Chemulator: Thermochemical emulator for hydrodynamical modeling
Abstract
The neural network-based emulator Chemulator advances the gas temperature and chemical abundances of a single position in an astrophysical gas. It is accurate on a single timestep and stable over many iterations with decreased accuracy, though performs less well at low visual extinctions. The code is useful for applications such as large scale ISM modeling; by retraining the emulator for a given parameter space, Chemulator could also perform more specialized applications such as planetary atmosphere modeling.
- Publication:
-
Astrophysics Source Code Library
- Pub Date:
- August 2021
- Bibcode:
- 2021ascl.soft08016H
- Keywords:
-
- Software