Sign up
Title:
Crane suspension control apparatus
United States Patent 3921818
Abstract:
A system for controlling a suspension type crane which moves transversely while suspending a load by a rope. The crane is accelerated at least two times to a predetermined maximum speed during an acceleration period, the swing of the rope is minimized when a predetermined maximum speed is reached, the crane is moved at the predetermined maximum speed for a predetermined interval, the crane is decelerated from the maximum speed at least two times during the deceleration period, and the crane is stopped when the swing of the rope is reduced to a minimum, and the areas of the acceleration and deceleration periods of the crane are made equal.


Application Number:
05/455906
Publication Date:
11/25/1975
Filing Date:
03/28/1974
Assignee:
Tokyo Shibaura Denki Kabushiki Kaisha
Primary Class:
Other Classes:
212/86, 212/328, 340/685, 388/815, 388/847, 388/904
International Classes:
B66C13/06; B66C13/22; (IPC1-7): B66C19/00
Field of Search:
318/384 212
View Patent Images:
US Patent References:
3850308APPARATUS FOR ACCOMMODATING THE PENDULUM ACTION OF A LOAD CARRIED BY A ROPE FROM A TRAVELLERNovember 1974Meyer et al.
3517830CRANESJune 1970Virkkala
3351213Control systemsNovember 1967Newman et al.
2806610Anti-swing craneSeptember 1957Goertz
Primary Examiner:
Spar, Robert J.
Assistant Examiner:
Johnson R. B.
Attorney, Agent or Firm:
Stevens, Davis, Miller & Mosher
Claims:
1. A control system for a suspension type crane running in the transverse direction, the improvement comprising means for providing a start command signal, means responsive to said start command signal for determining a maximum transverse running speed of said crane corresponding to the distance between the starting position and a predetermined target position of said crane, means for generating a deceleration command signal when said crane reaches a point a predetermined distance before said target position, which is determined by said maximum transverse running speed, and means responsive to said start command signal or said deceleration command signal for providing a predetermined acceleration-deceleration pattern signal corresponding to said maximum transverse running speed, whereby the transverse running speed of said crane is controlled so as to stop said crane at said target position.

2. The control system according to claim 1 wherein said crane includes a transversely running trolley operated by an electric motor and suspends a load by means of a rope, and said control system further comprises a speed controller responsive to said acceleration-deceleration pattern signal for controlling said motor.

3. The control system according to claim 2 which further comprises means for generating a speed signal proportional to the speed of said motor and means for negatively feeding back said speed signal to said speed controller.

4. The control system according to claim 3 which further comprises means for applying a signal corresponding to the swing angle of said rope at a point immediately prior to the stop of said trolley to said speed controller.

5. The control system according to claim 4 which further comprises means for applying to said speed controller a signal corresponding to the distance between said target position and the present position of said trolley.

6. The control system according to claim 1 wherein said deceleration command signal generating means includes means for comparing the distance S between a point at which deceleration of said crane is commenced and said target position with the deviation ΔL of the present position of said crane from said target position for generating said deceleration command signal when said deviation ΔL becomes equal to said distance S.

7. The control system according to claim 1 wherein said means for providing said acceleration-deceleration pattern signal comprises an integrator which is connected to integrate said maximum transverse running speed or a reference signal in accordance with the operation of said deceleration command signal generating means.

8. The control system according to claim 7 wherein said integrator includes means to reverse said pattern signal.

9. The control system according to claim 2 which further includes a switching time computer comprising means responsive to the operation of a controller for said trolley and the speed of said trolley driving motor for producing a signal │Vmax │ where Vmax represents a predetermined maximum transverse running speed of said trolley;

10. The control system according to claim 9 which further comprises means for detecting the polarity of a signal V1 - V2, where V1 represents a constant voltage start signal applied by said trolley controller and V2 the actual transverse running speed of said trolley; means responsive to the operation of said polarity detecting means for applying a signal +│α│ or -│α│ to one input of said third integrator, and means responsive to the operation of said first and second relay means for applying a signal 2 α to the other input of said third integrator.

Description:
BACKGROUND OF THE INVENTION

This invention relates to a method and system for controlling the positioning of a suspension type crane and more particularly to an improved method and system for suppressing swinging motions of a suspension rope of a trolley of the crane and for stopping the trolley at a correct target position when the swing of the rope is reduced to zero or substantially to zero.

When a suspension type crane is accelerated or decelerated during its transverse running, the rope suspending a load undergoes a pendulum motion. Such pendulum motion or swinging motion can be suppressed by the operation of the operator of the crane. Thus, when such swinging motion occurs the operator operates the controller of the crane for adjusting the transverse running speed to suppress the swinging motion. However, such adjustment cannot be made other than by a skilled crane operator and in most cases the adjustment of the transverse running speed becomes excessive or insufficient whereby a long time is required until the swinging motion is perfectly suppressed thus decreasing the cargo efficiency.

To obviate this difficulty, there has been proposed a method wherein the swinging angle θ of the rope and the angular velocity θ of the swinging motion are detected and signals corresponding to angle θ and angular velocity θ are negatively fed back to a transverse speed controller through a feedback circuit having a suitable gain for attenuating the swinging motion of the rope. With such a feedback system, if the gain of the feedback circuit were decreased for sufficiently suppressing the swinging motion the average transverse running speed would be decreased. Accordingly, a compromise method has been proposed in which an insensitive zone is provided for the feedback circuit for preventing the cargo efficiency from decreasing at the sacrifice of the accuracy of the swing suppression. Accordingly, such method is not satisfactory for such an application as a container crane which requires an extremely accurate swing suppression for the purpose of precisely lowering the load at a predetermined position.

