When Bifidelity Meets CoKriging: An Efficient Physics-Informed Multifidelity Method

In this work, we propose a framework that combines the approximation-theory-based multifidelity method and Gaussian-process-regression-based multifidelity method to achieve data-model convergence when stochastic simulation models and sparse accurate observation data are available. Specifically, the two types of multifidelity methods we use are the bifidelity and CoKriging methods. The new approach uses the bifidelity method to efficiently estimate the empirical mean and covariance of the stochastic simulation outputs, then it uses these statistics to construct a Gaussian process (GP) representing low-fidelity in CoKriging. We also combine the bifidelity method with Kriging, where the approximated empirical statistics are used to construct the GP as well. We prove that the resulting posterior mean by the new physics-informed approach preserves linear physical constraints up to an error bound. By using this method, we can obtain an accurate construction of a state of interest based on a partially correct physical model and a few accurate observations. We present numerical examples to demonstrate performance of the method.

[1]  James Demmel,et al.  The Componentwise Distance to the Nearest Singular Matrix , 1992, SIAM J. Matrix Anal. Appl..

[2]  Andy J. Keane,et al.  Engineering Design via Surrogate Modelling - A Practical Guide , 2008 .

[3]  Michael B. Giles,et al.  Multilevel Monte Carlo Path Simulation , 2008, Oper. Res..

[4]  Tiangang Cui,et al.  Multifidelity importance sampling , 2016 .

[5]  Ryan M. Weisman,et al.  Multi-fidelity orbit uncertainty propagation , 2019, Acta Astronautica.

[6]  Leo Wai-Tsun Ng,et al.  Multifidelity Uncertainty Quantification Using Non-Intrusive Polynomial Chaos and Stochastic Collocation , 2012 .

[7]  Dongbin Xiu,et al.  Multi-fidelity stochastic collocation method for computation of statistical moments , 2017, J. Comput. Phys..

[8]  Dan Yu,et al.  Surrogate Modeling of Computer Experiments With Different Mesh Densities , 2014, Technometrics.

[9]  K. A. Cliffe,et al.  Multilevel Monte Carlo methods and applications to elliptic PDEs with random coefficients , 2011, Comput. Vis. Sci..

[10]  Donald R. Jones,et al.  Efficient Global Optimization of Expensive Black-Box Functions , 1998, J. Glob. Optim..

[11]  Andreas C. Damianou,et al.  Nonlinear information fusion algorithms for data-efficient multi-fidelity modelling , 2017, Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[12]  Xiu Yang,et al.  Reweighted ℓ1ℓ1 minimization method for stochastic elliptic differential equations , 2013, J. Comput. Phys..

[13]  Wenxiao Pan,et al.  Optimizing Discharge Capacity of Li-O2 Batteries by Design of Air-Electrode Porous Structure: Multifidelity Modeling and Optimization , 2017 .

[14]  Loic Le Gratiet,et al.  RECURSIVE CO-KRIGING MODEL FOR DESIGN OF COMPUTER EXPERIMENTS WITH MULTIPLE LEVELS OF FIDELITY , 2012, 1210.0686.

[15]  Xiu Yang,et al.  Physics-Informed Kriging: A Physics-Informed Gaussian Process Regression Method for Data-Model Convergence , 2018, ArXiv.

[16]  Menner A Tatang,et al.  Direct incorporation of uncertainty in chemical and environmental engineering systems , 1995 .

[17]  Dongbin Xiu,et al.  A Stochastic Collocation Algorithm with Multifidelity Models , 2014, SIAM J. Sci. Comput..

[18]  Andreas Krause,et al.  Near-Optimal Sensor Placements in Gaussian Processes: Theory, Efficient Algorithms and Empirical Studies , 2008, J. Mach. Learn. Res..

[19]  Chi-Wang Shu,et al.  Strong Stability-Preserving High-Order Time Discretization Methods , 2001, SIAM Rev..

[20]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[21]  Andy J. Keane,et al.  Multi-fidelity design optimisation of a transonic compressor rotor , 2011 .

