Assessment of gas-surface interaction modelling for lifting body re-entry flight design

Abstract

My wish is that the reader enjoys this dissertation and that it brings him/her a better understanding on Gas-Surface Interaction (GSI) phenomena. Nonetheless, this journey through the VKI doctoral program has been long, and I wouldn't be able to address this topic without acknowledging people that were by my side during this research, because without them this thesis would not have been possible: A big thank to Olivier Chazot for giving me the opportunity to develop my thesis in the Plasmatron facility. His office was always open and ready for discussion. A special mention to his telephone, most of the time ringing during each one of our conversations. Also to Roberto Flores, director of this thesis in UPC, who allowed me to keep in contact with my university. Following the experimental nature of this dissertation, I would like to mention also the contribution made by Pascal Collin. He did an excelent job running the Plasmatron and installing the probes. I will always remember his desperation when water leaks appear when making the vacuum in the chamber. I thank also the rest of the Plasmatron team: Bernd Helber and Damien Le Quang as radiation experts, and Işil Şakraker as advisor for calorimetric measurements. Many CFD simulations appear in this dissertaion. For that, the advice of Vincent Van Der Haegen, my office mate, has been extremely useful to achieve a converged solution. Also, the flight extrapolation technique developed here deserves my gratitude to Fabio Pinna, whose expertise in CFD helped me to improve my IXV simulations and achieve the "Pinna Certification". I really enjoyed working with them. Other VKI colleagues I want to mention are Alessandro Turchi, who helped me understand many aspects of GSI phenomena, Francesco Panerai who set the basis to develop this dissertation, and Georgios Bellas-Chatzigeordis to whom I encourage to continue with the catalytic model development and implementation in VKI numerical tools. Last but not least, special thanks to my family. They encouraged me to pursue a doctorate degree and supported me in the most difficult moments.

Stochastic calibration of a carbon nitridation model from plasma wind tunnel experiments using a Bayesian formulation

“…In Table 1 the measured quantities are the dynamic pressure P d , the wall temperature T w , the chamber static pressure P s , the recession rate ṡ and the local CN densities ρ CN estimated from spectra measurements locally resolved. The experimental uncertainties are assumed independently distributed Gaussian distributions N(µ, σ) with σ taken considering the bounds given on the measurements as 2σ level of confidence [14][15][16].…”

“…Besides the likelihood function, we need to define prior distributions for all the parameters in q. In the available literature for tests in the VKI Plasmatron and stagnation line plasma flows [14,15,39,40], we can only establish some bounds a priori. In a maximum entropy set-up [41] with this a priori information, the resulting prior distributions are uniform over the prescribed range.…”

A surrogate-based optimal likelihood function for the Bayesian calibration of catalytic recombination in atmospheric entry protection materials

“…We assess the role of the auxiliary material used for testing, referred until now as "reference" material. As extending the testing methodology to include three different materials is already a possibility [44], we study the information gain with this methodology with synthetic data. In this section, we discuss how choosing different auxiliary materials and performing experiments with more probes impact the outcome of the inference.…”

“…3), one fundamental uncertainty in the way of rebuilding the TPS catalytic behavior is the fact that the boundary layer edge conditions cannot be estimated accurately if the auxiliary material behavior is not a priori well-known, which is the present case. We explore an assessment of this argument by assuming a high catalytic material (more than the conventional reference of copper) which resembles the catalytic response of a hypothetical probe made with silver [44]. We devise synthetic data where the heat flux that the auxiliary material would measure is higher than the previously considered copper, while still keeping the same wall temperature.…”