turntable ladder

Abstract

A turntable ladder, telescopic mast platform, or the like, with a telescopic ladder set or telescopic mast and optionally a lift platform attached thereto, which turntable ladder or telescopic mast includes a control system for the movement of the ladder sections or telescopic mast sections, designed such that vibrations of the ladder sections or telescopic mast sections are suppressed during the movement of the turntable ladder or telescopic mast platform by feeding at least one of the measured variables—bending of the ladder set or telescopic mast in the horizontal and vertical directions, erection angle, rotation angle, extension length, and torsion of the ladder set or telescopic mast—back to the control variables for the drives via a controller, characterized in that inertial sensors for detecting the bending state of the ladder set or telescopic mast are attached to the ladder set or telescopic mast and / or to the lift platform, wherein the lift platform and / or the end of the ladder set or telescopic mast,a car is attached to which a gyroscope is provided for measuring the angular velocity in various spatial directions and strain gauge sensors are attached to the ladder assembly or telescopic mast for detecting the bending state of the ladder assembly or telescopic mast, and further characterized by a feedforward control which, during the movement of the car, maps the idealized motion behavior of the ladder or telescopic mast in a dynamic model based on differential equations and calculates idealized control variables for the drives of the ladder sections or telescopic mast sections for a substantially vibration-free movement of the ladder or telescopic mast from the dynamic model, which dynamic model replicates a mass distribution of the ladder assembly or telescopic mast, wherein all state variables x1 to x4 of the dynamic model represent measured variables of the strain gauge sensors and the gyroscope.and which combines the measured quantities in an observer structure and estimates the amplitude components of the first two modes of the natural oscillation of the conductor or telescopic mast, which are fed back to the controller as a control variable.

Description

[0001] The present invention relates to a turntable ladder, a telescopic mast platform or the like, comprising a telescopic ladder set or telescopic mast and optionally a platform attached thereto, according to the preamble of claim 1.

[0002] Specifically, the invention relates to a turntable ladder, for example, a fire ladder with an articulated arm, or a similar device, such as articulated or telescopic mast platforms and aerial rescue platforms. Such devices are generally rotatable and erectable, mounted on a vehicle, and may have an articulated arm that can also be extended telescopically via a further axis. The control system is a path control system that guides the platform or lifting platform along a predetermined path within the working area of ​​the turntable ladder or lifting platform. Vibrations and pendulum movements of the platform or lifting platform are actively dampened.

[0003] Control systems for turntable ladders, elevated platforms, or similar equipment are known, for example, from DE 100 16 136 C2 and DE 100 16 137 C2. Vibrations of the ladder sections can be suppressed by feeding at least one parameter of the ladder assembly back to the control parameters for the drives via a controller. A feedforward control system models the idealized movement behavior of the ladder in a dynamic model based on differential equations and calculates idealized control parameters for the drives of the ladder sections to enable essentially vibration-free movement of the ladder. DE 10 2005 042 721 A1 discloses such a control system for a turntable ladder that has an articulated arm at the end of its ladder assembly, to which a platform is attached. The dynamic properties of the platform are included in the dynamic model for representing the properties of the ladder, so that appropriate control can be implemented.

[0004] Common articulated ladders and similar devices are controlled hydraulically or electro-hydraulically via hand levers. In the case of purely hydraulic control, the hand lever deflection is directly converted into a proportional control signal for the control block, which is designed as a proportional valve, via the hydraulic control circuit. Dampers in the hydraulic control circuit can be used to make the movements less jerky and smoother during transitions. However, these cannot be satisfactorily adjusted across the entire working range of extension length and tilt angle. Furthermore, this often leads to heavily damped settings with sluggish response.

[0005] Current track control systems actively counteract vibrations in the event of a conductor set oscillation. However, the oscillation is reconstructed solely from a strain gauge signal, and the underlying model only considers the fundamental frequency components. Higher modes of the oscillation are not taken into account in the vibration-damping control according to DE 100 16 136 C2 and DE 100 16 137 C2. Furthermore, the reconstruction of the bending state is based only on strain gauge signals or a reconstruction from the hydraulic pressure signal. For the present case of vibration excitation by harmonics, this is not always sufficient.

