Tenders Are Invited For Machine Learning For Atmospheric Impact Of Launchers Expro+

Tenders are invited for Machine Learning for Atmospheric Impact of Launchers Expro+ Objective: To develop a Machine Learning (ML) model to simulate the interaction of launcher exhaust gas with all atmospheric layers, in the frame of LCA and complementary to test bench and existing measurements (i.e. FIREWALL activity).Description: In the frame of ESAs Green Agenda objectives, most ESAs missions have been tasked to perform Life Cycle Assessment (LCA). However, LCA in the frame of space transportation is lacking data, especially when it comes to understanding the interaction of exhaust gas and species formed through ablation process during re-entry in the different layers of the atmosphere. This activity targets to model the interaction of the exhaust gas with all layers of the atmosphere using Machine Learning (ML). The activity aims at investigating whether AIcan generate a reliable dataset capturing exhaust plume and atmospheric interactions, which would serve as a reference or comparisonpoint during measurement campaigns. Having access to such tools would significantly enhance the efficiency of analysis once exhaustplume measurements become available. This activity complements the ongoing efforts on LCA and FIREWALL activity. The FIREWALL (Facilitate Inquiry of Rocket Emission Impact with Atmosphere Lower Layers) activity aims to better understand the interaction between the engine exhaust plume and the different atmospheric layers. It results with the assessment of several in-situ and remote sensing technics, to prepare a plume measurement campaign. The modelling focuses on the following four observations vectors and their respective types of observation: - Satellite: OH, H2O, PMC, rocket contrails and greenhouse gases (CO, CH4); - Balloons and sounding rockets: gas concentration, Ozone and temperature variation; - Ground measurements: aerosols, debris, accurate gas concentration, etc.; - Aircraft: reactive NOx, CO2, ozone loss, CIO, aerosol and CH4. This activity encompasses the following tasks: - Perform a state-of-the-art analysis on the ML methods used for atmospheric sciences; - Identify the potential modelling use cases: relevant exhaust species and corresponding altitude; - Identify the available dataset for each potential use cases; - Select the most promising use cases,including the exhaust species and the altitudes as well as the preferred ML approach; - Train and test the model; - Validate the model with upcoming measurement campaigns (e.g., FIREWALL Phase 2) and/or existing available data (e.g. from satellite observations, test benches exhaust plumes); - Define recommendations for future, parallel or complementary development. Software shall be deliveredunder an ESA Software Community License, so that any individuals or entities within ESA Member States can access it and can provideupdate to the community of users. Read less Tender Link : https://esastar-publication-ext.sso.esa.int/ESATenderActions/filter/open

Refer document

FileName	File Description
Global Tender Document	d1383a96-c4bb-4c97-a48c-13160a4a228c.htm