JPH08290893A - Control knowledge setting method and device and position control method and control device and decision method of standard value used for it in overhead crane control device - Google Patents

Abstract

PURPOSE: To provide a control knowledge setting device in an overhead crane control device capable of easily and safely correcting and setting a control parameter (knowledge) in accordance with a characteristic, etc., of each crane. CONSTITUTION: A basic knowledge formation means 20 receives a structural data about an overhead crane which is a processing object and a design data of a control target value, etc., comparatively rough adjustment is carried out by correcting a prototype data previously prepared in accordance with each of the received data, basic knowledge matching the crane is formed, and its result is transferred to a control part 13. The control part 13 is trially driven in accordance with given time, a driving state of the time is detected by a sensor and it is given to a knowledge correction part 22. Characteristic quantity is extracted from sensor output at the knowledge correction part 22, the driving state is evaluated in accordance with its characteristic quantity data, and control knowledge is adjusted so as to make an evaluated value better.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control knowledge setting method and apparatus in an overhead crane control apparatus, a position control method and control apparatus, and a reference value determination method used therein.

[0002]

2. Description of the Related Art An overhead crane is installed indoors or outdoors in a building as shown in FIG. 30, and when installed indoors, for example, a trolley is mounted on a traveling path (rail) 1 that is stretched between opposing walls. A (carriage) 2 is provided so as to be movable, and a hoist 3 is attached to the trolley 2 so as to be able to move up and down. And
The suspended load 4 conveyed to the hook 3 is lifted, the trolley 2 is moved in a predetermined direction, and then the hoist 3 is lowered to move the suspended load 4 to a predetermined position. ing. Then, the trolley 2 and the hoist 3 move in parallel (traveling movement and traverse movement moving in a direction orthogonal thereto) and ascending / descending with the rotation of the drive motor 6. Then, the control of the movement is executed by turning on / off the relay performed by the operator in the cab 7.

By the way, the suspended load 4 to be transported is 1-2.
There are things such as from about t to 500 t or more, which are very heavy in any case. For example, when the trolley 2 is being moved, if the relay is turned off at a predetermined timing, the trolley 2 stops following it, but the suspended load 4 Due to the moment of inertia of the load, the load does not stop immediately, but after proceeding in the traveling direction as it is, it returns to the stop position of the trolley 2 (deflecting back).
It swings like this and is very dangerous.

Then, for example, as shown in FIG. 31, the relay is temporarily turned off before the stop position (broken line in the figure). Then, the suspended load 4 further moves forward due to the inertia. Then, if the suspended load 4 reaches just before the target stop position (at this time, the inertial force is adjusted so as to be almost 0), the relay is re-minuted as shown by (two-dot chain line in the figure). The trolley 2 is moved forward for only the time and stopped at the target stop position. At this time, the suspended load 4 receives the force such as the inertial force and is positioned at the target stop position, and the timing is set so that the inertial force also becomes zero. Therefore, the suspended load stops at the target stop position without swinging back. In addition, in order to prevent the suspended load from swinging at the start of movement or during movement,
Similar to the above principle, the relay is turned ON / O at a predetermined timing.
FF control minimizes the shake and enables smooth transportation to a predetermined position.

[0005]

Although the swing-back can be suppressed by controlling the ON / OFF of the relay as described above, a skilled technique is required to perform the actual control. In particular, cranes (overhead cranes) have different characteristics for each device (characteristics of the device itself, environment of the installation location, etc.) and differ according to the weight of the suspended load and other conditions. In order to do so, we need to have many years of experience.

Therefore, if a person who is not a skilled operator performs ON / OFF control of the relay, the shaking cannot be suppressed by, for example, one or a small number of ON / OFF processes, and the relay is repeatedly turned ON / OFF. Since the process is repeated, not only is work efficiency poor, but there is also the danger of shaking, which is dangerous.

On the other hand, training skilled operators takes a long time. Therefore, in order to secure safety and improve work efficiency, it is desired to develop an automatic control device capable of suppressing shake during movement / stop as much as possible.

In the conventional control system developed in accordance with the above request, since the inverter motor is used as the drive motor for speed control, the existing equipment (control system by ON / OFF of the relay) is used as it is. Can not.

Further, since the characteristics and the like are different for each crane as described above, even if there is an automatic control device which is normally operating in a certain crane, it is diverted as it is as a control controller for another crane. Can't
Adjustment by trial and error is required at the installation site. However, such adjustment work requires specialized knowledge and skilled techniques, which not only makes it complicated but also makes the adjustment insufficient and dangerous.

On the other hand, in order to solve the above-mentioned problems, the present inventor has adopted the relay O
We have developed a method and device to automate the control system of an overhead crane by N / OFF, but when such automatic control is performed, the target object cannot be moved to the target stop position due to disturbance and various other factors. However, there is a possibility that a positional deviation will occur between the actual stop position and the target stop position.

Therefore, when performing automatic control, it is necessary to detect the position deviation in such a case and perform control to move the position again by a distance corresponding to the position deviation, but this position deviation cannot be feedback controlled. In many cases, it is so short. The higher the accuracy of the automatic control performed first, the shorter the position deviation.

As described above, the higher the accuracy and performance of the automatic control which is performed first after the improvement is made, the more the fine adjustment using the same control system cannot be performed when the position deviation occurs. The problem is not limited to the one in which the first invention of the present application controls the overhead crane by turning the relay ON / OFF, and the same problem also in the case of the conventional speed control using an inverter motor. However, it is also a problem common to various movement control devices that move a certain object to a target stop position.

Therefore, a search was made for a conventional technique capable of performing fine adjustment when the above-mentioned movement distance is so short that feedback control cannot be performed.
There was an invention relating to a stop position correction method in automatic driving as disclosed in Japanese Patent No. 223089.

In the invention disclosed in this publication, when there is a "difference" between the actual stop position where the object is moved and stopped by the automatic operation and the target stop position, it is determined according to the difference. It is moved again for a predetermined time. In the case of this prior art, it is unknown how the automatic control for moving to the target stop position performed first is performed. Therefore, this "difference" and the feedback controllable distance which is a problem of the present invention are Although the relationship is unknown, as a result, the present inventor, if a position deviation (distance where feedback control cannot be performed) occurs when the object is moved to a certain position by the feedback control, the time depending on the position deviation is generated. It was considered to make fine adjustment by turning on the relay for a certain period of time by control (no feedback control).

However, if the above technique is simply diverted, a problem of shake occurs. That is, as described above, in the case of the overhead crane, simply moving the position deviation by the amount of positional deviation causes the suspended load to shake due to the movement, which is dangerous and the shake cannot be suppressed.

Therefore, in order to achieve the suppression of the shake, the stop position accuracy is sacrificed to some extent (the fine adjustment based on the position deviation is not performed), or the fine adjustment is performed in order to accurately perform the stop position. However, at least during fine adjustment, it is said that the occurrence of shake is allowed, and it is difficult to obtain both shake suppression and improvement in stopping accuracy at the same time.
Further, such a problem is not limited to the above-described overhead crane, and commonly occurs in control devices in various moving devices in which vibration is generated when a normal crane moves or when an object moves.

The present invention has been made in view of the above background, and an object of the present invention is to perform operation control such as steady rest / positioning according to the crane without skilled skill or specialized knowledge. Can be performed by the controller, and optimum control parameters (knowledge) can be modified and set according to the characteristics of each crane, etc., and while the knowledge adjustment processing is being performed (overhead crane operation). (EN) Provided is a control knowledge setting method and device in an overhead crane control device capable of safely performing correction / adjustment processing without a large swing of a suspended load.

Further, as another object, even if the moving distance is short (to the extent that feedback control cannot be performed), it is possible to accurately stop at a target stop position while suppressing vibration, and control can be performed in a short time. An object of the present invention is to provide a position control method and device which are completed. Another object is to provide a determination method for accurately determining a reference value that is a reference for determining whether or not the movement distance is short.

