Abstract:
The main priorities in crop
production are increasing the yield and
decreasing the cost of production. Precision
farming is the best practice to approach
these goals. For real time measurement of
sugar beet yield, a yield monitor was
developed, and installed on the exterior side
of the harvester’s chassis. The advantage of
this arrangement over similar systems is the
location of the load cell and system’s frame
which prevents blockage by trash, mud or
plant roots. For measurement of weight, one
load cell on each side of the harvester
chassis was used. Conveyor and ground
speeds were measured using two proximity
sensors. Because vibrations of the harvester
can affect the output signals, it is necessary
to find the main bandwidth associated with
the weights moving on the conveyor. For
this purpose, three different masses were
placed on the moving conveyor and this
bandwidth was determined using signal
processing. Then, a suitable filter was
designed and undesirable frequencies acting
as noise were attenuated. After calibrating
all the sensors, final evaluation of the
system was performed in the field and the
mean and standard deviation of error were
6.48% and 1.52, respectively. Although the
error may seem to be somewhat high but the
low of standard deviation indicates that
there is a similar error in all tests. These
negative errors indicate that the weight is
systematically overestimated by the
monitor. Thus, the error can be reduced by
minor changes in conveyor shape or
modified by software means. By software
modification, the systematic error was
alleviated. The median sugar beet yield was
thus obtained to be 42.7 t/ha. Comparing
this with the actual mean yield of 41.8 t/ha,
it differs by only about 2%.
