粗皮桉木材力学性质的近红外光谱方法预测

Estimation of Wood Mechanical Properties of Eucalyptus pellita by Near Infrared Spectroscopy

以人工林粗皮桉木材为研究对象,采用常规力学测试方法和近红外光谱方法对其无疵小试样力学性质进行研究。用近红外光谱仪采集试样表面的近红外光谱,对采集的近红外漫反射光谱进行导数预处理并对不同波段光谱建立校正模型,以1/3试样作为预测集对校正模型进行验证。结果表明:二阶导数预处理、350～25000nm全光谱波段、径切面和弦切面平均光谱值对粗皮桉木材力学性质模型预测效果最好。抗弯弹性模量和抗弯强度、顺纹抗压强度的实测值与近红外光谱方法的预测值存在较好的相关性,相关系数均大于0.88,相对分析误差大于2.0,表明利用近红外光谱方法预测人工林粗皮桉木材力学性质效果较好。

In this study,collecting Eucalyptus pellita plantation as the research object,the mechanical properties of small clear wood samples were studied by national mechanical test methods and near infrared spectroscopy.After collecting the near infrared reflectance spectra of each sample,the spectra were pretreated and the calibration models were developed in different wavelength and validated by prediction set(one third of whole samples).The results showed that secondderivative pretreatment,350-25 000 nm full spectral wavelength,the mean of spectral value in radial and tangential section were the best in the prediction model for wood mechanical properties of Eucalyptus pellita.The relationships were good between the laboratory-determined mechanical properties values and NIR prediction values of mechanical properties,the correlation coefficients were more than 0.88,RPD were more than 2.0.The NIR analysis technique can realize the rapid prediction for the mechanical properties of small clear wood samples of Eucalyptus pellita according to better predicted effect.