[0019]

[Means for Solving the Problems]

* Technical idea devised to solve the problem In order to achieve the above-mentioned object, in the control knowledge setting method and device in the overhead crane control device according to the first invention, first, the characteristics of the target overhead crane, etc. Basic knowledge that suits the above is given, and the operation of the overhead crane is actually controlled based on the basic knowledge. Then, the operating state at that time is detected by a sensor or the like, the characteristic amount is extracted, and the control actually performed is evaluated. Then, each knowledge is modified to improve the evaluation, the modified knowledge is reset to the controller as temporary control knowledge, the overhead crane is operated again based on the given control knowledge, and the control is performed as necessary. Make corrections to your knowledge. Then, when an evaluation satisfying the specifications is finally obtained, the control knowledge at that time is set as the actual control knowledge. Claims 1 and 3 are based on this technical idea.

Further, the quality of the above basic knowledge is related to the subsequent adjustment processing. Then, there is a demand to generate good basic knowledge with the simplest input processing. Therefore, first, original data of knowledge relatively suitable for various overhead cranes is created in advance. And
The actual operator (the person who performs the correction process) inputs relatively well-known and clear data such as design data such as structural data and control target values for the ceiling crane to be processed, and based on the input data, the prototype The basic knowledge is generated by correcting the data (rough adjustment). By doing so, the actual input process is the above-mentioned clear data and can be easily performed. Claims 2 and 4 are based on this technical idea.

Considering the actual operating conditions, it is preferable to use fuzzy knowledge (rules and membership functions) as the control knowledge to be used, and further, as an overhead crane to be processed according to the present invention, for example, a relay crane. Some are operated by a motor driven by ON / OFF control.

Further, in the position control method and apparatus according to the second invention, for example, when the object is actually moved by performing automatic control based on the control knowledge set by the above-mentioned first invention. There is a possibility that the target stop position may be displaced due to such a cause. In this case, the invention is for finely adjusting the target object such as an object by moving the target object again without vibrating the target object. However, the second invention is a problem common to all position control devices that move a certain object to a predetermined target stop position, regardless of the position control of the overhead crane according to the first invention described above. The application target is not limited to the overhead crane.

Then, specifically, when the distance is to some extent and the control is performed by one continuous movement, vibration such as a shake of an object occurs. Therefore, the vibration itself caused by each movement is reduced by finely moving the movement twice, and the vibration generated by the first movement is absorbed by the second movement. In accordance with such a technical idea, the invention of the following method and apparatus was achieved.

Specifically, in a position control device for controlling the movement of an object to a target stop position, the object is controlled when the positional deviation from the existing position of the object to the target stop position is a reference value or less. A first movement control means (moving for the first time) for moving the object for a fixed time determined by using a timer, and the object generated by the movement of the object accompanied by the control by the first movement control means. The present invention has a basic configuration including a second movement control unit (moves a second time) that moves and stops the object again in a direction in which residual vibration is attenuated (claim 7).

Preferably, at least the present position of the object is detected, and a third movement control means for moving the object to the target stop position by feedback control, an absolute value of the position deviation and the reference value. Compare
A determination means for determining the position control using the first and second movement control means when the value is smaller than the reference value, and a third movement control means when the value is larger than the reference value. (Claim 9).

As an example of an object to which the invention is applied, there is a crane, and the object is a system including a crane and an object conveyed by the crane, and the vibration is a crane of the object conveyed. The present invention can be applied to a vibration with respect to (claim 8).

Further, the second invention can of course be applied to an overhead crane which was also an object of application in the above-mentioned first invention among the cranes, and as a concrete configuration, any one of claims 7 to 9 can be applied. The position control device described is mounted on a control device of an overhead crane that is drive-controlled by ON / OFF control of a relay. Then, the vibration is a shake of an object conveyed by the crane, and the first, second
Is to determine the ON time of the relay.

Further, in the method invention, the positional deviation from the existing position of the object to the target stop position is compared with a reference value. Then, when the positional deviation is small, the object is moved for a certain period of time using a timer. Then, after temporarily stopping, a timer is used to move the object for a certain period of time in a direction in which vibration remaining in the object caused by the movement of the object is attenuated at a predetermined timing.

In addition, preferably, as a general rule, it is moved twice as described above, and the detected position deviation is smaller than the reference value, and when the object is moved, vibration hardly occurs. For short distances,
You may make it move the said target object once for a fixed time using a timer.

Further, preferably, the absolute value of the position deviation is compared with the reference value, and when it is smaller than the reference value, the control according to claim 11 or 12 is performed. on the other hand,
When it is larger than the reference value, it is preferable to move the object to the target stop position by feedback control while monitoring the movement state.

Further, even if the fine adjustment by the feedback control is performed, the distance is further from the target stop position by the reference value or more, but if the distance is a distance in which the feedback control cannot be performed, a timer is used after the feedback control. The position may be adjusted (claim 14).

Further, in the present invention, the control based on the timer and the control based on the feedback described above may each be repeated a predetermined number of times, and by executing an appropriate number of times,
Highly accurate positioning can be performed.

The reference value serving as the above-mentioned determination reference differs depending on each device for moving the object, and if the reference value is not set correctly, the accurate position control can be performed by the feedback control originally. The position control based on the timer is performed (there is no problem in practical use). Conversely, although the feedback control cannot be performed, the position control based on the feedback control is tried instead of the control based on the timer. There is a risk of overshooting the stop position.

Therefore, in the present invention, as a method of determining the reference value, first, the moving means for moving the object is operated to move the object. Then, the movement of the object is monitored, and when the movement is detected, the operation of the moving means is stopped. Next, the movement distance from the movement start position to the stop position of the object is obtained, and the movement distance or a value obtained by adding a margin to the movement distance is determined as the reference value (claim 15).

* Definition of Terms in Claims The basic knowledge, as mentioned in the claims, refers to control knowledge that approximately matches the characteristics of the overhead crane to be processed. The term "approximately matched" is, of course, a case that is relatively close to the control knowledge finally determined (for example, obtained by rough adjustment based on structural data), but the present invention Without being limited thereto, if it does not match the characteristics of the overhead crane at all, it is dangerous if the suspended load largely shakes when it is operated for adjustment. Therefore, at least such knowledge is included. The tentative control knowledge includes both the basic knowledge initially set and the control knowledge corrected by the correction process (before the final determination).

The structural data refers to unique data concerning each crane itself which is apparent in the design stage, such as the weight of the trolley, the lift (suspension height), and the rated speed of traveling / traversing. Further, the design data is data for determining an operating environment such as a target value, which is performed using the crane, and includes, for example, an inching distance and a load of a suspended load operated by using the crane. In addition, the original shape data is data that is the basis of knowledge suitable for various overhead cranes, and based on the data unique to each overhead crane such as the above structural data and design data, the original shape data is processed and modified to obtain basic knowledge. Is generated. For example, the generally described membership functions and the like shown in FIGS. 5 to 11 and FIGS. 13 to 15 are the original data. Then, by inputting numerical values such as a, b, c, d, ... In each figure, basic knowledge is obtained.

The third for feedback control
Is a position control device of the present invention (for fine adjustment)
It is possible to apply the normal position control system as it is, that is, to use it as a normal position control (control for moving a relatively long distance), which is performed before the operation of the system. But of course

[0038]

First, the first invention (control knowledge setting method and device in an overhead crane control device) will be described.
Set the basic knowledge suitable for the crane to be processed in the controller. Then, the crane is actually operated while controlling the crane based on the basic knowledge. Then, the operating state (control state) at that time is detected, a predetermined feature amount is extracted,
Evaluate the basic knowledge, correct the knowledge so that the evaluation value becomes high, and reset it in the controller. Then, actually operate the crane based on the knowledge set in that way,
Re-evaluate. In this way, knowledge evaluation / correction and crane operation are repeated alternately, and if the desired specifications are finally satisfied, the correction process is terminated, and the control knowledge at that time is determined as the control knowledge to be used during actual operation. And set.

Temporary control knowledge set first (basic knowledge)
However, since it is almost suitable for the target overhead crane, the control knowledge correction processing can be performed in a short time, and even if the crane is operated for the correction, the suspended load does not shake significantly.