In order to have a better understanding of this invention, the problem involved in the control system for effecting suppression of the swinging motion in a shortest time will be analyzed hereunder.

In a diagram shown in FIG. 2, let m represents the mass of a load, T the tension of a suspension rope, g the acceleration due to gravity. Under a balanced condition of the horizontal component and the vertical component of the force acting upon the load, the following equations of motion hold: ##EQU1## There are the following relations among x, y, 1 (length of the rope), θ (angle of swing) and X (distance between the origin and the trolley)

By differentiating both sides of equations 3 and 4 with respect to time t, we obtain ##EQU2## By additionally differentiating both sides of equations 5 and 6 with respect to time, we obtain ##EQU3## Substituting equations 7 and 8 for the lefthand sides of equations 1 and 2, respectively, ##EQU4## When an equation

is operated, the second term in the lefthand side of equation 9 and the first term in the lefthand side of equation 10 cancel with each other, and the righthand side of equation 9 and the second term in the righthand side of equation 10 also cancel with each other, thus ##EQU5## By dividing the both sides of equation 11 by m and by sustituting a relation sin2 θ + cos2 θ = 1 ##EQU6## By putting ##EQU7## we obtain ##EQU8## Where θ is small, then cos θ ≉ 1 and sin θ ≉ θ, so that equation 12 can be rewritten as follows ##EQU9## Thus, equation 13 expresses the pendulum motion of the rope and the load.

If we assume that, the length 1 of the rope is constant, then ##EQU10## and equation 13 can be rewritten as follows. ##EQU11## Since there is a relation: ##EQU12## by putting ##EQU13## and by substituting this relation in equation 14, we obtain ##EQU14## By dividing the both sides of equation 15 by 1 and by putting ##EQU15## the following relation can be obtained ##EQU16## When the both sides of equation 16 are integrated with respect to θ, under an assumption that the acceleration α of the trolley is constant, the following equation will be obtained ##EQU17## where Co represents an integration constant. By multiplying the both sides of equation 17 by 2, ##EQU18## By modifying the term in the bracket and by substituting the result of the following equation 19 into equation 18, we obtain equation 20. ##EQU19##

When the length 1 of the rope is constant, the relationship between ωθ and θ or the phase plane locus corresponds to a circular motion rotating in the clockwise direction at a constant angular velocity ω on a circle having a center at ##EQU20## as shown in FIG. 3.

The speed of the motion around the circle can be obtained as follows. As shown in FIG. 4, since ##EQU21## where ##EQU22## Accordingly, the angular speed of the circular motion of a point (ω θ, θ) can be expressed as follows ##EQU23## where ##EQU24## From equation 14 ##EQU25## By substituting equation 24 into equation 23 the following equation can be derived ##EQU26## This equation shows that the point (ω θ, θ) rotates on a circle in the clockwise direction at a constant angular velocity ω, as has been pointed out hereinabove.

It can be readily understood that, during acceleration since α∠0, the center of the circle lies on the negative side of axis ω θ, whereas during decelration since α∠0, the center of the circle lies on the positive side, as shown by FIGS. 3a and 3b. Where the trolley is running transversely at a constant speed, α = 0 so that the center of the circle coincides with the origin 0 as shown in FIG. 3c. Further, the radius of the circle is determined by the initial conditions.

Assume now that the trolley is started from standstill at a constant acceleration, decelerated at a constant deceleration during an interval t1 - t2 and thereafter again accelerated at a constant acceleration during an interval t2 - t3 until a maximum speed is reached, as shown in FIG. 1. Under these conditions, the relationship between the swing angle θ and the angular velocity θ of the swinging motion will now be considered with reference to the phase phane locus described above.

Since the initial conditions are: t = to, θ = 0 and θ = 0, the phase plane locus starts from the origin 0 as shown in FIG. 5 so that the initial radius of the circle is equal to O,O. When the position of a state point (ω θ, θ) at t = t1 is denoted by P1, the time required for the point P1 to move from O to P1 is equal to (t1 - to), and since the angular velocity of the circular motion of the state joint P1 about the center O1 is expressed by ω, the following relation holds

During a period t expressed by t1 ≤ t≤ t2, α∠0 so that the phase plane locus becomes a circle having a center at θ2. At an instant t = t1, arc P1 P2 intersects arc UP1 at point P1 so that the radius of the latter circle will be O2 P1. By denoting the position of a state point (ω θ, θ) at t = t2 by P2, the time required for the state point to move from point P1 to point P2 will be (t2 - t1) and since the angular velocity of the circular motion about the center O2 is ω, the following relation holds,

During an interval expressed by a relation t2 ≤ t ≤ t3, since α > 0, the phase plane locus again assumes the circle with its center at O1. At t = t2, since arc P2 P3 intersects arc P1 P2 at point P2, the radius of the circle having a center at point O1 is equal to O1 P2. Further, since the time required for the state point to move from point P2 to state point P3 is equal to (t3 - t2) and the angular velocity of the circular motion about center O1 is ω, the following relation holds

During an interval wherein t2 ≥t3, as α = 0, the phase surface locus takes the form of a circle having its center at the origin and a radius OP3.

From the foregoing description, it will be clear that the phase surface locus varies when the time instants t1 and t2, FIG. 1, at which the acceleration is switched to deceleration or vice versa are varied.

