## Abstract

The predictive accuracy of the Navier–Stokes equations is known to degrade at the limits of the continuum assumption, thereby necessitating expensive and approximate solutions to the Boltzmann equation. While tractable in one spatial dimension, their high dimensionality increases the computational challenge of multidimensional, multiphysical Boltzmann calculations. It is therefore desirable to augment the Navier–Stokes equations for accuracy under these conditions. We present an application of a deep learning method to extend the validity of the Navier–Stokes equations to transition-continuum flows. The technique encodes the missing physics via a neural network, which is trained to reduce the error between the Navier–Stokes and Boltzmann solutions. While standard DL methods can be considered ad hoc due to the absence of underlying physical laws, at least in the sense that the systems are not governed by known partial differential equations, the DL framework leverages the a priori known Boltzmann physics while ensuring that the trained model is consistent with the Navier–Stokes equations. The online training procedure solves adjoint equations, constructed using algorithmic differentiation, which efficiently provide the gradient of the loss function with respect to the learnable parameters. The model is trained and applied to predict stationary, one-dimensional shock thickness in low-pressure argon.

Original language | English (US) |
---|---|

Pages (from-to) | 5484-5497 |

Number of pages | 14 |

Journal | AIAA journal |

Volume | 61 |

Issue number | 12 |

DOIs | |

State | Published - 2023 |

## ASJC Scopus subject areas

- Aerospace Engineering