Further, when the basic knowledge is generated by the basic knowledge generating means based on predetermined data, the input process can be easily performed because it is known data such as crane structure data and design data. Then, since the original shape data is corrected based on the information about the crane, the basic knowledge as a result of the correction is a relatively rough adjustment. Becomes

Next, the second invention (position control method and device) will be described. In principle, the fine movement is performed twice. Therefore, when the positional deviation is smaller than the reference value, the first movement control means is differentially operated, and the object is first moved for a fixed time determined by the timer according to the positional deviation. At this time, it vibrates with this movement.
Next, the second movement control means is made to differentially move the object again for a fixed time determined by the second timer at a predetermined timing so as to cancel the generated vibration. If the sum of the two movement distances is adjusted to be equal to the position deviation, the object can be moved to the correct position even if the distance is such that feedback control cannot be performed, and the residual vibration can be absorbed. Vibration subsides with time.

According to the ninth and thirteenth aspects, when the position deviation is smaller than the reference value, the fine adjustment is performed with the timer as the reference, and when it is larger than the reference value, the feedback is performed. Since precise fine adjustment is performed by control, by adopting a control method according to the situation, accurate position control can be performed in a short time and vibration does not occur (damping in a short time).

By the way, according to the present invention, the vibration is suppressed by the two movements and the vibration is moved to the predetermined position. However, when the movement distance is very short, even if the movement is performed once, the vibration is not so much. It does not occur and decays in a short time. Also,
To move in such a short distance in two steps,
Since the moving distance of one time becomes very short, the timer control is required to have high accuracy. Therefore, according to a twelfth aspect, when the position deviation is shorter than the reference value, and when the distance is very short, the movement to the target stop position is performed by only one timer control movement. To As a result, the object can be moved to the target stop position in a short time, and the vibration of the object hardly occurs originally, so there is no problem of vibration. Even in this case, if the positional deviation is such that vibration occurs, the movement is divided into two as a general rule.

[0044]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of a control knowledge setting method and apparatus, a position control method and control apparatus and a reference value determination method used therein in an overhead crane control apparatus according to the present invention will be described in detail with reference to the accompanying drawings. To do. FIG. 1 shows an embodiment of the present invention. In this embodiment, an O of a control relay that drives / stops a drive motor at a predetermined timing to reciprocate a trolley mounted on an overhead crane.
The example applied to the control device which controls N / OFF is shown.

First, the crane steadying process by the control device will be described prior to the description of the specific configuration. Focusing on the movement of the trolley among the operations of the crane, there are cases of moving a predetermined distance (long distance) to the vicinity of the target stop position and moving a short distance for positioning after temporarily stopping near the target stop position. There are two types. On the other hand, the swing of the suspended load occurs mainly at the start of movement and at the time of movement stop. The suspended load shakes even during long-distance movement, but the main reason is that such vibration occurs due to continuation (vibration) of the vibration generated at the start of movement. Therefore, the shake generated at the start of the movement is stopped by suppressing the shake and then moved for a long distance to suppress the occurrence of the shake during the movement (to suppress the slight shake). Further, the shake generated when the movement is stopped is adjusted by adjusting the timing of a minute movement for position adjustment. Specifically, it is as follows.

First, in the process including long-distance movement, the relay output is first turned on as shown in FIG. 2 (FIG. 2 (A)). Then, the trolley 2 moves first, and the suspended load 4 starts moving with a delay, so that a predetermined deflection angle θ occurs as shown by the solid line in FIG. If the trolley 2 moves a minute distance in this way, the relay is turned off. Then, the suspended load 4 moves with a delay due to the swing-back, and then is positioned below the trolley 2.

In this way, waiting for the swing-back, the trolley 2
When the suspended load 4 comes under, the relay is turned on again and the trolley 2 is restarted. As a result, the large shake generated at the start of the movement is eliminated, and during the long distance movement thereafter, the shake is small and less dangerous ((C) in the same figure). Then, the relay is temporarily turned off immediately before the target stop position.
Then, the suspended load 4 moves forward due to inertial force, so the relay is turned on with a little delay and the trolley 2 is moved forward, and when the suspended load is located right below, the relay is turned off. Then, the stop position at this time is controlled to be the target stop position. Accordingly, when the movement is stopped, the swing load largely swings forward, and the suspended load does not swing due to the swing back (or the swing amount can be reduced).

On the other hand, in the case of a short distance, there is a possibility that the target stop position will be passed if the above long distance movement is carried out, so the above-mentioned movement start at the time of long distance movement and the final movement stop processing are performed. 3 (A), (B)
Also, as shown in FIGS. 7C and 7D, the shakeback can be utilized to prevent the shake. In addition, the relay ON / O
In the case of control by FF, if the number of times of ON / OFF switching is large, the protection circuit operates and the power is turned off, but in the present embodiment, stable steady rest is performed by a small number of times (for example, once) of ON / OFF switching. As a result, there is no risk of the power supply going down.

Further, the steady rest will be described in detail. Starting from the time when the trolley travels right below in the traveling direction of the trolley, the timing at which the relay is turned on and the time during which the relay is turned on are standardized by the swing period. I will give it with the data. That is, the swing period T is given by the following equation, assuming that the swing motion of the crane is a single vibration and the swing angle is minute.

T = 2π (L / g) ^1/2 L: Suspended length, g: Gravitational acceleration On the other hand, FIG. 4 shows the relationship of the swing of the suspended load due to ON / OFF control of the relay. There is. As shown in the figure, when the relay is continuously in the ON state for moving a long distance, the suspended load vibrates slightly. Then, this vibration becomes a simple vibration according to a constant cycle according to the above equation. Turn off the relay in this state (t0)
Then, the suspended load largely shakes, and when the relay is turned on again for a short time to prevent the shake (t1), the shake is attenuated and becomes small, and the shake is stopped.

The feature quantities extracted at this time include "shake peaks 1, 2, 3" and "shake position features 1, 2." First, the deflection peak 1 is the data of the last peak amplitude value when the relay is ON, and it is set to 0 if the elapsed time from the detection of the last peak to the relay OFF is 1.5 times the detection cycle or more. . The shake peak 2 is absolute value data of the peak amplitude value that appears first during execution of the shake stop.
If it appears after the relay is turned off, the value is taken. The shake peak 3 is data of a peak amplitude value appearing in the opposite direction immediately after the shake is stopped. It can be said that the smaller this value is, the higher the steady-rest effect is, and therefore it can be used for evaluation. The amplitude data can be expressed by the sign of the amplitude and the absolute value of the amplitude data.

The shake position characteristic 1 is that the traveling / traverse relay is ON.
It is an absolute value (mm) of a value obtained by adding a predetermined offset to the data of the shake position when the speed of the trolley becomes half or less of the rated speed or the maximum speed after switching from OFF to OFF. In addition, the runout position feature 2 has a relay for steady run
It is data (mm) of the shake position when the speed of the trolley becomes half or less of the rated speed or the maximum speed after switching to OFF after setting to N (t1).

The relay ON timing is the timing for turning on the steady rest, and is the elapsed time from the zero crossing to the ON state with reference to the runout waveform, and is expressed as a ratio to 1/2 cycle. Data.

In addition to the above, the feature amount data to be extracted further includes the relay ON duration, which is the time during which the relay is turned on to prevent shake, which is a 1/2 cycle period based on the shake waveform. It consists of data expressed as a ratio.

Next, the control device will be described. The input processing part 11 receives various sensor outputs for detecting the state of each part constituting the overhead crane and a crane operation signal given by an operator via an input device such as a keyboard. At the same time, the received data is sent to the feature quantity computing unit 12. Then, the sensor information received includes, for example, the deflection angle of the crane suspended load, the suspension length (suspended amount of the suspended load), the moving direction and position of the trolley, the traveling (traverse) speed, and the like. Then, the shake angle is extracted by performing predetermined image processing from the image data obtained by imaging with a video camera, for example. Similarly, the suspension length may be extracted from image data obtained from a video camera, but the suspension length is calculated by detecting the rotation angle of the motor that controls the lifting and lowering of the suspended load. May be. The moving speed can be measured by a speedometer mounted on the trolley.

The feature amount calculation unit 12 provides the sensor information (crane swing angle, trolley position, moving speed, etc.).
Knowledge to perform feature control required for normal crane operation, such as crane runout period and crane (trolley) distance moved by steady stop operation, and crane control such as steady rest and positioning It has two functions of extracting the feature amount necessary for adjusting the.

