Plume diagnostics for electric propulsion systems

Current modelling and simulations require either generic assumptions to be made for fluid dynamic based modelling leading to inaccuracies between modelled and experimental data or, intense computational recourses for limited duration yet highly accurate particle-in-cell (PIC) modelling. Kingston University has developed a simulation model that narrows the gap between these two simulation regimes by harnessing and advancing the latest developments in AI Machine Learning.

The code developed (called Persius) runs approximately 20x faster than traditional (PIC) models and can scale bridging the gap between micro and macro regimes of simulation. The code needs to be verified and further refined with user cases against experimental data sets. The aim of this project is to collect data from plasma and combustion-based propulsion systems and analyses the effectiveness of Persius for the space launch and propulsion sectors and wider aerospace industry.

Student: Simonas Brasas

University Partner: Kingston University

Industry Partner: Pulsar Fusion