Secondments

The investigation of Health Monitoring (HM) methodologies is pivotal to advance the TRL of critical technologies for Reusable Launch Vehicles (RLVs). Among the functionalities of HM, anomaly detec-tion is the first step for promoting the enhancement of the reusability. Given the innovativeness of the topic, there is not clearly accepted SOTA and consequently, defining the performances of the al-gorithms for health analysis is crucial. In this sense, cooperation with DLR Lampoldshausen could be beneficial for both institutions in investigating the frontier of anomaly detection in a joint approach.
In detail, the activities include design and testing of algorithms for anomaly detection based on unsu-pervised machine learning approaches. Promising algorithms will be used as a basis for the analysis and modified for fitting the available data.
Final objective of this activity is to test anomaly detection algorithms on simulation data coming from liquid propellant engines.

The efficiency of reuse depends directly on the concept for RLV recovery, including the recovery strategy (propelled vertical landing scenarios including flow conditions derived from trajectory analysis). For this purpose, the Secondment in DEIMOS Space (Madrid) offers the opportunity to tailor these technologies to the problem of Missionization for a reference reusable launch vehicle.

In detail, the activities include numerical studies on atmospheric descent and subsonic retro-propulsion for the aerodynamic characterization of this vehicle in both conventional and advanced configurations. Additionally, this comes together with the definition of boundary conditions for the recovery in collaboration with ESR 14 and 15. More insights at system-integration level and impact on the overall launcher architecture follow, in collaboration with ESR 3 and 4.

Final objective of this activity is to deliver a method of analysis for main-stage recovery with advanced nozzles that aims to prove the applicability of these technologies to the future class of reusable launch vehicles.

Aerothermodynamics of Reusable winged first stages, ONERA, Toulouse, France

Reusable winged first stages attempt to minimize re-entry loads through the use of aerodynamic surfaces and high angles of attack to slowly decelerate higher in the atmosphere. A trade-off therefore exists between the flight trajectories, although mostly determined by the separation conditions, and the aerodynamic and aerothermal loads which the vehicle structure must withstand. To account for these at early design phase where computational power is limited, it is necessary to use fast methods to predict the corresponding aerothermodynamic loads at the critical vehicle locations and flight conditions. The secondment was focused on reviewing aerothermodynamics of first stages and performing analysis with basic design tools of the Spaceliner booster first stage.

The Uncertain lifespan of a space system, University of Strathclyde, Glasgow, UK

Space sustainability is an emerging area of research, especially considering the forecasted increase in launch rates, that combines the sustainable use of space with the sustainable use of resources on Earth to develop space systems. The latter can be enabled by the eco-design approach which evaluates the environmental impact of the whole life cycle of space systems. The evaluation of the environmental impact is currently performed using life cycle assessment (LCA) tools that heavily rely on databases collecting information on materials, processes and the associated generation of pollutants or exploitation of natural resources. Thus, a proper decision based on the value of the indicators needs to account for this uncertainty, which, in nature, is most often mixed with both epistemic and aleatory ones. This project aims at using modern Imprecise Probability Theories to define an uncertainty model for the Life Cycle Assessment of space systems. The ultimate goal of this project is to establish for the first time a proper framework for robust LCA (RLCA) that enables robust decision-making, enabling applications as the robust sustainable design of reusable launch vehicles when combined with state of art system engineering approaches as MDAO.

DLR, Lampoldshausen

For my PhD program there are two secondments foreseen. First of all, I already conducted a secondment at the German Aerospace (DLR) in Lampoldshausen from March to Mai 2022. During this secondment I had several tasks. One of my tasks was the assistance in Cold flow testing of flexible small scale rocket nozzles (TIC) to investigate oscillations and damping capabilities. However, this is a still ongoing investigation by DLR and served in the first place to get first practical hands-on experience of nozzle testing. This experience then helped me during my collaboration on Cold flow ED nozzles testing under vacuum condition with ASCenSIon colleague Giuseppe Scarlatella at Technical University in Dresden. The second task I had at DLR Lampoldshausen was conducting numerical simulations with the DLR inhouse solver DLR TAU Code that I could then also use for verification of the own numerical methods.

AVIO, Colleferro

The last planned secondment that was not yet conducted is in cooperation with AVIO in Colleferro. During this collaboration it is planned to concentrate more on the system aspects and system integration of the ED nozzles.