Further, the characteristic amount calculation unit 12 gives the extracted characteristic amount to the control unit 13 as a controller during the actual crane operation. Based on the control knowledge set by the knowledge setting unit 14, the control unit 13 is configured to perform two controls, that is, steady rest control and positioning control, according to a given feature amount required for the crane operation. The timing to turn ON / OFF the relay is determined, and the determination signal is sent to the output processing unit 15. Then, the output processing unit 15 is configured to turn ON / OFF the relay in accordance with the ON / OFF timing of the relay determined by the control unit 13 and send an ON / OFF command to the driver of the drive motor. Note that the steady rest control includes inching control that is performed when moving a short distance and normal control that is performed when moving a long distance.

Here, the present invention is provided with a knowledge setting device for setting and adjusting the control knowledge that is the basis of the control in the control section 13. In the present embodiment, the basic knowledge generation unit 20 that sets the basic knowledge that is the source (default) of the control knowledge that is the adjustment target in the knowledge setting unit 5,
A knowledge correction unit 22 that corrects the control knowledge in accordance with the operation status of the crane while controlling the operation of the crane based on the basic knowledge thus provided is provided. In this way, setting the basic knowledge in advance performs the correction process while actually operating the crane, so if the initially set control knowledge deviates greatly from the optimum one, for example, the suspended load will be large. Since it is dangerous to shake, the control knowledge that is close to the optimum knowledge to some extent (there is no problem even if executed) is initialized to prevent such danger.

Then, the basic knowledge generating section 20 has a basic structure input section 20a for taking in the structural data of the crane which is apparent at the design stage such as the weight of the target trolley, the lifting height (suspension height), and the rated speed of traveling / traversing. , A basic knowledge adjusting unit 2 for generating and outputting basic knowledge suitable for the crane based on the data taken in by the basic structure input unit 20a.
0b and.

The knowledge correction section 22 evaluates the current control knowledge on the basis of the feature quantity extracted by the feature quantity calculation section 12 in addition to the feature quantity calculation section 12, and the knowledge of the knowledge is calculated based on the evaluation result. A teacher data creation unit 22a that automatically creates teacher data and a parameter adjustment unit 22b that adjusts control knowledge based on the teacher data created by the teacher data creation unit 22a are provided.

Of the above-mentioned units, the input / output processing units 11 and 15 and the control unit 13 are constituted by a control panel equipped with a programmable logic controller (PLC), and the initial knowledge setting unit 20, the knowledge correction unit 22 and the knowledge setting unit. The unit 14 is composed of a host computer connected to the control panel. Then, the transmission and reception of data between the two are performed via communication.

Next, each part constituting the above-mentioned device will be described. The basic structure input unit 20a that constitutes the basic knowledge generation unit 20 includes a traveling speed, a traverse speed, a hoisting load rated current (or hoisting speed / hoisting motor input voltage / rated hoisting load required for approximation), and dimensions for traveling and traverse. A moving distance is given. Then, three types of short distance a1, medium distance a2, and long distance a3 are input as inching distances, respectively. The inching distance is given in [cm].

The basic knowledge adjusting unit 20b is first given knowledge as a prototype, and the knowledge of the prototype (FIG. 5 to FIG. 1) is given.
1, FIG. 13 to FIG. 15 and Tables 1 to 3) are modified based on the above-mentioned design data structure data to generate basic knowledge. Specifically, the membership function is the membership function of the target movement distance shown in FIG. 5, the membership function of the suspension load shown in FIG. 6, the membership function of the suspension length shown in FIG. 7, and the movement OFF shown in FIG. Membership function of specified distance, membership function of runout peak 1 shown in FIG. 9, FIG.
Membership function of the sign of deflection peak 1 shown in FIG.
2 is a membership function of shake position feature 1 shown in FIG.

Then, among the given design structure data, the membership of the target moving distance in FIG. 5 is based on the three values of the moving distance (inching distance) (obtained for traveling and traverse (the same applies below)). The values of a1 to a4 of the function are determined. Here, the target moving distance means the target stop position,
The relative distance from the current position of the trolley is expressed as a positive number, and the unit is [cm].

For a1 to a3, the inching distance data given through the basic structure input unit 20a is used as a default value as it is, and the maximum distance a4 is obtained based on this input (a4 = a3 × 2). However, when the value of a4 obtained by the execution of the calculation is larger than the default value (for example, 3 m), the default value is set. Then, the membership function as shown in FIG. 5 is generated from the input data and the result of the above calculation. At this time, however, the label shape may be a shape such as a singleton which does not have a fitness between 0 and 1. However, in such a case, the label shape as it is is adopted as basic knowledge.

The suspension load current [A] has the label shape shown in FIG. The suspension load refers to the maximum value of the motor current during winding (noise removal by moving average or the like), and the unit is [A]. Therefore, the basic structure input unit 20
Since the rated current is given from a, the given input value (rated current) is determined as the value of b7.

[0067]

[Equation 1] Then, each value is calculated based on the value b7. That is, b1 is set to 1/10 of the rated current, and b1 and b7 thereof are used as a reference, and the remaining values b2 to b are set so that the division becomes similar.
6, b8 is determined. Therefore, the specific calculation method is b
1 = b7 × 0.1, b2 = b7 × 0.15, b3 = b7
× 0.3, b4 = b7 × 0.4, b5 = b7 × 0.6,
Each value is obtained by executing b6 = b7 × 0.7 and b8 = b7 × 1.2.

For the suspension length membership function,
As shown in FIG. 7, the default value is adopted regardless of the basic setting data. And each value is c1 = 400, c
2 = 450, c3 = 600, c4 = 650, c5 = 9
0, c6 = 1000 [cm].

The movement OFF designated distance is given as a relative distance to the movement relay OFF in the inching parameter setting data, and the unit at that time is [cm]. The normal steady rest is set to 4095. Further, the respective values d1 to d3 that determine the label shape are obtained by d1 = a1 / 2 [cm] d2 = a2 / 2 [cm] d3 = a3 / 2 [cm].

The values e1 to e3 for determining the label shape of the shake peak 1 absolute value are as follows: e1 = a1 × 10/2 [mm] e2 = a2 × 10/2 [mm] e3 = a3 × 10/2 Calculated by [mm].

The shake peak sign is used to obtain the shake peak 1 together with the above shake peak absolute value, and is used as switching data for the shake stop. When the shake peak 1 is negative or 0, it becomes 0, and when it is positive, Set to 10. Further, it is set to 150 for inching and 4095 for positioning.

The fuzzy inference performed by the control unit 13 is as shown in FIG. 12, and is inferred based on the positioning knowledge from the suspension length and suspension load and the traveling (traverse) target moving distance in the inference condition unit. To determine the target stop position for traveling (traversing). In addition, based on the travel distance to the traveling (traversing) relay OFF and the absolute value of the traveling (traversing) shake peak 1, based on the steady rest knowledge for inching in the steady rest knowledge, the timing of the traveling (inverse) relay ON during inching is set. And relay ON
Find the duration. Furthermore, traveling (transverse) runout peak 1
Based on the absolute value of and the sign of the running (horizontal) runout peak and the running (horizontal) runout position feature 1 from the normal steadying knowledge among the steadying knowledge, running (running) during normal movement
Calculate the relay ON timing and relay ON duration.

The membership function of the stop start position, which is the output in the above inference, is as shown in FIG. 13, and the membership function of the relay ON timing is as shown in FIG. The membership function of the ON duration is as shown in FIG. The basic knowledge generation unit 20 also obtains each value that determines the label shape of the output membership function.

That is, the stop start position is h1 = a1 /
2, h2 = a2 / 2, h3 = a3 / 2, h4 = a4 / 2
Ask by Since the label ZR (h0) is not used in the initial knowledge rule, it is excluded from the target of initial knowledge and h0 = 0 is set. Similarly, h5 and h6 are also excluded, and h5 = h4 and h6 = h4.

Further, the default values of the relay ON timing and the relay ON duration are as shown in FIGS.
The value shown in is set as the default value of each label. Then, each of these values and the above-mentioned movement OFF designated distance (d1 to d
The correlation between 3) and each value of the shake peak 1 absolute value (e1 to e3) is shown in FIG.