[0006] JP 2003-221191 A discloses a lifting device whose state is determined, among other things, by angle sensors and gyro sensors.

[0007] The object of the present invention is therefore to provide a turntable ladder, a telescopic mast platform, or the like, according to the preamble of claim 1, in which the vibration states of the ladder or telescopic mast can be detected and reconstructed more precisely, so that actively occurring vibrations (either during movement or at rest, e.g., due to wind influences or load changes) can be dampened, or the ladder end with the platform or work platform can be guided along a predetermined path. This should make it possible not only to compensate for the fundamental vibration but also to effectively dampen the higher modes of the vibration.

[0008] According to the invention, this problem is solved by a turntable ladder or telescopic mast platform with the features of claim 1.

[0009] Preferred embodiments of the invention are set forth in the dependent claims.

[0010] As with the prior art, the railway control system according to the invention with active vibration damping is also based on the basic idea of ​​first mapping the dynamic behavior of the mechanical and hydraulic system of the turntable ladder or the telescopic mast platform in a dynamic model based on differential equations.

[0011] In contrast to applications DE 100 16 136 C2 and DE 100 16 137 C2, the dynamic model does not use an approach based on an elastic multibody model as an approximation for the distributed parametric model, but rather the distributed masses of the ladder set are modeled directly. Here, the mass of the elevator car can still be assumed to be a point mass.

[0012] A preferred embodiment of the invention will be explained in more detail below with reference to the drawing. Fig. Figure 1 schematically shows the mechanical structure of an embodiment of a turntable ladder according to the present invention; Fig. Figure 2 is a schematic representation to illustrate the degrees of freedom of the system; Fig. Figure 3 schematically shows the control loop for regulating the movement of the turntable ladder according to a first embodiment; Fig. Figure 4 shows a further control loop for controlling the movement behavior of a turntable ladder according to a second embodiment; Fig. Figure 5 is a diagram for representing eigenfunctions of the differential equation for describing the bending; and Fig. 6a and Fig. 6b are diagrams illustrating the vibration behavior as a function of time.

[0013] While Fig. 1 schematically represents the structure of the overall system, based on Fig. 2. The rotary motion of the turntable ladder according to the invention is explained by way of example. The following description of the invention relates only by way of example, but not exclusively, to a turntable ladder and is readily applicable to a telescopic mast platform or the like, which is equipped with a telescopic mast. The individual parts of this telescopic mast then correspond to the ladder sections of the ladder set of the turntable ladder described here. Furthermore, the invention is not limited to a turntable ladder with a platform, but is also applicable to ladders or telescopic mast platforms without a platform.

[0014] Fig. Figure 3 schematically shows a control loop for regulating the movement of the turntable ladder presented here. In the model-based evaluation, the gyroscope measurement data are first corrected for offset. The strain gauge signal is then analyzed to eliminate the gravitational influence of the ladder's own weight (even during movement). In the subsequent modal transformation, the first two modes of the natural oscillation are calculated from these signals. These can then be fed back in separate form via the controller feedback, compensating for the oscillation and thus damping its effect. Compared to the previous state of the art, this achieves not only damping of the fundamental oscillation but also damping of the first harmonic.

[0015] An alternative structure is in Fig. 4 shown. Unlike Fig. However, in step 3, only the fundamental frequency is extracted from the sensor signals, and the components of the higher modes are calculated and removed using a model-based observer. While this does not provide active damping of the first harmonic, it prevents the components of the higher modes from having a feedback effect on the vehicle, thus preventing destabilization.

[0016] The isolated fundamental oscillation is then actively damped and fed back to the control input via the feedback loop.