[22]  Gianluca Iaccarino,et al.  A low-rank control variate for multilevel Monte Carlo simulation of high-dimensional uncertain systems , 2016, J. Comput. Phys..

[23]  Ramakanth Munipalli,et al.  A Multifidelity Approach to Parameter Dependent Modeling of Combustion Instability , 2018, 2018 Joint Propulsion Conference.

[24]  Benjamin Peherstorfer,et al.  Optimal Model Management for Multifidelity Monte Carlo Estimation , 2016, SIAM J. Sci. Comput..

[25]  Alexander I. J. Forrester,et al.  Multi-fidelity optimization via surrogate modelling , 2007, Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[26]  P. L’Ecuyer,et al.  Random Number Generation and Quasi-Monte Carlo† , 2015 .

[27]  Jing Li,et al.  A unified framework for mesh refinement in random and physical space , 2015, J. Comput. Phys..

[28]  Roman Garnett,et al.  Bayesian optimization for sensor set selection , 2010, IPSN '10.

[29]  David A. Barajas-Solano,et al.  Physics-Informed CoKriging: A Gaussian-Process-Regression-Based Multifidelity Method for Data-Model Convergence , 2018, J. Comput. Phys..

[30]  Thomas J. Santner,et al.  Design and analysis of computer experiments , 1998 .

[31]  Hillary R. Fairbanks,et al.  Multi-fidelity uncertainty quantification of irradiated particle-laden turbulence , 2018, 1801.06062.

[32]  Roger Woodard,et al.  Interpolation of Spatial Data: Some Theory for Kriging , 1999, Technometrics.

[33]  Dongbin Xiu,et al.  High-Order Collocation Methods for Differential Equations with Random Inputs , 2005, SIAM J. Sci. Comput..

[34]  Robert W. Ritzi,et al.  Introduction to Geostatistics: Applications in Hydrogeology , 1998 .

[35]  A. Stein,et al.  Universal kriging and cokriging as a regression procedure. , 1991 .

[36]  D. Brus,et al.  A comparison of kriging, co-kriging and kriging combined with regression for spatial interpolation of horizon depth with censored observations , 1995 .

[37]  Robert M. Kirby,et al.  Fast predictive models based on multi-fidelity sampling of properties in molecular dynamics simulations , 2018 .

[38]  Michael S. Eldred,et al.  Multifidelity Uncertainty Quantification Using Spectral Stochastic Discrepancy Models. , 2015 .

[39]  Houman Owhadi,et al.  A non-adapted sparse approximation of PDEs with stochastic inputs , 2010, J. Comput. Phys..

[40]  Stéphane Zaleski,et al.  Order and complexity in the Kuramoto-Sivashinsky model of weakly turbulent interfaces , 1986 .

[41]  Dongbin Xiu,et al.  Computational Aspects of Stochastic Collocation with Multifidelity Models , 2014, SIAM/ASA J. Uncertain. Quantification.

[42]  Alberto Guadagnini,et al.  Uncertainty Quantification in Scale‐Dependent Models of Flow in Porous Media , 2017 .

[43]  G. Sivashinsky On Flame Propagation Under Conditions of Stoichiometry , 1980 .

[44]  J. Hyman,et al.  The Kuramoto-Sivashinsky equation: a bridge between PDE's and dynamical systems , 1986 .

[45]  Xiang Ma,et al.  An adaptive high-dimensional stochastic model representation technique for the solution of stochastic partial differential equations , 2010, J. Comput. Phys..

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

[47]  Paris Perdikaris,et al.  Machine learning of linear differential equations using Gaussian processes , 2017, J. Comput. Phys..

[48]  Yoshiki Kuramoto,et al.  Diffusion-Induced Chaos in Reaction Systems , 1978 .

[49]  P. Sagaut,et al.  Building Efficient Response Surfaces of Aerodynamic Functions with Kriging and Cokriging , 2008 .

[50]  P Perdikaris,et al.  Multi-fidelity modelling via recursive co-kriging and Gaussian–Markov random fields , 2015, Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences.

[51]  Xiu Yang,et al.  Adaptive ANOVA decomposition of stochastic incompressible and compressible flows , 2012, J. Comput. Phys..