JPS6146842B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an automatic control device in which a dead zone area is set.

Generally, this type of automatic control device is configured such that the controlled amount is automatically adjusted and the automatic adjustment ends when the amount falls within a preset dead zone area. However, with this type of control means, automatic adjustment ends even when the controlled quantity approaches the limit of the dead band area, so if the controlled quantity changes even slightly due to disturbance or other reasons, it will immediately go out of the dead band area. It becomes like this. In this case, even if automatic adjustment is started again, the automatic control will inevitably be extremely unstable.

To solve this problem, some devices are designed to narrow the dead zone when the controlled amount exceeds the dead zone. However, this is not a particular problem when the deviation output is always given as an input, but in control that obtains the deviation output by sampling, the deviation output becomes step-like, so it always falls within the narrowed dead zone. There is no such guarantee, and if the vehicle does not enter the narrow dead zone, hunting will continue to occur in the narrow dead zone.

An object of the present invention is to ensure stability of control in an automatic control device in which a dead zone area is set.

Embodiments of the invention will be described with reference to the drawings. The illustrated embodiment is an example in which the output rotation speed of the continuously variable transmission is automatically controlled to match a preset target rotation speed. In the figure, 1 is an input terminal to which input pulse a is applied, 2 is a frequency dividing circuit, 3 is an oscillation circuit that generates clock pulse CP ₀ , 4 is a setting device for digitally setting the target rotation speed, 5 is a multi-channel rate plier, 6 is an AND circuit, 7 is a reversible counter, 8 is a detector for detecting the contents of the reversible counter, 9 is a setting device for setting a dead zone, such as a digital switch, 10 is a comparator, 1
1 is a timing circuit, 12 and 13 are latch circuits (for example, a D-type flip-flop), and 14 and 15 are
1 is an AND circuit, and 16 is a setting device for presetting the reversible counter 7, such as a digital switch.

The input pulse a is a pulse generated from a rotary encoder (which may also be a pulse generator) driven by the output rotation of the continuously variable transmission. Here, to simplify the explanation, it is assumed that the rotary encoder used is configured to generate 60 pulses every time the continuously variable transmission rotates once. Of course, it is not limited to this. The setting device 4 sets the target rotation speed in rpm units. For example, if the target rotation speed is set as 1000 rpm, this 1000 is given to the multi-rate plier 5, which outputs a clock pulse CP ₀ to produce 1000 M pulses (M is 1 or more) per unit time (this is 1 second). ) is the number of pulses.
It will now be output as CP ₁ . in particular,
The oscillation frequency of clock pulse CP ₀ is 1MHz,
If M is 100, the frequency of pulse CP ₁ is 100KHz. The setting value N by the setting device 16 is arbitrarily set depending on the accuracy to be obtained, but in this example, the maximum value of the target rotation speed is 4 digits, so the setting value N is the same number of digits as 1000 (=N) It is as follows. The frequency dividing value of the frequency dividing circuit 2 is set to N/M based on the relationship between M and N described above. From the previous example, N=1000, M=100
Therefore, the frequency division value is 10. That is, the frequency dividing circuit 2 outputs a pulse b whose width is equal to 10 pulses of the input pulse a. Therefore, if the continuously variable transmission is rotating at 1000 rpm, then from the above example, the width of pulse b is N/M x 1/1000.
= 1/100 (second). Pulse b is applied to the timing circuit 11, and when the falling edge of pulse b is detected, pulse c is applied after some time, followed by pulse d.
issue. Pulse C is used as a latch signal for latch circuits 12 and 13, and also serves as a sampling signal. The pulse d is applied to the frequency dividing circuit 2, which is forcibly reset, and the next input pulse a and pulse b are set to H level again to repeat the previous frequency dividing operation. In terms of the original frequency dividing operation, this means that 10 pulses of input pulse a are divided into pulses b.
should be at the L level, but by forcibly resetting it as described above, the repetition period of the gate time, which will be described later, is accelerated.
The pulse d is also given as a preset read signal to the reversible counter 7, and this signal causes the setting device 1 to
A set value N (=1000) set to 6 is preset.

