U.S. patent application number 12/128112 was filed with the patent office on 2008-09-18 for method for computing a target setting value.
Invention is credited to Joachim Baumgarten, Willi Behnke, Sebastien Neu.
Application Number | 20080228361 12/128112 |
Document ID | / |
Family ID | 36481434 |
Filed Date | 2008-09-18 |
United States Patent
Application |
20080228361 |
Kind Code |
A1 |
Behnke; Willi ; et
al. |
September 18, 2008 |
METHOD FOR COMPUTING A TARGET SETTING VALUE
Abstract
In a method for computing a target setting value--which is
adapted to a harvesting process--for a control parameter of a
working unit of a harvesting machine, operating-result curves are
plotted for a plurality of different operating-result parameters as
a function of the related control parameter, a target setting value
of the control parameter is subsequently computed based on a
combination of the plotted operating-result curves, and a method
and a corresponding control unit for controlling a working unit of
a harvesting machine, and a harvesting machine with a control unit
of this type are also provided.
Inventors: |
Behnke; Willi; (Steinhagen,
DE) ; Baumgarten; Joachim; (Beelen, DE) ; Neu;
Sebastien; (Bad Laer, DE) |
Correspondence
Address: |
STRIKER, STRIKER & STENBY
103 EAST NECK ROAD
HUNTINGTON
NY
11743
US
|
Family ID: |
36481434 |
Appl. No.: |
12/128112 |
Filed: |
May 28, 2008 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11385100 |
Mar 21, 2006 |
|
|
|
12128112 |
|
|
|
|
Current U.S.
Class: |
701/50 |
Current CPC
Class: |
A01D 41/127
20130101 |
Class at
Publication: |
701/50 |
International
Class: |
G06F 17/00 20060101
G06F017/00 |
Foreign Application Data
Date |
Code |
Application Number |
May 24, 2005 |
DE |
10 2005 0014278.8 |
Claims
1-20. (canceled)
21. A control unit for controlling a working unit of a harvesting
machine, comprising a number of measured-value inputs for acquiring
operating-result measured values (MR, MK, MV) of various
operating-result parameters of the working unit; a curve
calculating unit for computing operating-result curves (KR, KK, KV)
for the various operating-result parameters, each of which is based
on a number of the operating-result measured values (MR, MK, MV) of
a particular operating-result parameter acquired at various setting
values of a certain control parameter (SG, SO, SU) of the working
unit; a target setting value detection unit for computing a target
setting value (ZG, ZO, ZU) adapted to a harvesting process for the
control parameter (SG, SO, SU) based on a combination of the
operating-result curves (KR, KK, KV) of the various
computed-operating result parameters; and a control parameter
output for controlling an operation selected from the group
consisting of controlling the working unit based on the computed
target setting value (ZG, ZO, ZU); offering the computed target
setting value (ZG, ZO, ZU) to an operator to use in controlling the
working unit, and both.
22. A control unit as defined in claim 21; and further comprising a
terminal interface for connection to a control terminal for
acquiring default values (VW) for computing a corresponding one of
the target setting values (ZG, ZO, ZU).
23. A harvesting machine, comprising a control unit as defined in
claim 21;
24. A combine harvester, comprising a control unit as defined in
claim 21.
25. (canceled)
Description
BACKGROUND OF THE INVENTION
[0001] The present invention relates to a method for computing a
target setting value--which is adapted to a harvesting process--for
a control parameter of a working unit of a harvesting machine, in
particular a combine harvester. The present invention further
relates to a method and a corresponding control unit for
controlling a working unit of a harvesting machine, and a
harvesting machine with a control unit of this type.
[0002] Modern agricultural harvesting machines, in particular
self-propelled harvesting machines such as combine harvesters,
forage harvesters, etc., include one or more adjustable working
units for processing various types of crops. With modern harvesting
machines, the individual units are equipped with adjusting
devices--which are usually remotely controllable from the driver's
cab--with which various control parameters of the working units can
be set. Typical working units of a combine harvester are, e.g., the
threshing mechanism, which usually includes a concave and one or
more cylinders, and a cleaning unit located downstream of the
threshing mechanism, the cleaning unit typically including a blower
and a plurality of sieves. Different types of crops and harvesting
conditions, such as moisture, crop height, ground conditions, etc.,
require that the individual units and/or their adjustable control
parameters be adjusted as exactly as possible to the specific,
on-going harvesting process, in order to obtain an optimum
operating result overall.
[0003] Despite the many setting aids offered to operators by the
manufacturers of harvesting machines--such as comprehensive
operator training, printed lists of setting values predetermined
for various harvesting situations that the operator can refer to,
and electronic tools such as electronic fieldwork information
systems preprogrammed with optimized combinations of setting values
for highly diverse harvesting situations for the operator to choose
from--it is still relatively difficult for operators to adjust the
machine such that it functions in an optimum manner in accordance
with the desired requirements. This is the case, in particular, for
inexperienced and/or untrained operators, particularly at the
beginning of a harvesting season. In many cases, therefore, the
harvesting machine and/or its working units are not adapted to the
current harvesting process in an optimum manner. As a result, the
available harvesting capacity of the machine is under-utilized,
poor operating results are obtained, or, in some cases, unnecessary
crop losses result.
[0004] To solve this problem, DE 101 47 733 A1 provides an
automated method for computing a setting for an agricultural
harvesting machine which has been adapted to the harvesting
process. With this method, one control parameter of the harvesting
machine is varied while the setting remains the same and the
harvesting conditions are the same. The operating results are
subsequently compared to select exactly that setting value for the
particular control parameter that delivered a better operating
result. Using this method, even inexperienced operators learn
relatively quickly whether, when and to what extent the varied
control parameter affects the operating result, and they can set
the control parameter accordingly. The setting can also be carried
out automatically, of course. The operating-result values can be
recorded, in particular, and, by referring to the recorded
operating results, a relationship between the varied setting
parameter and the operating result obtained can be identified.
Based on this relationship, an optimum setting parameter that leads
to the best operating result can then be selected.
[0005] Since a system is involved with most of the working units on
harvesting machines, however, setting one control parameter affects
highly diverse operating-result parameters. For example, setting a
blower speed--which is a single control parameter of the cleaning
unit of a combine harvester--influences not only the losses due to
cleaning, but also the total tailings and grain tailings. The
tailings are the crop material components that are returned to the
threshing unit to be threshed again. A distinction is made between
total tailings, which is the total quantity of tailings, and grain
tailings, which refers to the grain portion of the total tailings.
The losses due to cleaning are the portions of grain carried out of
the machine with the non-grain components as a loss. A main
objective of selecting the setting, of course, is to keep losses to
a minimum. Since tailings place an additional load on the threshing
unit, however, the quantity of tailings should also be a minimum,
in the ideal case. Unfortunately, it is not necessarily the case
that, when the blower speed is varied from a certain point outward
in a certain direction, that all of the various operating-result
parameters mentioned above, e.g., losses due to cleaning, total
tailings and grain tailings, are automatically improved, since the
minimum values of the various operating-result parameters are not
all located at the same blower speed. This example also applies for
other control parameters of the cleaning unit, e.g., the upper
sieve width setting and the lower sieve width setting, and for a
large number of other working units and their control parameters.
