JP3525466B2 - Tillage control device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a tilling depth control device for controlling a tilling depth of a tilling machine to be maintained substantially constant. 2. Description of the Related Art A cultivator mounted on a tractor so as to be able to move up and down freely detects a cultivation depth (hereinafter referred to as a cultivation depth) with a cultivation depth sensor, and according to the detection result, a cultivator. Is appropriately raised and lowered to control the tilling depth to be maintained substantially constant. In the case of a rotary tiller, a detection value of a tillage depth sensor is generally represented by a rear cover angle. The rear cover, which is a cover that covers the rear end of the tilling blade, is attached to the rotatable side so that the lower end always contacts the upper end of the field scene and the leveled field scene is leveled. Therefore, when the tillage depth is shallow, the rear cover will be in a standing state, and when the tillage depth is deep, the rear cover will be in a lying state, so know the tillage depth at that time by measuring the angle of the rear cover. You can do it. [0003] However, since the shape and the mounting state of the rear cover differ depending on the maker and the model, there is a problem when uniformly controlling various rotary cultivators. For example, when the tilling machine A with the rear cover as shown in FIG. 10 and the cultivator B with the rear cover standing down are compared, the rear cover detected by the tillage depth sensor is the same even if the amount of fluctuation of the tillage depth is the same. _A difference occurs in the amount of change in the angle (θ _A <θ _B ). For this reason,
When the dead zone H of the control is set according to the cultivator A,
When the work is performed using a cultivator, the detection value of the cultivation depth sensor deviates from the dead zone even with a slight change in the cultivation depth, so that the output for raising and lowering the cultivator is frequently output, and hunting is likely to occur. When the dead zone is set in accordance with the cultivator of B, when the work is performed using the cultivator of A, the up-and-down output of the cultivator is not output unless a considerably large variation in tillage depth occurs. become worse. The present invention has been made in view of the above circumstances, and it is an object of the present invention to perform appropriate tillage depth control on various rotary tillers having different rear cover shapes and mounting conditions. In order to solve the above problems, the present invention has the following configuration. That is, the tillage depth control device according to the present invention is a tillage depth control that appropriately raises and lowers the tillage device based on the detection result of the tillage depth sensor that detects the tillage depth of the tillage device, and maintains the tillage depth substantially constant. Equipment in Shenzhen
Sensor output is either above or below the target value.
If it changes within the box, the range is
Upward peak of detection output when changing on the upper side
If it is changing below the target value,
Detecting the downward peak of the detection output,
Control a value smaller than the tillage sensor value at the peak
And the detection output exceeds the target value
When the dead zone changes to the opposite side, reset the dead zone
And then set the dead zone the next time a peak is detected.
It is characterized by having such a configuration . [0006] [act] The tilling depth control device, which is configured to re-set the dead zone width in accordance with the control the size of the detected output of the tilling depth sensor, specifically, tilling depth target value Across
In the range of the side (the deeper side) or the lower side (the shallower side)
If the detection output is changing at
From the tillage depth sensor value at the peak.
Is set as the control dead zone width. on the other hand,
The detection output of the tillage depth sensor changes to the opposite
Resets the dead zone, then
When the peak is detected, reset the dead zone. The book
In the embodiment, the upper side of the cultivation depth target value is the side where the cultivation depth is deeper.
And the lower side of the target value indicates the shallower side.
Is clear from FIG. Thus, when the operating range of the operating portion for detecting the tillage such as the rear cover is large, the dead zone width becomes large, and when the operating range is small, the dead zone width becomes small, and the dead zone width suitable for the tiller is set. Will be done. Therefore, it is possible to maintain high control responsiveness and perform plowing depth control that does not cause hunting for various rotary tillers having different rear cover shapes and mounting states. An embodiment of the present invention will be described below in detail with reference to the drawings. FIG. 1 is a diagram showing a use state of a rotary cultivator, in which a cultivator 3 is mounted on a working machine connecting device 2 of a tractor 1. The working machine connecting device 2 in the illustrated example is a 3P hitch, 5 in the figure is a pair of left and right lower links, 6 is a single top link in the center of the left and right, and a working machine is provided at the rear end of these links 5, 5, and 6. 3 is attached. Reference numeral 7 denotes a lift arm which is driven by a hydraulic device provided in the tractor 1, and lower links 5 and 5 are suspended from rear ends of the lift arm via lift rods 8, 8, respectively. By rotating the rotary till vertically, the rotary tiller 3 moves up and down. In addition, one of the left and right (in the illustrated example, the right side) lift rod 8R is extendable, and by changing the length of the lift rod, the amount of left and right inclination of the tiller 3 with respect to the tractor 1 is adjusted. I have. The rotary cultivator 3 is configured to cultivate a field by rotating a cultivating blade 10 having a plurality of blades planted on a rotating shaft in a predetermined direction. A rear portion (rear cover) 12 of a rotary cover 11 that covers the periphery of the tilling blade 10 is attached to the rotary cover body so as to be rotatable up and down.
