Prediction of Solid Propellant Burning Rate Characteristics Using Machine Learning Techniques

When formulating a new solid propellant, one of the most important aspects its performance is burning rate's response to change in pressure. For this reason, it useful be able predict rate given propellant before formulation created such that experimental trade studies are minimized or reduced scale. While many theoretical and phenomenological models have been proposed explain various energetic material combustion, little work has made publicly available application machine learning predicting rates. To facilitate model creation, parameters for over 600 formulations collected into coherent data set. This utilizes large amount inform random forest (ML) prediction parameters. ML operates parameter space including ammonium perchlorate composite, plastic bonded high explosive, double-base propellants. The model's accuracy, adaptability, capabilities presented discussed. effects different materials on propellant's expected examined.