乳化粉状炸药制备的优选BP网络模型  

A BP net model of optimizing selection for preparation of powdery emulsion explosives

作　　者：段宝福[1] 汪旭光[2] 宋锦泉[2] 

机构地区：[1]辽宁工程技术大学工程爆破研究所 [2]北京矿冶研究总院炸药与爆破研究所,北京100044

出　　处：《爆炸与冲击》2003年第5期430-435,共6页Explosion and Shock Waves

基　　金：国家自然科学基金项目(50174008)

摘　　要：根据BP(Backpropgation)神经网络理论,建立了制备乳化粉状炸药模型,采用乳化粉状炸药研制过程中的有关数据,编写程序进行了模型的训练、用训练后的模型对预测样本进行了预测与优选。模型优选结果的性能价格比的预测值为2841.3(m/s)/($/kg),实验得到该结果的实际性能价格比为2767.4(m/s)/($/kg),相对误差为2.7%。模型还可对具有相同配方的炸药优选制备工艺参数,对部分组分不同的样品优选了工艺参数,模型对这些样品优选的最优工艺参数与实际工艺参数基本一致。An optimizing selection model for preparation of PEE(Powdery Emulsion Explosives) is proposed based on the BP(Back Propagation) net theory. The model has been trained using the data from PEE manufacturing practice. Then it is used for optimizing test samples. The optimized value of the capabilityto cost ratio is about 2841.3 (m/s)/($/kg), with a relative error of only 2.7% compared with the actual value, i.e., 2767.4 (m/s)/($/kg). The model can also be used to optimize the manufacturing parameters for explosives of the same prescription. The optimized parameters are very close to those from practice, demonstrating the capability of the model.

关 键 词：爆炸力学 优选模型 神经网络理论 乳化粉状炸药 制备 BP网络模型 

分 类 号：TQ564[化学工程—炸药化工] TQ560.6