In most cases, the various operating-result parameters are a not a
function of just one control parameter, but of a large number of
control parameters. Conversely, changing one control parameter
affects a plurality of operating-result parameters.
SUMMARY OF THE INVENTION
[0006] It is therefore an object of the present invention to create
an improved target setting value determination method, and a method
and control unit for controlling a working unit of a harvesting
machine that permit the most reliable, simple and automatable
selection of a target setting value, even when very complex setting
dependencies are involved, the target setting value being optimally
adapted to the particular harvesting process, and to therefore
permit optimized control of the working unit.
[0007] In keeping with these objects and with others which will
become apparent hereinafter, one feature of the present invention
resides, briefly stated, in a method for computing a target setting
value (ZG, ZO, ZU) which has been adjusted according to a
harvesting process, for a control parameter (SG, SO, SU) of a
working unit of a harvesting machine, the method comprising the
steps of plotting operating-result curves (KR, KK, KV) for a
plurality of different operating-result parameters as a function of
a related control parameter (SG, SO, SU); and, based on a
combination of the operating-result curves (KR, KK, KV), computing
the target setting value (ZG, ZO, ZU) of the control parameter (SG,
SO, SU).
[0008] Another feature of the present invention resides, briefly
stated, in a method for controlling a working unit of a harvesting
machine, comprising a first step including computing a target
setting value (ZG, ZO, ZU) for a control parameter (SG) of the
working unit as defined in claim 1; and subsequently controlling
the working unit based on a target setting value (ZE) that was
determined.
[0009] Still a further feature of the present invention resides in
a control unit for controlling a working unit of a harvesting
machine, comprising a number of measured-value inputs for acquiring
operating-result measured values (MR, MK, MV) of various
operating-result parameters of the working unit; a curve
calculating unit for computing operating-result curves (KR, KK, KV)
for the various operating-result parameters, each of which is based
on a number of the operating-result measured values (MR, MK, MV) of
a particular operating-result parameter acquired at various setting
values of a certain control parameter (SG, SO, SU) of the working
unit; a target setting value detection unit for computing a target
setting value (ZG, ZO, ZU) adapted to a harvesting process for the
control parameter (SG, SO, SU) based on a combination of the
operating-result curves (KR, KK, KV) of the various
computed-operating result parameters; and a control parameter
output for controlling an operation selected from the group
consisting of controlling the working unit based on the computed
target setting value (ZG, ZO, ZU); offering the computed target
setting value (ZG, ZO, ZU) to an operator to use in controlling the
working unit, and both.
[0010] According to the present invention, in order to compute an
optimum target setting value for a certain control parameter, the
first step is to plot operating-result curves for each of a
plurality of various operating-result parameters as a function of
the particular control parameters. The target setting value of the
control parameter is then computed based on a combination of the
operating-result curves that were plotted.
[0011] Combining the operating-result curves ensures that, even
when the various operating-result parameters have very complex
dependencies on the particular control parameters, an optimum
target setting value is still found for the current harvesting
conditions, thereby ensuring that the machine achieves an operating
result that is optimum overall for the given conditions.
[0012] In the method according to the present invention for
controlling a working unit of a harvesting machine, this target
setting value--which was computed as described above--is used as
the "setpoint" for the particular control parameters, in order to
control the working unit.
[0013] A control unit according to the present invention requires,
among other things, a number of measured-value inputs in order to
acquire operating-result measured values for various
operating-result parameters of the working unit. This control unit
also requires a curve calculating unit to plot operating-result
curves for the various operating-result parameters, each curve
being based on a number of operating-result measured values for the
particular operating-result parameter acquired at various setting
values of a certain control parameter of the working unit. A
control unit of this type must also include a target setting value
computation unit in order to compute a target setting value--which
has been adapted to a harvesting process--for the control parameter
based on a combination of the plotted operating-result curves for
the various operating-result parameters. Finally, the control unit
requires a control-parameter output to control the working unit
directly based on the computed target setting value, or to at least
offer an operator the setting values of the particular control
parameter so he can make a selection for control purposes. With a
control unit of this type, the particular control parameter can be
automatically adapted to the current harvesting conditions in an
optimum manner, and the operator need not have extensive experience
in doing this.
[0014] A control unit of this type can be designed in the form of a
programmable microprocessor, in particular, the curve calculating
unit and the target setting value computation unit being
implemented in the form of software on this processor. It is also
possible to design an existing programmable control unit of a
harvesting machine according to the present invention by
implementing units realized in the form of software modules,
provided this control unit includes an appropriate number of
measured value inputs for acquiring the required operating-result
measured values and the corresponding control-parameter outputs.
The required software components and/or all required program code
means can be loaded directly into the memory of the programmable
control unit, e.g., using a data memory, as a computer program
product, in the form of an update in particular.
[0015] The method for controlling a working unit can also be
refined in accordance with the method for computing a target
setting value, and vice versa. The control unit can also be refined
in accordance with the dependent method claims.
[0016] To plot an operating-result curve, operating-result measured
values are preferably acquired for a number of various setting
values of the control parameter. A mathematical function is then
adapted to the operating-result measured values as a function of
the setting values, the mathematical function ultimately forming
the operating-result curve.
[0017] The measured values can be preferably acquired by measuring
the particular control parameter alternately at high and low
setting values. In this manner, the situation can be prevented in
which, during extended operation in a certain working range of the
control parameter, systematic measurement errors are prevented from
forming and being added accumulatively. This also prevents the
possibility of the working units becoming overloaded when, e.g.,
operating-result measured values must be acquired in an extreme
working range of the control parameter.
[0018] The number and scattering of measured values, i.e., the
working range across which the setting values of the control
parameter for recording the measured values vary, depends on the
circumstances of the particular measurement, the type of control
parameter, the type of operating-result parameter, and, possibly,
on the mathematical function to be adapted, including, in
particular, any advance knowledge of the curve to be expected. A
fixed number of setting values can be specified in advance, for
example. It can also be specified in advance that exactly certain
setting values of the control parameter must be applied to acquire
the operating-result measured values. It is also possible to select
the number and position of the setting values as a function of
current conditions and/or on the basis of advance knowledge of
previous optimization cycles, etc., especially for use in the
current computation of the target setting value. It must be taken
into account that a large variance, i.e., the broadest possible
range of variation of the control parameter, has the advantage that
it increases the level of certainty with which a mathematical
function that describes the actual curve as exactly as possible can
be graphed. On the other hand, performing a measurement within a
small range of variation has the advantage that measuring time is
shortened and it is not necessary to work in extreme loss ranges
while the measurement is being performed, provided, e.g., that
losses must be measured as operating results. As a result of the
method according to the present invention for computing the setting
value, the losses that occur during the optimization process itself
are less substantial.
[0019] Highly diverse fit methods can be used to calculate a
mathematical function that is adapted to the operating-result
measured values depending on the type of scattering of the
operating-result measured values and the shape of the curve that is
expected. It can be ensured that, regardless of the form,
deviations of the measured values from the curve are minimized, but
"outliers" among the measured values do not matter very much.
[0020] With a particularly preferred exemplary embodiment, the
operating-result measured values are subjected to a regression
analysis for this purpose. In order to describe a linear dependency
of the operating-result measured value on the control parameter, a
linear regression method can be used, for example, to describe a
parabolic dependency, i.e., a quadratic regression. To ensure that
a physically meaningful assertion can be made, at least four
measured values should be applied in a quadratic regression.
Particularly preferably, operating-result measured values are
plotted for five different setting values of the control parameter,
however. This is a very good compromise between minimizing the
measuring time and obtaining the necessary number of test points in
order to generate a parabolic operating-result curve with
informative value. Preferably, all operating-result measured values
for various operating-result parameters are determined
simultaneously as a function of the varied control parameter. This
means, e.g., the control parameter is set for a certain measurement
setting value, then the operating-result measured values are
acquired in parallel for all operating-result parameters that are
dependent on this measurement setting value. Measurement time is
shortened considerably in this manner. All operating-result
parameters can also be measured independently, of course, provided
this would be a meaningful approach for certain reasons in a
specific case.
[0021] Preferably, when a target setting value is computed, a
default value specified by the operator, for example, can be taken
into account. The operator of the harvesting machine can therefore
determine whether a certain operating-result parameter is more
important than other operating-result parameters for the on-going
harvesting process. For example, the option to select either
"increased cleanliness" or "increased cleaning output" can also be
provided among the settings for a cleaning unit on a combine
harvester. If increased cleanliness is selected via the default
value, a narrower sieve setting than the target setting value can
be selected, for instance. If increased cleaning output is
required, a somewhat wider sieve opening than the target setting
value is selected.
[0022] There are highly diverse methods for computing the target
setting value based on a combination of the plotted
operating-result curves.
[0023] With a preferred exemplary embodiment, a curve-specific
target setting value or a curve-specific target setting value
range, e.g., a range below or above a certain threshold value
capable of being determined using the curve, is initially computed
separately for each of the operating-result curves. These
curve-specific target setting values or target setting value ranges
are then linked with each other in a suitable manner.
[0024] With a preferred variation, the extreme values and/or
inflection points of the operating-result curves are calculated in
order to determine the target setting value of the control
parameter, and they are linked according to a predetermined rule.
In other words, the curve-specific target setting values or target
setting value ranges are defined in this case by the extreme values
and/or inflection points.
[0025] With cleaning-loss curves, tailings curves, or curves for
other operating-result parameters, the purpose of which is to keep
the measured values as low as possible, the minimum values of the
operating-result curve are calculated, for example, and linked
according to a predetermined rule. With operating-result parameters
defined to attain the highest results possible, e.g., in terms of
the quantity of crop material that is conveyed, the maximum values
of the particular operating-result curves can be used.
[0026] A link of this type can be carried out, e.g., by calculating
the mean of the extreme values or inflection points of the
operating-result curves. A weighted mean can also be
calculated.
[0027] With a preferred exemplary embodiment, the extreme value
and/or the inflection point of at least one of the operating-result
curves is acted upon with an offset value, e.g., before the mean is
calculated. This offset value can be selected, e.g., by
compensating once more for systematic errors that are unavoidable,
e.g., due to the position of sensors with which the
operating-result measured values are recorded, or for other
reasons. An offset value of this type can also be used to weight
various curves with respect to each other.
[0028] In particular, an offset value of this type can also be
selected as a function of the default value entered in advance by
the operator, in order to obtain a target setting value that
corresponds to the requirement, e.g., for increased cleanliness or
increased cleaning output in the case of a cleaning unit, for
example.
[0029] The offset value can be selected, preferably, as a function
of the slope of the operating-result curve in the particular range.
As such, the manner in which the application of the offset affects
the particular operating result can be taken into account.
[0030] As an alternative to the linking of extreme values or
inflection points described above, it is also possible (as
mentioned briefly, above) to determine a threshold value based on
an initial operating-result curve and to use this threshold value
to compute the target setting value for a second operating-result
curve in a supplementary manner. The target setting value then
depends primarily on the second operating-result curve, but it does
not lie above or below the threshold value specified via the first
operating-result curve.
[0031] If the working unit includes a cleaning unit, or if the
working unit is a cleaning unit, separating curves are preferably
plotted as operating-result curves for the cleaning losses and/or
grain tailings and/or the total tailings. In this case, the blower
speed and/or the upper-sieve opening width of the cleaning unit are
then preferably set based on all three separating curves. The
lower-sieve opening width of the cleaning device is preferably set,
however, based only the separating curves for the grain tailings
and the total tailings.
[0032] Since, with working units such as a threshing mechanism
and/or a cleaning unit in particular, the operating-result measured
values depend, to a great extent, on the throughput, which, in turn
depends primarily on the crop quantity and ground speed, these
harvesting conditions are held constant within a certain tolerance
range over a predetermined measurement period in order to acquire
the most unequivocal operating-result measured value possible. It
is therefore preferably ensured that, once the measurement has
started, the driver holds the speed as constant as possible while
the measurement is being carried out. An automated method of
holding the harvesting conditions constant can also be provided, of
course.
[0033] Advantageously, measurement of an operating-result measured
value is automatically interrupted when the harvesting machine is
driven out of a field to be harvested, and is automatically
restarted when the harvesting machine is driven back onto the field
to be harvested. Various possibilities for keeping the harvesting
conditions as constant as possible over the predetermined
measurement period, for automatically interrupting it when the
harvesting machine is driven out of a field to be harvested and
restarting it when the harvesting machine is driven back onto the
field to be harvested will be described in greater detail
below.
[0034] The target setting values computed in a manner described
according to the present invention are used, as mentioned above, in
a method according to the present invention for the automated
control of a working unit of the harvesting machine by controlling
the working unit based on the computed target setting value.
[0035] According to this method, target setting values for various
control parameters of the working unit are preferably computed in
succession. An initial target setting value for a first control
parameter is computed, then the working unit is initially
controlled based on the computed target setting value. A further
target setting value is then computed for a further control
parameter of the working unit, and it is also set. This method is
continued until all control parameters have been set in an optimum
manner. Instead of the working unit being controlled immediately
and automatically with the computed target setting value, the
computed target setting value can first be offered to the operator
for selection, e.g., in a display. The operator can accept the
value, e.g., by entering a confirmation command.
[0036] If a cleaning unit is controlled using the method according
to the present invention, a target setting value for a blower speed
is preferably computed in a first step, a target setting value for
the upper-sieve opening width is computed in a second step, and a
target setting value for a lower-sieve opening width of the
cleaning device is computed in a third step. The particular
components are then preferably controlled immediately using the
computed target setting value.
