Prediction the Detonating Velocity of Elemental Explosives by Support Vector Machine

Aiming at the better explosive performance prediction for new energetic materials,the support vector machine( SVM) method was applied to establish a predictive model for the detonation velocity of elemental explosive. The relative molecular mass of explosives( M),loading density( ρ),and oxygen balance the percentage composition of carbon,hydrogen,oxygen,nitrogen( wC、wH、wO、wN),were selected as inputs of the support vector machine models. The results of the model indicate that,the prediction results are in good agreement with the experimental values. The most relative error is 4. 79% and the least is 0. 04%. The prediction accuracy of the model is good,enabling a certain extrapolation. This method can also provide guidance to predict other detonation parameters of explosives,and has practical applications.