Integration of experimental and computational methods for identifying geometric, thermal and diffusive properties of biomaterials
 
More details
Hide details
1
Faculty of Agriculture and Bioengineering, Institute of Biosystems Engineering, Poznań University of Life Sciences, Wojska Polskiego 28, 60-637 Poznań, Poland
 
2
Faculty of Wood Technology, Department of Mechanical Engineering and Thermal Techniques, Poznań University of Life Sciences, Wojska Polskiego 28, 60-637 Poznań, Poland
 
 
Publication date: 2016-04-21
 
 
Int. Agrophys. 2016, 30(2): 253-260
 
KEYWORDS
ABSTRACT
Knowledge of physical properties of biomaterials is important in understanding and designing agri-food and wood processing industries. In the study presented in this paper computational methods were developed and combined with experiments to enhance identification of agri-food and forest product properties, and to predict heat and water transport in such products. They were based on the finite element model of heat and water transport and supplemented with experimental data. Algorithms were proposed for image processing, geometry meshing, and inverse/direct finite element modelling. The resulting software system was composed of integrated subsystems for 3D geometry data acquisition and mesh generation, for 3D geometry modelling and visualization, and for inverse/direct problem computations for the heat and water transport processes. Auxiliary packages were developed to assess performance, accuracy and unification of data access. The software was validated by identifying selected properties and using the estimated values to predict the examined processes, and then comparing predictions to experimental data. The geometry, thermal conductivity, specific heat, coefficient of water diffusion, equilibrium water content and convective heat and water transfer coefficients in the boundary layer were analysed. The estimated values, used as an input for simulation of the examined processes, enabled reduction in the uncertainty associated with predictions.
eISSN:2300-8725
ISSN:0236-8722
Journals System - logo
Scroll to top