DE3933527A1 - Crane load oscillation damping with strategic set point - involves electronic determn. of correction to target position from actual speed integral and angle of swing - Google Patents

Abstract

The target position signal (x(t)) from a strategic swing damping device (1) is applied to a position controller (2) and a recognition unit (3) which operates also on output from a unit (7) giving the actual angle of swing (alpha), a corrected integral (6) of the calculated actual speed (4) and a measure of the length (L) of swinging load. An electronic unit (8) determines the position correction set point (z(t)) to be applied to the position controller (2). USE/ADVANTAGE - Esp. in automatic crane. Travel and transfer times are minimised irrespective of type and amplitude of swings occurring prior to movement.

Description

The invention relates to a control method for damping Pendulum vibrations according to the preamble of claim 1.

Such a procedure is from the German published application 32 28 302 A1 known.

Here, the target force for acceleration and braking the Trolley connected to it by a drive motor standing computer, which for a computer program to solve the for Dual mass vibration system formed from trolley and load body applicable differential equation works, controlled so that the load no longer commutes at the destination. The computer is available on the input side a pendulum length transducer to determine the distance of the Load body connected by the trolley.

The disadvantage here is that this known method for determining the Control command is not of the prediction type. Furthermore, this takes place no special consideration of one before the start of the movement applied pre-oscillation of the load instead.

A consideration of any disruptions in driving is from the German patent specification 35 13 007 C2 known. The Control commands based on an estimation method that is based on the theoretical calculated tax command are scattered after a complicated mathematical selection process selected. The disadvantage here is correspondingly great effort and the time required to carry it out of the tax process.

The invention is therefore based on the object of the method of to further develop the type described above in such a way that with the help of any strategic pendulum damping method, which consists of Additional pendulum vibrations dampen driving movements also dampen the pre-vibrations during the driving process, so that no additional time is required.

The solution to this problem is based on the characteristic Features of claim 1. Advantageous embodiments of this Invention are the subject of dependent claims 2 and 3.

The advantages achieved with the invention are that independent minimal driving times or the type and size of the pre-oscillations Loading times, especially when used in automatic cranes, reached will.

An embodiment of the invention is shown in the drawing. It shows

Fig. 1 block diagram: complete overview.

Fig. 2 block diagram: position correction setpoint determination.

Fig. 1 shows a block diagram of the embodiment, with the arrangement and linking of individual controller components that work according to the method and are linked.

The position command value x (t) present at the output of a strategic pendulum damping device 1 is used as an input variable. This position setpoint x (t) is passed on the one hand to a position controller 2 and on the other hand to a unit 3 for detecting the position setpoint x (t). The position controller 2 calculates the target speed from the position target value x (t) and a feedback with a position correction target value x (t). This speed setpoint V _s (t) is fed to a speed controller 4 , which is part of a drive controller 5 of the traction motors.

The actual speed value V _I (t) obtained is fed on the one hand to a computing unit 6 and on the other hand to a pendulum unit 7 , 6 and 7 representing the vehicle (cat or crane bridge) with the load hanging on the rope. The pendulum unit 7 delivers the real pendulum angle α (t) from this actual speed value V _I (t). The arithmetic unit 6 supplies the value of the actually traveled distance y (t) by the driving movement (= integration of the actual speed value V _I (t) over time). This value y (t) is fed to a summation point a at which the value of the position correction setpoint z (t) is also present in feedback with the position controller 2 . The difference between the two values y (t) -z (t) arises at this summation point a. This value is fed to a unit 3 for recording the corresponding value. In a further step, the pendulum length L is now measured and also recorded in this unit 3 . The detected values α (t), y (t) -z (t), L and x (t) are fed to an electronic device 8 for determining the position correction setpoint z (t). The position correction setpoint z (t) determined therefrom is superimposed on the position setpoint x (t) by feeding to the position controller 2 .

The principle of the electronic unit for determining the position correction setpoint is shown in the block diagram in FIG. 2.

The pendulum angle α (t) is fed to a multiplier 10 which sets a time window via a synchronization pulse I · Δt and initiates real-time simulation of the pendulum angle α (t). On the one hand, this initial value is led via a differentiating element 13 to a first integrator 14 and, on the other hand, in parallel directly to a second integrator 15 . The input variable y (t) -z (t) is fed to a double differentiator 19 . The output signal is connected to a division element 21 , to which the pendulum length L is supplied. The signal obtained from this is present at the output, ie at the summation point b, which is connected to integrator 14 . The input variable position setpoint x (t) is supplied to a double differentiator 20 , the output of which is connected to a division element 22 , to which the pendulum length L is also supplied. The output of the division element lies above the summation point c at the input of the integrator 17 . The input variable pendulum length L is fed to a computer 23 which outputs the value of the damping variance function ϑ (ω, t) and the value of the angular frequency ω² together with the value of the pendulum length L itself at the output. The damping variance function is fed to a multiplier 11 , and the angular frequency ω² is fed via multipliers 12, 12 'to the two real-time simulations of the real and the strategic pendulum angle.

In both integration elements 14, 15 and 17, 18 the respective real-time model is created at the output. Both output signals ϕ, ψ come together at the summation point d. The quantity obtained there, ie the difference between ϕ (t) and ψ (t) gives the pre-oscillation component ε (t). The subsequent multiplication by the damping variance function ϑ (ω, t) and the pendulum length L is fed to an integrator 16 , which then supplies the position correction setpoint z (t) as an output variable.

