POD and Galerkin-based reduction of a wood chip drying model

Abstract

The drying of wood chips can be modeled with coupled PDEs that describe water diffusion and heat conduction. Models of this type are coupled to discrete element codes that simulate the motion of wood chips in industrial rotary dryers in state-of-the-art solvers. Because many wood chips need to be modeled, the computational effort quickly becomes prohibitive. Reduced order models for the drying process of wood chips are obviously of interest in this context. We apply proper orthogonal decomposition and Galerkin projection to a model of a wood chip drying process. Gauss’ theorem is applied in the Galerkin projection step so that the boundary conditions appear explicitly in the reduced model. Our computational experiments indicate the wood chip drying process may be controlled with the ambient conditions in the rotary dryer.