TY - JOUR
T1 - A dynamic grain flow model for a mass flow yield sensor on a combine
AU - Reinke, Ryan
AU - Dankowicz, Harry
AU - Phelan, Jim
AU - Kang, Wonmo
N1 - Funding Information:
This material is based on work supported by a grant from Deere & Co.
PY - 2011/10
Y1 - 2011/10
N2 - A model is developed to describe the flow of grain through a clean grain elevator system on a combine in order to facilitate accurate mass flow rate estimation. The relationship between mass flow rate and impact force described by the model depends upon machine operational characteristics, mechanical interactions of the grain and the machine geometry, and material properties of the grain. The model was designed to be adaptable to varying grain conditions, such as those influenced by moisture content, by allowing free parameters of the model to be estimated through a nonlinear regression algorithm. Simulations were performed using discrete element modeling software and data was obtained from experiments conducted on a clean grain elevator system at the University of Kentucky Combine Yield Monitor Test Facility to determine the ability of the model to accurately estimate mass flow rate. The model estimated mass flow rate with a normalized root mean squared residual (NRMSR) less than 2% for discrete element modeling simulations. For experiments involving machine components, NRMSR values were less than 3% for corn at 14% moisture, less than 3% for corn at 21% moisture, and less than 5% for corn at 26% moisture.
AB - A model is developed to describe the flow of grain through a clean grain elevator system on a combine in order to facilitate accurate mass flow rate estimation. The relationship between mass flow rate and impact force described by the model depends upon machine operational characteristics, mechanical interactions of the grain and the machine geometry, and material properties of the grain. The model was designed to be adaptable to varying grain conditions, such as those influenced by moisture content, by allowing free parameters of the model to be estimated through a nonlinear regression algorithm. Simulations were performed using discrete element modeling software and data was obtained from experiments conducted on a clean grain elevator system at the University of Kentucky Combine Yield Monitor Test Facility to determine the ability of the model to accurately estimate mass flow rate. The model estimated mass flow rate with a normalized root mean squared residual (NRMSR) less than 2% for discrete element modeling simulations. For experiments involving machine components, NRMSR values were less than 3% for corn at 14% moisture, less than 3% for corn at 21% moisture, and less than 5% for corn at 26% moisture.
KW - Discrete element modeling
KW - Experiments
KW - Flow sensor
KW - Impact plate
KW - Nonlinear regression
UR - http://www.scopus.com/inward/record.url?scp=80052299510&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=80052299510&partnerID=8YFLogxK
U2 - 10.1007/s11119-010-9215-0
DO - 10.1007/s11119-010-9215-0
M3 - Article
AN - SCOPUS:80052299510
SN - 1385-2256
VL - 12
SP - 732
EP - 749
JO - Precision Agriculture
JF - Precision Agriculture
IS - 5
ER -