[0017] In both structures, it is highly advantageous that the desired trajectories are guided via model-based feedforward control and thus adapted to the system's dynamics. This prevents the excitation of vibrations by the system's reference variables. Unlike the control systems described in DE 100 16 136 C2 and DE 100 16 137 C2, this method is applicable to the generally nonlinear case presented here.

[0018] This will now be explained using the example of straightening and tilting. The procedure is directly transferable to the direction of rotation, since the influence of gravity is neglected for the considerations of the ladder's dynamic behavior. For the feedforward control design, a model with lumped parameters is again used. The equations of motion for the straightening / tilting direction are: (I)φ¨A=−1τAφ˙A+kAτAu(II)0=m1v¨z+m1Lφ¨A+d44v˙z+cvvz+(Lφ˙D+v˙y)22Lsin(2x1(t)+2Lx3(t))

[0019] Legend: L(t) time-varying parameter conductor length m1 moving mass (dynamic component of the total mass consisting of ladder set, possibly articulated part and car or lifting platform) c v (L(t)) Stiffness as a function of conductor length d 44 (L(t)) Attenuation coefficient of the conductor as a function of the conductor length τ A Time constant of the hydraulic drive k A Hydraulic drive amplification factor φ̇ D (t) time-varying parameter rotational speed v̇ y (t) Time-varying parameter: Deflection at the conductor end in the horizontal direction

[0020] The model equations according to Eq. 1 are now transformed into the generally nonlinear state form. x_˙(t)=f(x_(t), u(t))y=g(x_(t)), with x_(t)=[φA(t)φ˙A(t)vz(t)v˙z(t)] and u(t)=φ˙A,Soll

[0021] The control input is the voltage at the proportional valve of the hydraulic system u(t), which is used as the target velocity φ̇. A,Soll can be interpreted. Using Eq. 1, one then obtains the following nonlinear state equation with the output equation for the car position, taking the bending into account: x_˙(t)=[x2(t)−1τAx2(t)+kAτAu(t)x4(t)LτAx2(t)−cvm1x3(t)−d44m1x4(t) −(Lφ˙D+v˙y)2Lsin(2x1(t)+2Lx3(t))−LKAτAu(t)]y(t)=x1(t)+sin(x3(t)L)

[0022] For further system analysis, the relative degree of the system with respect to the chosen output is determined. The relative degree of y(t) from Eq. (3) is 2. This is smaller than the system order (n = 4). Therefore, a differentially flat input with relative degree r = 4 is chosen. z(t)=x1(t)+x3(t)L≈y(t)∀x3(t)L≤0.05

[0023] Since it corresponds to the real controlled variable (car position taking bending into account) to a first approximation, the solution of the residual dynamics for generating the reference trajectories is omitted.

[0024] The flatness-based analysis and design are fully incorporated in accordance with Rothfuß R.: Application of flatness-based analysis and control of nonlinear multivariable systems, VDI-Verlag, 1997.

[0025] Will φ̇ D = 0 and v̇ y Assuming = 0, the following parameterization results based on the flat output of the car position: z(t)=x1(t)+x3(t)Lz˙(t)=x2(t)+x4(t)Lz¨(t)=cvx4(t)m1Lz⃛(t) =cvx4(t)m1Lz⃜(t)=−cvτAm12L(m1(Lx2(t)−τAcvx3(t)−m1LkAu(t))

[0026] The system's new entrance will be equipped with v(t)=z⃜(t) designated and the inverse tax law for the input tax follows from this: u(t)=1cvkA(cvz˙ref(t)+cvτAz¨ref(t)+m1z⃜ref(t)+τAm1v(t))

[0027] This allows an ideal control parameter for the valve position to be generated using target trajectories over time for the car position and their derivatives as input information, without exciting natural oscillations when the ladder moves. Residual oscillations occur due to model inaccuracies and external disturbances (loading and unloading in the car, wind influence) and are damped via the feedback mechanism.

