Anabel del Val

Research activities

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

Anabel del Val, Olivier P. Le Maître, Olivier Chazot, Thierry E. Magin, Pietro M. Congedo

  • Engineers still grapple with understanding the physical phenomena involved during the atmospheric entry of space capsules. So far, this severe lack of knowledge has been overcome in practice by over-doing safety measures, not too engineering-like if you ask me. Since the 60s, the community has been trying to gather knowledge on what happens to the space vehicle’s surface when it enters dense planetary atmospheres. We still don’t have a comprehensive answer but we are better suited to understand it. The problem is that many things happen at the same time. It is difficult (at least with current technology) to test separately each little phenomenon occurring during atmospheric entry or even to design tests where you isolate one effect: thermal behavior of the gas, chemistry, etc.

    When we test to learn about different phenomena, we can only measure quantities that are a final effect of a bunch of different things. If you want to learn about chemical reactions in the gas, you have to assume many things about that gas, the experimental data is not enough. This way, you will always get partial understanding and highly dependent on the assumptions made and conditions tested. The work I present here contains 3 different contributions for the community:

    1) The 2 most immediate causes that could explain the experiments on reusable protection materials are considered not known at the same time, even though we only want to learn about one of them. We need this in order not to bias the result with a hard assumption.

    2) The cause we are not interested in (contribution of the convective heat flux to the overall heat flux the protection material experiences) is limited to its most likely values thanks to the mathematical framework we developed, reducing the uncertainty on the cause we are interested in (contribution of the diffusive heat flux).

    3) We show that it’s possible to improve the experiments in a plasma wind tunnel to increase our certainty about the value of the parameter of the cause we are interested in (catalytic efficiency), vital for the community as it’s embedded in all CFD simulations of this kind.

    This contribution is only an idea in a very promising research direction that we are exploring in my PhD. Stay tuned!

Experimental methodology for the accurate stochastic calibration of catalytic recombination affecting reusable spacecraft thermal protection systems.

Anabel del Val, Diana Luís, Olivier Chazot

  • In the previous work, we asked ourselves the question: can we find the needle in this huge haystack? The needle being catalytic recombination parameters and the haystack being all the different physical processes affecting the heat flux experienced by the material exposed to the plasma flow. We investigated a Bayesian methodology to be able to confidently pinpoint the effect of catalytic recombination on the observations (heat flux). Now we pose the following question: can we increase the confidence in knowing we found the good needle by changing the haystack where the needle is hidden? In other not-so-metaphoric words, can we condition the physics of the experiments so much that we can be very certain about the parameters obtained? To do this, we looked at experimental methodologies that add certainty and constraints to our models of choice together with Bayesian analysis tools.

    The work I present here contains 2 different contributions for the community:

    1) We guide the design of the experimental methodology with stochastic analyses, looking for reduction of uncertainty on the sought out parameters. This is a first in the aerothermodynamics literature where deterministic methods are still used to solve inverse problems. The deterministic nature of the problem prevents the evaluation of uncertainty on the results and, as a consequence, the information coming from the experiments cannot be properly assessed, resulting in experimental cases that bring little or no information to the catalytic parameters. In other words, the haystack is so large that anything can resemble the needle we are looking for.

    2) This work delivers the first catalysis database with accurate uncertainty estimates in the aerothermodynamics literature. Furthermore, it guides the production of high-quality wind tunnel data for validation purposes by greatly increasing the certainty on the computed parameters such that we do not need to make assumptions about them. The more informative our experiments are, the more useful are the data to increase the confidence in our tools or, conversely, invalidate our models, finding the limits of validity with high certainty.

List of publications

Book chapters

  • A. del Val, O. Chazot, T. E. Magin and P. M. Congedo (2020). A Bayesian perspective on TPS catalysis phenomena: learning from experiments and proposing new ones, Review of the VKI Doctoral Research 2019-2020, Rhode-St-Genese, Belgium.

  • A. del Val, O. Chazot and T. E. Magin. Uncertainty Treatment Applications: High-Enthalpy Flow Ground Testing, In Optimization Under Uncertainty with Applications to Aerospace Engineering, Ed. M. Vasile, Springer Nature 2020

  • A. del Val, O. Chazot, T. E. Magin and P. M. Congedo (2019). Towards a Systematic Framework for the Design of Experiments: Application to Catalytic Phenomena in Plasma Wind Tunnels, Review of the VKI Doctoral Research 2018-2019, Rhode-St-Genese, Belgium.

  • A. del Val, O. Chazot, T. E. Magin and P. M. Congedo (2018). Stochastic Inference of the Catalytic Properties of Thermal Protection Materials from Plasma Wind Tunnel Experiments, Review of the VKI Doctoral Research 2017-2018, Rhode-St-Genese, Belgium.

