| BP神经网络与近红外光谱定量预测杉木中的综纤维素、木质素、微纤丝角 |
| |
| 引用本文： | 丁丽,相玉红,黄安民,张卓勇.BP神经网络与近红外光谱定量预测杉木中的综纤维素、木质素、微纤丝角[J].光谱学与光谱分析,2009,29(7):1784-1787. |
| |
| 作者姓名： | 丁丽 相玉红 黄安民 张卓勇 |
| |
| 作者单位： | 1. 首都师范大学化学系,北京,100037
2. 中国林业科学院木材研究所,北京,100091 |
| |
| 基金项目： | 北京市教育委员会科技发展计划，国家科技支撑计划，中央级公益性科研院所基本科研业务费专项基金 |
| |
| 摘 要： | 利用近红外光谱（NIR）技术结合BP神经网络定量预测了杉木中的综纤维素、木质素和微纤丝角。首先对杉木的原始近红外光谱数据进行卷积（Savitzky-Golay）平滑和二阶导数处理，然后利用小波变换压缩，将由171个数据点组成的近红外光谱压缩为86个数据点，最后用BP神经网络建模，采用Leave-n-out交叉验证法对模型进行验证，并讨论了隐含层神经元个数、学习速率、动量因子和学习次数对所建BP网络的影响。用所建的网络模型预测了测试集中杉木样本的综纤维素、木质素和微纤丝角 ，预测的相关系数R2值分别为0.91, 0.90, 0.87，预测均方根误差RMSEP分别为：0.86%, 0.33%, 4.99%。结果表明该方法快速，无损，基本能满足定量分析的要求。
|
| 关 键 词： | 近红外光谱 BP神经网络 综纤维素 木质素 微纤丝角 |
| 收稿时间： | 2008/3/19 |
Quantitative Prediction of Holocellulose,Lignin,and Microfibril Angle of Chinese Fir by BP-ANN and NIR Spectrometry |
| |
| DING Li,XIANG Yu-hong,HUANG An-min,ZHANG Zhuo-yong.Quantitative Prediction of Holocellulose,Lignin,and Microfibril Angle of Chinese Fir by BP-ANN and NIR Spectrometry[J].Spectroscopy and Spectral Analysis,2009,29(7):1784-1787. |
| |
| Authors: | DING Li XIANG Yu-hong HUANG An-min ZHANG Zhuo-yong |
| |
| Institution: | DING Li1,XIANG Yu-hong1,HUANG An-min2,ZHANG Zhuo-yong1 1.Department of Chemistry,Capital Normal University,Beijing 100037,China 2.Chinese Academy Forestry,Beijing 100091,China |
| |
| Abstract: | The amount of holocellulose,lignin,and microfibril angle of Chinese fir was predicted by using back-propagation neural network(BP-ANN) combined with near infrared(NIR) spectrometry.First,the data of original spectra were pretreated by Savitzky-Golay smoothing algorithm and the second derivative,then the data of near infrared spectrometry with 171 points were compressed to 86 points by using wavelet transform,and finally,the models were established by using BP-ANN.The models were validated using leave-n-out ... |
| |
| Keywords: | Near infrared spectroscopy BP-ANN Holocellulose Lignin Microfibril angle |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
