Prediction of burning rate of HTPB propellant by using support vector regression

The performance of Hydroxyl-Terminated Polybutadiene (HTBP) propellants will be changed by filling with nano-structure catalysts. In this study, a novel regression approach, the support vector regression (SVR) approach combined with particle swarm optimization (PSO) was introduced to investigate the influence of the additives on burning-rate of HTBP. The SVR model was trained and tested with an experimental dataset of RDX/AP/Al/HTPB propellants containing nano-structure ns-Fe₂O₃ catalyst. The prediction performance of SVR model was compared with that of artificial neural network (ANN) model. The results demonstrate that the prediction ability of SVR is superior to that of ANN. This investigation indicates that SVR-based modeling is a practically useful tool in prediction of the burning-rate of HTPB affected by three different factors (mass fraction of ns-Fe₂O₃, mass fraction of HTPB and pressure).