Artificial Neural Network Simulation of Average Particle Size for Preparing Energetic Materials in Spray Drying
- 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:为了提高喷雾干燥制备含能材料的实验效率,采用前馈反向传播神经网络(FFBPNN)、串级正反向传播神经网络(CFBPNN)、埃尔曼正反向传播神经网络(EFBPNN)、递归神经网络(LR)和非线性自回归神经网络(NARX)5种类型的神经网络对含能材料平均粒径进行预测,结果表明LR预测性能最优;并采用Levenberg-Marquardt(L-M)算法、动量梯度下降和自适应学习率算法(GDX)、遗传算法(GA)和粒子群算法(PSO)4种算法对LR进行优化,结果表明GA-LR的计算精度最高,更适用于对喷雾干燥制备含能材料的平均粒径的预测。
Abstract:In order to improve the experimental efficiency of preparing energetic materials in spray drying, different types of artificial neural networks were used to predict the average particle size, such as Feed-forward back propagation neural network (FFBPNN), Cascade-forward back propagation neural network (CFBPNN), Elman-forward back propagation neural network (EFBPNN), Layers Recurrent neural network (LR) and Nonlinear autoregressive neural network (NARX), of which LR has the best performance. Levenberg-Marquardt (L-M) algorithm, Momentum gradient descent and adaptive learning rate algorithm (GDX), Genetic algorithm (GA) and Particle swarm optimization (PSO) were used to optimize LR. The results show that GA-LR has the highest calculation accuracy and is more suitable for predicting the average particle size of energetic materials prepared by spray drying.
var swiper = new Swiper('.swiper_xq', {
navigation: {
nextEl: '.swiper-button-next',
prevEl: '.swiper-button-prev',
},
});
$(function(){
$('img').bigic();
});
jQuery(".slideTxtBox").slide({ trigger: "click" });