CALIBRATING A HEAT TRANSFER MODEL FOR BALED SWITCHGRASS USING HEAT GENERATION

This study characterizes heat transfer within baled switchgrass through the development of a two-dimensional heat conduction model and an exponential drying approach to predict moisture content. The model was simulated using the finite difference method while accounting for the effect of internal heat generation to improve prediction accuracy. Model calibration was performed using rectangular bales of switchgrass (102 x 46 x 36 cm) stored in a controlled environmental chamber for 60 days at a fixed temperature and relative humidity. The heat and mass transfer model was implemented as an inverse model to empirically estimate heat generation and dry matter loss (DML). Internal heat generation and dry matter losses were also assessed using previously published models describing the aerobic respiration rate of switchgrass. Both heat generation models were in close agreement at a nominal moisture content of 10% w.b. (R² = 0.9542) but deviated at nominal moisture contents of 20% to 40% w.b. (R² < 0.7543). DML simulated using the numerical model closely agreed with the DML measured in the storage evaluation (R² > 0.9060), indicating sufficient DML estimates through the conduction model and exponential drying approach. DML simulated through the published aerobic respiration model deviated to a greater extent, particularly at the highest moisture treatment (R² = 0.4340), indicating an insufficient consideration of the various biochemical processes occurring at elevated levels of moisture and density. The results of this study provide a framework for the development of post-harvest quality models to optimize the storage, drying, and bioconversion of baled biomass feedstocks.