Modeling the compressibility behavior of hard red wheat varieties

Authors: TURNER, AARON - UNIVERSITY OF KENTUCKY; MONTROSS, MICHAEL - UNIVERSITY OF KENTUCKY; MCNEILL, SAMUEL - UNIVERSITY OF KENTUCKY; SUMA, MICHAEL - UNIVERSITY OF KENTUCKY; CASADA, MARK; BOAC, JOSEPHINE - KANSAS STATE UNIVERSITY; BHADRA, RUMELA - KANSAS STATE UNIVERSITY; MAGHIRANG, RONALDO - KANSAS STATE UNIVERSITY; THOMPSON, SIDNEY - UNIVERSITY OF GEORGIA

Submitted to: Transactions of the ASABE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/15/2016
Publication Date: 6/1/2016
Citation: Turner, A.P., Montross, M.D., McNeill, S.G., Suma, M.P., Casada, M.E., Boac, J.M., Bhadra, R., Maghirang, R.G., Thompson, S.A. 2016. Modeling the compressibility behavior of hard red wheat varieties. Transactions of the ASABE. 59(3):1029-1038. https://doi.org/10.13031/trans.59.11432.
DOI: https://doi.org/10.13031/trans.59.11432

Interpretive Summary: Stored grain compacts in a storage structure due to the overburden pressure created by the cumulative weight of overlying grain, thereby increasing the stored grain bulk density. The mathematical relationship between overburden pressure and bulk density must be determined experimentally with uniaxial compression tests to accurately predict compaction in storage bins with existing models based on the differential form of Janssen’s equation. In this comprehensive study, uniaxial compression tests were conducted on twenty-seven different varieties of hard red wheat, at three moisture levels, over the range of pressures typically encountered in storage structures (0-20 psi). Several regression equations for bulk density were evaluated to describe the resulting pressure/density relationship. A modified Page equation fit the compressibility data best of the tested models. When the better-fitting models were applied to the field data a slight bias was observed in steel and concrete bins, but several of the models, including the modified Page and polynomial equations, produced an average error of less than 2% from the reported mass. This new compressibility data set is more robust than previously published data and should lead to improved predictions of compaction in grain bins, particularly with larger bins with higher internal pressure, which is more accurately characterized with the new data.

Technical Abstract: The bulk density of grain in a storage structure varies vertically caused by the overburden pressure created by the cumulative weight of the overlying material. As the overburden pressure increases, the stored material compacts. This compaction is believed to be caused by rearrangement of kernels along with higher inter-granular stress between particles. This compaction is of primary concern when estimating the amount of grain in a storage structure. Previous research has shown that the differential form of Janssen’s equation, assuming a variable bulk density, can be used to estimate the density increase and total mass of stored grain in a bin. In previous works, bulk density equations were developed empirically based on uniaxial compression tests using a limited number of wheat samples. In this comprehensive study, uniaxial compression tests were conducted on twenty-seven different varieties of hard red wheat, at three moisture levels, over the range of pressures typically encountered in storage structures (0-138 kPa). Mathematical models using the prior, modified, and new forms of the bulk density equation were evaluated to describe the resulting pressure/density relationship. The best performing regression models were subsequently applied to field measurements from thirty-five concrete and sixteen steel storage structures. A modified version of the Page equation had the lowest RMSE at 4.67 kg/m3, while several others, including the original polynomial equation used in the WPACKING program, were between 6-7 kg/m3. When applied to the field data a slight bias was observed in steel and concrete bins, but several of the models, including the modified Page and polynomial, produced an average error of less than 2% from the reported mass.