JP3011275B2 - Home position learning method of actuator - Google Patents

Description

The present invention relates to a home position learning method for an actuator such as a stepping motor.

[Prior Art] For example, in a fuel injection system of a diesel engine, a stepping motor is used as an actuator for controlling a fuel injection amount. This stepping motor rotates by the number of steps determined by the number of pulses output from the control means. Once the home position has been determined, the position control can be performed accurately without the need for a feedback signal. Can be.

FIG. 6 shows an example of a fuel injection system for a diesel engine. In this example, the operation of the control member of the fuel injection pump 1 is controlled by the stepping motor 2. The output shaft of the stepping motor 2 is connected to a potentiometer 3 for learning the home position and detecting step-out. The potentiometer 3 outputs a voltage corresponding to the rotation amount of the stepping motor 2, and its output signal is input to the control unit 4.

In addition, the reference numeral 5 indicates an engine speed detecting device,
Reference numeral 6 denotes a water temperature sensor, and the control unit 4 controls the water temperature sensor based on input information from these detection elements.
A control signal is output to control the stepping motor 2.

In such a system, the home position of the stepping motor 2 is first learned when the start switch is turned on. The method in that case is usually performed as follows. That is, while monitoring the output of the potentiometer 3, the number of input steps to the stepping motor 2 is increased or decreased, a position where the output of the potentiometer 3 matches a certain learning reference value Q is set as a home position, and the position is set to, for example, 100 steps. Position, and so on.

By the way, in a system in which the rotational position of the stepping motor 2 is monitored by the potentiometer 3 as described above, since mechanical coupling means, for example, a link mechanism or the like is interposed between the stepping motor 2 and the potentiometer 3, There is play that cannot be avoided mechanically. Therefore,
Depending on the operation direction of the stepping motor, a certain deviation may occur between the rotational position of the stepping motor (this is proportional to the number of input steps, and hence will be appropriately referred to as “number of input steps”) and the output of the potentiometer. Such a relationship is called a relationship having hysteresis. In particular, in the case of a system having a large deviation, in other words, a system having a large hysteresis, the following problem occurs when learning the home position.

[Problems to be Solved by the Invention] First, before describing the problem, the features of this system will be described. As shown in FIG. 7, in this system, the number of input steps and the potentiometer A relationship having a hysteresis as indicated by a dotted line is established between the output and the output. That is, the output P of the potentiometer changes along the lower curve m as the number of steps increases, that is, as the stepping motor rotates in the forward direction. Further, as the number of steps decreases, that is, as the stepping motor rotates in the reverse direction, it changes along the upper curve n. Further, when the rotation direction of the stepping motor is switched, there is a certain area where the potentiometer does not completely follow the stepping motor at the beginning of the operation.

In a system having such a relationship, when learning the home position, there is a problem that the home position to be learned differs depending on the rotation direction of the stepping motor. That is, while the stepping motor is rotated in the forward direction, the potentiometer output P
Is different from the learning reference value Q, and the home position is different between the case where the potentiometer output P matches the learning reference value Q while rotating the stepping motor in the reverse direction. I will.

This will be specifically described with reference to FIG. In this example, the potentiometer output P during normal use is 1 V to 3 V
, And the home position learning reference value Q is 3V.

In this system, when the initial value X of the potentiometer output P when the switch is turned on is smaller than the learning reference value 3V (in the case of -1, -2), the stepping motor is rotated in the forward direction, and the potentiometer output becomes The position A when the voltage becomes 3V is set as the home position.

When the initial value X is a value indicated by -3, the position B is set as the home position. Further, when the initial value X is larger than the learning reference value 3V (in the case of -4 and -5),
By rotating the stepping motor in the reverse direction, the position C when the output P of the potentiometer becomes 3V is set as the home position.

As described above, according to the conventional method, since the rotation direction of the stepping motor at the time of home position learning is not unified, especially in the case of a system with a large hysteresis, the home position differs depending on how the initial value appears, There was a problem that accurate home position learning could not be performed.

The present invention has been made in view of the above circumstances, and has as its object to provide a home position learning method of an actuator that can accurately learn a home position even in a system having a large hysteresis.

[Means for Solving the Problems] The method of the present invention solves the above problems, and the features thereof are as follows.