For example, when state point P3 is made to coincide with the origin O as shown in FIG. 6 by a suitable selection of acceleration-deceleration switching points t1 and t2, when the state point P3 is reached or when the trolley attains the maximum running speed, both swing angle θ and angular velocity θ of the swinging motion become zero so that it will be clear that during the succeeding interval in which the trolley runs at a constant speed the swing angle of the rope is always maintained zero.

Let us now consider a case wherein the trolley running at the maximum speed is to be stopped. Consider now a speed pattern as shown in FIG. 7 wherein the deceleration of the trolley is commenced at time t4, acceleration is commenced at time t5 and thereafter deceleration is commenced again at time t6. The phase plane locus in this case is shown in FIG. 8. As has been pointed out hereinabove, as it is possible to make zero both the swing angle θ and the angular velocity of the swinging motion θ during the interval in which the trolley is running at a constant speed, if the state point coincides with the origin at t = t4, the phase plane locus shown in FIG. 8 would originate from the origin O.

Since α<0 during an interval t4 ≤ t ≤t5, the phase plane locus will become a circle having its center at point O2 and a radius of O2 O.

Further, since α>0 during an interval t5 ≤ t≤t6, the phase plane locus will become a circle having its center at point O1 and a radius of O1 P5.

During an interval t6 ≤ t≤t7 in which α<0, the phase plane locus will again become the circle having its center at point O2 and a radius O2 P6 as determined by the intersecting condition of the locus at state point P6.

At a time t = t7 the trolley stops and thereafter since α = 0, the phase surface locus would be a circle having its center at the origin O and a radius of OP7.

In this manner, by the suitable selection of the acceleration-deceleration switching points t5 and t6 it is possible to make the phase plane locus as that shown in FIG. 9. At a time t = t7 at which the trolley stops, since both the swing angle θ of the rope and the angular velocity θ of the swinging motion are zero and since α = 0 during the period t ≤ t7, the rope will be maintained in a condition in which its swing is zero.

For this reason, where the speed pattern from start to stop of the trolley is selected to be equal to that shown in FIG. 10 and where the switching paints t1, t2 - t6 are determined such that a phase plane locus as shown in FIG. 11 can be provided, it is possible to make zero the swing of the rope both during the period t3 - t4 in which the trolley runs at a constant running speed and at time t7 at which the trolley stops.

It will thus be clear that it would not be necessary to vary the speed pattern during the intervals to - t3 and t4 - t7 in accordance with the transverse running distance or stroke of the trolley if the interval t3 - t4 in which the trolley runs at the constant speed were varied in accordance with the transverse stroke of the trolley when it is controlled by the speed pattern as shown in FIG. 10. The fact that the swing angle θ of the rope is kept at zero during the interval t3 - t4 of the constant speed running is advantageous from the standpoint of safeness of the cargo operation. The locus P4 - P5 - P6 - P7 shown in FIG. 11 can be obtained by tracing the locus O - P1 - P2 - P3 in the opposite direction.

Consequently, following equations hold.

Let us now consider the acceleration-deceleration switching points which are so selected that the phase plane locus shown in FIG. 11 can be obtained.

From FIG. 10, the following equation holds

and from FIG. 11

so that from equations 26 and 28 we obtain

With reference to an arc OP1 shown in FIG. 11 since OO1 = O1 P1 ##EQU27## with reference to an arc P1 P2 having a center at O21 ##EQU28## However ##EQU29## By substituting equation 37 into equation 36, we obtain ##EQU30## From equations 35 and 38, we obtain ##EQU31## By the concurrent solution of equations 32, 34 and 39, (t1 - to), (t2 - t1) and (t3 - t2) can be obtained.

From the foregoing description, it can be noted that it is possible to make zero the swing of the rope at the time of stopping the trolley when the trolley is controlled according to the speed pattern shown in FIG. 10 and when the acceleration-deceleration switching points which satisfy equations 32, 34 and 39 are selected. However, as equation 39 is a complicated equation in terms of implicit functions including complicated trigonometrical functions, a complicated and expensive electronic computor is necessary for the simultaneous solution of equations 32, 34 and 39. Incorporation of such an expensive computer into the control system of a crane increases the cost thereof so that at present the control system is not provided with such computer but merely depends upon a mathematical analysis.

The inventor has solved equations 32, 34 and 39 with an electronic computer utilizing the data regarding the rope length and the transverse running speed of the trolley and found that a high accuracy sufficient for the practical use can be obtained from the following equation 40 in which interval (t1 -to) is approximated as the explicit functions of Vmax, and 1.

where a, b, and c represent constants.

Accordingly, t2 - t1 and t3 - t2 can be obtained as follows from equations 32 and 34. ##EQU32##

As can be noticed from FIG. 10, time t3 (or t7) represents an instant at which the transverse running speed of the trolley reaches a predetermined ultimate value and at which the difference between the ultimate speed commanded by the trolley controller and the actual running speed of the trolley reduces to substantially zero. Accordingly, by terminating the acceleration or deceleration by detecting this condition it will be not necessary to calculate t3 - t2 by using equation 42. In other words, it is sufficient to calculate (t1 - to) and (t2 - t1) alone by using equations 40 and 41.

The straight lines shown in FIG. 12 show the relationship between the switching time t1 - to and the rope length obtained by solving equations 32, 34 and 39 for the rope length of from 7.5m to 22.5m and the trolley running speed of from 31.25 m/min. to 125m/min. Straight lines shown in FIG. 12 show the solution of equation 40. Thus, FIG. 12 shows that even when the switching time is calculated according to equation 40 of approximation, it is possible to realize sufficiently high practical accuracy for the ranges of the rope length variation and the trolley speed variation encounted in the actual use.