[0028] In the case of the structure of Fig. 3. Feedback is achieved via a modal decomposition of the measurement signals from the strain gauge and the gyroscope, as well as feedback of the separated vibration signals.

[0029] Modeling the ladder as a distributed mass and the car as a point mass at the end of the ladder as follows: the boundary conditions of the dynamics of the concentrated mass must be satisfied. Although the oscillation of the ladder in the vertical plane is considered (where it may be tilted by a certain angle of inclination), the influence of gravity on the concentrated mass is neglected, since the mathematical model can be assumed to be linear for small angles and the steady-state solution of the problem takes the influence of gravity into account. Therefore, for the task of vibration damping, this influence can, in principle, be eliminated beforehand. The ladder is moved in the plane of rotation by a moment M(t) at z = 0 of the ladder ( Fig. 2) L is the length of the conductor, θ(t) corresponds to the angle of rotation. w(z,t) is the bending angle. M p and J pis the mass of the car and possibly the articulated arm, or the moment of inertia of the car converted into the moment of inertia with respect to the center of mass of the point mass. Assuming that the angular velocity θ̇(t) is small and that Coriolis forces can be neglected, the ladder equation can be described as a distributed mass over a Bernoulli bending beam by the following partial differential equation, where the rotation of the ladder equation by the angle of rotation θ(t) is additionally taken into account in all acceleration terms characterized by the second derivative with respect to time: EI∂4w(z,t)∂z4+ρS∂2[θ(t)z+w(z,t)]∂t2=0z∈(0,L), t>0, w(0,t)=0, t>0, ∂w(0,t)∂z=0, t>0. −EI∂2w(l,t)∂z2=JPd2dt2(θ(t)+∂w(L,t)∂z), t>0, EI∂3w(L,t)∂z3=Mpd2dt2(θ(t)L+w(L,t)), t>0.

[0030] Eq. (7) is the partial differential equation that describes the bending. E is the modulus of elasticity, I is the area moment of inertia of the conductor theorem, ρ is the density, and S is the (equivalent) cross-sectional area of ​​the conductor theorem.

[0031] Eqs. (8)-(9) are the boundary conditions corresponding to a clamped beam end at the beginning of the ladder theorem at z = 0.

[0032] Eqs. (10) and (11) are the boundary conditions corresponding to the transition condition between distributed and concentrated mass, where Eq. (10) describes the balance of moments and Eq. (11) describes the balance of forces.

[0033] To simplify matters, the following variable is introduced, describing the movement of individual points of the ladder theorem in inertial space: V(z,t)≜θ(t)z+w(z,t)

[0034] The system can therefore be written according to equations 7-11 as EI∂4V(z,t)∂z4+ρS∂2V(z,t)∂t2=0, V(0,t)=0, ∂V(0,t)∂z=θ(t), −EI∂2V(l,t)∂z2=JPd2dt2(∂V(L,t)∂z), EI∂3V(l,t)∂z3=MPd2V(L,t)dt2.

[0035] For simplification, the following is introduced as an input variable: u(t)≜θ(t)=∂V(0,t)∂z.

[0036] The dynamics of the drive system are not considered. However, the torque M(t) can be calculated directly from the above input variable according to the following relationship: M(t)=Jhd2θ(t)dt2−EI∂2w(0,t)∂z2, J h is the moment of inertia of the ladder gear (15).

[0037] The solution according to Eq. (12)-(16) can be represented in the following form. V(z,t)≜VH(z,t)+VI(z,t), where V1(z, t) is the part of the general solution V(z,t) that only needs to satisfy the inhomogeneous boundary conditions according to Eq (13)-(16), while V H (z, t) must satisfy the homogeneous boundary conditions as well as Eq. (12). The following is chosen as the approach for V1(z, t): VI(x,t)≜f(z)u(t) wheref(z)≜A+Bz+Cz2+Dz3+Ez4+Fz5,z∈[0,L]

[0038] From (13) and (14) it follows: A=0,B=1.

