Proximate Analysis of Sawdust Using Near Infrared Spectroscopy and Locally Weighted Partial Least Squares

In recently years, biomass as a renewable and widely available source of energy has increasingly been used in power generation industry and the concept of bio-power is widely accepted. However, conventional methods for proximate analysis are time consuming and can only be performed in laboratory. In this paper, 110 biomass samples are collected and near infrared spectroscopy (NIRS) technology is applied to predict proximate analysis of samples. The data show that NIRS combined with the locally weighted partial least squares (LW-PLS) obtained better prediction results comparing to conventional methods like principal component regression (PCR) and partial least squares (PLS).