Equations 29, 30 and 31 also show that the stroke of the trolley (the area of the lefthand shaded portion in FIG. 1) during interval to - t3, in which the trolley has accelerated to a maximum speed Vmax after starting is equal to the stroke (the area of the righthand shaded portion in FIG. 1) during interval t4 - t7 in which the trolley has decelerated from Vmax to standstill. This method of operation is the result of approximation of the above described analysis in terms of the maximum speed and the length of the rope.

From this it can be understood that it is possible to terminate the swinging motion of the rope when the trolley stops by measuring or calculating the distance S over which the trolley travels from starting until the maximum speed is reached and by issuing a deceleration initiation command signal when the trolley reaches a point spaced from a target stopping position by a required distance.

From the foregoing description, it will be clear that according to the control system described hereinabove, it is possible to substantially reduce to zero the swing of the rope when the trolley is accelerated to a predetermined maximum speed Vmax and when the trolley is brought to stop. With this system, however, as no signal is given as to when the deceleration should be commenced at time t4, the crane operator must determine by himself such time by relying upon his skill. Accordingly, it is not always possible to correctly stop the trolley at the target position at time t7.

SUMMARY OF THE INVENTION

It is an object of this invention to provide a novel method and system for controlling a suspension type crane capable of suppressing to substantially zero the swing of the load suspending rope while the crane is running at a constant speed.

Another object of this invention is to provide a novel method and system for controlling a suspension type crane capable of initiating the deceleration at a correct time for stopping it at a predetermined target position without any swinging motion of the rope.

Still another object of this invention is to provide a novel method and system of controlling a suspension type crane capable of operating the same with a minumum time without permitting any swing to the rope while the crane is running at a constant speed and when the crane is stopped, thereby increasing the cargo efficiency. A further object of this invention is to provide a novel acceleration-deceleration pattern signal generating circuit suitable for use in this invention.

According to one aspect of this invention there is provided a method of controlling a suspension type crane which is moved transversely while suspending a load by means of a rope wherein the crane is accelerated at least two times at spaced points to a predetermined maximum speed during the acceleration period, the swing of the rope is minimized when the predetermined maximum speed is reached, the crane is run at the predetermined maximum speed for a predetermined interval, the crane is decelerated from the maximum speed at least two times at spaced points during the deceleration period, and the crane is stopped when the swing of the rope is reduced to a minimum, characterized in that the areas of the acceleration and deceleration periods of the crane are made equal.

According to another aspect of this invention there is provided a control system for a suspension type crane running in the transverse direction, characterized by comprising means for providing a start command signal, means responsive to the start command signal for determining a maximum transverse running speed of the crane corresponding to the starting position and a predetermined target position of the crane, means for providing a deceleration command signal when the crane reaches a point a predetermined distance before the target position, which is determined by the maximum transverse running speed, means for generating a deceleration command signal, and means responsive to the start command signal or the deceleration command signal for providing a predetermined acceleration-deceleration pattern signal corresponding to the maximum transverse running speed, whereby the running speed of the crane is controlled so as to stop the crane at the target position.

BRIEF DESCRIPTION OF THE DRAWINGS

In the accompanying drawings:

FIG. 1 is a diagram showing a typical transverse running speed pattern of the trolley of a suspension type crane which can be realized by the control system of this invention;

FIGS. 2 to 11 inclusive are diagrams useful to explain the principle of this invention;

FIG. 12 is a graph showing the relationship between the switching time and the rope length calculated for various rope lengths and trolley speeds which are used actually;

FIG. 13 is a block diagram of one embodiment of the novel control system of this invention;

FIG. 14 is a block diagram of a modified embodiment of this invention;

FIG. 15 shows a modified speed pattern;

FIG. 16 is a block diagram of a crane control system;

FIG. 17 is a block diagram showing one example of the acceleration-deceleration switching time operating circuit utilized in this invention;

FIG. 18 is a diagram for explaining the operation of the operating circuit shown in FIG. 17;

FIG. 19 shows a block diagram of the speed reference generating circuit controlled by the operating circuit shown in FIG. 17; and

FIG. 20 is a diagram for explaining the operation of the speed reference generating circuit shown in FIG. 19.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

FIG. 13 shows the construction of one embodiment of the control system of this invention which comprises a deceleration command signal generator A which generates a deceleration commandsignal in accordance with the deviation Δ L of the present position L from the target position Lo for providing a transverse running speed pattern as shown in FIG. 1, a maximum transverse running speed determining unit B which determines the maximum transverse running speed Vmax in accordance with a deviation Δ L corresponding to the distance Lo to the target position and rope length 1 (for the reason to be described later, rope length is not taken into consideration at the present stage of the description), an acceleration-deceleration pattern generator C connected to receive the output from the maximum transverse running speed determining unit B when the deceleration command signal generator A operates for forming the transverse running speed pattern shown in FIG. 1, and a speed controller D for controlling the speed of a motor M for driving the trolley in accordance with the output from the acceleration-deceleration pattern generator C. These component elements will be described in detail in the following.

The deceleration command signal generator A will firstly be described. The distance S over which the trolley which has been running at the maximum speed Vmax should travel before it is stopped in accordance with the speed pattern shown in FIG. 1 can be derived out from equations 29, 30 and 31, thus

Intervals (t1 - to) and (t2 - t1) can be obtained from the following equations. ##EQU33## and ##EQU34## Instead of using equation 41, an approximate value of distance s can be derived out from equations 42, 44 and the following equation 45 which is an equation of approximation expressed by an explicit function of the maximum speed Vmax and the rope length 1

where a, b and c are constants.

This also corresponds to the distance of running during interval t7 - t6 shown in FIG. 1 but this distance of running can be obtained by storing the running distance during interval t8 - t3. This is because the running distances during intervals t4 - t7 and to - t3 are equal as has been mentioned hereinbefore. In any way, the distance S required to stop the trolley has already been determined by the time at which the trolley attains its maximum speed Vmax. Such measurement or calculation is required to be made only once during the operation of the crane, and the result is given to the deceleration command signal generator A.

Thus, the deceleration command signal generator A stores a signal corresponding to distance S and operates to compare the deviation Δ L (= Lo - L) of the present position L of the trolley from the target postion Lo, with signal S for producing a deceleration command signal when Δ L becomes equal to S. The deceleration command signal can be generated by switching the speed command for the acceleration-deceleration pattern generator C from Vmax to 0, as shown in FIG. 13.

The maximum transverse running speed determining unit B will now be described. While in the foregoing description it was explained that the maximum transverse running speed Vmax is prescribed, as can be noted, from equation 43 where the maximum speed Vmax and rope length 1 are given it is possible to determine acceleration and decleration intervals t1 - to, t2 - t1, t3 - t2, t5 - t4 , t6 - t5 and t7 - t6.

Accordingly, where the values of Vmax and 1 are given, the distance over which the trolley runs between starting and completion of acceleration, and the distance over which the trolley runs from the maximum speed until it stops will also be determined. For this reason, even when a deceleration command signal is generated at an instant t3 at which acceleration has been completed thus making t3 = t4, the trolley runs a ddistance 2S, that is, the sum of the distance S from start to the completion of acceleration and the distance S from the maximum speed to the stop. Accordingly, the running distance Lo is shorter than 2S, so that it is necessary to suitably decrease the maximum speed.

The purpose of the maximum transverse running speed determining unit B is to determine such an optimum maximum transverse running speed. The maximum speed Vmax can be derived from equations 42, 43 and 44 by putting ##EQU35## (In lieu of equation 44, equation 45 can also be used). For this reason, in FIG. 12 the distance between the starting position and the target position is designated by Lo /2. As shown in FIG. 12, since the maximum speed Vmax does not vary so much with the rope length 1, it is possible to simplify the control device by ignoring the effect of length 1. FIG. 13 shows such simplified construction wherein a signal representing 1 is not applied to the maximum transverse running speed determining unit B.

Turning now to the acceleration-deceleration pattern generator C, it is comprised essentially of integrators and is constructed and operated to generate a predetermined acceleration-deceleration pattern signal as will be described later in detail in connection with FIGS. 17 to 20. At this stage of description, it is merely pointed out that the deceleration command signal generator A switches the input to the acceleration-deceleration pattern generator C from signal Vmax to a reference signal O at time t4. Further, a signal representing the rope length 1 is also applied to the pattern generator C for compensenting for the variation in the value of the maximum speed Vmax caused by the variation in the rope length 1.

Upon reception of these input signals the acceleration-deceleration pattern generator C generates a pattern signal Vref (as shown in FIG. 13) which is applied to the speed controller D. In response to this pattern signal Vref, the automatic speed control circuit ASC provided for the speed controller D controls the speed of trolley driving motor M. The speed control circuit ASC is provided with a negative feedback circuit including a tachometer generator TG coupled to motor M.

The crane control system shown in FIG. 13 operates as follows. For the sake of description it is assumed herein that the crane is installed in a container yard at a warf for transporting containers between a container ship alongside the warf and the container yard. When an operator provides a command signal for the "Off board commencement", the maximum transverse running speed determining unit B forms a maximum transverse running speed signal Vmax corresponding to deviation Δ L (= Loo) of the present position of the trolley from the target position. This signal Vmax is applied to the acceleration-deceleration pattern generator C in response to the operation of the deceleration command signal generator A. At the same time, rope length signal 1 is applied to pattern generator C.

In response to these signals the acceleration-deceleration pattern generator C produces a pattern signal in accordance with equation 45. This pattern signal produces a speed pattern which causes the rope swing to decrease to zero at the time t3 of completing acceleration, that is the speed pattern during the interval t2 - t3 in FIG. 1.

Since this speed pattern is applied to the transverse speed controller D as a speed reference the trolley is accelerated to the maximum speed Vmax according to this speed pattern and the swing of the rope is decreased to zero when the trolley attains the maximum speed. Whereupon, the deceleration command signal generator A operates to compare distance S which is necessary for stopping the trolley and can be derived as described above with the deviation ΔL of the present position of the trolley from the target position, thus applying speed command signal O to the acceleration-deceleration pattern generator C when Δ L becomes equal to S. Then, the integration operation is performed in the reverse direction as has been pointed out before thereby producing a speed pattern that causes the swing of the rope to reduce to zero at and after a time t7 at which the trolley should be stopped, or the speed pattern during interval t4 - t7 in FIG. 1. As this speed pattern is applied to the speed controller D, the trolley is caused to run and stop in accordance with this speed pattern and at and after time t7, the swing of the rope will be decreased to zero. At this time, the trolley and hence the load will be correctly positioned at the target position.

FIG. 14 shows a block diagram of a modification of the control system shown in FIG. 13 in which means for compensating for external disturbances and various errors are added to the control system shown in FIG. 13. A so-called feed forward control system not provided with a feedback circuit as shown in FIG. 13 can be used only in a case wherein the rope swings as expected and there is no error. Actually, however, such a case does not exist. More particularly, measuring errors of the rope length, the error in the computation of the distance S desired for correct stopping and external distrubances such as the effect of wind are inevitable. For this reason, it is often impossible to reduce to zero the rope swing at the time of stopping the trolley thereby mispositioning the load.

In the control system shown in FIG. 14, a feedback control is incorporated into a feed forward control system. To control the residual swing of the rope to be within a permissible range, a signal Kθ corresponding to the rope swing angle θ immediately prior to the stopping of the trolley is fed back to the transverse running speed controller D. Further, for the purpose of controlling the position error caused by various errors described above to be within a permissible range, a signal, KL produced by amplifying at a suitable gain a quantity corresponding to the difference Δ L between the present position of the trolley when it reaches a point close to the target position of the trolley, and the target position is fed back to the transverse running speed controller D. In this connection, if the swing angle signal Kθ and position deviation signal KL were not properly related, the feedback system would become unstable. Generally speaking, in order to stabilize the feedback system the swing angle signal Kθ and the position deviation signal KL should be positive.

In FIG. 14, a contact e is arranged to be closed while the trolley is running in the transverse direction at a uniform speed and during a relatively short interval including the stopping point of the trolley but excluding an interval between t4 and a point close to t7, whereas a contact f is arranged to be closed immediately before the stopping point of the trolley. With this arrangement, from the starting point to a point at which the acceleration is completed a feed forward control is provided similar to the control system shown in FIG. 13 and when the trolley attains a substantially uniform speed, the feedback control of the rope swing angle is effected thus correcting the swing preventing operation. As will be described later in more detail, upon occurrence of a deceleration command signal the control is returned back to the feed forward control to decelerate the trolley according to a predetermined speed pattern and when the trolley approaches the target stopping position the rope swing angle signal Kθ and the position deviation signal ΔL are fed back thereby providing a correction operation so as to limit the error in the position of the trolley and hence of the load to be within a permissible range.

FIG. 15 illustrates another example of the speed pattern which is different from that shown in FIG. 1 in that t1 - to ≠ t3 - t2 and t5 - t4 ≠ t7 - t6. Even with such a modified pattern it is only necessary to make equal the shaded areas on the left and righthand sides. This modified speed pattern is suitable for a case where the period of the pendulum motion is shorter than ##EQU36##

FIG. 16 shows a block diagram of a crane control system to which the control system of this invention is applicable. A load m suspended by a rope 1' is hoisted or lowered by a hoist motor MM and the length 1 of the rope is detected by a synchro transmitter SY, for example. The hoist motor MM is energized by a variable voltage source SCR1 under the control of a speed controller S1 which is controlled by a hoist controller 101. The trolley driving motor M is energized by another variable voltage source SCR2 under the control of a speed controller S2 (corresponding to speed controller D shown in FIGS. 13 and 14). There are also provided a switching time computor 102 for computing said equations 42 and 43 to determine acceleration-deceleration switching points in response to a transverse running speed signal V1 determined by a transverse running controller 100, a speed signal V2 generated by a tachometer generator TG2 coupled to motor M and representing the actual speed of the trolley and a signal l generated by the synchro transmitter SY and representing the length of rope 1', and a reference speed signal generator 103 responsive to the output from the switching time computor 102.

The detail of the switching time computor 102 will be described with reference to FIGS. 17 and 18. At time to (FIG. 18) when the transverse running controller 100 is operated to the OFF BOARD side while the trolley is at a standstill or under a condition of V2 = 0, then V1 = Vmax and this signal will be applied to a signal variation detector Z and to a first addition circuit A1. In response to the output from the signal variation detection circuit Z, relay Ryo detects the operation of the transverse running controller 100, as shown by a graph "Ryo " in FIG. 18. To the other input of the addition circuit A1 is applied the speed signal V2 generated by the tachometer generator TG2 as an inversion input, and the output (V1 - V2) from the addition circuit A1 is applied to an absolute value circuit X to obtain the absolute value │V1 - V2 │ which is applied to a relay amplifier Ry A1 and to a holding circuit H via the normal open contact Ry Oa of relay Ry O. As shown by a graph "Ry 4" in FIG. 18, relay Ry 4 connected to the output of relay amplifier X is deenergized when the output from the absolute value circuit X becomes zero. For a short interval in which the controller 100 is operated, since V2 - 0 and a signal │V1 - V2 │ = │Vmax │ is applied to the holding circuit H through contact Ry Oa the holding circuit operates to hold this signal │Vmax │. As the contact Ry O2a of relay Ry O is closed, relay Ry 1 is operated which is held energized through contacts Ry 4a and Ry 1a until relay Ry 4 is deenergized, as shown by graph "Ry 1" in FIG. 18. The output │Vmax │ from the holding circuit H is applied to a setter S1 which sets the constant a, and the output a │Vmax max│ is applied to the first input of a second addition circuit A2. A setter S2 responsive to the rope length signal 1 from synchro transmitter SY sets the constant b and its output b1 is applied to the second input of the addition circuit A2. A variable resistor RH1 is provided to set the constant c which is applied to the third input of the addition circuit A2. Accordingly, addition circuit A2 produces an output (a│ Vmax│ + b1 + c) which is applied to the inversion input of a third addition circuit A3 and to the multiplying terminal of the fourth addition circuit A4. Energization of relay Ry 1 applies a definite voltage produced by a variable resistor RH2 upon an integrator I1 via its contact Ry 12a to vary the output of integrator I1 as shown by graph "I1 " shown in FIG. 18. The output of integrator I1 is applied to one input of addition circuit A3. When the output from integrator I1 becomes equal to the output (a│ Vmax│ + b1 + c) at time t1, a relay amplifier Ry A2 operates to energize relay Ry 2 as shown by a graph "Ry 2" shown in FIG. 18. Since a signal corresponding to (a│ Vmax│ + b1 + c) is applied to the inversion input of addition circuit A3, by suitably selecting the integration gain of integrator I1 it is possible to make equal the interval (t1 - to) from the time of operation of relay Ry 1 to the time of operation of relay Ry 2 and (a│ Vmax│ + b1 + c). Upon energization of relay Ry 2, integrator I2 begins to integrate the constant voltage provided thereto by variable resistor RH2 through contact Ry 12a and the normal open contact Ry 2a of relay Ry 2, thus producing an output as shown by a graph I2 in FIG. 18 which is applied to the noninversion input of the fifth addition circuit A5. Output ##EQU37## of setter S3 is applied to the inversion input of the fourth addition circuit A4 whereas the signal (a │ Vmax│ + b1 + c) is applied to the noninversion input of the addition circuit A4 and multiplied by 2. Accordingly, addition circuit A4 produces an output ##EQU38## which is applied to the inversion input of the fifth addition circuit A5. Consequently, when the output of integrator I2 becomes equal to 2 ##EQU39## at an instant t2 (FIG. 18) relay amplifier Ry A3 operates to energize relay Ry 3. Where the integration gain of integrator I2 is suitably selected, the interval (t2 - t1) from the operation of relay Ry 2 to the operation of relay Ry 3 becomes equal to ##EQU40##