[0037] The initial setting values to be used in an optimization of
this type are preferably grain-dependent setting values
predetermined by the manufacturer of the harvesting machine for
certain crops under certain harvesting conditions, or that were
computed in advance by the operator of the harvesting machine under
similar conditions for the crop being harvested. In particular,
crop-dependent setting values stored in an electronic fieldwork
computer system can be used as the initial setting values. In the
optimization process, these initial setting values are first
entered for the various control parameters One of the first control
parameters is then varied, the associated measured values are
plotted, and an optimum setting value for this control parameter is
then identified in a manner according to the present invention. A
second control parameter is then optimized in this manner, etc.,
until all of the desired control parameters have been optimized.
The operator can determine which of the control parameters to
optimize. Normally, all control parameters are optimized, to the
extent this is possible.
[0038] Preferably, after a certain period of time and/or when a
predetermined event occurs, a new target setting value for a
control parameter of the working unit is computed, and the working
unit is controlled based on the new setting value. It is also
possible, of course, to recompute an entire chain of target setting
values for various control parameters of the working unit, for
example, as described above.
[0039] The certain time period can be selected such that
optimization is carried out Whenever it is expected that the
harvesting conditions have changed. For example, with a harvesting
process that takes an entire day to complete, reoptimization can be
carried out in the morning, in the afternoon and in the evening,
because it is possible that the straw moisture in the crop material
changed over the course of the day.
[0040] The events that could make it necessary to reoptimize the
target setting values can include, in particular, a change in
throughput, e.g., if harvesting is carried out at a speed that
differs from the speed that existed when the target setting values
were determined. The event can also be a change to a control
parameter of another working unit of the harvesting machine. It can
be assumed, for example, that the cleaning load in a combine
harvester changes considerably if the threshing mechanism was
repositioned to a considerable extent. Events can also be
predetermined via other measurement sensors, so that the target
setting values are recomputed, e.g., depending on the crop, i.e.,
when a changed property of the crop, such as grain moisture, is
measured.
[0041] When a recomputation of the target setting values is
started, i.e., for the re-start of optimization, the target setting
values computed in the previous optimization are preferably used as
the default setting for the measurement.
[0042] The novel features which are considered as characteristic
for the present invention are set forth in particular in the
appended claims. The invention itself, however, both as to its
construction and its method of operation, together with additional
objects and advantages thereof, will be best understood from the
following description of specific embodiments when read in
connection with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0043] FIG. 1 shows a schematic cross section through a combine
harvester,
[0044] FIG. 2 shows a schematic depiction of a control unit for
controlling a cleaning unit of a combine harvester that includes a
connected control terminal with a user interface, in a first
process state,
[0045] FIG. 3 shows a depiction of a user interface of the control
terminal according to FIG. 2, in a second process state,
[0046] FIG. 4 shows a depiction of a user interface of the control
terminal according to FIG. 2, in a third process state,
[0047] FIG. 5 shows a flow chart of a possible sequence of steps in
an optimization of the cleaning unit of a combine harvester,
[0048] FIG. 6 shows a diagram of a possible sequence of an
optimization of one of the control parameters within a method
sequence according to FIG. 5,
[0049] FIG. 7 shows a flow chart for computing an optimized target
setting value for a control parameter of a cleaning unit,
[0050] FIG. 8 shows a diagram that depicts the plotting of
operating-result curves based on operating-result measured
values,
[0051] FIG. 9 shows a diagram that depicts the identification of an
optimized target setting value for the blower speed of a cleaning
unit,
[0052] FIG. 10 shows a diagram that depicts the identification of
an optimum target setting value for the upper-sieve opening
width,
[0053] FIG. 11 shows a diagram that depicts the identification of
an optimum target setting value for the lower-sieve opening
width.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0054] The exemplary embodiment of the present invention shown in
FIG. 1 is a self-propelled combine harvester 1 with a tangential or
cross-flow threshing mechanism 4 and a plurality of shakers 9
located behind it, as the separating unit. The separating unit is
composed of a plurality of tray-type shakers 9 with a plurality of
shaking speeds. A cleaning unit 10 is located beneath shaker 9,
which is composed of a plurality of sieves 13 located one on top of
the other, and a blower 11.
[0055] The mode of operation of a combine harvester 1 of this type
is as follows:
[0056] Using a reel of the cutting disc, the crop material is
placed on mowing unit 2 and is cut using knives. The crop material
is then conveyed via a header auger and a feed rake in a feeder
housing 3 to the inlet of threshing mechanism 4.
[0057] A feed and/or pre-acceleration cylinder 5 is located at the
inlet of threshing mechanism 4. Located behind threshing mechanism
4, in the direction of crop flow, is a cylinder 6 with an axis of
rotation positioned transversely to the direction of crop flow,
i.e., transversely to the longitudinal axis of the combine
harvester. Located beneath cylinder 6 is a concave 8 which is
shaped to encompass cylinder 6. The crop material coming out of
feeder housing 3 is grasped by pre-acceleration drum 5 and pulled
further by cylinder 6 through the threshing gap between cylinder 6
and concave 8. The crop material is threshed, i.e., beaten and/or
crushed, by the beater bars of cylinder 6, a grain-chaff mixture
falling downward through concave 8 and being subsequently guided to
cleaning unit 10 in order to separate the grains from the
admixtures, i.e., stalk and chaff parts.
[0058] From threshing mechanism 4, the threshed crop flow is
directed by impeller 7 to tray-type shaker 9, via which the grain
and any short straw and chaff located in the crop flow is separated
out. The grain, short straw and chaff also reach cleaning unit 10,
where the grain is separated from the short straw and chaff.
[0059] The grain is separated from the non-grain components in
cleaning unit 10 in a manner such that wind is blown through the
sieve openings (holes, mesh, slits) into sieves 12, 13--which are
driven in an oscillating manner--using blower 11, the wind
loosening the crop material directed over sieves 12, 13 and
ensuring that the specifically lighter chaff and short-straw
portions are separated out, while the heavy crop grains fall
through the sieve openings. An upper sieve 12 and a lower sieve 13
are located one on top of the other in certain areas such that the
crop material is sifted with different levels of fineness at the
various levels.
[0060] The grain that passes through both sieves 12, 13 of cleaning
unit 10 falls to a first capture and guide floor and is conveyed to
a grain-delivery auger. The grain is then conveyed by an elevator
15 into a grain tank 19 of combine harvester 1, from where it can
be transferred to a trailer as necessary using a tank unloading
conveyer.
[0061] The particles in cleaning unit 10 that initially fall, at
the rear end, through the sieve openings of upper sieve 12 are
typically heavier particles, i.e., particles that contain a grain
particle that has not been fully separated from other components of
the grain. These particles fall, behind lower sieve 13, onto a
second capture and guide floor located beneath and somewhat behind
the first capture and guide floor, and are returned to threshing
mechanism 4 as tailings via a tailings elevator 14.
[0062] Components that do not fall through upper sieve 12 are
discarded as a loss. The straw and a certain percentage of waste
grain also travel via tray-type shaker 9 to the rear end of combine
harvester 1, from where they are ejected.
