Research on the Multiple Linear Regression Model of Color Difference and Physicochemical Properties of Thermal Treated Biomass

This study assesses the relationship between the color changes and physicochemical properties of thermally treated biomass feedstocks, using colorimetric measurements to study the color difference values of straw, forest, and grass undergoing thermal treatments at 120~200 °C. We establish a multiple linear regression model to correlate the physicochemical properties of the treated solid products with three-dimensional color coordinates (L*, a*, b*). The results indicate that as the treatment temperature increases, the color difference ($∆E_{ab}^{\mathrm{*}}$) value also increases. Meanwhile, the number of conjugated structures in the chromophore groups increases, causing the color of the solid products to tend toward black. The ash, volatile, fixed carbon, carbon, hydrogen, oxygen, nitrogen, higher calorific value, solid yield, energy yield, bulk density, and water contact angle of the thermally treated biomass feedstocks have a statistical measure of R² ≥ 0.90 with the three-dimensional color coordinates, indicating a good correlation. Therefore, it is possible to quickly predict the basic physicochemical properties of thermally treated biomass feedstocks based on their surface color, providing a theoretical basis for the rapid quality assessment of solid products using CIELAB color changes in industrial applications.