A “look-ahead” sensor that converted the bending load of napiergrass to a measure of yield was one of four yield-sensing approaches developed by University of Illinois researchers. The study was conducted in Lorida, Florida and funded by the Energy Biosciences Institute (EBI).
Napiergrass, also known as elephant grass, resembles sugarcane in stature and in methods of propagation. The grass is emerging as a candidate bioenergy crop, but there are limited studies available for napiergrass yield sensing, a technology that could play an important role in implementing precision agriculture and reducing harvesting cost. Alan Hansen, a professor in the U of I Department of Agricultural and Biological Engineering, and Sunil Mathanker, a postdoctoral researcher in the department, worked with colleagues from John Deere and BP Biofuels to field test the four yield-sensing approaches and document their correlation to napiergrass yield.
In this study, a stem-bending yield sensor was developed to fit a John Deere 3522 sugarcane billet harvester. Four load cells were fitted between two parallel pipes to form a push bar. The push bar was installed between the crop dividers about 1.2 meters above the ground and 1.5 meters ahead of the basecutter. The study also investigated the hydraulic pressures of basecutter, chopper, and elevator drives as indicators of yield. Three pressure sensors were fitted to the inlets of the hydraulic motors operating the basecutter, chopper, and elevator on the John Deere harvester.
The sensor that measured stem-bending force was the most accurate among the four methods tested. “What’s particularly good about this sensor,” said Hansen, “is that you’re able to measure yield at the point of entry. This is somewhat unique. In combine harvesters, for instance, you’re monitoring a yield sensor at a point much farther along in the flow of material, where the grain is about to enter the tank at the top of the combine. The delay between when the grain comes in and when it reaches the point of measurement creates a potential for error, and we have to come up with an estimate in relation to the time lag. So having this look-ahead sensor right up front is of significant value.”
While the look-ahead sensor showed the best correlation with yield, Mathanker said there are issues, such as crop lodging, harvester speed, and the ability of critical components to respond to sudden changes in ground speed, that pose a challenge for this sensing approach. Varietal characteristics, harvest time, moisture content of the stems, soil conditions, sensor height, and physical properties of the stems could also influence the bending force experienced on a push bar.
Among the three hydraulic pressure-sensing approaches, the chopper pressure showed the highest correlation with yield. A reasonable correlation was found between the basecutter pressure and yield, although in addition to yield, it was expected that the basecutter pressure would depend on cutting height. Chopper and elevator pressures were less affected by factors other than yield compared to basecutter pressure.
“Based on the results of this study,” Mathanker said, “the stem-bending yield sensor showed potential for real-time napiergrass yield prediction. It can also be used to control operating parameters of the harvester [such as travel speed] and to generate yield maps for precision agriculture. We believe this stem-bending force sensing approach can be extended to other thick-stemmed crops.”
Hansen and Mathanker published their findings in Computers and Electronics in Agriculture 111 (2015). Co-authors of the paper were H. Gan (Department of Agricultural and Biological Engineering, University of Illinois at Urbana-Champaign), J.C. Buss (John Deere, Thibodaux, LA), and J.F. Larsen (BP Biofuels North America, Houston, TX.)
The Energy Biosciences Institute is a public-private collaboration in which bioscience and biological techniques are being applied to help solve the global energy challenge. The partnership, funded with $500 million for 10 years from the energy company BP, includes researchers from the University of California, Berkeley; the University of Illinois at Urbana-Champaign; and the Lawrence Berkeley National Laboratory.
Source: ACES