欢迎访问《兵工学报》官方网站,今天是

兵工学报 ›› 2006, Vol. 27 ›› Issue (2): 244-247.

• 论文 • 上一篇    下一篇

选择性神经网络二次集成在火药近红外分析中的应用研究

施彦1,黄聪明2   

  1. 1.北京工商大学信息工程学院,北京100037;2.北京理工大学化工与环境学院,北京100081
  • 收稿日期:2005-07-08 上线日期:2014-12-25
  • 通讯作者: 施彦

The two-Level Selective Neural Network Ensembles Applied to Quantitative Analysis of Propellants

SHI Yan1, HUANG Cong-ming2   

  1. 1.College of Information, Beijing Technology and Business University, Beijing 100037, China; 2. School of Chemical Engineering and Environment, Beijing Institute of Technology, Beijing 100081, China
  • Received:2005-07-08 Online:2014-12-25
  • Contact: SHI Yan

摘要: 针对建立近红外光谱定量分析的神经网络校正模型时,存在变量数过多以及容易出现过拟合等问题,采用小波变换对近红外光谱进行预处理,用以消除噪声,减少变量个数;并在此基础上,提出一种新的神经网络校正模型一基于改进贪心法的选择性神经网络二次集成,来提高神经网络的泛化能力。实验结果表明:在建立火药近红外分析的校正模型中,该模型不仅建立过程简单而且具有较好的泛化能力。

关键词: 人工智能 , 近红外光谱 , 火药 , 小波变换 , 选择性神经网络二次集成

Abstract: During quantitative analysis of propellant based on near infrared spectroscopy (NIR),neural networks are good tools for establishing calibration model, but there are still some problems to be con?sidered such as too much variables and “ over-fitting ”. To solve these problems, wavelet transform (WT) was used to preprocess the NIR,and the two-level selective neural network ensembles based on greedy algorithm with first improvement strategy,were proposed. Based on the ensembles, a cali?bration model for the quantitative analysis of propellant was established. Experiment results show that the model is easy to be built and promotes the generalization ability of neural network system.

Key words: artificial intelligence , near infrared spectroscopy , propellant , wavelet transform , two-lev?el selective neural network ensembles

中图分类号: