Article Details

This study investigated the drying kinetics of Anacardium occidentale L. wood using convective drying method and comparative modeling approach. Wood samples of 3 mm and 5 mm thickness were dried at 70, 80, and 100 °C, and their moisture ratios monitored. Traditional thin-layer drying models, including the Logarithmic, Page, and Henderson models, and artificial intelligent model (that is, artificial neural network) were utilized to predict the drying characteristics with statistically evaluations. Results showed that drying progressed in single falling rate period, while higher temperature and reduced thickness increased drying rate. The effective moisture diffusivity increased with temperature, while the activation energy was found to be 22.9 kJ/mol for 5 mm samples and 10.1 kJ/mol for 3 mm samples. The Logarithmic model, achieving an R² of up to 0.9995 and RMSE as low as 0.0056, performed best amongst traditional thin layer models. Artificial neural network model yielded an R² value of 0.9998, thereby demonstrating overall superior predictive accuracy. These results demonstrated the potential of integrating AI-based models to optimize drying processes in wood processing applications.