Within this research project, Melissa Lantelme spent three months in 2021 at ArianeGroup SAS in Les Mureaux, France, to perform calculations on the THEMIS demonstrator with the FLUSEPA code and a further three months in 2022 at the German Aerospace Center (DLR) in Göttingen to perform calculations on the SpaceLiner and RETALT2 configurations with the TAU code. Those secondments provided the option to study further reusable launch vehicles and create a heterogenous database necessary for the development of the surrogate model.

ULB, Brussels

During my PhD, I worked for several months together with Riccardo Gelain (ESR 9) at the Université Libre de Bruxelles in 2022. We conducted the first big experimental test campaign on the use of steps in HREs there using two different motors. Together with the motor at ONERA, this means we tested the idea of stepped geometries on three different motors. In the picture below, you can see Riccardo and me preparing the rocket engine for an experiment.

OHB Systems, Bremen

Finally, in order to collect all the findings of the experimental and numerical study into a complete design, I will spend a short period in the propulsion department of OHB System in Bremen in order to investigate the possible design solutions on a system level. Because all research aside – in the end we do not only academic ideas, we need solutions that can be translated to real-world applications.

DLR, Lampoldshausen

I did a first secondment at the German Aerospace Center (DLR) in Lampoldshausen, where I studied how to improve the quality of the combustion visualization in hybrid rocket slab burners and implemented a method to measure the regression rate of the fuel from the high-speed videos.

ONERA, Toulouse

I was then hosted by the French Aerospace Lab (ONERA) in Toulouse, where I improved some ballistics reconstruction techniques to model the burning process of the fuel grain over time from the acquired experimental data.

ArianeGroup, Bremen

During the six months secondment performed at ArianeGroup Bremen, several activities were carried out in the frame of the following projects:

• A6 ULPM 
• FLPP LunaNova (LN) kick stage:
  • Tradeoff assessments at system and subsystem level (propulsion)
  • LN functional analysis in Capella

Università di Pisa, Pisa

The Aerospace department of the University of Pisa has great experience in the development of turbomachinery. Relaying on this expertise, a feasibility study for the development of a small centrifugal pump will be carried out. The design comprises several features that make this study particularly interesting. Centrifugal pumps in the space domain are traditionally developed for bigger applications and this fact poses several challenges regarding their efficiency when scaled down. Moreover, the designed pump shall be compatible with highly concentrated hydrogen peroxide, fact that leads to additional constraints in terms of material compatibility and hence design complexity.

I have had the opportunity to do two secondments during my project.

DLR, Lampoldshausen

I worked for 6 months at the German Aerospace Agency, DLR, where I have taken part to experimental campaigns, and I have learnt the testing procedures and analysis.

D-Orbit, Fino Mornasco

I worked for 4 months at D-Orbit, a company focused on systems with the final purpose of in-orbit servicing.

During the ongoing secondment at the Space Debris Office at ESA, the research is focused on the PMD of rocket bodies. The access to specialized tools and the possibility to discuss about my research and ideas with experts in this field have been of incredible value for my Individual Research Project (IRP). The details on the outcome will come at the end of this secondment!

OHB, Bremen

Related to the industrial interest of the PhD topic, especially regarding possible missions of interest and realistic definition of them, it was decided to perform a secondment at OHB in Bremen, Germany, within the Mission Analysis Department. The scope was to identify industrial demands within the mission definition to be used as case-scenarios, and a preliminary analysis of the optimization tool using these. The outcome of this secondment was a publication in the 73rd International Astronautical Congress in Paris, in which a specific case of an industrially interesting and realistic multi-payload multi-orbit delivery is analysed.

Deimos, Madrid

In the more technical part of the thesis, the experience of Deimos in terms of trajectory definition was interesting in order to help the development of the more complex and detailed trajectory optimization related to the multi-payload multi-orbit injection problem. As such, the output preliminary mixed-integer nonlinear optimization problem was to be used as first guess for a more precise nonlinear optimal control solver which can provide with a reference trajectory to the system. The help of the company has been crucial to identify the problems and characteristics related to developing a more complex optimizer. The output is a tool that allows not only to have the specific trajectory and control law necessary to achieve all the transfers, but to also include different perturbances and dynamic models of different accuracies to study the sensitivity of the solution to changes in the problem framework.