Pulse b and pulse CP ₁ are input to the AND circuit 6, and while pulse b is at H level, pulse CP ₁
is output as pulse CP ₂ . Therefore, the number of pulses CP ₂ per unit time (during the period when pulse b is at H level) is equal to the target rotation speed determined by the setting device 4.
If Sn and the output rotation speed of the continuously variable transmission are Rn, then N/M×1/Rn×Sn×M=NSn/Rn. If the output rotational speed matches the target rotational speed Sn, the number of pulses _CP2 per unit time will be N, which will match the preset value.

When the reversible counter 7 reaches zero,
Output f. When the detector 8 receives this output f, it outputs an output g meaning an instruction to increase or decelerate. This output g is reset when the pulse d enters the detector 8 as a reset signal. The output g is given to the reversible counter 7 for determining the reversible mode. That is, if the output g is at the L level, the reversible counter 7 is in the down mode, and when the output f is generated and the output g is at the H level, the reversible counter 7 is in the up mode. Therefore, the reversible counter 7 has a content of N
If the pulse CP ₂ from the AND circuit 6 is input from the time when the output is preset to
is not output, the output g is at L level, and the reversible counter 7 is in down mode, so the pulse CP ₂
is subtracted from the preset value N. If the target rotation speed and the output rotation speed match, the number of pulses from the AND circuit 6 per unit time is N as described above, so when the unit time has elapsed, the contents of the reversible counter 7 becomes zero. Even if the output f is produced at this time, the pulse d is subsequently produced, so the output g is still at the L level. When the output rotation speed is lower than the target rotation speed, the width of pulse b becomes wider, so that the number of pulses CP ₂ becomes N or more during one pulse period of pulse b, and by the time the reversible counter 7 is preset next time, Its content becomes zero. When this becomes zero, the output g causes the reversible counter 7 to enter the up mode, and the subsequent pulses _CP2 are counted up.

The absolute value of the difference between the preset value of the reversible counter 7 and the number of pulses CP ₂ is outputted from the reversible counter 7 as an output h and inputted to the comparator 10. The comparator 10 outputs a set value i for setting the dead zone area.
and is compared with the output h. Assume that the dead zone is set to ±1%. Preset value N
If is 1000, 1% of it is 10, so the output i
has a value of 10. Then, the comparator 10 outputs an output j indicating a shift instruction, which goes to the H level in a range where the output h is larger than the output i, and goes to the L level in the opposite range. Therefore, when the target rotation speed and the output rotation speed match, or when the output rotation speed is within the dead band region with respect to the target rotation speed,
The time when pulse c is generated is at L level. Further, as described above, the output g is at the L level at this time. As a result, since the stored values in both the memory circuits 12 and 13 are "0", no output is output from the AND circuits 14 and 15. This means that the continuously variable transmission does not perform any acceleration or deceleration operations.

As mentioned above, when the output rotation speed is lower than the target rotation speed, the output f is output within the width of one pulse of pulse b, the reversible counter 7 becomes the up mode, and the subsequent pulse CP ₂ is counted up. go. Since the output h is the absolute value of the difference between the preset value N and the pulse CP ₂ , if the output rotation speed is outside the dead zone region, the output j will be H at the time when the pulse c is generated.
It should be at the level. Also, at this time, the output g is at H level. Therefore, when the pulse c is output, the H level of the outputs g and j is stored in the memory circuits 12 and 13, and the outputs k and m are output from this. Therefore, an output n is output from the AND circuit 14. This output is used as a signal to adjust the continuously variable transmission to increase speed. Specifically, the output n is used to rotate the speed change handle of the continuously variable transmission in the speed increasing direction.