[0039] To satisfy the boundary conditions (15) and (16), the following system of equations must be solved: [26L12L220L32L3L24L35L40624L60L2L2L3L4L5][CDEF]=[0−10−L]. Eq. (20) has a unique solution: C=−4L−1,D=6L−2,E=−4L−3,F=L−4

[0040] The solution approach according to Eq. (19) is now substituted into Eqs. (12)-(16). This results in the following system of equations: EI∂4VH(z,t)∂z4+ρS∂2VH(z,t)∂t2==−[EI∂4VI(z,t)∂z14+ρS∂2VI(z,t)∂t2], VH(0,t)=0, ∂VH(0,t)∂z=0, ∂2VH(L,t)∂z2−μd2dt2(∂VH(L,t)∂z)=0, ∂3VH(L,t)∂z3−κd2V(L,t)dt2=0.

[0041] For the sake of simplicity, the following were substituted: μ≜−JpEI, κ≜MpEI.

[0042] The following approach is chosen for the homogeneous solution: VH(z,t)≜Z(z)T(t).

[0043] Substituting (27) into the homogeneous part of Eq. (21) yields: EI Z(IV)(z)T(t)+ρSZ(z)T¨(t)=0.

[0044] In the following, it is assumed that a set of functions {Zn(z)}n≥1, {Tn(z)}n≥1 Eq. (28) and the boundary conditions Eqs. (22)-(25) can be satisfied. Eq. (28) can then be written as Zn(IV)(z)Zn(z)=−ηT¨n(t)Tn(t)≜λn, where η≜ρS / EI λ n is the corresponding eigenvalue. From (29) we can then determine the amplitudes of the time functions T. n (t) the following differential equation can be formulated. T¨n(t)+ωn2Tn(t)=0, where ωn=λn / η to the eigenvalue λ n is.

[0045] Using (29) and substituting the approach (27) into (22)-(25), the following boundary value problem for the position-dependent eigenfunctions Z can be solved. n(z) formulate. Zn(IV)−λnZn=0, Tnzn(0)=0, TnZn'(0)=0, TnZn''(L)=μT¨nZn'(L), TnZn'''(L)=κT¨nZn(L).

[0046] Using equation (24), the remaining time functions T can be determined. n (t) in (34)-(35). From the boundary conditions (32)-(35) we obtain: Zn(0)=0, Zn'(0)=0, Zn''(L)+λn(μη)Zn'(L)=0, Zn'''(L)+λn(κη)Zn(L)=0.

[0047] The solution leads to the following relationships. Since the characteristic polynomial (31) contains the roots {−λn4,λn4,−jλn4,jλn4} If the general solution is available, it can be written in the following form. Zn(z)=An cosh(λn4z)+Bn sinh(λn4z)+ +Cn cos(λn4z)+Dn sin(λn4z).

[0048] The boundary conditions (36)-(37) at the boundary z=0 immediately lead to Cn=−An,Dn=−Bn and: Zn(z)=An[cosh(λn4z)−cos(λn4z)]++Bn[sinh(λn4z)−sin(λn4z)]

[0049] The boundary conditions (38)-(39) at the boundary z=L lead to the following system of linear algebraic equations Anλn(cosh+cos)+Bnλn(sinh+sin)++Anλn4λn(μη)(sinh+sin)++Bnλn4λn(μη)(cosh−cos)=0, An(λn4)3(sinh−sin)+Bn(λn4)3(cosh+cos)++An(κη)λn(cosh−cos)+Bn(κη)λn(sinh−sin)=0.