Abstract about the secondment in ONERA, the French Aerospace Lab, Toulouse, France (from 10/01/2022 to 08/04/2022 – finished):

The aim of this first secondment was to join the Department of Multi-Physics for Energy at ONERA towards the development of an accurate aerodynamic model in the RLV re-entry dynamics simulator. To carry out this research, the following aerodynamic codes have been analysed and then used: open access S/HABP (Supersonic/Hypersonic Arbitrary Body Program), and ONERA ARES (Atmospheric Re-Entry Software).

Therefore, the aerodynamic model of the simulator was improved by developing an aerodynamic database which allows to obtain the aerodynamic coefficients and the center of pressure as function of the Mach number and the aerodynamic angles (angle of attack, sideslip angle) using S/HABP program. To do that, the geometry of the first-stage rocket is modelled as a cylinder which is meshed by the program. Then, the aerodynamic parameters are computed by calculating the pressure coefficient of each element using the modified Newtonian method. Using this routine, several simulations have been carried out for a first-stage vehicle configuration with Mach number going from Mach 5 to Mach 0.8 and angles of attack ranging from 0 to 180 deg. This aerodynamic database was compared with available data in the literature and the existing uncertainties quantified. Robustness and accuracy of S/HABP solver were compared to ARES solution and validated.

The outcomes of this secondment are summarised in the IAC conference paper [1]. Once the database implemented in the simulator, the guidance method was enhanced to consider the varying effects. Numerical results have shown that the G&C methods implemented well managed to consider the aerodynamic effects and therefore to enable an atmospheric re-entry and pinpoint landing. However, adding uncertainties on the aerodynamic parameters points out that the robustness of the simulator must still be improved. Steerable planar fins and a more robust control method such as H-infinity would be a good candidate solution.

[1] Alice De Oliveira and Michèle Lavagna. “Assessment of Reusable Launch Vehicles Re-entry Dynamics Control Effectiveness with Enhanced Aerodynamics Modelling”. Proceedings of the 72nd International Astronautical Congress (IAC), Paris, France, 18-22 September 2022.

DLR, Bremen

During the period at DLR, the work is focused on the development of a tool which aims at estimating the aerodynamic coefficients of a re-entry vehicle by interpolating multiple aerodynamic database (AEDB) depending on some geometrical parameters of the vehicle itself. The tool developed within this period aims at estimating the longitudinal aerodynamic coefficients and building a representative database for a wide range of vehicles that can be used for the preliminary phases of the design of a re-entry mission. The tool is based on multiple AEDBs of several vehicles which are interpolated onto the geometrical features space by means of interpolating functions. The number of interpolating functions is defined by categorizing various re-entry vehicles. These can be subdivided into four categories: Winged Re-entry Vehicles (WRV), like the Space Shuttle Orbiter, Winged Lifting Body (WLB), such as the X38, No-Winged Lifting Body (LB), for instance Space Rider, and Vertical Landing-Reusable Launch Vehicle (VL-RLV), like the New Glenn first stage. The AEDBs are built by exploiting the baseline vehicles for the different categories by varying the determined geometrical parameters. The computation is done with semi-empirical and simplified-theories methods implemented in the CAC and HOTSOSE tools provided by the German Aerospace Center DLR. This method aims at achieving a representative AEDB of a vehicle taking into account a limited number of parameters, that can be used in the preliminary phases of the mission design and for multidisciplinary analysis. The results obtained are validated against the data computed with more accurate methods available in literature and their reliability is critically discussed.  The tool developed within this research will be part of the missionisation tool for re-entry vehicles.

Politecnico di Milano, Milan

During the period at Politecnico di Milano, the work is focused on the development and implementation of the missionisation layer of the missionisation tool. The missionisation layer is constituted by an optimization routine that exploits the Multidisciplinary Design Analysis (MDA) framework for the optimal tuning of the design parameters both for a missionized solution and for evaluating the mission capabilities during the preliminary mission design phase. The missionisation layer implies Multidisciplinary Design Optimization (MDO) techniques to solve the problem. By using this methodology, the goal is to evaluate different design options to perform trade-off studies and identify the optimal design, especially when a multi-objectives approach is considered and nondominated optimal Pareto solutions are obtained. Within the MDO process, indeed, the solution space domain is explored through the variation of the design parameters. Several MDO approaches have been studied, from the classical methods, such as Multi-Objectives Particle Swarm Optimization and Sequential Quadratic Programming, to more novel Metamodel-based techniques. These last methodologies are promising for reducing the computational cost related when expensive evaluations of the MDA are needed. Both solutions have been analysed and traded off in terms of accuracy and computational effort to select the most suitable methodology.