Modeling of the energy requirements of a non-row sensitive corn header for a pull-type forage harvester
With the constant diversification of cropping systems and the constant increase in farm size, new trends are observed for agricultural machinery. The increase in size of the machinery and the increasing number of contractors has opened the market to selfpropelled forage harvesters equipped with headers that can harvest row crops in any direction, at any spacing. High-capacity pull-type forage harvesters are also in demand but no commercial model offers non-row sensitive corn headers. The objectives of this research were to collect data and develop models of specific energy requirements for a prototype non-row sensitive corn header. The ability to better understand the processes involved during the harvesting and the modeling of these allowed the formulation of recommendations to reduce the loads on the harvester and propelling tractor. Three sets of experiments were performed. The first experiment consisted of measuring specific energy requirements of a non-row sensitive header, in field conditions, and to compare them with a conventional header. The prototype tested was found to require approximately twice the power than a conventional header of the same width, mostly due to high no-load power. Some properties of corn stalk required for the modeling of the energy needs, that were not available in literature, were measured in the laboratory. Those include the cutting energy with a specific knife configuration used on the prototype header and the crushing resistance of corn stalk. Two knife designs were compared for required cutting energy and found not to be significantly different with values of 0.054 J/mm2 of stalk cross-section area and 0.063 J/mm2. An average crushing resistance of 6.5 N per percent of relative deformation was measured. Three mathematical models were developed and validated with experimental data to predict and understand the specific energy needs of the non-row sensitive header. An analytical model was developed based on the analysis of the processes involved in the harvesting. A regression model was developed based on throughput and header speed and a general model suggested in literature was also validated with the data. All three models were fitted with coefficient of correlation between 0.88 to 0.90.
DegreeMaster of Science (M.Sc.)
DepartmentAgricultural and Bioresource Engineering
ProgramAgricultural and Bioresource Engineering
CommitteeLaguë, Claude; Hertz, P. Barry; Crowe, Trever G.
Copyright DateDecember 2003