Further, the fuzzy rules used when executing inference using the membership function are as shown in Tables 1 to 3 below. Then, Table 1 is a stop start position rule, Table 2 is an inching steady rest rule (relay ON timing, relay ON continuation time), and Table 3 is a normal rest rest rule. Then, the basic knowledge for the crane, which is obtained by calculating a4 and b1 to b6, b8 from the knowledge of the original shape and corrected, is output from the basic knowledge adjusting unit 20b and sent to the knowledge setting unit 14.

[0077]

[Table 1]

[0078]

[Table 2]

[0079]

[Table 3] The feature amount calculation unit 12 that constitutes the knowledge correction unit 22 has each feature amount shown in FIG. 4, that is, each shake peak and shake position feature, and the drive motor as information necessary for correcting (tuning) the basic knowledge. Relay ON timing and relay ON obtained based on ON / OFF information of
The continuation time is obtained and given to the teacher data creation unit 22a. Table 4 shows the specific data structure to be transferred.

[0080]

[Table 4] As for the movement OFF designated distance in the table, the flag data 4095 is stored when normal (long distance) knowledge is used, and the shake position feature 1 is flag data 4095 when inching (short distance) knowledge is used.
To store.

The teacher data generator 22a corrects the steady rest parameters (relay ON timing, ON duration) based on the given feature amount data based on the following criteria.

It is judged whether or not the shake peak 3 is larger than the shake amplitude upper limit required to suppress shake. If the shake peak 3 is large, it is determined that the adjustment is not sufficient. Increases the relay ON timing by 1/16 cycle and decreases the relay ON duration by 1/16 cycle. On the other hand, when the shake position feature 2 is positive, conversely to the above, the relay ON timing is reduced by 1/16 cycle and the relay ON duration time is increased by 1/16 cycle.

Then, the corrected input feature amount and the steady rest parameter, and if the steady rest has been achieved (below the swing amplitude upper limit), the input feature amount and steady rest parameter at that time are saved as teacher data.

The above-mentioned processing is the processing for correcting the steady rest knowledge. As the processing for correcting the stop position control knowledge, the data is divided for each operating condition of the crane, and for each condition, from the movement OFF to the idling distance and The total movement distance required for steady rest is calculated, and the total value is stored as the distance required for stop.

Each data stored as described above is given to the parameter adjusting unit 22b in the next stage. Parameter adjusting unit 22
In b, regarding adjustment of steady rest knowledge, the given teacher data is first divided into inching knowledge adjustment data and normal knowledge adjustment data based on the knowledge used at the time of steady rest,
The area is divided using each data.

* Adjustment of inching (short-distance) knowledge Extraction of teacher data with shake position feature 1 of 4095 and setting of a feature amount space area in which the abscissa is the movement OFF designation distance and the shake peak 1 absolute value is the ordinate Then, the extracted teacher data is expanded in the area (see FIG. 17).

The vertical axis and the horizontal axis are each divided by three fixed straight lines to divide the feature amount space into 16. At this time, for each straight line, a value is selected so that the number of different parameters existing in the same area after the area division is minimized. Then, when there is one type of parameter existing in each area, that parameter is determined as a new parameter for the area. Further, when a plurality of parameters exist in the same area, the most existing parameter is determined as a new parameter of the area (for example, e1 to e2 and d).
In the area partitioned by 1 to d2, the parameter of "x" mark becomes a new parameter. Furthermore, the parameter before adjustment is determined as it is in the area where no data exists.
In this way, parameters are defined for 16 areas, and the knowledge correction for inching is completed.

* Normal (long-distance) knowledge adjustment Similarly, in normal knowledge adjustment, the movement OFF specified distance is 40.
The teacher data of 95 is extracted, the feature amount spatial area in which the shake position feature 1 is on the horizontal axis and the shake peak 1 is on the vertical axis is set, and the extracted teacher data is expanded in the area (see FIG. 18).

The vertical axis and the horizontal axis are divided by a constant straight line ± k and a straight line of feature quantity = 0, and the feature quantity space is divided into 16. At this time, for each straight line, a value is selected so that the number of different parameters existing in the same area after the area division is minimized. Since there is no data in which the shake position feature 1 is larger than the absolute value of the shake peak 1, no parameters are set in those areas (four in total). Therefore, the parameters are reset for the 12 areas excluding those areas.

That is, when there is one kind of parameter existing in the area to be processed, that parameter is determined as a new parameter for the area. When a plurality of parameters exist in the same area, the most existing parameter is determined as a new parameter for that area. Furthermore, the parameter before adjustment is determined as it is in the area where no data exists. Further, by using the ± k thus determined, each value for determining each label shape of the membership function of the shake position feature 1 is adjusted as shown in FIG. In this way, parameters are defined for 16 areas, and the normal knowledge adjustment is completed.

* Positioning Knowledge Adjustment (Adjustment of Input Membership Function) Adjustment for positioning knowledge is performed as follows. First, regarding the membership function of the lifting load, the lightest load data among the collected data is set as the hoisting current value b1 when there is no load (no hanging load), and the current value when the heaviest load is set is b7. . Then, like the initial knowledge, b
It is standardized by 1 and b7. That is, b1 to b7 are equally divided into six parts, b2 to b6 are set at the respective division points, and a value larger than b7 by the equal intervals is set to b8 (see FIG. 20).

Similarly, regarding the suspension length membership function, the shortest suspension length in the collected data is c1.
And the longest suspension length is c5, and other values are c1,
Normalize based on c5.

* Adjustment of positioning knowledge (adjustment of output variable label, output membership function) The adjustment of the output variable label of the rule is first performed from the data collected, and the data which can be steady (swing peak 3
Data having an absolute value of 10 cm or less) is extracted. Then, sort the data that was able to prevent steady by the target travel distance,
Each inching distance (3 types) and movement distance are divided into four blocks of automatic or manual operation data that is a4 or more.

The maximum value (MAX) and the minimum value (MIN) of the distance required for stopping are extracted from the data of each block, and when the difference is larger than 0.1 m, the median of MAX and MIN is obtained. The data is divided into two by the value, and each data group is set as follows. Further, the difference between MAX and MIN is 0.
When the length is 1 m or less, the group is not divided into two and remains as a data group according to the inching distance. That is, in the corresponding group, the even-numbered data group in Table 5 below does not exist.

[0095]

[Table 5] Next, for each data group classified as described above, the average value of the distances required for stopping is determined, and the value is set as the distance Si (i = 1 to 8) required for stopping each data group. Then, the divided data group is 7 or less, that is, even one is MAX.
When there is a difference between MIN and MIN of 0.1 m or less, Si as the representative value is assigned to the labels (ZR, VS, S, ...) Of the conclusion part shown in FIG. 13 in order from the smallest representative value. Therefore, if there are seven data groups, Z
If all are set from R to VB and there are less than 7, V
In this case, the labels are unassigned in order from B. In that case, the label of the unassigned portion is made equal to the maximum value of the assigned label (the same as when M, B, and VB were made equal to h4 in initial knowledge). On the other hand, when all the data of S1 to S8 exist, the two representative S with the smallest difference among Si
j and Sj + 1 are extracted, the average value of both is taken, and it is set as a new representative value. As a result, the number of representative values becomes 7, so that the labels are assigned in ascending order.

* Setting of output label of rule (data block in which data exists at the time of adjustment) The following processing is performed on the divided data group shown in Table 5 to set the label of the output membership function. To do. That is, first, the data range of each suspension load is extracted with respect to the larger distance and the smaller distance required for stopping (stop start position). Then, the label of the output membership function of the rule is set according to the data having the larger distance required for stopping.

Thus, for example, when the same input membership function area contains data with a short stop distance and data with a long stop distance, as shown in FIG. The output label of the data group having a large stopping distance with respect to the range of the goodness of fit 1 is used. In addition, i after FIG. 21 takes a value of 1 to 4.

Further, as shown in FIG. 22, when there is no overlapping range, the output label set in each rule is used. Further, as shown in FIG. 23, when the data range of the suspension load of the data having a large stopping distance and the data having a small stopping distance are separated, the stopping distance is set for the input membership function region which is not included in either of them. Use output labels for large data sets.