The two real-time models ϕ, ψ work according to the Differential equations of the corresponding pendulum systems. First is the real pendulum angle α (t) by the synchronization with a  Real-time model coupled and simulated by the same. This has to Consequence that all from a static deformation of the structure resulting interference are eliminated. This coupling must mathematically convert the function α (t) into the function ϕ (t). The same applies to the derivation of the corresponding functions.

There are then basically three structurally identical ones Vibration systems:

[ϕ]: The real load pendulum system consists of a real-time model that simulates the pendulum oscillation including the pre-oscillation.
[ψ]: The strategic pendulum system consists of a real-time model, whose pendulum deflection ψ simulates that of the α, but without pre-oscillation.
[ε]: Exists only theoretically, arises from the subtraction

[ϕ]: - [ψ]: = [ε]:

and represents the pre-oscillation component ε in ϕ.

The systems obey the following differential equations:

In Laplace transform the equations are:

As a prerequisite for the mathematical solvability of the imagined Systemes [ε]: it is assumed that it is an ideal overall Controller acts. This means that the setpoints as actual values too can be achieved.

The pre-oscillation component ε then results from the following difference:

The differential equations 1, 2, 3 are identical in structure and describe undamped harmonic vibration processes. The task accordingly, the pre-oscillation component ε in α or durch should pass through Damping to disappear. For this purpose, the Differential equation 3 introduced a first-order term ϑ ε.

 + ϑε + ω²ε = 0 (Eq. 4)

respectively.

From the relationship [ε]: + [ψ]: = [ϕ] one obtains the differential equation for ϕ, with damped pre-oscillation component ε.

The differential equation is through two integrations in the To transfer local space.

The object of the invention, the location goal even to be damped Reaching pre-vibration without delay leads to the following Condition:

∫ϑε (t) dt → 0 (Eq. 7)

The fulfillment of this requirement depends on the course of time of the Attenuation variance function ϑ ε (t). This becomes the characteristic Polynomial P (s) viewed from equation 4a.

P (s) = s² + ϑs + ω² = 0 (Eq. 8)

The two roots are:

The following then applies to the damped harmonic oscillation:

With the introduction of the damping parameters (β, γ) <1 this equation can now be used represent:

The following solution results from 5:

The above requirement specified in equation 7 can then be met, if the damping variance function ϑ remains time-variant.

ϑ → ϑ (ω, t) or ϑ (ω, t) = 2ωβ (ω, t)

For the position correction setpoint z (t):

z (t) = 2Lω∫β (ω, t) ε (t) dt → 0 at the location target

Now the time course of β (ω, t) has to be determined so that the Position correction setpoint goes to zero. This is in math closed form and therefore practically not electronically possible. This damping function is therefore determined by Tests on the real system.

Of course, it is possible to use the part of the invention that the Determination of the position correction setpoint z (t) by a To replace computer program.

Position list

Signals, measured and control variables
x (t) = position setpoint
z (t) = position correction setpoint
V _s (t) = speed setpoint
V _I (t) = actual speed value
Y (t) = distance actually traveled, actual position value
L = pendulum length
ω = pendulum angular frequency

Fig. 1, overall view (block diagram)

1 = strategic pendulum damping device
2 = position controller
3 = registration unit
4 = speed controller
5 = drive controller
6 = arithmetic unit
7 = pendulum unit
8 = electronic device for determining z (t)
a = summation point

Fig. 2, position correction setpoint determination

10-12, 12 ′ = multipliers
I · Δt = synchronization pulse
13 = differentiator
14-18 = integrators
19-20 = two differentiators
21, 22 = division members
23 = calculator
b, c, d = summation points

Claims (3)

1. Control method for damping load oscillations from driving movements in crane systems with trolleys, the driving movement being controlled by determining state variables of the crane-load system and calculating setpoints so that a load hanging on the crane no longer oscillates at the destination, characterized in that that are not mitgedämpft from the driving movement resulting preshoots during the driving process, by determining a position reference value (x (t)) a corresponding position correction target value (z (t)) and the nominal position value (x (t)) is superimposed, wherein in the following Process steps are controlled:

• a) Determination of the strategic position setpoint (x (t)) and determination of a real pendulum angle (α (t)) as well as the rope length (L),
• b) representation of the real pendulum angle (α (t)) by means of a pendulum angle (ϕ (t)) simulated from a first real-time model and synchronized in time with (α (t)),
• c) Representation or ongoing calculation of a strategic pendulum angle (ψ (t)) resulting only from the driving movement using a second real-time model.
• d) Forming the difference (ε (t)) = (ϕ (t)) - (ψ (t)), where (ε (t)) is the real pendulum angle (α (t)) or (ϕ (t) ) is the pre-oscillation component,
• e) Determination of a position correction setpoint (z (t)) dependent on (ε (t)) from (x (t)), (y (t)), (ϕ (t)), (ψ (t)) and (L ) with the help of a determined damping variance function (ϑ (ω, t)), which is adapted to the respective pendulum situation,
• f) Superposition of the position setpoint (x (t)) with the position correction setpoint (z (t)) at the input of the position controller ( 2 ) and transfer of the resulting output variable (v (t)) as setpoint to the speed controller ( 4 ).

2. Control procedure for damping load oscillations after Claim 1 characterized, that the real-time model (ϕ (t)), before or at the start of the movement by a synchronization pulse (I · Δt) a short time Coupling with the determined real pre-oscillation (α (t)) and is synchronized with it in time.   3. Control method for damping load oscillations according to Claims 1 and 2, characterized, that the damping variance function (ϑ (ω, t)) has a profile, which is the pendulum angle difference (ε (t)) and the Position correction setpoint (z (t)) practical at the end of the driving process makes zero.