Technical notes

  • A. del Val, T. E. Magin, O. Chazot. Uncertainty Assessment on Experimental Data, ESA TRP Characterization of High Enthalpy Facilities and Streamlining of Calibration and Tests (CHEF)

Conference proceedings

  • A. del Val, O. P. Le Maître, O. Chazot, P. M. Congedo and T. E. Magin. Inference methods for gas/surface interaction models: from deterministic approaches to Bayesian techniques. Uncertainty Quantification & Optimization 2020 conference proceedings

  • D. Luis, A. del Val and O. Chazot. Characterization under Uncertainty of Catalytic Phenomena in Ceramic Matrix Composites Materials for Spacecraft Thermal Protection Systems, 8th European Conference for Aeronautics and Aerospace Sciences (EUCASS), Madrid, June 2019. DOI: 10.13009/EUCASS2019-257. Shortlisted for best student paper award

  • B. Arizmendi, T. Bellosta, A. del Val, G. Gori, J. Reis, M. Prazeres. On Real-time Management of On-board Ice Protection Systems by means of Machine Learning, AIAA AVIATION Forum, Dallas, June 2019

  • A. del Val, T. E. Magin, O. Chazot. Uncertainty Assessment on the Characterization of Testing Conditions in Arc-jet Facilities, 33rd AIAA Aerodynamic Measurement Technology and Ground Testing Conference, AIAA AVIATION Forum, Denver, June 2017

  • A. del Val, T. E. Magin, B. Dias, O. Chazot. Characterization of Ground Testing Conditions in High Enthalpy and Plasma Wind Tunnels for Aerospace Missions, 8th European Symposium on Aerothermodynamics for Space Vehicles, Lisbon, March 2015

Journal articles

  • A. del Val, O. P. Le Maître, O. Chazot, T. E. Magin, P. M. Congedo. A surrogate-based optimal likelihood function for the Bayesian calibration of catalytic recombination in atmospheric entry protection materials. Applied Mathematical Modelling (in press)

  • A. del Val, D. Luís, O. Chazot. Experimental methodology for the accurate stochastic calibration of catalytic recombination affecting reusable spacecraft ther- mal protection systems. (Submitted to Experimental Thermal and Fluid Science)

  • A. del Val, O. P. Le Maître, P. M. Congedo, T. E. Magin. Inference of nitridation reaction efficiencies of graphite in nitrogen plasma flows using a Bayesian formulation. (To be submitted).

  • A. del Val, M. Capriati, O. P. Le Maître, P. M. Congedo, T. E. Magin. Assessment and selection of surface balance models through Bayesian evidence for the calibration of nitridation reaction efficiencies under model form uncertainties. (In preparation).

Theses

  • A. del Val. “Bayesian calibration and assessment of gas-surface interaction models and experiments for atmospheric entry plasmas”. PhD thesis. Institut Polytechnique de Paris and von Karman Institute for Fluid Dynamics, 2021.

  • A. del Val. “Characterization of Ground Testing Conditions in High Enthalpy and Plasma Wind Tunnels for Aerospace Missions”. Master thesis. VKI and Universidad Politécnica de Madrid, Madrid, Spain

International meetings, invited seminars, presentations, and posters

  • ESA ACT Science Coffee, European Space Agency Advanced Concepts Team, Online, March 2021: "Surviving the fire: how games of chance are helping us understand spacecraft atmospheric entry."

  • Uncertainty Quantification & Optimization Conference, Online, Nov 2020: "Inference methods for gas/surface interaction models: from deterministic approaches to Bayesian techniques."

  • Optimization in Space Engineering (OSE5), Ljubljana, November 2019: "On the Inference of Chemical Model Parameters for Tomorrow’s Space Journeys: an Overview on Reusable and Ablative Space Systems."

  • Stats4Grads, Durham University, Durham (UK), 6 Nov 2019: "Bayes goes to Space: inferring chemical model parameters for tomorrow's Space journeys."

  • International Congress on Industrial and Applied Mathematics (ICIAM), Valencia University, Valencia, Spain, 15-19 July 2019: "Bayesian Calibration of Gas/Surface Interaction Models for Thermal Protection Materials under Spacecraft Reentry Conditions."

  • 8th European Conference for Aeronautics and Aerospace Sciences (EUCASS), Madrid, June 2019: "Characterization under Uncertainty of Catalytic Phenomena in Ceramic Matrix Composites Materials for Spacecraft Thermal Protection Systems."

  • Uncertainty Quantification & Optimization Conference, Sorbonne University, Paris, France, 18-20 March 2019: "Robust calibration of the catalytic properties of thermal protection materials: Application to plasma wind tunnel experiments."

  • Fluid Mechanics Seminar Series, ETSIAE, UPM, Madrid, Spain, 1 Oct 2018: "Characterization of spacecraft reusable heatshield materials from plasma wind tunnel experiments: a Bayesian inference approach."

  • Entry Systems and Technology Division Seminar, NASA Ames Research Center, Moffet Field, CA, 30 August 2018: "Stochastic inference of the catalytic properties of thermal protection materials from plasma wind tunnel experiments."

  • 7th European Conference on Computational Fluid Dynamics (ECCOMAS), Glasgow, UK, 11-15 June 2018: "Stochastic inference of the catalytic properties of thermal protection materials from plasma wind tunnel experiments."

Involvement in contracts and projects

  • Team Member (VKI): “CHEF: Characterization of High Enthalpy Facilities and Streamlining of Calibration and Tests.” Task 7: Uncertainty Assessment on Experimental Data, ESA AO/1-7205/12/NL/CP.