[0050] To simplify matters, the argument was λn4L The trigonometric and hyperbolic functions have been omitted. The homogeneous system (42) has non-trivial solutions if the determinant of the coefficients is 0. [λn(cosh+cos)+λn4λn(μη)(sinh+sin)]︸≜A11(λn)××[(λn4)3(cosh+cos)+(κη)λn(sinh−sin)]−− [λn(sinh+sin)+λn4λn(μη)(cosh−cos)]︸≜A12(λn)××[(λn4)3(sinh−sin)+(κη)λn(cosh−cos)]=0

[0051] This transcendental function can be described with respect to the eigenvalues ​​λ. nIt can only be solved numerically. (The number λ = 0 corresponds to a trivial solution to the problem and is therefore not an eigenvalue.) The eigenfunction corresponding to the eigenvalue λ n can now be written as Zn(z)=−A12(λn)A11(λn)[cosh(λn4z)−cos(λn4z)]++[sinh(λn4z)−sin(λn4z)].

[0052] In Fig. 5 are the first three eigenfunctions of the problem in normalized form (Z 10 , i=1,2,3) is shown as an example. Based on these results, the modal representation of the distributed parameter system is now derived. The part of the solution that solves the homogeneous boundary conditions as well as Eq. (21) can now be written as VH(z,t)≜∑k=1+∞VHi*(t)Zk0(z).

[0053] Substituting (43) into Eq. (21) yields ∑k=1+∞VHk*(t)λkZk0(z)+η∑k=1+∞d2VHk*(t)dt2Zk0(z)==− ∑k=1+∞ƒk(IV)*Zk0(z)u(t)−η∑k=1+∞ƒk*Zk0(z)d2u(t)dt2, ƒk*,ƒk*(IV) are the coefficients of ƒ(z)=∑k=1+∞ƒk*Zk0(z), ƒ(IV)(z)=∑k=1+∞ƒk(IV)*Zk0(z).

[0054] With (44) all time modes (amplitudes of the eigenfunctions) in the frequency domain can be represented via the Laplace transform in the following form. V^Hk*(s)=−ƒk*s2+1ηƒk(IV)*s2+ωk2u^(s)++ss2+ωk2VHk*(0)+1s2+ωk2V˙Hk*(0).

[0055] In the frequency domain, one therefore obtains with (19) V^(z,s)=V^H(z,s)+V^I(z,s)=∑k=1+∞V^Hk*(s)Zk0(z)++(z−4L−1z2+6L−2z3−4L−3z4+L−4z5)u^(s).

[0056] For simulation purposes, N modes (i.e., the first N summands of the infinite series (45)-(46)) can now be used. For the following controller design, the modal representation is chosen as follows: Vk*(t)=VHk*(t)+VIk*(t), k≥1, where VIk*(t)=fk*u(t) out of VI(z,t)=f(z)u(t) is determined. Using (44) and (47) one then obtains d2Vk*(t)dt2+ωk2Vk*(t)=(ωk2fk*−fk(IV)*η)u(t).

[0057] Only the first two modes will be considered as examples. The state variables are then introduced. x1(t)≜V1*(t), x2(t)≜x˙1(t)=V˙1*(t),x3(t)≜V2*(t), x4(t)≜x˙3(t)=V˙2*(t). This results in the state space representation. [x˙1x˙2x˙3x˙4]=[0100−ω12000000100−ω220]︸≜A[x1x2x3x4]++[0ω12f1*−f1(IV)*η0ω22f2*−f2(IV)*η]︸≜Bu(t).

[0058] Based on this representation, the controller design is now carried out. For this purpose, the operating point is determined. V0(z)≡0,∀z∈[0,L] Considered. For the technical implementation, it has proven sufficient to consider only the first two modes of the system in the controller design, since the cutoff frequency of the hydraulics is around 3-4 Hz, while the third mode is approximately 6.5 Hz. Therefore, the higher modes are neglected. To stabilize the system (48) with two conjugate complex pole pairs using state feedback that lies on the imaginary axis (since there is no damping in the presented model), all state variables x1 to x4 must be measurable. A strain gauge sensor on the lower part of the conductor and a gyroscope at the conductor tip are available as measurement parameters. The idea is to estimate the amplitude contributions of the modes to the oscillation from these sensor values ​​in an observer structure that also combines (i.e., fuses) both measured values.The strain gauge sensor at the mounting point z=z1 provides a bending value which can be written in the present notation as follows:. m1(t)=∂2w(z1,t)∂z2=∂2V(z1,t)∂z2

[0059] The measurement from the gyroscope (mounted at point z=z2) provides m2(t)=∂2V(z2,t)∂z∂t.