Conversely, when the output rotation speed is higher than the target rotation speed,
Since the width of the pulse b becomes narrower, the output f is not produced within one pulse width of the pulse b, and the reversible counter 7 is always in the down mode. If the pulse b is not completed before the difference between the preset value N and the number of pulses _CP2 becomes 10 or less, the output j is still at the H level. When the pulse c is output here, the H level of the output j is stored in the storage circuit 13, and an H level output m is output. At this time, the output g is at L level, so the memory circuit 12 outputs L.
Since the output that is inverted from the level output k is output as an H level, the output from the AND circuit 15 is 0.
coming out. Contrary to the above, this output is used as a signal for decelerating the continuously variable transmission. As a result of the above, the output rotation speed is controlled so as to reach the target rotation speed (actually, so as to reach within the dead zone region).

Next, the operation of the above configuration will be explained based on specific numerical values. Now assume that N is 1000, M is 100, 60 pulses are given as input pulse a every time the output rotating shaft rotates once, and the dead zone is set to ±1%. When the target rotational speed is 1000rpm, if the output rotational speed is 1000rpm, the output h will be zero. Therefore, the continuously variable transmission is not increased or decreased in any way. However, for example, the output rotation speed is 900 rpm.
Assuming that, the width of pulse b is 1/900×N/M=11.11 (m seconds). The number of pulses CP ₁ that enter within this time is 1/900×N/M×1000×M=1111 pulses.
Therefore, by the time pulse b falls, reversible counter 7
There are times when the content of becomes zero, and from this time on, it becomes up mode. Furthermore, since the output h when the pulse b falls is 1111-1000=111, the output j is at the H level, so the output n is output and the speed increase adjustment is started. Conversely, if the output rotation speed is 1050 rpm, the width of pulse b is 1/1050 x N/M = 9.52 (m seconds) The number of pulses CP ₂ that enters during this period is 1/1050 x N/M x 1000 x M = 952 pulses, output h is 1000-952 = 48, and when pulse b falls, output j is at H level, and output g of detector 8 is at L level, so pulse c is output. At this time, an output O is output from the AND circuit 15 and deceleration adjustment is started. If the output rotational speed is within the dead band region, for example 990 rpm, the width of pulse b is 10.10 (ms) and the number of pulses CP ₂ during this period is 1010 pulses. Therefore, the output h is 1010
Since -1000=10, the output j remains at L level. Therefore, neither outputs n nor O are output.
Similarly, when the output rotation speed is 1010 rpm, the width of pulse b is 9.90 (m seconds), and the pulse CP ₂ that falls between this
The number of pulses is 990. Therefore, the output h is
1000-990=10, and the output j remains at the L level at this time as well. In other words, if the difference between the output rotation speed and the target rotation speed is within ±1% of the target rotation speed, no increase/deceleration adjustment is performed, and the difference between the output rotation speed and the target rotation speed is within ±1% of the target rotation speed.
If so, the number of pulses CP ₂ in one pulse of pulse b is 1010 pulses, or if it is -1%, it is 990 pulses.

Next, assume that the target rotation speed is set to 100 rpm under the same setting conditions as above. This is the case when the gear ratio of the continuously variable transmission is 10:1. If the dead zone area at this time is also ±1% as before, then
The output i given from the setting device 9 to the comparator 10 is 10
It is. Assuming that the output rotation speed is now 100 rpm (that is, the rotation is equal to the target rotation speed), the width of pulse b is 1/100 x N/M = 100 (msec) Pulse CP ₂ that enters during this period The number of pulses is 1/100×N/M×100×M=1000 pulses. As mentioned above, no increase/deceleration adjustment is performed at this time.

Since the dead band area is assumed to be ±1%, if the current output rotation speed is 99 rpm, pulse b
The width of is 101.01 (msec), and the number of pulses CP ₂ that enters this period is 1010 pulses. This corresponds to the number of pulses CP ₂ in the example when the target rotation speed is 1000 rpm and the output rotation speed is 990 rpm, which is 1% lower. Therefore, no increase or deceleration is performed at this time. If the output rotational speed becomes even lower than this, the output n will be produced, which is exactly the same as in the above example. Conversely, if the output rotation speed is 101 rpm, the width of pulse b is 99.01 (ms), and the number of pulses CP ₂ that enters this period is
990 pulses. This case also matches the number of pulses CP ₂ when the target rotation speed is 1000 rpm and the output rotation speed is 1010 rpm, which is 1% higher than the target rotation speed in the previous example. Therefore, at this time as well, no increase or deceleration is performed, and if the output rotational speed increases further, the output will be 0, just as in the above example. In this way, no matter how the target rotation speed is changed, the set dead band areas will all have the same percentage, and there will be no setting error, and even if the target rotation speed is set over a wide range, the same accuracy will be obtained in all cases. It becomes like this.

