|PRICE, RANDY - Kansas State University|
|LARSEN, JOHN - Us Sugar Corporation|
|PETERS, ALEX - Us Sugar Corporation|
Submitted to: Transactions of the ASABE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/6/2010
Publication Date: 3/9/2011
Citation: Price, R.R., Johnson, R.M., Viator, R.P., Larsen, J., Peters, A. 2011. Fiber optic yield monitor for a sugarcane chopper harvester. Transactions of the ASABE. 54(1):31-39.
Interpretive Summary: Sugarcane producers in the U.S. continue to search for ways to increase yields and profitability. One way to increase profitability would be to accurately predict and map cane yields at harvest so that transportation costs could be minimized and in-field variability could be more effectively managed. In a cooperative research effort, Sugarcane Research Unit scientists and United States Sugar’s Corporation Engineers worked with a Kansas State University Agriculture Engineer to develop and test an optical yield monitor mounted to the elevator of a sugarcane chopper harvester to predict cane yields under field harvest conditions. Results demonstrated that there was a linear relation between the optical sensor response and actual cane yields. The average observed prediction error was 7.5%; however, the magnitude of the error decreased as the total harvested area (tonnage) increased, with an error of 0.03% for an estimate of 63.7 tons. The yield monitor was not influenced by variety or harvester speed and appeared to require minimal maintenance. Field testing in Florida and Louisiana under a number of conditions to include: rainfall, variety, distance traveled, harvester speed, and lodging totaled more than 557 hrs of operation and indicated that the system was very robust, maintenance free, and for the most part self-cleaning. This technology will allow sugarcane producers to map within field yield variability as a means of identifying areas requiring additional inputs to insure yield consistency throughout the field and ultimately improve sugarcane yields.
Technical Abstract: A fiber optic yield monitoring system was developed for a sugarcane chopper harvester that utilizes a duty-cycle type approach with three fiber optic sensors mounted in the elevator floor to estimate cane yield. Field testing of the monitor demonstrated that there was a linear relationship between the optical sensor response and the actual cane yields with an R-square of 0.98 and zero intercept. The average observed prediction error on 0.5 to 1.6 metric ton estimates was 7.5%; however, the magnitude of the error decreased as the harvested area (tonnage) increased, with an error of 0.03% for an estimate of 57.8 metric tons. Factor testing indicated that the duty cycle reading was not affected by cane variety; harvester speed, harvested distance, or direction of cut (lay of the cane). Field testing across several locations in the U.S. totaled more than 557 hrs of operation and indicated that the system was very robust, maintenance free, and for the most part self-cleaning as some obstruction of the fiber optic sensors did occur in wet, muddy soils. The problem of eliminating the lens’s obstruction during wet harvesting conditions was solved by relocating the fiber optics closer to the bottom of the elevator and leaving holes on each side of the sensors to enhance cleaning and scouring. This monitoring system compares well with all previously researched methods and is more durable and easy to install.