[0060] If only two modes are considered, the two measured values ​​can be expressed via the state variables as follows: m1(t)=∂2V(z1,t)∂z2=x^1(t)Z10''(z1)+x^3(t)Z20''(z1),m2(t)=∂2V(z2,t)∂z∂t=x^2(t)Z10''(z2)+x^4(t)Z20'(z2).

[0061] To reconstruct all state variables, 2 further signals are needed, which are obtained through integration or real differentiation. dm1(t)dt=∂3V(z1,t)∂z2∂t=x^2(t)Z10''(z1)+x^4(t)Z20''(z1 ),∫0tm2(τ)dτ=∂V(z2,t)∂z=x^1(t)Z10'(z2)+x^3(t)Z20'(z2).

[0062] Then the amplitudes of the modes can be estimated from the solution of the following system of algebraic equations. [Z10''(z1)0Z20''(z1)00Z10'(z2)0Z20'(z2)0Z10''(z1)0Z20''(z1)Z10'(z2)0Z 20'(z2)0][x^1(t)x^2(t)x^3(t)x^4(t)]==[m1(t)m2(t)dm1(t) / dt∫0tm2(τ)dτ].

[0063] After inverting the matrix with the eigenfunctions Z, a direct functional relationship is established between the measured values ​​at the strain gauge and the gyroscope and the amplitudes for the first two modes. This allows for the direct design of a pole-setting controller. u(t)=K[x^1(t)x^2(t)x^3(t)x^4(t)]T, where the amplification matrix is ​​derived from the determinant of A−BK is determined by specifying the zeros of the characteristic equation. Fig. 6a and Fig. Figure 6b shows an example of the damping characteristics of the control system. Fig. 6a shows the bending b (L, t) plotted against time, in Fig. 6b the control variable. In the case shown here, the control is switched on after ten seconds.

[0064] The following section will describe the alternative structure of the Fig. Section 4 explains how the fundamental frequency is separated from the measurement signals in a model-based observer. This section will now be described in more detail below.

[0065] The disturbance observer for sensor data fusion from gyroscope measurements at the basket mounting and strain gauges at the clamping point of the conductor is intended to separate the fundamental wave of the bending vibration from its dominant harmonics in order to largely exclude harmonic amplification in the feedback loop.

[0066] For the observer, simple vibration differential equations are used as the model equations. Since the gyroscope signal is superimposed with a significant offset, an integrator perturbation model is included in the model equations to compensate for this influence. K0ω02φ¨A(t)+v¨0z(t)+2D0ω0v˙0z(t)+ω02v0z=0…fundamental waveK1ω12φ¨A(t)+v¨1z(t)+2D1ω1v˙1z(t)+ω12v1z=0…1. Harmonic v¯¨z=0…trival dynamics of the offset of the gyroscope measurement (φ˙A(t,L)≙v˙z)

[0067] The parameters ω i , D i and K i are determined through experimental process analysis. In the state representation, this then corresponds to: x_˙(t)=Ax_(t)+bu(t)y=Cx_(t), with x_(t)=[φ0z(t)φ˙0z(t)v1z(t)v˙1z(t)v¯(t)] and u(t)=φ¨A with x_˙(t)=[01000−ω02−2D0ω00000001000−ω12−2D1ω1000000]x_(t)+[0−K0ω020−K1ω120]u(t) ym(t)=[1000001011]x_(t)≙[DMSGyroscope]

[0068] The first component of the output vector corresponds to the strain gauge signal, the second component to the gyroscope measurement.

