EP4077929A1

EP4077929A1 - Method and device for determining a change in a mass distribution of a rotor blade on a wind turbine

Info

Publication number: EP4077929A1
Application number: EP20823795.8A
Authority: EP
Inventors: Daniel Brenner
Original assignee: Weidmueller Monitoring Systems GmbH
Current assignee: Weidmueller Monitoring Systems GmbH
Priority date: 2019-12-19
Filing date: 2020-12-10
Publication date: 2022-10-26
Also published as: WO2021122319A1; DE102019135108A1

Abstract

The invention relates to a method for determining a change in a mass distribution of a rotor blade (41) of a wind turbine (1), said method having the following steps: − determining first natural frequency states of the rotor blade (41) on the wind turbine (1) at a first mass distribution of the rotor blade (41); − capturing vibrations of the rotor blade (41) at a second mass distribution, which may differ from the first mass distribution; − determining second natural frequency states of the rotor blade (41) from the vibrations captured; − determining changes between the first and the second natural frequency states; and − deriving a distribution of an additional mass which corresponds to the difference between the second and the first mass distribution. The invention further relates to a device for determining a change in a mass distribution of a rotor blade (41) on a wind turbine (1), comprising at least one sensor (61) for recording a vibration in the rotor blade (41) and an evaluating unit (63) for implementing a method of this kind.

Description

Method and device for determining a change in a mass distribution of a rotor blade of a wind turbine

The invention relates to a method for determining a change in a mass distribution of a rotor blade of a wind turbine. The invention further relates to a device for carrying out the method, the device having vibration sensors on the rotor blade and / or on components of the wind turbine connected to it, as well as an evaluation device for signals from the vibration sensors.

The evaluation of vibrations of a rotor blade of a wind turbine either via vibration sensors that are arranged in the rotor blade itself and / or via vibration sensors that are attached to components that are connected to the rotor blade, such as a drive train or a nacelle of the wind turbine, is an effective means of detecting damage to rotor blades or the accumulation of additional mass. Such additional masses can result, for example, from the accumulation of dirt and, in particular, of ice. Ice can be deposited in large quantities on rotor blades, up to the 10 or 100 kilograms (kg) range. To avoid the dangers of falling and thrown ice, knowledge of the accumulated mass of the ice is of great interest.

The document DE 10 2016 124 554 A1 describes a method for recognizing an accumulation of ice on a rotor blade of a wind turbine, in which a change in a natural frequency is used to deduce ice accumulation. In order to achieve reliable detection results, this indirect method is combined with measurement results from sensors, with the aid of which it is possible to immediately deduce ice growth. Sensors with which this is possible are, for example, conductivity sensors on the surface of the rotor blade or optically or acoustically operating sensors that can determine the layer thickness of an incremental ice. The disadvantage of these sensors is that they can only detect ice locally in the immediate vicinity of the sensor.

A method is known from the document DE 10 2006 009 480 B1 in which a plurality of such direct sensors are provided for determining a thickness profile in order to measure the thickness of an accumulated ice. Layer to be able to detect over a larger spatial area on the rotor blade. Due to the large number of sensors, such an arrangement is complex to install and also to maintain and / or troubleshoot.

Particularly with regard to the growth of ice, however, the question of the distribution of the additional mass on the rotor blade is relevant, since ice that is arranged further out on the rotor leads to a larger static imbalance than an arrangement of an equally large additional mass in the area of the blade root. A large static imbalance represents a load on the bearings of the rotor, which can lead to damage to the position or at least a reduction in the life of the bearing.

In addition, ice growing on the tip of the rotor blade runs the risk of being thrown far away from the location of the wind turbine. Ice accumulation in the area of the blade root is less critical insofar as it would generally not be thrown outside the rotor radius even when the wind turbine is rotating. This area is usually marked as a danger area, where, on the other hand, ice thrown further away from the location of the wind turbine can lead to property and personal danger.

It is an object of the present invention to provide a method and a device of the type mentioned above, with which an additional mass growing on a rotor, for. B. an ice mass, can not only be detected, but also a distribution of the additional mass along the rotor blade can be determined.

This object is achieved by a method and a device with the respective features of the independent claims. Advantageous Ausgestaltun conditions and developments are the subject of the dependent claims.

