Predicting unit energy consumption during industrial veneer drying via data-driven approaches

Abstract Veneer drying usually consumes a significant amount of energy including heat and electricity. The soaring energy price as well as the substantial social-environmental concerns regarding energy use have urged veneer manufacturers to adapt and become more efficient in energy consumption. Different from the physics-based methods commonly seen in the literature, this research embraced a data-driven approach to analyze and predict unit gas and electricity consumption during industrial veneer drying. Both linear regression and random forest (RF) algorithms were deployed for prediction. Based on cross-validation evaluations, the RF model with all explanatory variables slightly outperformed two linear models regarding almost all accuracy metrics, although linear models had the advantage of providing an easy-to-interpret solution.