Development of Physical Models for Launch Vehicle Dynamics Simulation

Abstract

This thesis presents the development and integration of a set of physical models for launch vehicle dynamics simulation within an existing simulation framework used at Spacebel. The work focuses on improving the representation of the physical environment and the associated forces acting on the vehicle, with particular emphasis on atmosphere, aerodynamics, and gravity.

A modular modeling approach is adopted, in which new environment and force computation blocks are implemented and coupled with the vehicle dynamics. The developments include the implementation of an atmospheric model with stochastic turbulence based on the Dryden formulation, the computation of aerodynamic angles and loads, and improvements in the handling of kinematics, reference frames, attitude dynamics, and gravity modeling. The implemented models are verified through unit tests and consistency checks in order to ensure their physical validity and numerical robustness.

In a second step, the impact of atmospheric disturbances on the vehicle response is investigated. Deterministic sensitivity analyses are first carried out to assess the influence of key parameters such as turbulence intensity and mean wind magnitude. Then, Monte Carlo simulations are performed to quantify the statistical dispersion of critical aerodynamic quantities, including the angle of attack, sideslip angle, dynamic pressure, and lateral aerodynamic force.

The results show that the turbulence intensity mainly affects fluctuation-related metrics, while steady wind components and stochastic realizations can significantly modify the flight conditions at which peak loads occur. In particular, lateral aerodynamic quantities exhibit a very large variability across realizations, and rare but more severe cases are observed. These results highlight the strongly nonlinear nature of the coupled atmosphere--aerodynamics--dynamics system.

Overall, this work provides a consistent extension of Spacebel's simulation framework and demonstrates its capability to support physically sound modeling and uncertainty-aware analyses for launch vehicle dynamics simulation.