Rich Propellant Fixed Static Burning Rate

The burning rate of the fuel-rich propellants is the parameter which is important for Ducted Rocket design and performance analysis and is main inspection criterions of propellant prescriptions. Therefore, it is a focus of researchers to predict the dynamic burning rate with the static burning rate by doing research on the correlation of the dynamic burning rate and the static burning rate of the same propellant.This paper focuses on the correlation of static burning rate and dynamic burning rate of the fuel-rich propellant. To get rid of analyzing various factors on burning rate, combustion processes is not considered. In this paper, the NN and genetic algorithm are applied together, based on both of the NN (Neural Network) which has advantage in nonlinear function mapping and genetic algorithm which has full field research ability in artificial intelligence fields. By above method, optimizations for the primary weights values and threshold values are done, the GA-BP algorithm is designed and analysis models for the correlation of the dynamic burning rate and static burning rate are made. Appliance shows that the method is steady in performance, good in reliability and strong generalization property, the prediction of the Network model is more accurate than the prediction of the linearity regression, and the convergence rate of the Network model is related with the correlative coefficient, but the accuracy is not sensitive to the correlative coefficient. Therefore, this method is feasible in the research of the dynamic burning rate and static burning rate relation.The method developed in this paper can be used in the research field of the Ducted Rocket and fuel-rich propellant, to reduce the experimental cost effetely and shorten the research period greatly.