[0069] For the design of the observer, a method based on representation as an observer normal form can be chosen, for example. The advantage is that simple design equations for the observer feedback matrix H can then be derived via the specified poles. pi, i∈ℕ∧[1, 5] can be derived. After transformation to the observer normal form of the second kind for multi-quantity systems, equation 51 becomes... ABNF 2.Type=[0−ω020001−2D0ω00000(1−4D02)ω04+4D0D1ω03ω1−ω02ω120000(4 D0−8D03)ω03+(8D02D1ω1−2D1ω1)ω02−2D0ω0ω1210−ω120001−2D1ω1]bBNF 2. Art=[−K0ω020−(ω12−ω02+4D02ω0−4D0D1ω0ω1)K0ω022(D0ω0−D1ω1)K0ω02−K0ω02−K1ω12]CBNF 2. Art=[0100000001]⇒HBNF 2. Type=[−ω0+p1p20−D0ω0−p1−p20(1−4D02)ω04+4D0D1ω03ω1−ω02ω12−p3p4p5(4D0−8D0 3)ω03+(8D02D1ω1−2D1ω1)ω02−2D0ω0ω12−ω12+p3p5+p3p4+p4p50−2D1ω1−p3−p4−p5]

[0070] The observer is thus able to generate a reconstructed estimated signal for the fundamental frequency from the disturbed (by harmonics etc.) measurement signals of strain gauges and gyroscope, which then dampens the conductor vibrations via the feedback.

[0071] Basically, it should be noted that all approaches presented so far can be applied analogously to the direction of rotation of the conductor.

Claims

[1] Turntable ladder, telescopic mast platform or the like, with a telescopic ladder set or telescopic mast and, if appropriate, a car attached thereto, which turntable ladder or telescopic mast comprises a control for the movement of the ladder parts or telescopic mast parts, which is designed in such a way that, when the turntable ladder or telescopic mast platform is moved, vibrations of the ladder parts or telescopic mast parts are suppressed by at least one of the measured variables bending of the ladder set or telescopic mast in the horizontal and vertical direction, erection angle, rotation angle, extension length and torsion of the ladder set or telescopic mast is fed back to the control variables for the drives via a controller, characterized bythat inertial sensors are mounted on the ladder set or telescopic mast and / or on the car to detect the bending state of the ladder set or telescopic mast, wherein a gyroscope is provided on the car and / or on the end of the ladder set or telescopic mast to which the car is mounted to measure the angular velocity in different spatial directions, and strain gauge sensors are mounted on the ladder set or telescopic mast to detect the bending state of the ladder set or telescopic mast, and furthermore characterized bya feedforward control which, when the car is moving, maps the idealised movement behaviour of the ladder or the telescopic mast in a dynamic model based on differential equations and calculates idealised control variables for the drives of the ladder sections or telescopic mast sections for an essentially vibration-free movement of the ladder or the telescopic mast from the dynamic model, which dynamic model simulates a mass distribution of the ladder set or telescopic mast, whereby all state variables x1 to x4 of the dynamic model represent measured variables of the strain gauge sensors and the gyroscope, and which combines the measured variables in an observer structure and estimates the amplitude components of the first two modes of the natural vibration of the ladder or the telescopic mast, which are fed back to the controller as a control variable. [2] Turntable ladder or telescopic mast platform according to claim 1, characterized bythat further inertial sensors for measuring acceleration in different spatial directions are provided on the car and / or on the end of the ladder set or telescopic mast to which the car is attached. [3] Turntable ladder or telescopic mast platform according to claim 1 or 2, characterized by that the result of the sensor data fusion of a modal transformation is carried out to calculate the first two modes of the natural vibration of the ladder set or telescopic mast. [4] Turntable ladder or telescopic mast platform according to one of the preceding claims, characterized bythat a path planning module is provided for generating the path of movement of the ladder or the telescopic mast in the working space, which transmits the path of movement in the form of time functions for the car position, car speed, car acceleration, car jerk and, if applicable, derivation of the car jerk to a pilot control block which controls the drives of the ladder sections or telescopic mast sections.

Images