First, in the system to which the method of the present invention is applied, a potentiometer is directly or indirectly connected to the output of the actuator, and there is a relationship having a certain hysteresis between the input to the actuator and the output of the potentiometer. In the initial stage of switching the operation direction of the actuator,
In this system, there is a region where the potentiometer imperfectly follows the operation of the actuator, and the width of the output change of the potentiometer corresponding to the maximum width S of the region is measured as T.

Then, for this system, the output P of the potentiometer
When the actuator is operated such that the output P of the potentiometer is equal to the learning reference value Q while monitoring the position, the position where the output P of the potentiometer matches the learning reference value Q is learned as the home position of the actuator. Execute as follows.

First, in the present invention, the operation direction of the actuator for home position learning is set to the forward direction to approach the learning reference value Q from the side where the output P of the potentiometer is small, or to the side where the output P of the potentiometer is large when the direction is reversed. Is determined in advance to make it closer to the learning reference value Q.

(I) When it is determined that the output value P of the potentiometer approaches the learning reference value Q from the smaller side, the following is executed depending on the magnitude of the initial value X of the potentiometer.

When X ≦ (Q−T), the actuator is operated in the forward direction so that the output P of the potentiometer becomes large.

If (QT) <X, the actuator is operated in the reverse direction until the potentiometer output P becomes equal to or less than (QT), and then the actuator is operated in the forward direction.

(Ii) When it is determined to approach the learning reference value Q from the side where the output P of the potentiometer is large, the following is executed depending on the magnitude of the initial value X of the potentiometer.

When (Q + T) ≦ X, the actuator is operated in the reverse direction so that the output P of the potentiometer becomes smaller.

When X <(Q + T), the actuator is operated in the forward direction until the potentiometer output P becomes equal to or more than (Q + P), and then the actuator is operated in the reverse direction.

Thereby, the output P of the potentiometer is made to coincide with the learning reference value Q, and the coincident position is learned as the home position of the actuator.

[Operation] In the present invention, since the actuator is finally operated in the unified direction regardless of the initial value of the potentiometer, a constant home position can always be learned.

Embodiment An embodiment in which the present invention is applied to the fuel injection control system shown in FIG. 6 will be described below with reference to FIGS. 1 to 3.

In the system targeted here, a potentiometer is connected to the output shaft of the stepping motor via a link mechanism, and between the number of input steps to the stepping motor and the output of the potentiometer, in a normal use range, A relationship having hysteresis as shown by a dotted line in FIGS. 1 and 2 is established.

In this system, when the stepping motor is switched from the reverse rotation to the forward rotation and vice versa, the output P of the potentiometer is normal to the rotation of the stepping motor only in the initial region of the operation. , And follows the area in a regular, directly proportional relationship. Now, let S be the maximum width of the incompletely following area. Further, the width of the change in the output of the potentiometer corresponding to the area S measured in that case is T. In this example, the normal output range of the potentiometer is 1 V to 3 V, and T = 0.2 V has been observed.

To learn the home position of the stepping motor for such a system, the output P of the potentiometer is monitored while the output P of the potentiometer is adjusted to the learning reference value Q (here, Q = 3V is the learning reference value). The stepping motor is operated so that
In the present invention, first, learning is performed by operating the actuator in the forward direction, or learning is performed by operating the actuator in the reverse direction. In this embodiment, it is determined that learning is performed by operating in the forward direction.

Then, the learning operation is performed as follows according to the initial value X of the potentiometer when the switch is turned on.

(I) As shown in FIG. 1, X ≦ (Q−T), that is, X
When ≤2.8 V, the stepping motor is rotated in the forward direction so that the output P of the potentiometer becomes large. In the case of the route on the figure, it becomes, '→→.

(Ii) As shown in FIG. 2, when (Q−T) <X, that is, 2.8 V <X, the output P of the potentiometer is temporarily set.
Is rotated in the reverse direction until the voltage of the stepping motor becomes 2.8 V or less (indicated by the path on the drawing, ′ →), and then the stepping motor is rotated in the forward direction (→).

Finally, in any case, while the stepping motor is rotated in the forward direction, the position where the output P of the potentiometer reaches the learning reference value 3V is set as the home position, and the number of steps at this time is set to, for example, 100.

Thereafter, by increasing or decreasing the number of steps based on this position, the rotation control of the stepping motor is performed.

