Prediction for the Burning Rate of Igniting Mixture by Artificial Neuron Network
- 摘要
|
- 图/表
|
- 访问统计
|
- 参考文献
|
- 相似文献
|
- 引证文献
|
- 资源附件
|
- 文章评论
摘要:以某混合点火药各组分含量作为药剂性能描述，利用误差反传神经网络（BP网络）算法，通过对9个配方药剂的训练建立了燃速与组成之间的定量关系模型，并对另外9个配方混合点火药的燃速进行了预测。结果表明，模型很好地反映了配比与燃速之间的关系，预测值与实际测量值比较接近，相对误差小于12%。该方法为混合药剂的研究和开发提供了一条新的途径。
Abstract:A computational paradigm is presented for making rapid and accurate estimations of burning rates for igniting mixtures by back-propagation neuron networks. Quantitative relational model between burning rate and combination has been established by training for 9 mixtures. Prediction has also been conducted for the other 9 mixtures. The results show that the neuron network is capable of efficiently formulating the correlations necessary to make accurate predictions and the prediction error less than 12%. This method has been proven to be efficient for the development of igniting mixture.
var swiper = new Swiper('.swiper_xq', {
navigation: {
nextEl: '.swiper-button-next',
prevEl: '.swiper-button-prev',
},
});
$(function(){
$('img').bigic();
});
jQuery(".slideTxtBox").slide({ trigger: "click" });
