复合固体推进剂燃烧性能模拟计算的神经网络方法

A NEURAL NETWORK FOR MODELING CALCULATIONS FOR COMPOSITE PROPELLANTS

在总结已有燃烧模型的基础上，重点考虑压强、氧化剂的重均粒径、氧化剂的质量浓度三种主要影响因素，提出了一种基于误差反传（ＢＰ）神经网络的复合固体推进剂燃烧性能模拟计算方法，计算结果和实验值吻合较好，这为推进剂配方的计算机辅助设计提供了一种新方法。

he traditional modeling of composite propellant combustion, which strives to representthe combustion process in mathematical terms, is restricted by the knowledge of the combustionmechanism. In this paper, a new scheme has been proposed to calculate the combustion of compositepropellants without considering the specific combustion processes by applying the back-propagation(BP) neural network, which has the capability to 'learn' system characteristics through nonlinearmapping. The computed results are in good agreement with the experimental data, which shows thatthe scheme offers a new way for CAD of propellant formulation.