Here, Figure 2 shows the case when the target rotation speed and output rotation speed match, Figure 3 shows the case when the target rotation speed is higher than the output rotation speed, and Figure 4 shows the case when the target rotation speed is the output rotation speed. Each time chart is shown below. The embodiment shown in the figure uses a reversible counter, but instead of this, a down pulse counter and an up pulse counter are prepared, N is preset in the down pulse counter, and the input pulse from the AND circuit 6 is The overflow signal of the down pulse counter may be first supplied to the down pulse counter, and when this overflows, the subsequent input pulses may be counted by the up pulse counter.In this case, the overflow signal of the down pulse counter is supplied to the output The count value of the up pulse counter may be used as the output h.

In the embodiment shown in the figure, the target rotation speed is treated as a fixed value set by the setting device 4, but it can be changed independently of the controlled variable, for example, by following the rotation output of another continuously variable transmission. Control is also possible, in which case the rotational speed of the other continuously variable transmission becomes the set target value. Therefore, the pulses per unit time are determined from the pulses from the pulse generator or rotary encoder driven by this rotation, and this is supplied to the multi-rate pliers 5 instead of the output from the setting device 4 to set the target. It is sufficient to configure a value setting section. Further, the number of pulses per revolution of the pulse encoder driven by the another continuously variable transmission, and the number of pulses per revolution of the pulse encoder driven by the continuously variable transmission generating the input pulse a. If the ratio with the number of pulses is set to 1:N according to the above example, the pulse from the pulse encoder driven by the other continuously variable transmission is directly used as pulse CP ₁ and the AND circuit is used. 6, and the target value setting section may be configured using this input.

The above embodiments are all examples of automatically controlling the output rotation speed of a continuously variable transmission, but the invention is not limited to this, and voltage, current, electric power, flow rate, etc. are the controlled variables, and these are It is possible to perform automatic control to adjust to a set target value.

Now, according to the above configuration, when the output rotation speed is within the dead zone region with respect to the target rotation speed, neither outputs n nor O are output. For example, if the target rotation speed is
At 1000 rpm, the dead band area is ±1%, and the output i
is ±10, when the continuously variable transmission is increased to 990rpm or more, or when it is decelerated.
When the speed becomes 1010 rpm or less, speed increase adjustment or deceleration adjustment can no longer be performed. This is because the output j will no longer be output. However, even if the output rotation speed slightly enters the lower sensing zone and becomes stable, if it remains stable near the limit of the range of the dead zone, then due to disturbances etc. may go up or down, which may cause control to become unstable. To avoid this, even if a dead zone area is set,
Once speed increase or deceleration is started, the output rotation speed may be stabilized at a value close to the center of the dead zone region.