[0063] With the exemplary embodiment below it is assumed that,
according to the present invention, the objective is to compute
target setting values ZG, ZO, ZU for various control parameters SG,
SO, SU of cleaning unit 10 of a combine harvester, e.g., for the
upper-sieve opening width, the lower-sieve opening width, and the
blower speed. The method according to the present invention has
already been proven to be effective for setting a cleaning device
10 of this type, and it can therefore be used particularly
advantageously. The method according to the present invention
and/or the corresponding control unit can also be used, of course,
to set other working units, e.g., the rotational speed of the
cylinder or the width of the concave on any other harvesting
machine. For reasons of completeness, reference is also made to the
fact that the present invention can also be used very well to
control cleaning units on other types of combine harvesters.
[0064] To compute target setting values ZG, ZO, ZU adapted to the
harvesting process for the various control parameters SG, SO, SU of
cleaning unit 10, various operating-result measured values MR, MK,
MV for different operating-result parameters must be measured and,
based on these operating-result measured values MR, MK, MV,
operating-result curves KR, KK, KV must be plotted.
[0065] With the exemplary embodiment shown, the losses due to
cleaning and the total tailings and grain tailings are observed as
the operating-result parameters used to compute the target setting
values for the various control parameters. The grain tailings are
the grain components contained in the total tailings.
[0066] Different measuring units 16, 17, 18 are located in various
locations in the combine harvester for this purpose.
[0067] A cleaning-loss measuring unit 16 is located directly
beneath the rear end of upper sieve 12 and is used to measure the
losses due to cleaning, the cleaning-loss measuring unit 16
typically being designed as a knock sensor. The signal detected by
knock sensor 16 is a measure of how many components fall directly
behind upper sieve 12. Based on this information, the total loss
can be estimated relatively well.
[0068] The total tailings are measured with the aid of a total
tailings-measuring unit 18 located in tailings elevator 14. It
measures the total quantity conveyed, e.g., via the weight conveyed
by tailings elevator 14 or optical and/or capacitive measurements
etc. The grain portion of the total tailings, i.e., the grain
tailings, is measured using a grain-tailings measuring unit 17
located on the second capture and guide floor behind lower sieve
13. Grain-tailings measuring unit 17 is also preferably a knock
sensor, the output signal of which is a measure of the amount of
grain that falls behind lower sieve 13 into the tailings.
[0069] All of these measuring units 16, 17, 18 are connected to a
control unit 30. A control terminal 40 is also connected to control
unit 30, control terminal 40 having a display with a user interface
41 with which a driver can operate and/or program control unit 30.
Control terminal 40 is located inside driver's cab 20. The
connection of individual measuring units 16, 17, 18 and control
terminal 40 with control unit 30, and control unit 30 itself, are
not shown in FIG. 1, to prevent the figure from becoming overly
complex. Instead, a somewhat more detained depiction is shown in
FIG. 2, to which reference is made for the explanations to
follow.
[0070] In this case, control unit 30 includes three measured-value
inputs 31, 32, 33, to which measuring units 16, 17, 18 are
connected. Cleaning-loss measured values MR are transmitted by
cleaning-loss measuring unit 16 to input 31 of control unit 30,
grain-tailings measured values MK are transmitted by grain-tailings
measuring unit 17 to input 32 of control unit 30, and
volume-tailings measured values MV are transmitted by
volume-tailings measuring unit 18 to control unit 30.
[0071] Control unit 30 also has three control parameter outputs 35,
36, 37, via which the setting values for the control parameters
"blower speed" SG, "upper-sieve width setting" SO and "lower-sieve
width setting" SU are transferred as setpoints to the particular
components of cleaning unit 10. Using appropriate (not shown)
sensors, control unit 30 can check to determine whether the desired
setting values were actually attained.
[0072] Control terminal 40 is connected to control unit 30 via a
terminal interface 34. In this case, control terminal 40 is
designed as a touchpad which the operator can use to press on
certain regions of user interface 41 to enter certain input
commands.
[0073] In the upper region of user interface 41, three
setting-value display fields 45 for the blower speed, upper-sieve
setting and lower-sieve setting (from top to bottom) are depicted,
one below the other. The value of control parameter SG, SO, SU that
was originally set for the individual components, and the new
target setting value ZG, ZO, ZU are displayed in setting-value
display fields 45 during operation.
[0074] Two default fields 43, 44 are located in the next row down
in the display. By pressing default fields 43, 44, the operator can
select a default value VW, which is transmitted to control unit 30.
Default value VW is displayed in a default value display 42
directly below default fields 43. When the operator presses left
default field 44, default value VW is reduced. As a result, the
optimization process ensures that, in the harvesting process,
greater emphasis is placed on "increased cleanliness" than on
cleaning output. Conversely, when the operator presses on right
default field 43, default value VW is increased, so that the
optimization process places greater emphasis on the criterium
"increased cleaning output" than on cleanliness. A start field 46
is located under default-value display 42. Start field 46 is
pressed to start the optimization process.
[0075] It should be noted that user interface 41 can also have a
completely different design, of course. In particular, it can also
be part of a larger control terminal 40 where even more regions for
setting other components are displayed, and where additional
information for the operator is displayed. It is also possible to
use another form of user interface other than a touchpad.
[0076] In this case, control unit 30 is designed in the form of a
programmable microprocessor, on which components which are
essential to the present invention are implemented in the form of
software modules, the components including, e.g., a curve
calculating unit 38 which uses input signals MR, MK, MV to
calculate operating-result curves KR, KK, KV, and a target setting
value computation unit 39 which uses curves KR, KK, KV and default
value VW set via control terminal 40 to respectively calculate the
target setting values for the various control parameters SG, so,
SU.
[0077] A control unit 30 that serves only to control cleaning unit
10 is shown in FIG. 2. It is clear that a control unit 30 of this
type can also control other working units, e.g., the threshing
mechanism of combine harvester 1, and that the control units for
highly diverse types of working units can be located in the form of
modules in a master control unit of combine harvester 1. It is also
clear that a control unit of this type can also include further
measured-value inputs and control-parameter outputs. For example,
combine harvester 1 can also include sensors in the feeder housing
for measuring the height of the crop layer, and/or further sensor
units in the grain tank and/or at the outlet of the grain elevator,
such as a yield measuring device for determining the total quantity
of grain, or grain breakage detectors, with which damaged and/or
broken grains can be detected, or sensors located at the end of the
tray-type shakers for determining the straw walker losses.
[0078] None of these components are depicted in the exemplary
embodiment shown in FIG. 2, however, to prevent the figure from
becoming overly complex.
[0079] Reference is made to FIG. 5 in the explanation of the
sequence of steps that take place in a complete optimization
process of cleaning unit 10.
[0080] The process starts when the operator enters a default value
VW, as described above. The operator then starts the optimization
process. To start the optimization process, the operator touches
start field 46 on control terminal 40 (refer to FIG. 2). In the
next step, a check is carried out to determine whether the
necessary starting conditions are given, i.e., whether the
threshing units and cleaning unit 10 have been set. If they have
not, the process is halted immediately.
[0081] If they have, optimization of the blower speed is started.
Setting values specified by the electronic fieldwork system for the
particular type of crop are first selected as the starting values
for the individual control parameters.
