论文部分内容阅读
采用人工神经网络进行了液体叠氮燃料密度预测。通过将已知的特征分子结构编码作为输入参数,设计了神经网络代码,并得到了预测的密度数据。结果表明,密度预估值和文献值相比偏差为–1.8%~2.69%,人工神经网络对液体叠氮燃料的密度预测结果具有一定参考价值。
Artificial neural network was used to predict the density of liquid azide fuel. By coding known molecular structure as input parameter, the neural network code is designed and the predicted density data is obtained. The results show that the deviation of density estimation from the literature value is -1.8% -2.69%. Artificial neural network is of some reference value for the density prediction of liquid azide fuel.