Three dimensional transient heat, mass, momentum and species transfer stored grain ecosystem model using the finite element method

• Main Author: Lawrence, Johnselvakumar
• Other Authors: Stroshine, Richard L.
• Format: Text
• Language: English
• Published: Purdue University 2010
• Subjects: Food Science|Agricultural engineering; DML
• Online Access: https://docs.lib.purdue.edu/dissertations/AAI10160044

The objective of this research was to develop a three-dimensional transient heat, mass, momentum and species transfer model using the finite element method to predict grain temperature, moisture content, interstitial air velocity and gas (fumigant, CO2) concentration in the stored grain mass. The physical, chemical and biological processes for stored grain ecosystems were represented as partial differential governing equations (PDE). Effects of boundary conditions including solar radiation, internal heat generated by insects and molds, and wind speed and direction on the ecosystems are three dimensional. Thus, two-dimensional ecosystem models developed by various researches are not adequate. Dry matter loss (DML) and insect population were calculated as post-processing using the predicted grain temperature and moisture content. The developed 3D stored grain ecosystem model was validated using data collected in two locations: PHERC Bin12 at Purdue University, West Lafayette, IN for corn and the SPREC bin at Oklahoma State University, Stillwater, OK for wheat. Different combinations of models such as conduction, convection and internal heat generation were studied along with linear and quadratic elements. The conduction plus convection model predicted grain temperatures that closely followed the observed grain temperatures during the non-aeration period. The standard error of prediction was in the range of 0.9-3.6°C for wheat and 1.0-3.1°C for corn. The predicted and observed grain moisture contents varied with an error of 0.1-1.28%. Validation of the 3D model required formulation of improved headspace, plenum and wall models into systems of ordinary differential equations (ODEs) using energy and mass balance principles which were solved by the Fourth Order Runga Kutta Method. There were nine headspace air temperatures and relative humidities, nine plenum air temperatures and relative humidities and forty eight wall temperatures formulated which is unique compared to published literature. The predicted headspace ...