For this purpose, in addition to the setting device 16, a preset value (N
-a) and the setting device 16a (N+a) for setting
A setting device 16b for setting is prepared. Here, a is set within a range smaller than the output i. To give a concrete example, if N is 1000 and i is ±10, then a
For example, set to 5. Therefore, the setting device 16
The setting value of a is 955, and the setting value of setting device 16b is 1005.
becomes. The setter 16a is triggered by the output n and its set value is preset into the reversible counter 7, and the setter 16b is triggered by the output O and its set value is preset into the reversible counter 7. The output nO is input to a NOR circuit 17, and when both inputs are absent, the output from the circuit 17 triggers the setter 16 to preset the set value in the reversible counter 7. Suppose now that the output rotation speed is lower than the target rotation speed, so that an output n is output and the engine enters a speed increasing state. At this time, the setter 16a is triggered and the preset value of the reversible counter 7 becomes 995 in accordance with the above example. Therefore, the output rotational speed is gradually increased, and the output f is produced earlier by the difference in preset values (1000-995). In other words, the output h will increase by the amount added to the difference in preset values, and the error will appear to be large, and the error will be corrected by the amount added to the difference in preset values than the normal dead band area. It turns out. Therefore, since the output h is larger than the output i from the comparator 10, the output j continues to be output even at H level. Conversely, if the output rotation speed is higher than the target rotation speed and the deceleration state is entered, the setting device 16b is triggered by the output O, and the preset value of the reversible counter 7 is set to
It becomes 1005. Therefore, even if the output rotation speed has decreased to 1010 rpm, the deceleration will not stop.
The deceleration is stopped only when the speed drops to 1005 rpm. Once the increase/deceleration is stopped, the setter 16 is triggered and the reversible counter 7 is preset to 1000 for the first time. Thereafter, control will be performed according to the dead zone area set by the output i. With such control, extremely stable control can be expected.

In addition, in this kind of control, there is a method in which the dead band area is narrowed, and when the output rotation speed enters the narrowed dead band area range, it is returned to the original dead band area.According to this, Due to the narrowing of the dead zone area, the corrective action may be performed excessively during the next correction, thereby passing through the dead zone area and deviating from the dead zone area, which may result in hunting. To prevent this, the dead zone region cannot be made too narrow, and therefore it becomes extremely difficult to achieve the original control stability.

In this regard, if the entire dead zone area is moved in the positive or negative direction by changing the preset value of the reversible counter 7 without changing the width of the dead zone area as in the present invention, it is possible to move the entire dead zone area in the positive or negative direction. Since the value remains at the initial value, even if the correction is made excessively, there is less risk of the correction passing into the dead zone region. Therefore, the original control can be carried out stably, which is convenient.

As detailed above, according to the present invention, when the adjustment of the controlled quantity is started, the dead zone area is simply moved without changing the width of the dead zone area, and the controlled quantity is moved. When entering the dead zone area, the dead zone area is moved again to return to the dead zone area set from the beginning, so that stable control can be expected.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the invention, and FIGS. 2 to 4 are time charts for explaining the operation. 1... Input terminal, 2... Frequency divider circuit, 4... Target value setter, 5... Multi-rate pliers, 6...
AND circuit, 7... Reversible counter, 8... Detector, 9... Dead band setter, 10... Comparator, 11
...timing circuit, 12,13...memory circuit,
16, 16a, 16b...preset value setter,
17...Noah circuit.

Claims (1)

[Claims] 1 a dead zone setting device, an arbitrarily set first value N, and a second value that is smaller than the first value within a certain range set by the dead zone setting device and is also smaller than the first value. a counter device that is selectively preset to a larger third value; a pulse generator that generates a number of pulses per unit time that is M times the target value (where N>M); and a pulse generator that generates pulses corresponding to the controlled quantity. A frequency dividing circuit that takes as input and divides this pulse into N/M, and a pulse from the frequency dividing circuit as input, and from when this input pulse ends, the next pulse corresponding to the controlled amount is generated. a timing circuit for generating a latch signal and a subsequent reset signal for resetting the counter device; and a timing circuit for generating a latch signal and a subsequent reset signal for resetting the counter device; an input circuit that inputs 0 to the counter device, and a preset number of pulses output from the input circuit during the period when the pulse from the frequency divider circuit is output at the time of generation of the latch signal at which sampling is performed. a circuit for issuing a command to change a controlled quantity when the difference between the first value N and the first value N exceeds a range of a dead band region set by the dead band setting device; Based on this, when the latch signal is generated, if the number of pulses from the input circuit is in a state before reaching the preset value N, a reduction command is issued to decrease the controlled amount, and the preset value N is a circuit that issues an increase command to increase the value when the value exceeds the value; and a circuit that presets the second value in response to the increase command and the third value in response to the decrease command in the counter device. Automatic control device consisting of.