The above method is actually a control unit (6th
The microcomputer shown in FIG. 4 automatically executes the processing. The control flow in that case will be described with reference to FIG. First, the program starts when the switch is turned on. First, it is determined whether or not the initial output value of the potentiometer is 2.8 V or less (step 10).
1). If YES, the number of input steps is increased, and the stepping motor is rotated in the forward direction until the output of the potentiometer reaches the learning reference value of 3 V (step 10).
2).

If NO, the stepping motor is rotated in the reverse direction until the potentiometer output P becomes 2.8 V or less in step 103, and when it becomes 2.8 V or less, the process proceeds to step 102 and the stepping motor is turned on until the potentiometer output becomes 3 V. Rotate in the forward direction. And when it reaches 3V,
The number of steps at that time is set to 100 steps (step 104). This ends the learning.

As described above, in the learning method according to the present embodiment, since the direction approaching the learning reference value Q is finally unified in the positive direction, the same position can always be learned as the home position even when the hysteresis is large.

In the above embodiment, the case where the learning is performed by operating the stepping motor in the forward direction has been described. However, the learning may be performed by operating the stepping motor in the reverse direction. An example in that case will be described with reference to FIGS.

In this case, the learning operation is performed as follows according to the magnitude of the initial value X of the potentiometer.

(I) As shown in FIG. 4, when (Q + T) ≦ X, that is, when 3.2V ≦ X, the stepping motor is rotated in the reverse direction so that the output P of the potentiometer becomes small as it is. In the route on the figure,, '→→
Becomes

(Ii) As shown in FIG. 5, when X <(Q + T), that is, X <3.2 V, the output P of the potentiometer is temporarily set.
Until the voltage becomes 3.2 V or more (indicated by the path on the drawing, ′ →), and then the stepping motor is rotated in the reverse direction (→).

Finally, in any case, while the stepping motor is rotated in the opposite direction, the position where the output P of the potentiometer reaches the learning reference value 3V is set as the home position, and the number of steps at this time is set to, for example, 70.

In the above embodiment, an example in which the present invention is applied to home position learning of a stepping motor has been described. However, the present invention can of course be used for learning the home position of another actuator.

[Effects of the Invention] As described above, according to the learning method of the present invention, the home position of the actuator can be accurately learned even in a system with hysteresis.

[Brief description of the drawings]

1 and 2 are explanatory views of an embodiment of the present invention, FIG. 3 is a flowchart of an apparatus for executing the embodiment, and FIGS. 4 and 5 are diagrams of another embodiment of the present invention. FIG. 6 is a diagram showing the relationship between the mechanisms of the system in question here, and FIG. 7 is a diagram for explaining a conventional problem. X: initial value, Q: learning reference value, T: maximum change width of the potentiometer in the incomplete tracking area.

Claims (1)

(57) [Claims] A potentiometer is connected directly or indirectly to an output portion of an actuator, and a relationship having a certain hysteresis exists between an input to the actuator and an output of the potentiometer, and the operation direction switching of the actuator is performed. At the beginning of the time, there is a region where the potentiometer imperfectly follows the operation of the actuator, and in a system where the width of the output change of the potentiometer corresponding to the maximum width S of the region is measured as T, the output P of the potentiometer is While monitoring, the actuator is operated so that the output P of the potentiometer becomes the learning reference value Q. When learning the position where the output P of the potentiometer matches the learning reference value Q as the home position of the actuator, the home position learning is performed. Direction of the actuator for It is determined in advance whether to approach the learning reference value Q from the side with the smaller output P of the potentiometer or to approach the learning reference value Q from the side with the larger output P of the potentiometer in the opposite direction. If it is determined to approach the learning reference value Q from the smaller side of the output P, and if the initial value X of the potentiometer is X ≦ (Q−T), the actuator is moved in the forward direction so that the output P of the potentiometer becomes larger. When (QT) <X, once the actuator is operated in the reverse direction until the potentiometer output P becomes equal to or less than (QT), the actuator is subsequently operated in the forward direction (ii) If it is determined that the output P of the potentiometer approaches the learning reference value Q from the larger side, the initial value X of the potentiometer becomes When (Q + T) ≦ X, the actuator is operated in the reverse direction so that the output P of the potentiometer becomes small. When X <(Q + T), the actuator is moved in the forward direction until the potentiometer output P becomes (Q + P) or more. Then, the actuator is operated in the reverse direction, and the output P of the potentiometer is changed to the learning reference value Q.
And learning the matched position as the home position of the actuator.