* Setting of output label of rule (data block where data does not exist at the time of adjustment) For each divided data group shown in Table 5,
The process of setting the label of the output membership function is performed according to the rule of. If the rule before adjustment has the same output membership function for the same data block, the label of the same output membership function after adjustment of the input membership function becomes the output label of all data in the data block. To do. If there are two output membership function labels for the same data block rule based on the knowledge before adjustment, the minimum input value with a goodness of fit of 1 is extracted from the rules whose output membership function labels are large. The value is set to A, the maximum input value with a goodness of fit of 1 is extracted in the rule with a small label of the output membership function, and the value is set to B (see FIG. 24). Then, the extracted A is used as the minimum input value of the data having a long moving distance, and the B is replaced with the maximum input value of the data having a small moving distance to reset the label of the output membership function of the rule. As a result, FIG.
As shown in 5, the membership function shown by the broken line is adjusted (reset) as shown by the solid line. Note that, as shown in the figure, A and B do not always have the conformance of 1 for each label.

New control knowledge (rules, membership functions) obtained by performing the above-described various processes is the knowledge setting unit 14
And is reset by the control unit 13 via the knowledge setting unit 14. In addition, this data is the teacher data creation unit 22a.
Will also be saved in and used as reference data for the next adjustment.

Next, an embodiment of the setting method according to the present invention will be described using the above apparatus. First, design value data, structural data, etc. of the crane are given to the basic knowledge generating unit 20, and the basic knowledge adjusting unit 20b in the generating unit 20 is executed to generate basic knowledge. Then, the basic knowledge thus generated is set in the controller via the knowledge setting unit 14.

Next, based on the basic knowledge, the control unit 13
Actually operate while controlling the crane at. Then, the operating state (control state) at that time is detected by a sensor, a predetermined feature amount is extracted by the feature amount calculation unit 12, and the teacher data creation unit 2 is based on the extracted feature amount data.
Evaluation is performed in 2a. Then, the basic knowledge currently set by the parameter adjusting unit 22b (or the temporary control knowledge corrected thereafter) is corrected so that the evaluation value becomes high,
The corrected provisional control knowledge is reset in the control unit 13 via the knowledge setting unit 14.

After that, the evaluation / correction of the above knowledge and the operation of the crane are alternately repeated, and if the desired specifications are finally satisfied, the correction process is terminated, and the control knowledge at that time is used as the control knowledge to be used during the actual operation. And set it.

In the above-mentioned embodiment, the teacher data is automatically extracted and generated by using the feature amount computing unit, but the present invention is not limited to this, and may be separately provided from the outside.

Next, a control device (movement control device) for operating an actual overhead crane will be described using the control knowledge set as described above. FIG. 26 shows a preferred embodiment of the position control device according to the present invention. As shown in FIG. 26, the basic configuration for performing normal position control, which is the premise of the present embodiment, is based on the above control knowledge. The system part that controls the operation status that was used when setting is applied as is. That is, the target stop position, which is one of the operator operation signals, the weight (suspension load) of the suspended load that is the object, and the suspension length are provided to the feature amount calculation unit 12 via the input processing unit 11. There is. Then, the sensor output obtained from the crane 25 (data used as input data at the time of knowledge setting, the swing angle of the suspended load, the position where the trolley exists, the moving direction, etc.) is also given to the feature amount calculation unit 12.
Then, the feature amount calculation unit 12 performs necessary data processing (absolute value of shake angle, detection of sign associated with shake direction (positive for shake in the same direction as the traveling direction), etc., and extracts. Alternatively, the sensor output is converted into data that can be processed by the control unit 13, and the obtained feature amount is given to the control unit 13.

In the control unit 13, the knowledge setting unit 14 gives the control knowledge (fuzzy knowledge) necessary for automatically controlling the operating state (position control for moving to a predetermined position without causing shake). Based on the feature amount sent from the feature amount calculation unit 12, fuzzy inference is performed by using the above knowledge to determine the ON / OFF timing of the relay, the position of the trolley, the moving direction, and the swing angle of the suspended load. While monitoring etc., the relay is turned on at the determined timing.
A control command is sent to the output processing unit 15 to turn it off / off.

The output processing unit 15 turns on / off the driver (actuator 26) of the drive motor based on the ON / OFF timing given from the control unit 13.
The F command is sent, and the actuator 26 turns on / off the drive voltage for the motor by turning on / off the relay based on the command to operate the crane 25. More specifically, the trolley is moved to a desired position while suppressing swinging of the suspended load.

By the way, the setting process of the control knowledge that accompanies the initial setting is performed accurately, and the control knowledge is finally determined as described above until the desired specifications are finally satisfied. Therefore, the determined control knowledge becomes knowledge suitable for the overhead crane, and highly accurate positioning accuracy is secured. Since the above-mentioned basic configurations 11 to 13 and 15 are used for monitoring the operating state and determining whether or not the specifications are satisfied until the control knowledge is determined, in the normal case,
When the actual position control of the crane is performed by executing the control device of the basic configuration, the suspended load (trolley) that is the object (object) moves to the target stop position with almost no position deviation (within the allowable range). Can be stopped.

However, in practice, the suspension load is changed when the initial setting is performed, but since it is not possible to make adjustments covering all conditions, the load of the suspended load, which is the actual moving object, is initially set. If it is different from the one used for setting,
The accuracy may decrease, and the entered conditions (suspension load, suspension length, etc.) may differ from the actual ones.
Due to various reasons such as disturbance, the actual stop position may deviate significantly from the target stop position, and the position deviation may fall outside the allowable range.

Therefore, in the present invention, even if the position deviation is outside the allowable range, the suspended load is moved so as to be within the allowable range by fine adjustment, and the swing of the suspended load is suppressed during the movement. A position control device for enabling (in particular, suppressing the swing-back when stopped) is added to the above basic configuration.

Specifically, a stop deviation judging section 27 as a judging means and a fine movement operating section 28 are provided. Stop deviation determination unit 27
Is obtained by acquiring the target stop position and the actual stop position data from the feature amount calculation unit 12, obtaining the position deviation, and whether or not the distance of the absolute value of the position deviation is within the allowable range a,
In addition, when it is out of the range, it is determined whether or not it is equal to or larger than a predetermined reference value α, and the position deviation calculated based on the determination result is given to the input processing unit 11 or the fine movement operation unit 28.

An embodiment of the method of determining the reference value α according to the present invention for determining the reference value α will be described. First, the feedback control, that is, the limit distance (shortest distance) capable of positioning control by inching operation performed by using the above-described basic component is set as the reference value α. That is,
When the relay is turned on, a voltage is applied to the drive motor and the trolley moves accordingly. At this time, due to static friction, there is a response delay from the time the relay is turned on to the time when the relay starts moving, and the movement of the trolley is detected instantly after the movement due to the demand for resolution of the sensor that detects the position (movement) of the trolley. Instead of doing so, it may be possible to detect after moving a certain distance, and when that is detected, the relay is turned on.
Even in FF, the drive motor continues to rotate due to inertial force, and even if the drive motor stops, the trolley and the suspended load are very heavy, so the above movement was detected because of a certain braking distance. It moves a certain distance from the position until it finally stops.

Therefore, if the distance from the first stop position to the final stop position is shorter than the above, feedback control cannot be performed. Therefore, the distance α is set as a reference, and when the position deviation (absolute value) is larger than the reference value α, the absolute value of the position deviation is given to the input processing unit 11 as the next target movement distance, Again, the movement control (inching operation) using the basic configuration is performed. Then, the reference value α is obtained by an experiment, and is determined by executing the flow shown in FIG. 27 based on the principle described above.

In the illustrated process step, the movement start position α0 and the stop position α1 are obtained, and the reference value α is obtained by obtaining the difference between the two. Of course, the moving distances after the relays are turned off may be respectively obtained and both may be added, and various kinds of concrete arithmetic processing can be applied. Further, although α obtained by the above process is used as it is as the reference value, a constant margin may be added to the calculated value. By doing so, it is possible to suppress as much as possible the control by the inching operation at a distance where the feedback control cannot be performed.

Therefore, if the position deviation (absolute value) is greater than or equal to the reference value α, the position deviation judgment unit 27 sends the obtained position deviation to the input processing unit 11 and the position deviation is set as a new movement target value. As a result, a predetermined feature amount is sent to the control unit 13 via the feature amount calculation unit 12, and an inching operation is performed. That is, in the present embodiment, the characteristic amount calculation unit 12 and the control unit 13 serve as the third movement control means in the present invention.