It is urged downward by a spring or the like. By moving the lower part of the rear cover 12 while being pressed against the field scene as the aircraft advances, the field scene after tilling is leveled and leveled. The cultivator 3 is provided with a tillage depth sensor 15 represented by a rear cover angle. The tractor 1 is provided with a tillage depth setting device 16. Then, the actual measured value of the plowing depth detected by the plowing depth sensor 15 is compared with the target value of the plowing depth set by the plowing depth setting device 16, and the plowing depth control is performed so that the actual measured value approaches the target value. The tillage depth control device for controlling the tillage depth is configured as shown in FIG. That is, signals from the tillage depth sensor 15 and the tillage depth setting device 16 are input to the CPU 19 via the input interface 18, the CPU 19 performs processing described later based on these data, and the lift arm via the output interface 20. Lift valve 21 for vertical movement
And a valve opening / closing signal is output to the descending valve 22. The data processing in the CPU 19 is performed in the order shown in the flowcharts of FIGS. The "deviation peak" in the figure refers to the tillage depth sensor 1
The displacement point at which the direction of the detection output 5 changes is detected. When this deviation peak is detected, the dead zone width of the control is calculated by the following two calculation methods. One is by fuzzy inference.
Deviation and input variables the rate of change of the deviation, the dead zone width deviation peak side target value as a reference W _FP, and deviation peaks opposite the dead zone width W of the tilling depth sensor value and the tilling depth target value
Fuzzy inference is performed using _FA as an output variable. The dead zone width calculation method based on this fuzzy inference will be described later. Another method is to calculate a value obtained by multiplying the deviation peak value by a coefficient (for example, 0.9) smaller than 1 to obtain a dead zone width W on the deviation peak side.
_M is the method. [0015] Then, by comparing the absolute value of the dead zone width W _M of the deviation peak side determined from the dead zone width W _FP and deviation peak value of the deviation peak side determined by fuzzy inference, the larger dead zone numerically deviation peak side of Applied as width W. For the dead zone width on the side opposite to the deviation peak, the dead zone width W _FA obtained by fuzzy inference is applied as it is. For example, when the sensor value is larger than the target value (the tillage depth is too deep), in other words, when there is a deviation peak on the “(tilling cultivator) raising request side”, the dead zone width of the “raising request side” is W And set
To set the W _FA to the dead zone width of "lowering demand side". Also,
On the side where the sensor value is smaller than the target value (the plowing depth is too shallow),
In other words, when there is a deviation peak on the "reduction request side",
Set the W _FA to the dead zone width of the "up request side", to set the W in the dead zone width of "lowering demand side". When a new deviation peak is detected, the dead zone width is obtained in the same manner as described above, and the dead zone is reset with the dead zone width based on the new deviation peak. Thus, every time a deviation peak is detected, the dead zone is sequentially changed, and the cultivator is moved up and down in accordance with the dead zone at that time, so that for the same plowing depth, a large dead zone and a narrow operation for the rear cover having a wide operation range. For those having a range, a small dead zone is set. That is, since the dead zone is set according to the ratio of the operation range of the rear cover, smooth control with excellent responsiveness can be performed. Further, when the sensor value reaches the deviation peak value when it is near the tillage target value, the dead zone based on fuzzy inference is applied, so that a dead zone having a certain width is set, so that hunting and the like can be prevented. . Further, if the deviation changes in the opposite direction,
The dead zone up to that point is reset, and the dead zone is set again when a deviation peak is detected next. By resetting the dead zone at the time of deviation reversal, if the previous setting dead band was small, the output immediately after deviation reversal would be output immediately, or if the previous setting dead band was large, the output after deviation reversal would be difficult. We prevent that we do not come out. Next, how the dead zone changes when the tillage depth sensor value changes as shown in FIG. 5 will be specifically described. First, the deviation peaks at point a. However,
At this time, it is not clear whether or not the deviation peak is present, and it is clear that the point a is the deviation peak at the point b where the sensor value is reduced to 9/10. Then, determine the dead zone width in a manner based on the method and the deviation peak value by fuzzy inference, it sets the dead zone H _1. In this case, since a W _M (absolute value)> W _FP (absolute value), the dead zone width deviation peak side is set to W _M. The dead zone width of the deviation peak on the opposite side is the W _FA. Thereafter, maintains its dead band, then the greet deviation peak c, then the sensor value is set a new dead zone H ₂ at d point was reduced to 9/10. In this case, since W _M (absolute value) <W _FP (absolute value), the dead zone width on the deviation peak side is W _FP . The dead zone width of the deviation peak on the opposite side is the W _FA. The direction of the deviation is reversed at point e, and the dead zone up to that point is reset. And
When the first deviation peak f is reached after the deviation inversion, the dead zone H ₃
Is set. Next, a method of calculating a dead zone width by fuzzy inference will be described. As described above, the deviation D of the cultivation depth sensor value from the cultivation depth target value and the rate of change △ D are input variables (conditional part membership function), dead zone widths W _FP and W _FP .