When the transverse running speed V2 of the trolley becomes equal to the maximum value Vmax corresponding to the speed V1 given by the transverse running controller 100 at time t3, the output of absolute value circuit X becomes substantially zero thereby deenergizing relay Ry 4 whereby its contact Ry 4a is opened to deenergize relay Ry 1. Deenergization of relay Ry 1 opens its normal open contact Ry 12a and closes its normal close contact Ry 1b thus reducing to zero the output from integrator I1. As a result, the output from the third addition circuit A3 becomes zero to deenergize relay Ry 2. Thus, the integrator I2 is reset thereby deenergizing relay Ry 3. In this manner, concurrently with the deenergization of relay Ry 4, relays Ry 2 and Ry 3 are also deenergized.

Assume now that the transverse running controller 100 is returned to the "STOP" position while the trolley is running at a speed corresponding to the output V1 from the controller 100 in the direction of OFF BOARD. Under these conditions, when the signal V1 varies from Vmax to "0", the signal variation detector Z operates to momentarily energize relay Ry O. Because, since V1 = 0 and V2 = Vmax, again the relation │V1 - V2 │ = │Vmax│ can be obtained. Thereafter the control system operates in the same manner as described above in connection with the operation when the controller 100 is moved to OFF BOARD position from the STOP position.

For this reason, the interval (t1 - to) between the operations of relays Ry 1 and Ry 2 is equal to (a│ Vmax│ + b1 + c) whereas the interval (t2 - t1) between the operations of relays Ry 2 and Ry 3 is equal to ##EQU41## When the trolley stops, relay Ry 4 is deenergized and integrators I1 and I2 are reset.

When the transverse running controller 100 is moved to the ON BOARD position while the trolley is stopped, V1 = - Vmax and V2 = 0 so that again the relation │V1 - V2 │ = │Vmax │ holds and the integration of the switching time is performed in the same manner as the foregoing case in which the controller 100 was moved to the OFF BOARD position.

As has been described above, at all times the switching time computer oprates to make equal (that is satisfies equation 40) the interval (t1 -to) between the operations of relays Ry 1 and Ry 2 and (a │Vmax│ + b1 + c) and to make equal the interval (t2 - t1) between the operations of relays Ry 2 and Ry 3 and ##EQU42## (which satisfies equation 43).

Accordingly, by controlling the speed reference of the trolley with the output of the switching time computer 102 shown in FIG. 17 it is possible to control the speed of the trolley according to the pattern shown in FIG. 1.

FIG. 19 is a block diagram showing one example of a speed reference generator wherein the speed reference of the trolley is controlled by the output from the switching time computer shown in FIG. 17. The transverse running speed controller 100, addition circuit A1 and tachometer generator TG are identical to those shown in FIG. 17. The output of the addition circuit A1 is applied to a relay amplifier Ry A5 through a diode D1 and to a relay amplifier Ry A6 through a diode D2 which is poled oppositely with respect to diode D1. Relay Ry 5 is energized when the addition circuit A1 produces a positive output, whereas relay Ry 6 is energized when the addition circuit A1 produces an output of the negative polarity. A variable resistor RH3 is provided to set a variable speed and the output thereof is applied to the noninversion input of an addition circuit A6 through the normally closed contact Ry 5a of relay Ry 5 and to the inversion input of the addition circuit A6 through the normally opened contact Ry 6a of relay Ry 6. The output of the addition circuit A6 is applied to the noninversion input of an integrator I3 via the normally opened contact Ry 13a of relay Ry 1 shown in FIG. 17 and to the inversion input of the integrator I3 via the normally opened contact Ry 22a of relay Ry 2 and the normally closed contact Ry 3b of relay Ry 3. The inversion input "2" of the integrator I3 shows that the inverted input signal is doubled, and the output of the integrator I3 is applied as a speed reference signal to the speed controller S2 which controls the trolley driving motor M as shown in FIG. 16.

