当前位置: 首页 > 文章 > 基于近红外光谱分析的多模型建模方法研究 林业科技 2014,39 (2) 24-28
Position: Home > Articles > Research on Multi-model Modeling Method using Near Infrared Spectral Analysis Forestry Science & Technology 2014,39 (2) 24-28

基于近红外光谱分析的多模型建模方法研究

作  者:
刘胜;范雅婷
单  位:
北京林业大学理学院
关键词:
近红外光谱;多模型方法;相思树
摘  要:
以相思树的α-纤维素含量为研究对象,用一种多模型方法建立了相思树α-纤维素含量的近红外光谱分析模型。模型预测值的平均相对误差为0.97%,实验值与预测值之间的相关系数为0.963 1,模型的拟合优度为0.924 5。研究结果表明,使用的光谱数据量越大,模型的预测效果一般会越好。此外还发现了子模型中待定常数的个数与所使用光谱数据量之间的关系:建模时使用的光谱数据量越大,每个子模型中待定常数的个数一般应该越小。该结果有助于今后使用该方法建立其它近红外光谱分析模型。所建模型可用于快速测定相思树的α-纤维素含量,并有望用于其它树种某些化学成分含量的预测。
译  名:
Research on Multi-model Modeling Method using Near Infrared Spectral Analysis
作  者:
LIU Sheng;Beijing Forestry University;
关键词:
Near infrared spectroscopy;;Multi-model method;;Acacia confusa
摘  要:
Based on the research object of α- cellulose content of Acacia confusa,the near infrared spectral analysis model of α- cellulose content of Acacia confusa was built by a multi- model method. The mean relative error of the prediction value of the model was 0. 97%. The correlation coefficient between the experiment values and the predicted values was 0. 9631. The goodness of fit of the model was 0. 9245. The results show that the greater the amount of the spectral data that were used,the better the prediction effect of the model. Besides,the relationship between the number of undetermined constants of sub models and the amount of the spectral data by used was found: The greater the amount of the spectral data used in modeling process,the smaller the number of undetermined constants of each sub model usually. These results is helpful for constructing other near infrared spectral analysis models by this method. The model constructed in the paper can be used to determine the alpha cellulose content of Acacia confusa rapidly. It is possible that the method of modeling in the paper can be used to predict the contents of some chemical components of other kind of trees.

相似文章

计量
文章访问数: 10
HTML全文浏览量: 0
PDF下载量: 0

所属期刊

推荐期刊