_{Let FA} be the output variable. When performing fuzzy inference, fuzzy control rules shown in Table 1 are employed. The horizontal row is the deviation D, the vertical row is the change speed ΔD, and the dead zone widths W _FP and W _FA are shown in the table. Note that the upper part is the deviation peak side dead zone width W.
_FP and the lower part are the dead band width W _FA opposite to the deviation peak. Here NB is greater in the negative, N S is small in the negative, Z0 zero,
PS means positive and small , and PB means positive and large. The deviation D is represented by the angle of the rear cover 12, the intermediate point of the rotation angle of the rear cover is ZO, and the rear cover becomes more horizontal as PS and PB, and the rear cover becomes closer as NS and NB. [Table 1] From Table 1, for example, for one row and one column,
"If the rear cover is standing (NB) and the rear cover is rotating downward at a large speed (NB), the dead zone width on the deviation peak side is widened (PB) and the opposite side of the deviation peak is Narrower dead zone width (NS)
Please. ". Regarding the second row and the first column, "If the rear cover is standing (NB) and the rear cover is rotating downward at a small speed (NS), the dead zone width on the deviation peak side is also the deviation peak. The width of the dead band on the other side is slightly wider (P
S) Do it. ". Regarding 5 rows and 1 column, "If the rear cover is standing (NB) and the rear cover is rotating upward at a high speed (PB), the dead zone width on the deviation peak side is also the deviation peak. Make the opposite dead zone wider (PB). " Regarding 1 row and 5 column, "If the rear cover is horizontal (PB) and the rear cover is rotating downward at a large speed (NB), the dead band width on the deviation peak side and the opposite side of the deviation peak are shown. Is maintained at the reference width (Z
O) Do it. ". Regarding 5 rows and 5 columns, "If the rear cover is facing sideways (PB) and the rear cover is turning upward at a high speed (PB), the dead zone width on the deviation peak side is narrowed (NB). At the same time, widen the dead zone width opposite to the deviation peak (PB). " The deviation D, the change speed ΔD, and the dead zone widths W _FP and W _FA are classified according to their sizes as shown in FIGS. 6 to 8 and are represented as membership functions. Assuming that the deviation D is d and the change rate ΔD is Δd, the dead zone width W _FP on the deviation peak side in this case is determined (see FIG. 9). At this time, the membership value of the deviation D is a ₁ given by NS and a ₂ given by ZO, and the membership value of the change rate ΔD is b ₁ given by PS and b given by ZO.
₂ The membership areas U ₁ and U ₂ of the dead zone widths W _FP and W _FA are obtained according to the fuzzy rules, and the position of the center of gravity of these areas in the horizontal axis direction is set as a correction amount C. Dead zone width W _FA opposite the deviation peaks in a similar manner can also be determined. As described above, in the tillage depth control device according to the present invention, the dead zone width of the control is set according to the operation range of the tillage depth detecting operation part such as the rear cover. For various rotary cultivators with different rear cover shapes and mounting conditions, maintain high control responsiveness, and
It became possible to control plowing depth without hunting.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a side view of an agricultural working machine according to one embodiment of the present invention. FIG. 2 is a block diagram of a tillage depth control device. FIG. 3 is a first flowchart of the tillage depth control device. FIG. 4 is a second flowchart of the tillage depth control device. FIG. 5 is a time chart of a tillage depth sensor value and a dead zone. FIG. 6 is a diagram illustrating a deviation membership function. FIG. 7 is a diagram showing a membership function of a change speed of a deviation. FIG. 8 is a diagram showing a membership function of a dead zone width. FIG. 9 is a diagram illustrating a procedure of fuzzy inference. FIG. 10 is a time chart of a tillage depth sensor value. [Description of Signs] 1 Tractor 2 Work implement coupling device 3 Rotary cultivator (tiling work machine) 7 Lift arm 12 Rear cover 15 Tillage depth sensor 16 Tillage depth setting unit 19 CPU (dead zone width setting means)

Claims (1)

(57) [Claims] [Claim 1] Cultivation that appropriately raises and lowers the tillage device based on the detection result of the tillage depth sensor that detects the tillage depth of the tillage device, and maintains the tillage depth approximately constant. in depth control device, plowing
If the output of the depth sensor is above or below the target
If it changes within the range, the range is
If it changes above the upper limit of the detection output,
Changes below the working depth target
Detects each downward peak of the detection output and
Value smaller than the tillage sensor value at the peak
Set as the dead zone width of the control, and the detection output
When it changes to the opposite side after passing, the dead zone
Set, then set the dead zone when the next peak is detected
A tillage depth control device characterized in that it is configured to re- drill.