Machine Learning in Combustion: Optimization of Fuel Combustion of Rockets

Abstract

Application of Machine Learning (ML) Techniques in research has seen a drastic increase in past few years and the main aspect of it in the rocket propulsion is to understand the combustion process and finding various methos to optimize it. This paper provides an overview of the relationship between machine learning and combustion with a specific focus on optimizing the rocket fuel combustion. The introduction presents an overview of Combustion and various ways which the ML is associated with it. Subsequently, the paper also discusses about the various ML algorithms which extends its discussions with the supervised, unsupervised and semi-supervised learning techniques and some of its types. An overview of different types of rocket engine is presented for the understanding of the characteristics, advantages and disadvantages of the commonly used rocket engines such as solid, liquid and hybrid propellant ones. Focusing on rocket fuel combustion, the discussion extends to various methods of optimization of the combustion process. Finally, the paper presents comprehensive results and discussions derived from the studies conducted on rocket fuel combustion optimization.