[0082] To measure operating-result curves KR, KK, KV,
operating-result measured values KR, KK, KV are then acquired for
various setting values M.sub.1, M.sub.2, M.sub.3, M.sub.4, M.sub.5
(also referred to below as "test points") of the control parameter
to be optimized. Since the objective in this case is to first
optimize the blower, operating-result measured values MR, MK, MV
are measured at various measuring points M.sub.1, M.sub.2, M.sub.3,
M.sub.4, M.sub.5 of the blower speed. While performing a
measurement of this type, the other parameters must not be changed,
and any other harvesting conditions must be held as constant as
possible. The applies to throughput in particular.
[0083] All of the steps involved in the measurement procedure up to
the point at which optimum target setting value ZG is determined
are depicted schematically in FIG. 6. This figure shows a "UML"
state diagram which describes the procedure for measuring the
blower speed (UML=Unified Modeling Language; in this UML diagram,
the symbols "/" mean "condition and/or action", and the symbols
"&&" mean "and".)
[0084] First, the system "learns" a constant ground speed. As
mentioned above, a uniform flow of crop material during the
measurement procedure is an important requirement for performing an
optimization. The throughput quantity depends to a considerable
extent on the speed, however. It must therefore be ensured that the
average speed be held as constant as possible during the entire
optimization process. This state, during which the current ground
speed is being "learned", is displayed to the operator, as shown in
FIG. 3, in user interface 41 of control terminal 40. In this
procedure, the working units use the crop-dependent default values
in the electronic fieldwork system as the starting values. These
are the basic settings for the units. The starting value of the
particular control parameter is displayed in setting-value display
fields 45. In this case, the starting values are a blower speed of
1,200 rpm, an upper-sieve setting of 15 mm, and a lower-sieve
setting of 9 mm, which could be used as the starting values when
harvesting wheat, for example.
[0085] As soon as a mean ground speed is reached, this is also
displayed on user interface 41 (refer to FIG. 4). On user interface
41, the current speed within a tolerance range is also displayed,
in a speed-display field. The driver must then ensure that the
current speed within this tolerance range remains in the middle--as
depicted in the display--to the greatest extent possible.
[0086] The Stop field is now displayed instead of the Start field.
The operator can use the Stop field to stop the optimization
process at any time. When he does so, the machine returns to the
starting values. Likewise, the operator can use the symbols
displayed next to setting-value display fields 45 to select which
of the control parameters, "blower speed" SG, "upper-sieve width"
SO, or "lower-sieve width" SU to optimize. According to the
standard procedure, as shown in FIG. 5, the blower is optimized
first, followed by the upper sieve and then the lower sieve.
[0087] Once the desired ground speed has been reached, the first
measurement setting value is applied for the parameter to be
optimized, i.e., the blower speed in this case. The system itself
is then initially in a waiting state until a start-up phase has
been completed, in which the parameters have stabilized once the
ground speed and starting values have been set. A fixed delay time
of, e.g., a few seconds, can be specified for this. Once the
start-up phase has been completed and the current target harvesting
conditions have been achieved, measurement of the first measured
value can be started.
[0088] Various measuring units 16, 17, 18 then acquire various
operating-result measured values MR, MK, MV at the predetermined,
first test point M.sub.1. This means, e.g., a measured value MR for
the losses due to cleaning, a measured value MK for the grain
tailings, and a measured value MV for the total tailings are
acquired when a relatively low speed has been set. This is shown in
FIG. 8. In FIG. 8, various measured values MV, MR, MK are shown
plotted to the far left for first test point M.sub.1 with respect
to the blower speed. When measured-value acquisition has been
completed, the next test point, M.sub.5, is applied. After the
start-up phase has been completed, the further measured values for
losses due to cleaning, grain tailings and total tailings are
acquired. These measured values are also plotted in FIG. 8.
[0089] In this case, it is preferably not the next higher test
point, M.sub.2, that is applied, but rather a test point M.sub.5
located at the other end of the range to be measured. This means,
e.g., the measurement is first carried out at the lowest blower
speed to be measured, and then at the highest blower speed to be
measured. Measurements are then carried out at the second-lowest
blower speed, followed by the second-highest blower speed, etc. The
advantage of performing measurements in an alternating manner, at
high and low extreme values, is that, since the losses and tailings
are typically greater in these ranges, the units will not be
overloaded, and systematic measurement errors that could result
from accumulatively added disturbances are prevented.
[0090] If the target harvesting conditions stop being met during
measured-value acquisition, e.g., because the machine has driven
out of the field to be harvested, the measurement is interrupted
and, e.g., the starting values specified by the fieldwork
information system can be applied, and the measured value which has
already been measured can be stored. The machine then remains in
the waiting state until the target harvesting conditions are
attained again. Measurement setting value M.sub.1, M.sub.2,
M.sub.3, M.sub.4 M.sub.5 at which the current measurement is to be
carried out is then set again and, after the start-up phase has
been completed, measured-value acquisition is continued.
[0091] This interruption of the measurement procedure can take
place automatically, e.g., with the aid of sensors used to detect
the field to be harvested. Two sensors are preferably used for this
purpose. A crop-layer height can be determined in the feeder using
a first sensor. When the machine is driven off of the field, the
height of the straw in the feed rake decreases, practically without
any time delay. This sensor can be used to determine when the
machine leaves the field to be harvested, and the measurement can
therefore be interrupted immediately. A separate measuring unit,
e.g., a grain-throughput measuring unit, is preferably used to
restart the measuring procedure. It checks to determine whether the
grain throughput has climbed above a minimum threshold again. Since
this sensor, which registers the quantity being conveyed in the
upper region of the grain elevator, for example, has a delayed
reaction relative to the cleaning unit, the layer of crop material
on the upper sieve of the cleaning unit has formed completely, even
when this sensor reading is low, thereby indicating with certainty
that the target harvesting conditions are in place again.
[0092] Once all measurements have been completed, the optimum
target setting value ZG is determined in a subsequent step, in the
manner according to the present invention. Reference is made to
FIGS. 7 and 8 in the explanation of this procedure.
[0093] In a first step I, mathematical functions are adapted for
operating-result measured values MR, MK, MV, in order to obtain
operating-result curves KR, KK, KV. Since it can be expected, due
to the physical conditions, that curves KR, KK, KV are parabolic in
shape, the best-fit mathematical functions are determined using
quadratic regression based on the recursive least squares method.