The control system shown in FIG. 19 operates as follows. When the transverse running controller 100 is moved to the OFF BOARD position from OFF position at times t2, FIG. 20, where the trolley is not operative that is V2 = 0, then V1 - V2 > 0, and relay Ry 5 is energized to close its normally opened contact Ry 5a. As described above, since relay Ry 1 is energized when the controller 100 is operated, its normally opened contact Ry 13a is also closed. Accordingly a signal +│α│ is applied to the noninversion input of the inverter I3 so that the output thereof increases linearly with an acceleration of +│α│. At time t1 relay Ry 2 is energized as described above to close its normally opened contact Ry 22a. At this time, relay contact Ry 3b has been closed so that the output from the addition circuit A6 is also applied to the inversion input of the integrator I3. Thus, its overall input will be +│α│- 2│α│ = - │α│ . Accordingly, the output from the integrator I3 decreases with a deceleration of -│α│ during an interval between the closure of normally opened contact Ry 22a and the opening of normally closed contact Ry 3b. When relay Ry 3 is energized to open its normally closed contact Ry 3b at time t2, input +│α│ alone will be applied to the noninversion input of the integrator I3 whereby the output thereof increases again with the acceleration of +│α│. When the running speed of the trolley becomes equal to the speed V1 commanded by controller 100 at time t3, relay Ry 5 is deenergized to open its normally opened contact Ry 5a thus removing the input from integrator I3. Consequently, the output thereof does not vary but maintains a defenite value. At this time, since relays Ry 1 - Ry 3 are denergized, their normally opened contacts Ry 13a and Ry 22a are opened and normally closed contact Ry 3b is closed.

When the trolley is stopped at t4, then V1 = 0 and V1 - V2 = Vmax∠0. Accordingly, relay Ry 6 is energized to close its normally opened contact Ry 6a. Concurrently therewith, relay Ry 1 is also energized as described above to close its normally opened contact Ry 13a. As a result, a signal -│α│ is impressed upon the integrator I3 so that the output thereof decreases with a deceleration of -│α│. At time t5, relay Ry 2 is energized to close its normally opened contact Ry 22a whereby the output from the integrator I3 increases with the acceleration of +│α│. At time t6 relay Ry 3 is energized to open its normally closed contact Ry 3b. Then the input to the integrator I3 becomes -│α│ and the output thereof decreases again with the deceleration - α . As the trolley stops, V2 = 0, so that relay Ry 6 is deenergized and relays Ry 1 to Ry 3 restore their original states.

Since t1 - to, FIG. 20, is equal to (a │Vmax│ + b1 + c), t2 - t1 is equal to ##EQU43## and since t1 - to = t3 - t2, by utilizing the output of integrator I3 as the speed reference for the trolley driving motor it is possible to make the rope swing to become zero when the trolley attains the maximum speed and when the trolley comes to stop.

In the foregoing description while the speed reference signal applied to the trolley driving motor was assumed to have a rectangular waveform as shown in FIG. 20, where the control system is incorporated with a speed limiting circuit or an armature current limiting circuit, the speed reference signal will have a stepped waveform.

Further, the control system described above is of the feed forward control type wherein the acceleration-deceleration switching points were determined by taking into consideration the speed command signal given by the transverse running controller, the transverse running speed of the controller, and the rope length. Actually however, there are such external disturbances as the wind pressure acting upon the load, spreader, crab and rope, measuring errors in the rope length and trolley speed, and errors of the computer. Where the effect of such external disturbances is substantial, with the feed forward control, there may be a case wherein the swing of the rope exceeds a predetermined limit when the trolley attains the constant speed or when the trolley comes to stop.

Such residual rope swing can be eliminated by adding to the feed forward control system a well known feedback system wherein the swing angle or swing angular velocity of the rope is detected when the trolley attains the uniform speed or stops for negatively feeding back the detected quantity to the transverse speed reference as has been described in connection with FIG. 14.

Two integrators I1 and I2 utilized in the circuit shown in FIG. 17 may be combined into a single integrator in which case the inversion input to the addition circuit A4 should be multiplied by a factor of 3.

Further it will be clear that various relays can be substituted by semiconductor elements, or another type of contactless relay means.

As described hereinabove this invention provides an automatic control system for a suspension type crane running in the transverse direction wherein the maximum transverse running speed is determined corresponding to the distance between the starting position and a predetermined target position at which the crane is to be stopped and, a deceleration command signal is provided when the crane reaches a point a predetermined distance before the target position, which is determined by the maximum transverse running speed and a predetermined acceleration-deceleration pattern signal corresponding to the maximum transverse running speed. Accordingly, the operator is required to initiate only the start signal and thereafter the crane is operated according to the acceleration-deceleration pattern so as to run the crane at said maximum speed while maintaining at zero or substantially at zero the swing of the rope and to correctly stop the crane at the target position when the swing angle of the rope is reduced to zero or substantially to zero thereby greatly reducing the load on the crane operator. Further, this invention provides a subtime optimal control system because the negative feedback is applied only when the control deviates from a prescribed pattern, thus not only improving the accuracy of the control without increasing the time required for the crane to reach the target position but also improving the cargo efficiency.

It will be clear that the function of at least one element can be performed by an electronic computer, that the control system can be applied to the control of the carriage of a crane and that the acceleration-deceleration pattern may be different from those illustrated. For example, the acceleration and deceleration during the acceleration and deceleration periods may be made three or more than three times.