Influência da idade na geração de modelos de espectroscopia nir, para predição de propriedades da madeira de Eucalyptus spp.

Abstract

The objective of this work was the development of NIRS models for physical and chemical properties of Eucalyptus wood at the ages 3-7 years. A total of 136 samples were collected for laboratory analyzes of basic density, lignin and extractives. Chip samples were ground in a cyclone mill to collect NIR spectra. The models were developed by partial least squares (PLS) regression and tested mathematical transformations. For model validation were used external samples of different ages. The developed models resulted in significant coefficients of determination (R2) for basic density (from 0.84 to 0.89) and lignin (from 0.72 to 0.88), but the best R2 values were found for the extractives content (from 0, 92-.98). By the values of ratio of performance to deviation (RPD) were assessed that samples of different ages used in development of the models did not show good predictions. But the basic density showed, for models specifically designed for the ages of 3 and 7 years, satisfactory predictions using samples at the ages 4- 6 years.