The basic form of a quadratic function of this type is:
y=a.sub.2x.sup.2+a.sub.1x+b (1)
[0094] The three coefficients a.sub.1, a.sub.2, b for this equation
are determined in a regression analysis based on measured values
MR, MK, MV that were obtained. The following equations are used for
this purpose, in order to calculate a factor k and auxiliary
variables A through F:
k = n ( x i 2 ) ( x i 4 ) + 2 ( x i ) ( x i 2 ) ( x i 3 ) - ( x i 2
) 2 - n ( x i 3 ) 2 - ( x i ) 2 ( x i 4 ) ( 2 ) A = [ ( x i 2 ) ( x
i 4 ) - ( x i 2 ) 2 ] 1 k ( 3 ) B = [ n ( x i 4 ) - ( x i 2 ) 2 ] 1
k ( 4 ) C = [ n ( x i 2 ) - ( x i ) 2 ] 1 k ( 5 ) D = [ ( x i 3 ) (
x i 2 ) - ( x i ) ( x i 4 ) ] 1 k ( 6 ) E = [ ( x i ) ( x i 3 ) - (
x i 2 ) 2 ] 1 k ( 7 ) F = [ ( x i ) ( x i 2 ) - n ( x i 3 ) ] 1 k (
8 ) ##EQU00001##
[0095] Using factor k and auxiliary variables A through F, the
individual coefficients of the regression polynomial can be
determined, as follows:
b=A.SIGMA.y.sub.1+D.SIGMA.x.sub.iy.sub.i+E.SIGMA.x.sub.i.sup.2y.sub.i
(9)
a.sub.1=D.SIGMA.y.sub.i+B.SIGMA.x.sub.iy.sub.i+F.SIGMA.x.sub.i.sup.2y.su-
b.i (10)
a.sub.2=E.SIGMA.y.sub.i+F.SIGMA.x.sub.iy.sub.i+C.SIGMA.x.sub.i.sup.2y.su-
b.i (11)
[0096] In the equations shown above, n is the number of test
points, x.sub.i represents the values of the individual test
points, and y.sub.i represents the operating result-measured values
measured at test points x.sub.i for the particular
operating-results parameter, and i is an index variable that counts
from 1 to n. Addition is performed accumulatively from i=1 through
n.
[0097] Curve KR for the losses due to cleaning, curve KK for the
grain tailings, and curve KV for the total tailings are shown in
FIG. 8. All of the curves decrease initially as the blower speed
increases, and they subsequently start to rise as the blower speed
increases. The reason for this is that, when blower speeds are too
low, an excessively thick layer forms on sieves 12, 13, and
cleaning unit 10 can no longer operate effectively. If the blower
speed is increased too much, the wind causes an excessive quantity
of particles to be carried out of the machine, thereby causing the
losses to increase significantly. Likewise, an increasing quantity
of grains that should drop through lower sieve 13 are carried into
the total tailings, thereby causing grain-tailings curve KK and
volume-tailings curve KV to increase.
[0098] Each of the minimum values RM, KM, VM of curves KR, KK, KV
is the ideal setting value for the blower with respect to the
particular curve KR, KK, KV. Unfortunately, however, minimum values
RM, KM, VM are not aligned directly one above the other, which
means an optimum target setting value ZG must be determined that
takes all result parameters into account in a suitable manner. To
this end, minimum values RM, KM, VM of individual curves KR, KK, KV
are linked in a suitable manner. This step is preceded, however, by
a few inquiries to determine the extent to which the individual
plotted curves KR, KK, KV have informative value.
[0099] This process is shown in FIG. 7. Step I is the quadratic
regression, which is carried out for all three operating-results
parameters, i.e., the losses due to cleaning, the grain tailings,
and the total tailings, in order to plot the three curves KR, KK,
KV.
[0100] In parallel with this step, curve-specific target setting
values ZR, ZK, ZV are initially determined for all three
operating-result curves KR, KK, KV. The following inquiries are
carried out for this purpose (the explanations below relate to the
handling of the cleaning-loss curve KR, as an example):
[0101] For curve KR, a check is initially carried out in Step IIa
to determine whether the shape factor, which is a measure of the
curvature of the parabola, is sufficiently great. This means that
the shape factor of measured curve KR is compared with a threshold
value and, only if this is the case, the minimum value RM of
plotted curve KR is accepted as curve-specific target setting value
ZR for the blower speed in terms of losses due to cleaning (Step
VIIa). If the shape factor is too low, the minimum value RM is not
unequivocal, and the informative value of curve KR is very low.
[0102] In a further query step, IIIa, a query is therefore made as
to whether the signal change is sufficiently great. To this end, a
check is carried out to determine whether the difference between
the operating-result value in minimum RM of curve KR and an
operating-result measured value at a maximum test point of the
particular control parameter--at a blower speed test point in this
case--exceeds a certain value. If so, it can be assumed that the
curve is indeed distinct enough. As a result, in Step VIIa, the
minimum value on curve KR is accepted as the optimim curve-specific
target setting value ZR for the blower in terms of the losses due
to cleaning.
[0103] If not, a check is carried out in Step IVa to determine
whether a signal change can even be detected. To this end, the
difference--described above--between minimum RM of operating-result
curve KR and an operating-result measured value is applied once
more at a higher test point of the particular control parameter and
compared with a further, lower threshold value. If the difference
is not below this threshold value, it is assumed that the curve
does not have informative value, and the crop-dependent starting
value specified by the electronic fieldwork system is applied as
the curve-specific target setting value ZR of the control parameter
in terms of the losses due to cleaning (Step Va). If the
differential value is below the threshold value, however, i.e., if
a signal change cannot be detected, this starting value plus an
offset value is used, i.e., the blower speed is increased by a
certain value, and this value is used as curve-specific target
setting value ZR. Since the losses do not increase significantly in
this case when the blower speed is increased, it makes sense in
terms of losses to select curve-specific target setting value ZR,
since the cleaning output is improved as a result without having to
put up with higher losses.
[0104] In Steps IIb through VIIb, the same method is carried out in
parallel for the grain tailings, and in Steps 1c through VIIc for
the total tailings. The only difference between the two is that, in
Step VIIb and VIIc, identified minimum values KM, VM of particular
curves KK, KV are also acted upon with an offset value OK, OV in
order to determine the particular curve-specific target setting
values ZK, ZV. Offset values OK, OV are used to assign top priority
to the losses due to cleaning in the computation of optimum target
setting value ZG, i.e., they are used to place greater weight on
cleaning-loss curve KR, since grain tailings and total tailings
should play a somewhat lesser role in daily operation compared with
the losses due to cleaning.
[0105] Offset OK, OV is applied such that the blower speed is
increased with respect to minimum value KM) of grain tailings curve
KM, and the blower speed is reduced with respect to minimum value
KV of total tailings curve KV. A fixed offset value with respect to
the operating-result parameter is specified in the system. A
displacement of this type along the measured curve around a fixed
operating-result value, i.e., in the y-direction of particular
curve KK, KV, makes it possible to dynamically adapt offset OK, OV
with respect to the control parameter, i.e., in the x-direction.
This means the actual offset value is a function of the slope of
curve KK, KV within the minimum range. If curve KK, KV has sharp
curvature, displacement is slight. If curves are relatively flat
and the differences in tailings are therefore slight, the
displacement is greater.
[0106] A displacement of the optimum target setting value of this
type with respect to a certain control parameter can be calculated
based on the solution of the quadratic equation (1), as
follows:
[0107] In this case, Y.sub.Min is the operating-result value in
minimum value KM, VM of particular operating-result curve KK, KV,
and X.sub.Neu is the curve-specific target setting value ZK, ZV for
the control parameter with respect to this curve KK, KV, which is
used instead of identified minimum value KM, VM of curve KK, KV.