On the other hand, when the absolute value of the position deviation is smaller than the reference value α and the feedback control cannot be performed and exceeds the permissible range a, the stop deviation judging section 27 obtains the fine movement operating section 28. Sent position deviation.

The fine movement operation unit 28 selects a required relay ON time based on the given position deviation, and sends a relay ON signal to the output processing unit 15 while observing the swinging state of the suspended load as necessary. It is like this. Then, the swinging state of the suspended load is acquired from the feature amount computing unit 12.

The fine movement control in the fine movement operating section 28 is as follows.
The built-in timer is used to control the relay to be turned on for a certain period of time. In principle, the relay is turned on twice. That is, as shown in FIG. 28, when the relay is turned on for a certain period of time, the trolley 2 first moves, and the suspended load 4 starts to move with a delay due to inertial force or the like, so that a slight shake occurs. Then, if the trolley 2 is stopped after being moved by a predetermined distance as it is, the suspended load 4 makes a pendulum motion with the position of the trolley 2 as a fulcrum (swing back).

Therefore, in order to suppress such shake, the relay is turned on once for a fixed time, and the trolley is moved for a predetermined distance.
After that, by turning on the relay twice at a predetermined timing, such as turning on the relay again at a predetermined timing, as described above, the shake caused by the movement by the first relay ON continues (residual vibration ) That 2
It is absorbed by the movement of the relay ON for the second time, and when the movement of the trolley by the relay ON of the second time is stopped, the suspended load 4 stops without shaking. It should be noted that this principle is the same as the suppression of the shake-back during the inching operation described above. The specific timing for turning on the relay for the second time is to turn on the relay for the second time when it swings forward with respect to the traveling direction (θ> 0, the swing angle is positive), as shown in the figure. To start. That is, the fine movement operation unit 28 in the present embodiment also serves as the first movement control means and the second movement control means in the present invention.

The position deviation is very small (however, a
In this case, the position adjustment is performed only by the fine movement operation by turning ON the relay once by the timer. That is, when the position deviation is very small, the distance traveled when the relay is turned ON is further reduced in consideration of the braking distance traveled by inertia after the relay is turned OFF. In such a small movement, the suspended load caused by the movement is reduced. The shake is small enough to be ignored. That is, the same effect can be exhibited without suppressing the shake by turning ON the relay twice as described above. Therefore, in order to correct the position with a simple control in a short time, when the position deviation is very small, the correction is performed by the fine movement operation by turning ON the relay once. In the present invention, the relay may always be turned on twice.

Then, the ON time of the relay controlled by the above-mentioned two times (or once) timer is set to the preset relay O.
Access the N table (see Table 6 below) to access the first relay ON time (relay ON time 1) and the second relay O
It is determined by extracting N hours (relay ON time 2). Then, the table is stored in the storage unit in the fine movement operation unit 28.

[0122]

[Table 6] Further, specific numerical values of the table are also obtained in advance by experiments. That is, first, the time to turn on the relay is variously changed based on the timer, and the moving distance at that time is obtained. An example of the result of this experiment is shown in Table 7 below.

[0123]

[Table 7] Then, based on the relationship between the relay ON time and the travel distance obtained above, a table is created by filling in the fields in the relay ON table according to the following three criteria (created according to this rule is shown in Table 6 above). Is). (1) In principle, by turning the relay ON twice,
Finely move to the specified position. Therefore, each time relay ON
The combination is such that the sum of the movement distances with time (see Table 7) becomes the target total movement distance. (2) When the moving distance is short (3 cm or less in the example of Table 7:
In order to detect a distance that does not swing much by experiment), move the relay to the target position by turning ON the relay only once. (3) First relay ON time, second relay ON
Try to be over time. This is the second relay O
This is because if the N time is longer, the shake may be amplified rather (the force in the direction of canceling the residual vibration becomes too large).

Next, an embodiment of the position control method according to the present invention will be described based on the above embodiment. First, a normal controller, that is, basic configurations 11 to 13 and 15
The suspended load is moved to the target stop by the position control by (ST0). Then, the actual crane stop position (trolley, suspended load stop position) at that time is acquired from the sensor output (ST1). Note that the sensor output is actually acquired as the feature amount data extracted via the input processing unit 11 and the feature amount calculation unit 12.

At the same time, the target stop position is also acquired, the difference between the two is calculated, and the position deviation is calculated (ST2). And
It is judged whether or not the absolute value of the position deviation is within the allowable range a (ST3), and if it is within the range, the movement process is terminated without finely adjusting the position.

On the other hand, when the absolute value of the position deviation exceeds the allowable range, it is judged whether or not the absolute value of the position deviation is equal to or greater than the reference value (the feedback controllable critical distance) α (ST4). When the value is equal to or larger than the reference value, the fine adjustment is determined by the inching operation, and when it is less than the reference value, the fine adjustment is determined by the fine moving operation by the timer control, and a predetermined signal is output. The processing from step 1 to this step is performed by the stop deviation determination unit 27.

If it is determined that the inching motion is the new deviation target value (distance), the inching motion in step 0 is performed in accordance with the same principle, and the inching steady is executed to obtain the predetermined position. (ST5).

If it is determined in step 4 that the absolute value of the position deviation is less than the reference value α, the position deviation is given to the fine movement operation unit 28 to perform the fine movement operation. Then, the fine movement operation unit 28 subtracts the absolute value of the allowable error from the absolute value of the position deviation (ST6), and determines the next movement target distance.

Then, the relay ON time is determined by referring to the relay ON table based on the obtained moving target distance (ST7). Then, the timer is operated for the determined first relay ON time to operate the crane. Also, in the case of a normal pattern in which the relay is turned on twice,
The swing of the moving load is monitored by the first relay ON, and the second relay is turned ON when the swing becomes positive.
After turning on the timer and holding the ON state for a certain period of time, the relay is turned off (ST8). In this way, fine movement control based on the timer is performed.

After the fine adjustment by the inching operation or the fine moving operation, the procedure returns to step 1 and it is judged whether or not the vehicle can be stopped at the target stop position (within the allowable error range), and within the allowable range. The above process is repeated until the position is reached.

The above-described embodiment shows an example applied to a control device for a ceiling clay, and the normal movement control performed in step 0 of FIG.
The present invention is not limited to this, but is applied to the control of an overhead crane in which the first normal position control is performed by conventional speed control using an inverter motor, for example. May be. Further, the present invention is not limited to an overhead crane, and can be applied to a control device in various moving devices in which vibration is generated when a normal crane moves or when an object moves.

[0132]

As described above, in the control knowledge setting method and device in the overhead crane control device according to the present invention, the basic knowledge is set, and the basic knowledge is actually used as the starting point of "operation control → evaluation → knowledge correction". By repeating "," the control knowledge is automatically optimized. Then, the control knowledge finally determined in this manner is based on the crane state (the characteristics of the crane itself and the operating environment).

Also, the basic knowledge that is the standard for the first operation for adjustment is suitable for the crane, and the accuracy gradually improves based on that, so the adjustment process of such knowledge is performed. It is possible to perform the correction / adjustment process safely without causing a large swing of the suspended load during the operation (during the operation of the overhead crane).

Therefore, when the crane is actually operated by using such control knowledge, the controller performs optimum control, so that the steady rest / positioning etc. according to the crane can be performed even if there is no particular skill or specialized knowledge. Operation control can be performed.

Further, in the position control method and apparatus according to the present invention, even when the position deviation is smaller than the reference value and the feedback control cannot be performed, the position is slightly moved in two steps and the second movement is started. By adjusting the timing, the object can be moved to the desired target stop position. Moreover, by dividing into two times, the moving distance of one time becomes shorter, the generated vibration itself becomes smaller, and the residual vibration generated by the first movement is attenuated by the second movement, so that it can be done in a short time. It is possible to suppress the vibration, and it is possible to make the vibration hardly occur when the target stop position is reached.

Further, in the method of determining the reference value according to the present invention, the reference value is obtained by actually conducting an experiment,
A reference value suitable for the system to which the application is applied can be determined.

