A multi-fidelity and multi-disciplinary approach for the accurate simulation of atmospheric re-entry

Evaluating the on-ground casualty risk assessments due to controlled/uncontrolled re-entry is highly sensitive to accurately predicting fragmentation and thermal demise events. The current state-of-the-art re-entry analysis tools approximate the prediction of the fragmentation process by assuming a fixed altitude break-up model. However, this model is based on limited experimental data, introducing uncertainty into the re-entry analysis. Moreover, the presence of multiple bodies and complex geometries results in complex flow features such as shock-shock and shock-surface interactions, influencing the localised aerothermodynamic loads during hypersonic re-entry. Low-fidelity re-entry analysis tools that use analytical expressions for various flow regimes do not consider the effects of such complex flow features and introduce uncertainties in the process. A possible way to reduce some uncertainties is to use high-fidelity modelling tools for the aerothermal and structural aspects of re-entry analysis. These high-fidelity analysis methods are computationally expensive to utilise at every point along the re-entry trajectory. Thus, a multi-fidelity analysis approach is needed to balance the required complexity and computational time. This paper proposes a multi-fidelity and multi-disciplinary framework that combines low- and high-fidelity aerothermodynamics, thermal analysis, flight dynamics, and structural analysis in a modular approach to achieve a favourable trade-off between cost and accuracy. The novelty in the current study is two-fold: one is to simulate a more natural destructive re-entry process without using a prescribed altitude trigger for fragmentation, and the other is to implement automatic fidelity switches between high- and low-fidelity models based on the shock-envelope approximation of Billig's formulation. For the high-fidelity flow modelling, the open-source SU2-NEMO code is used to solve the slip to continuum regimes; the SPARTA-DSMC solver is used for transitional and free-molecular regimes. To estimate the fragmentation altitude, a linear structural analysis of objects modelled as joints are continually carried out using the FEniCS finite elements solver. A temperature-dependent von Mises yield criterion is used to identify failure in joints. The proposed framework will be implemented as TITAN: Transatmospheric Flight Simulation code, and a few realistic re-entry test case scenarios are utilised to test the effect of the current multi-fidelity framework in reducing the uncertainties associated with the destructive re-entry process.