Depending on the sign of the square root in the quadratic formula
in equation (12), the shift is to the left (for total tailings) or
to the right (for grain tailings).
[0108] Default value VW--which the operator entered at the
beginning of the optimization process--can also be applied to the
offsets to attain "increased cleaning output" or "increased
cleanliness". Accordingly, the offsets of grain tailings curve KK
and total tailings curve KV are shifted to the left or right. As an
alternative, of course, an offset in one direction or the other can
also be applied to target setting value ZG computed overall,
depending on the default value.
[0109] Curve-specific target setting values ZR, ZK, ZM computed in
Steps Va through VIIa, Vb through VIIb and Vc through VIIc are then
averaged in Step VIII. This mean is optimized target setting value
ZG for the particular control parameter, which is blower speed ZG
in this case.
[0110] This procedure is depicted graphically in FIG. 9. FIG. 9
shows operating-result curves KR, KK, KV for the losses due to
cleaning, the total tailings, and grain tailings. The minimum
values RM, KM, VM are labeled on each of three curves KR, KK, KV.
In addition, offsets OK, OV and resultant curve-specific target
setting value ZK, ZV are plotted for grain tailings curve KK and
total tailings curve KV. Optimum target setting value ZG is also
labeled; it is computed from the mean of curve-specific target
setting values ZR, ZK, ZV, which corresponds to the minimum value
on cleaning-loss curve KK. It is also labeled with an arrow in the
figure.
[0111] After optimized target setting value ZG has been determined
for blower speed SG, the blower can be set with target setting
value ZG (refer to FIG. 6).
[0112] This procedure is followed by optimization of the upper
sieve, as shown in FIG. 5, in which case optimized target setting
value ZG of blower 11 is used. In this case, the starting values
for blower speed SG are therefore previously-determined optimized
target setting value ZG and, for all further parameters, the
crop-dependent starting values taken from the electronic fieldwork
information system.
[0113] Subsequently, as was the case for blower optimization,
measured values are determined for the losses due to cleaning,
total tailings and grain tailings for various settings of
upper-sieve width SO. The procedure is exactly the same as the
procedure used to compute target setting value ZG for blower speed
SG, as was explained with reference to FIGS. 6 and 7, i.e., the
same method steps are taken, but this time they are used to set the
upper sieve width. Curves KR, KK, KV are also plotted for all three
operating-result parameters using quadratic regression (refer to
Step I in FIG. 7), and the minimum values are subsequently linked
in a suitable manner.
[0114] The only difference in terms of determining target setting
value ZG for blower speed SG is that an offset OV is set only for
total tailings. Minimum value RM, KM of curves KR, KK are used as
curve-specific target setting value ZR, ZK for the grain tailings
and losses due to cleaning. This is depicted graphically in FIG.
10. In this case as well, minimum values RM, KM, VM of all three
curves KR, KK, KV are indicated and offset OV and curve-specific
target setting value ZV are also indicated for total tailings-curve
KV. Target setting value ZO for the upper sieve opening which
results from the individual values is also shown.
[0115] As shown in FIG. 5, after optimization is carried out for
the upper sieve, optimization is carried out for the lower sieve.
The individual operating-result values are also measured, in this
case, as a function of the setting for the lower sieve width in a
manner similar to that depicted in FIG. 6.
[0116] Optimum target setting value ZU for lower-sieve opening SU
(refer to FIG. 11) is computed only as a function of grain-tailings
curve KK and volume-tailings curve KV, since the lower sieve does
not affect the losses due to cleaning. In addition, for the total
tailings, a quadratic curve is not adapted to the measured values.
Instead, linear regression is used to determine a straight line as
characteristic KV, since total tailings decrease the wider the
opening of lower sieve 13 becomes, and it cannot subsequently
increase.
[0117] Target setting value ZU is therefore determined primarily
based on minimum value KM of grain-tailings curve KK, an offset
being applied in one direction or another to curve-specific target
setting value ZK determined there (which corresponds to minimum KM
in this case), depending on default value VW. This is done to
attain either increased cleanliness or increased cleaning output.
If the objective is to increase cleaning output, lower sieve 13 is
opened somewhat wider. If the objective is to increase cleanliness,
the width of the lower sieve opening is reduced.
[0118] In this case, the two operating-result curves KK, KV are
therefore not linked by linking minimum values of the two curves
KK, KV, but rather by selecting a threshold value SW by referring
to a curve KV--volume-tailings curve KV, in this case--that may not
be fallen below in the computation of an optimum target setting
value ZU using another curve, i.e., grain-tailings curve KK in this
case. In this manner it is ensured that, even though a value for
lower-sieve opening width SU is obtained that is ideal with respect
to grain tailings, the total tailings are not so great that
threshing mechanism 4 is overloaded, which would reduce the total
output of machine 1.
[0119] After optimization of the lower sieve has been carried out,
a wait ensues to determine whether an event will occur that would
require that optimization be repeated (refer to FIG. 5). An
automatic restart of the optimization process can take place in a
time-dependent manner, for example, when it can be assumed that the
harvesting conditions have changed. The restart can take place in a
throughput-dependent manner if, e.g., the harvesting speed has
changed significantly. The restart can take place in a
setting-dependent manner if other settings on the combine harvester
are changed that indicate that the cleaning load has changed
accordingly. The restart can take place in a crop-dependent manner
if, e.g., a change in a crop property such as grain moisture is
measured. If optimization must indeed be repeated, then the
crop-dependent setting values are not taken from the electronic
fieldwork information system to be used as the default values for
the start, but rather target setting values ZG, ZO, ZU that were
computed in the first optimization procedure. Target setting values
ZG, ZO, ZU can also be entered in the electronic fieldwork
information system, of course, in which case the harvesting
conditions are also preferably recorded, to the extent this is
possible, so that target setting values ZG, ZO, ZU computed in an
optimization process can also be used as the starting values in a
subsequent harvesting process in which the harvesting conditions
are relatively similar to those that prevailed when target setting
values ZG, ZO, ZU were computed.
[0120] For safety reasons, the system is designed such that the
driver can manually override one or all of the machine parameters
that were set, at any time during a harvesting operation. Finally,
it is pointed out once more that the combine harvester shown in the
figures, and the control and the specific method described in
conjunction therewith are merely exemplary embodiments that could
be modified in a variety of ways by one skilled in the art, without
leaving the framework of the present invention.
[0121] It will be understood that each of the elements described
above, or two or more together, may also find a useful application
in other types of methods and constructions differing from the
types described above.
[0122] While the invention has been illustrated and described as
embodied in a method for computing a target setting value, it is
not intended to be limited to the details shown, since various
modifications and structural changes may be made without departing
in any way from the spirit of the present invention.
[0123] Without further analysis, the foregoing will so fully reveal
the gist of the present invention that others can, by applying
current knowledge, readily adapt it for various applications
without omitting features that, from the standpoint of prior art,
fairly constitute essential characteristics of the generic or
specific aspects of this invention.
* * * * *