Comparison of five commonly used thin-layer moisture transfer models in fitting the re-wetting data of barley

Abstract

Five commonly cited thin-layer rewetting models, including Page, Diffusion, Approximate form of diffusion, Exponential, and Polynomial were compared for their abilities to the fit the experimental re-wetting data of barley based on the root mean square error (RMSE) and mean relative error (MRE) of estimate of the measured and simulated moisture contents.  The comparison shows the Page model is the most suitable model (average, RMSE = 0.176% d.b. and MRE = 0.713% d.b.) followed by the Diffusion model (average RMSE = 0.199% d.b. and MRE = 0.862% d.b) to fit the re-wetting experimental data of barley.  The Approximate form of diffusion, the Exponential and the Polynomial models have less fitting ability then the Page and Diffusion models for the entire period (> 4 days) of re-wetting of 33 tests at different combinations of temperatures (5.7- 46.3⁰C) and relative humidity (48.2%-88.6%).  The Page and Diffusion models were found to be the most suitable equations, to describe the thin-layer re-wetting characteristics of barley over a typically five day re-wetting.  These two models can be used for the simulation of deep-bed re-wetting of barley occurring during ventilated storage.

Keywords: Thin-layer, barley, re-wetting parameters, temperature, relative humidity