Predictive modelling to determine the attainable moisture content of Alstonia boonei wood using a solar kiln dryer

Abstract

Modelling of wood drying is important for predicting the performance and efficiency of solar dryers and to optimise the drying process during each season of the year. The attainable moisture content (MC) of Alstonia boonei wood was studied when dried in a laboratory-scale solar kiln. Meteorological data (temperature, relative humidity, wind speed/direction and solar radiation) were observed over a period of 31 days. For the purpose of this study, three of the variables: temperature, relative humidity and solar radiation—were used for mathematical modelling of the drying process. The average attainable MC observed over a 31-day drying period was divided into a 70:30 dataset, representing calibrating and validating sets. Several regression models were formulated using the calibrating set. The optimal model was selected based on higher values of R 2 and R, and lower standard error, after which validation was done using the remaining dataset (validating set) by performing tests of bias and percentage bias and a student’s t-test. To meet the required criteria for a suitable model, values of the validating parameters must be low and have p-values that denote significance. The log polynomial model MC = −16 + 3.99 ln(SR2) + 1.49 ln(T2) + 5.80 ln(H2) was judged best for computing the attainable MC of A. boonei wood using a solar dryer across the whole year in the study area (Ondo State, Nigeria). The computational results showed fair agreement between the predicted and measured MC, which established the validity of the model and its suitability for application when drying low-density wood in the range of 340–370 kg/m3.