[Brief description of drawings]

FIG. 1 is a block diagram showing a preferred embodiment of the present invention.

FIG. 2 is a diagram illustrating the principle of steady rest.

FIG. 3 is a diagram illustrating the principle of steady rest.

FIG. 4 is a diagram illustrating a feature amount to be extracted.

FIG. 5 is a diagram showing a membership function of a target moving distance which is an input condition of control knowledge.

FIG. 6 is a diagram showing a membership function of a suspension load which is an input condition of control knowledge.

FIG. 7 is a diagram showing a membership function of a suspension length which is an input condition of control knowledge.

FIG. 8 is a diagram showing a membership function of a movement OFF designated distance which is an input condition of control knowledge.

FIG. 9 is a diagram showing a membership function of a shake peak 1 which is an input condition of control knowledge.

FIG. 10 is a diagram showing a membership function of the code of shake peak 1 which is an input condition of control knowledge.

FIG. 11 is a diagram showing a membership function of shake position feature 1 which is an input condition of control knowledge.

FIG. 12 is a diagram showing a relationship between input and output variables of fuzzy inference performed by a control unit.

FIG. 13 is a diagram showing a membership function of a stop start position which is one of inference outputs.

FIG. 14 is a diagram showing a membership function of relay ON timing which is one of inference outputs.

FIG. 15 is a diagram showing a membership function of relay ON duration which is one of inference outputs.

FIG. 16 is a diagram showing a correlation in a default value of a relay OFF timing and a relay ON continuation time with respect to a movement OFF designated distance and a shake 1 absolute value.

FIG. 17 is a diagram illustrating a function of a knowledge correction unit (parameter adjustment unit).

FIG. 18 is a diagram illustrating a function of a knowledge correction unit (parameter adjustment unit).

FIG. 19 is a diagram illustrating a function of a knowledge correction unit (parameter adjustment unit).

FIG. 20 is a diagram illustrating a function of a knowledge correction unit (parameter adjustment unit).

FIG. 21 is a diagram illustrating a function of a knowledge correction unit (parameter adjustment unit).

FIG. 22 is a diagram illustrating a function of a knowledge correction unit (parameter adjustment unit).

FIG. 23 is a diagram illustrating the function of a knowledge correction unit (parameter adjustment unit).

FIG. 24 is a diagram illustrating a function of a knowledge correction unit (parameter adjustment unit).

FIG. 25 is a diagram illustrating a function of a knowledge correction unit (parameter adjustment unit).

FIG. 26 is a diagram showing an embodiment of the position control device according to the present invention.

FIG. 27 is a flowchart showing an embodiment of a reference value determining method according to the present invention.

FIG. 28 is a diagram illustrating an operation principle of a fine movement operation.

FIG. 29 is a flowchart showing an embodiment of the position control method according to the present invention.

FIG. 30 is a diagram showing an example of an overhead crane.

FIG. 31 is a diagram illustrating a steady rest when stopped.

[Explanation of symbols]

 12 feature amount calculation unit (third movement control unit) 13 control unit (controller, third movement control unit) 14 knowledge setting unit 20 basic knowledge generation unit 20a basic structure input unit 20b basic knowledge adjustment unit 22 knowledge correction unit 22a Teacher data creation unit 22b Parameter adjustment unit 27 Stop deviation determination unit (determination means) 28 Fine movement operation unit (first and second movement control means)

 ─────────────────────────────────────────────────── ─── Continuation of front page (72) Naohito Shiki, No. 10 Hanazono Dodo-cho, Ukyo-ku, Kyoto City, Kyoto Prefecture Omron Co., Ltd. Co., Ltd. (72) Inventor Takashi Inoue No. 10 Hanazono Dodo-cho, Ukyo-ku, Kyoto-shi, Kyoto Omron Co., Ltd.

Claims (15)

[Claims] 1. A method for setting control knowledge in an overhead crane control device for controlling operating conditions such as steady rest and positioning of an overhead crane, which is provided with basic knowledge that is approximately adapted to the characteristics of the overhead crane. The operating condition of the overhead crane is controlled based on the basic knowledge, and the specified feature amount is extracted from the operating condition of the overhead crane at that time. Based on this feature amount, the current control knowledge is evaluated, and the evaluation value improves. Control knowledge setting method in an overhead crane control device for adjusting the control knowledge as described above. 2. The basic knowledge is obtained by giving at least one of structural data about a ceiling crane to be processed and design data such as a control target value, and modifying original data prepared in advance based on each of the given data. The control knowledge setting method in the overhead crane control device according to claim 1, wherein 3. A setting device for setting control knowledge in an overhead crane control device for controlling operation states such as steady rest and positioning of an overhead crane, and controlling the overhead crane by using given provisional control knowledge. A controller, a sensor that detects the state of the object controlled by this controller, a feature amount calculation means that calculates the feature amount using the output of this sensor, and an operating state is evaluated based on this feature amount. Control knowledge setting device in an overhead crane control device, which comprises knowledge correction means for adjusting provisional control knowledge so as to improve performance. 4. Basic knowledge is obtained by receiving at least one of structural data and design data such as a control target value for an overhead crane to be processed, and correcting the original shape data prepared in advance based on each received data. 4. The basic knowledge generating means for generating the basic knowledge is set, the basic knowledge generated by the basic knowledge generating means is set in the controller, and the knowledge initially set as the temporary control knowledge is set as the basic knowledge. A control knowledge setting device in the overhead crane control device described in. 5. The control knowledge and the basic knowledge are fuzzy rules and membership functions.
A control knowledge setting device in the overhead crane control device described in. 6. The control knowledge setting device in the overhead crane control device according to claim 3, wherein the overhead crane to be processed is drive-controlled by ON / OFF control of a relay. 7. A position control device for performing control to move an object to a target stop position, wherein a timer for the object is set when a position deviation from a position where the object exists to the target stop position is a reference value or less. A first movement control means for moving the object for a fixed time, and the target again in a direction to damp the vibration remaining in the object generated by the movement of the object accompanied by the control by the first movement control means. The second to move and stop the object
And a movement control means of the position control device. 8. The system according to claim 7, wherein the object is a system including a crane and an object transported by the crane, and the vibration is vibration of the transported object with respect to the crane. Position control device. 9. A third movement control means for detecting at least the existing position of the object and moving the object to the target stop position by feedback control; an absolute value of the position deviation and the reference value. In comparison, it is determined that the first and second movement control means are used when the value is smaller than the reference value, and the position control is performed using the third movement control means when the value is larger than the reference value. The position control device according to claim 7 or 8, further comprising: a determination unit. 10. The position control device according to any one of claims 7 to 9 is mounted on a control device of an overhead crane driven and controlled by ON / OFF control of a relay, and the vibration is conveyed by the crane. A position control device which is a shake of an object, wherein the first and second movement control means determine an ON time of a relay. 11. A position control method for moving an object to a target stop position, comparing a position deviation from a position where the object exists to the target stop position with a reference value, and when the position deviation is small,
After moving the object for a certain period of time using a timer and then temporarily stopping it, the object is moved for a certain period of time in a direction in which vibration remaining in the object caused by the movement of the object is attenuated. A position control method characterized by the above. 12. When the position deviation is smaller than the reference value and the distance is such that vibration hardly occurs when the object is moved, the object is set to 1 for a certain period of time by using a timer. The position control method according to claim 11, wherein the movement is attempted to reach the target stop position by being moved once. 13. The absolute value of the position deviation is compared with the reference value, and when the difference is smaller than the reference value, the control according to claim 11 or 12 is performed, and when it is larger than the reference value, the moving state is set. The position control method is characterized in that the object is moved to the target stop position by feedback control while monitoring. 14. The method according to claim 11, wherein the object is moved to the target stop position by the feedback control according to claim 13, and then, when the absolute value of the position deviation becomes smaller than the reference value. A position control method characterized in that the control described in (1) is performed. 15. A method of determining the reference value, comprising moving a moving means for moving the object to move the object, and stopping the operation of the moving means when the movement of the object is detected, A method of determining a reference value, wherein a movement distance from a movement start position to a stop position of the object is obtained, and the movement distance or a value obtained by adding a predetermined margin to the movement distance is used as a reference value.