McCulloch R, Meng X-L (2000) Modeling covariance matrices in terms of standard deviations and correlations, with application to shrinkage. Stat Sin 10:1281–1311
O’Malley AJ, Zaslavsky AM (2008) Domain-level covariance analysis for multilevel survey data with structured nonresponse. Journal of the American Statistical Association 103(484):1405–1418
Kennedy MC, O’Hagan A (2001) Bayesian calibration of computer models. J R Stat Soc Ser B (Stat Methodol) 63(3):425–464
Rasmussen CE (2003) Gaussian processes in machine learning. In: Summer school on machine learning. Springer, pp 63–71
Matérn B (1960) Spatial variation, vol 36 of. Lecture notes in statistics
Bjørnstad ON, Falck W (2001) Nonparametric spatial covariance functions: estimation and testing. Environ Ecol Stat 8(1):53–70
Fortmann RS (2012) Understanding the bias-variance tradeoff. I. http://scott.fortmann-roe.com/docs/BiasVariance.html (hämtad 2017-09-29)
Tomé CN, Lebensohn RA (2009) Manual for code visco-plastic self-consistent (VPSC). Los Alamos National Laboratory, New Mexico
Voce E (1948) The relationship between stress and strain for homogeneous deformation. J Inst Metals 74:537–562
Voce E (1955) A practical strain hardening function. Metallurgia 51:219–226
Wang H, Wu PD, Tomé CN, Huang Y (2010) A finite strain elastic–viscoplastic self-consistent model for polycrystalline materials. J Mech Phys Solids 58(4):594–612
Lebensohn RA, Kanjarla AK, Eisenlohr P (2012) An elasto-viscoplastic formulation based on fast fourier transforms for the prediction of micromechanical fields in polycrystalline materials. Int J Plast 32:59–69
Lebensohn RA, Castañeda PP, Brenner R, Castelnau O (2011) Full-field vs. homogenization methods to predict microstructure–property relations for polycrystalline materials. In: Computational methods for microstructure-property relationships. Springer, pp 393–441
Zhao P, Low TSE, Wang Y, Niezgoda SR (2016) An integrated full-field model of concurrent plastic deformation and microstructure evolution: application to 3d simulation of dynamic recrystallization in polycrystalline copper. Int J Plast 80:38–55
Roters F, Eisenlohr P, Kords C, Tjahjanto DD, Diehl M, Raabe D (2012) Damask: the düsseldorf advanced material simulation kit for studying crystal plasticity using an fe based or a spectral numerical solver. Procedia Iutam 3:3–10
Roters F, Eisenlohr P, Hantcherli L, Tjahjanto DD, Bieler TR, Raabe D (2010) Overview of constitutive laws, kinematics, homogenization and multiscale methods in crystal plasticity finite-element modeling: Theory, experiments, applications. Acta Mater 58(4):1152–1211
Forrester AIJ, Sóbester A, Keane AJ (2007) Multi-fidelity optimization via surrogate modelling. Proc R Soc Math Phys Eng Sci 463(2088):3251–3269
Acknowledgements
S.R.N and D.E.R. received support from The Air Force Research Laboratory under Award FA8650-17-1-5277 “Ensemble predictions of material behavior for ICMSE for additive structures.” S.R.N. and O.A.C. received seed funding for this work through the Center for Emergent Materials: An NSF MRSEC (Chris Hammel PI, NSF Award DMR-0820414). Computing resources for the Ohio Supercomputer Center were provided through the Simulation Innovation and Modeling Center (SIMCenter) at Ohio State.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1: MCMC Sampler Implementation
Appendix 2: Derivation of Full-Conditionals
Derivation of full conditional distributions over \(\delta ^{2}, {\varPsi }, R\) and \({\varDelta }\).
Independent Error Precision,\(\delta ^{s}\)
Full Error Precision Matrix,\({\varPsi }\)
Random Effects Precision (\({\varLambda }\)) Decomposed Correlation, R
Discrepancy,\({\varDelta }\)
Appendix 3: Parameter Correlation
Appendix 4: Additional Figures
Marginal posterior distributions: Fig. 12 compares the distributions between the overall (dashed lines) and random effects (solid lines).
Rights and permissions
About this article
Cite this article
Ricciardi, D.E., Chkrebtii, O.A. & Niezgoda, S.R. Uncertainty Quantification Accounting for Model Discrepancy Within a Random Effects Bayesian Framework. Integr Mater Manuf Innov 9, 181–198 (2020). https://doi.org/10.1007/s40192-020-00176-2
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s40192-020-00176-2