近红外光谱法快速测定速生桉木材性的研究  

Study on Rapid Prediction of Wood Properties of the Fast-growing Eucalyptus by Near Infrared Spectroscopy Technique

作　　者：吴珽 房桂干[1] 梁龙[1] 崔宏辉 邓拥军[1] 

机构地区：[1]中国林业科学研究院林产化学工业研究所江苏省生物质能源与材料重点实验室国家林业局林产化学工程重点开放性实验室生物质化学利用国家工程实验室,江苏南京210042

出　　处：《造纸科学与技术》2016年第1期72-76,共5页Paper Science & Technology

基　　金：国家林业局948项目(2014-4-31)

摘　　要：用常规方法测定了速生桉木样品的化学成分和水分含量。运用近红外光谱法结合偏最小二乘法（PLS）,建立桉木样品综纤维素、木素、苯醇抽出物、水分含量的校正模型。四个校正模型的决定系数R2值分别为0.9631、0.9792、0.9894、0.9992;校正标准偏差RMSEC分别为0.0852%、0.0993%、0.0717%、0.2615%;交叉验证过程中的校正标准偏差RMSECV值分别为0.1097%、0.1211%、0.0831%、0.3395%,模型质量好。校正模型用于独立验证的偏差范围分别为-0.40%~0.32%、-0.52%~0.54%、-0.22%~0.21%、-1.03%~0.99%。结果表明,近红外光谱法可以对速生桉木的材性进行快速测定。A study was conducted to use the traditional method to analyze chemical composition and moisture content of the fast- growing eucalyptus samples. Then the near- infrared（ NIR） spectroscopy technique combined with partial least squares（ PLS） were used to build the calibration models of eucalyptus samples for holocellulose,lignin,benzene ethanol extractive and moisture content. The results show good performance of the four calibration models for the coefficient of determination（ R2） are 0. 9631,0. 9792,0. 9894,0. 9992,while the RMSEC values are 0. 0852%,0. 0993%,0. 0717%,0. 2615% and the RMSECV values are 0. 1097%,0. 1211%,0. 0831%,0. 3395%. The deviation ranges of the four calibration models are- 0. 40% ~ 0. 32%,- 0. 52% ~ 0. 54%,- 0. 22% ~ 0. 21%,- 1. 03% ~ 0. 99% in the independent verification. The results from this study show that the wood properties of the fast- growing eucalyptus and can be rapidly predicted by NIR spectroscopy technique.

关 键 词：近红外光谱法 速生桉木 偏最小二乘法 化学成分 水分含量 

分 类 号：O657.3[理学—分析化学] TS721[理学—化学]