heyoka.model.nrlmsise00_tn

Contents

heyoka.model.nrlmsise00_tn#

heyoka.model.nrlmsise00_tn(geodetic: Iterable[expression], f107: expression, f107a: expression, ap: expression, time_expr: expression) → expression#

Produces the expression of the thermospheric density as a function of geodetic coordinates and weather indices.

Added in version 4.0.0.

The expression is approximated by an artificial neural network (a thermoNET) trained over NRLMSISE00 data.

Note

The thermoNET parameters are published in the work: Izzo, Dario, Giacomo Acciarini, and Francesco Biscani. “NeuralODEs for VLEO simulations: Introducing thermoNET for Thermosphere Modeling” (arXiv preprint).

Note

A tutorial showcasing the use of this function is available.

A few checks are run on the input arguments. Specifically, the number of geodetic variables (i.e., the length of geodetic) must be three.

Parameters:

  • geodetic – expressions for the Geodetic components. [h, lat, lon] with h in km and lat in \(\left[ -\frac{\pi}{2}, \frac{\pi}{2} \right]\).

  • f107 – the F10.7 index for the day before time_expr.

  • f107a – the 81-day average of the F10.7 index centred on the day of time_expr.

  • ap – the average of the Ap indices on the day of time_expr.

  • time_expr – number of fractional days passed since 1st of January.

Returns:

the thermospheric density in [kg / m^3] as predicted by the NRLMSISE00 thermoNET model.

Raises:

ValueError – if one or more input arguments are malformed, as explained above.

Contents