Improving the accuracy of estimating grain weight by discriminating each grain impact on the yield sensor

Improving the accuracy of estimating grain weight by discriminating each grain impact on the... This study was aimed at accurately estimating total weight of harvested grain on a combine by simply attaching a small yield sensor in the grain tank and by processing the output of the sensor. The yield sensor was first installed in a grain tank of a 1.2 m-swath Japanese-style (head-feeding or jidatsu) combine, and the weight was estimated from individual impulses received at each rotation of a grain-releasing device i.e. an auger blade. A non-linear relation was assumed between the weight of grain released and the impulse received, and the parameters of the non-linear model were optimized to minimize the sum of squares between the estimated and actual weight of grain accumulated at each run of the combine. A threshold for the output discriminated between actual release and no release of the grain from the auger blade. The appropriate range of the threshold was 4–6 times the root-mean squared output of the sensor without throughput (F rms ) of grain. The aim was to enhance the accuracy of the estimation of grain weight by disregarding signals that did not relate to the accumulation of grain in the tank. Two methods of calculating the impulses were proposed after the discrimination: “successive addition” and “interval addition”, and two non-linear models of converting impulses into the weight of grain: “odd function model” and “positive function model”. The use of the odd function model with the impulse calculated by the interval addition was the most robust, and root-mean squared relative errors of calibration and validation were both stable and around 2.5 % at a threshold of 5F rms . In the confirmatory experiment with a larger 1.8 m-swath Japanese-style grain combine equipped with the same sensor, the odd function model with the interval addition achieved root-mean squared relative error of 3.6 % at calibration and 4.4 % at validation at a threshold of 5F rms . http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Precision Agriculture Springer Journals

Improving the accuracy of estimating grain weight by discriminating each grain impact on the yield sensor

Loading next page...
 
/lp/springer_journal/improving-the-accuracy-of-estimating-grain-weight-by-discriminating-FFhtyaWGf0
Publisher
Springer US
Copyright
Copyright © 2013 by Springer Science+Business Media New York
Subject
Life Sciences; Agriculture; Soil Science & Conservation; Remote Sensing/Photogrammetry; Statistics for Engineering, Physics, Computer Science, Chemistry and Earth Sciences; Atmospheric Sciences
ISSN
1385-2256
eISSN
1573-1618
D.O.I.
10.1007/s11119-013-9327-4
Publisher site
See Article on Publisher Site

Abstract

This study was aimed at accurately estimating total weight of harvested grain on a combine by simply attaching a small yield sensor in the grain tank and by processing the output of the sensor. The yield sensor was first installed in a grain tank of a 1.2 m-swath Japanese-style (head-feeding or jidatsu) combine, and the weight was estimated from individual impulses received at each rotation of a grain-releasing device i.e. an auger blade. A non-linear relation was assumed between the weight of grain released and the impulse received, and the parameters of the non-linear model were optimized to minimize the sum of squares between the estimated and actual weight of grain accumulated at each run of the combine. A threshold for the output discriminated between actual release and no release of the grain from the auger blade. The appropriate range of the threshold was 4–6 times the root-mean squared output of the sensor without throughput (F rms ) of grain. The aim was to enhance the accuracy of the estimation of grain weight by disregarding signals that did not relate to the accumulation of grain in the tank. Two methods of calculating the impulses were proposed after the discrimination: “successive addition” and “interval addition”, and two non-linear models of converting impulses into the weight of grain: “odd function model” and “positive function model”. The use of the odd function model with the impulse calculated by the interval addition was the most robust, and root-mean squared relative errors of calibration and validation were both stable and around 2.5 % at a threshold of 5F rms . In the confirmatory experiment with a larger 1.8 m-swath Japanese-style grain combine equipped with the same sensor, the odd function model with the interval addition achieved root-mean squared relative error of 3.6 % at calibration and 4.4 % at validation at a threshold of 5F rms .

Journal

Precision AgricultureSpringer Journals

Published: Nov 6, 2013

References

  • Grain yield mapping: Yield sensing, yield reconstruction, and errors
    Arslan, S; Colvin, T
  • A review of combine sensors for precision farming
    Reyns, P; Missotten, B; Ramon, H; Baerdemaeker, J

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off