Data-Driven Autoencoder Numerical Solver with Uncertainty Quantification for Fast Physical Simulations

Traditional partial differential equation (PDE) solvers can be computationally expensive, which motivates the development of faster methods, such as reduced-order-models (ROMs). We present GPLaSDI, a hybrid deep-learning and Bayesian ROM. GPLaSDI trains an autoencoder on full-order-model (FOM) data and simultaneously learns simpler equations governing the latent space. These equations are interpolated with Gaussian Processes, allowing for uncertainty quantification and active learning, even with limited access to the FOM solver. Our framework is able to achieve up to 100,000 times speed-up and less than 7% relative error on fluid mechanics problems.

[1]  Youngsoo Choi,et al.  gLaSDI: Parametric Physics-informed Greedy Latent Space Dynamics Identification , 2022, J. Comput. Phys..

[2]  Youngsoo Choi,et al.  LaSDI: Parametric Latent Space Dynamics Identification , 2022, Computer Methods in Applied Mechanics and Engineering.

[3]  C. Earls,et al.  PDE-READ: Human-readable Partial Differential Equation Discovery using Deep Learning , 2021, Neural Networks.

[4]  Christopher J. Earls,et al.  Bayesian Deep Learning for Partial Differential Equation Parameter Discovery with Sparse and Noisy Data , 2021, J. Comput. Phys. X.

[5]  Youngsoo Choi,et al.  A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder , 2020, J. Comput. Phys..

[6]  Huiyue Dong,et al.  Review of digital twin about concepts, technologies, and industrial applications , 2020 .

[7]  Yang Liu,et al.  Physics-informed learning of governing equations from scarce data , 2020, Nature Communications.

[8]  Jason Yon,et al.  Characterising the Digital Twin: A systematic literature review , 2020, CIRP Journal of Manufacturing Science and Technology.

[9]  Steven L. Brunton,et al.  Data-driven discovery of coordinates and governing equations , 2019, Proceedings of the National Academy of Sciences.

[10]  Kookjin Lee,et al.  Model reduction of dynamical systems on nonlinear manifolds using deep convolutional autoencoders , 2018, J. Comput. Phys..

[11]  B. Edmonds,et al.  Computational modelling for decision-making: where, why, what, who and how , 2018, Royal Society Open Science.

[12]  J. Nathan Kutz,et al.  Deep learning in fluid dynamics , 2017, Journal of Fluid Mechanics.

[13]  Steven L. Brunton,et al.  Data-driven discovery of partial differential equations , 2016, Science Advances.

[14]  Emil M. Constantinescu,et al.  Well-Balanced, Conservative Finite Difference Algorithm for Atmospheric Flows , 2016 .

[15]  S. Brunton,et al.  Discovering governing equations from data by sparse identification of nonlinear dynamical systems , 2015, Proceedings of the National Academy of Sciences.

[16]  T. Belytschko.  The Finite Element Method: Linear Static and Dynamic Finite Element Analysis: Thomas J. R. Hughes , 2008 .

[17]  Chi-Wang Shu,et al.  Efficient Implementation of Weighted ENO Schemes , 1995 .

[18]  M. Safonov,et al.  A Schur Method for Balanced Model Reduction , 1988, 1988 American Control Conference.

[19]  A. Patera,et al.  Reduced basis approximation and a posteriori error estimation for affinely parametrized elliptic coercive partial differential equations , 2007 .

[20]  Alex Schmid,et al.  Modeling and Simulation: the Computer Science of Illusion by Stanislaw Raczynski , 2007, J. Artif. Soc. Soc. Simul..

[21]  Geoffrey E. Hinton,et al.  Supporting Online Material for Reducing the Dimensionality of Data with Neural Networks , 2006 .

[22]  P. Holmes,et al.  The Proper Orthogonal Decomposition in the Analysis of Turbulent Flows , 1993 .