According to a method according to the invention, first natural frequency states of the rotor blade of the wind energy installation are determined. This can be done using measurements or using models if a mass distribution of the rotor blade and information about the stiffness of the rotor blade are known. When the wind power plant is in operation, the rotor blade vibrates with an unknown second mass distribution, which may differ from the first mass distribution. Second natural frequency states are determined from the vibration measurements. On the basis of a The deviation of the second natural frequency states from the first natural frequency states can be read off whether and how the second mass distribution differs from the first mass distribution, i.e. whether a dirt or ice build-up has occurred. A distribution of additional masses that corresponds to the difference between the second and the first mass distribution is then derived from the measured changes in the natural frequency states.

The invention is based on the knowledge that by considering at least two natural frequency states, not only the presence of an additional mass on the rotor blade can be detected, but statements about the position or distribution of the additional masses can also be derived. In the context of the application, resonance frequencies and / or resonance amplitudes of oscillation states of different orders are understood as natural frequency states.

The first natural frequency states can be determined on the basis of measurement data in that vibrations of the rotor blade are determined during the first mass distribution and the first natural frequency states of the rotor blade are determined from the recorded vibrations. For this purpose, the vibrations are preferably recorded using measurement signals from sensors which, for example, can be arranged in or on the rotor blades, with time-dependent vibration deflections derived from the measurement signals being recorded for a certain period of time.

From the oscillation deflections recorded in the time domain, an amplitude spectrum can then be determined by transformation into the frequency range and the natural frequency states can be determined on the basis of maxima of the amplitude spectrum. Natural frequency states are characterized, for example, by a frequency and / or an oscillation amplitude.

In an advantageous embodiment of the method, the second mass distribution is determined on the basis of known dependencies of the frequencies of the eigenfrequency states of mass distributions. Knowledge gained empirically or derived from model calculations about the effects of additional masses on the natural frequency states can be used to determine the unknown second mass distribution. The second mass distribution is preferred on the basis of known dependencies of the ratio of the frequencies of different natural frequency states of mass sensor distributions determined. This embodiment is based on the knowledge that the frequency ratio of different natural frequencies changes very characteristically with certain changes in the mass distribution. In particular, a distinction can be made between ice or dirt deposits that are evenly distributed along the rotor blade and more localized.

In a further advantageous embodiment of the method, the second mass distribution is determined in that natural frequency states of assumed mass distributions are determined using mathematical models of the rotor blade and compared with the measured second natural frequency states. The mathematical models preferably take into account the first natural frequency states, e.g. to adapt free parameters of the model to the actual measurements.

A device according to the invention for determining a change in a mass distribution of a rotor blade of a wind turbine has at least one sensor for recording a vibration of the rotor blade and an evaluation unit, the device being set up to carry out such a method. The advantages mentioned in connection with the process result.

The invention is explained in more detail below using an exemplary embodiment with the aid of figures. The figures show:

Fig. 1 is a schematic sectional view of part of a wind power plant;

2 shows a diagram for the representation of natural frequency states in the case of a rotor blade of a wind turbine;

3 shows an illustration of an amplitude spectrum of Schwingungszu states of a rotor blade of a wind turbine;

4 shows a diagram to illustrate the influence of different mass distributions on different natural oscillation states of the rotor blade; and

5 shows a flow diagram of a method for determining a mass distribution, for example an ice accumulation on a rotor blade. In Fig. 1 is an example of a sectional drawing of part of a wind power plant 1 is shown, which has a device 6 for determining a change egg ner mass distribution of a rotor blade. The wind energy installation 1 shown in FIG. 1 is therefore suitable and set up for carrying out a process according to the application for determining a change in a mass distribution of a rotor blade.

The wind energy installation 1 has a nacelle 3 which is rotatably mounted on a tower 2 and which carries a rotor 4. The rotor 4 has at least one rotor blade 41 which is connected to a rotor shaft 51 at a hub 42. The area of the hub 42 and the approach of the rotor blades 41 is covered by a spin ner 43.

In Fig. 1, two rotor blades 41 shown cut to length are exemplary shows ge. This is purely by way of example; wind turbines often have three rotor blades 41.

Said rotor shaft 51 is part of a drive train 5. It transmits the rotational movement of the rotor 4 to a gear 52. This in turn is coupled via a gear shaft 53 and a clutch 54 to a generator 55, which converts the mechanical energy of the rotor 4 into electrical energy converts. The representation of the wind energy installation 1 with gear 55 is also purely by way of example. The device according to the application and the method according to the application can just as easily be implemented with a gearless wind energy installation.

The device 6 for determining a change in a mass distribution of a rotor blade, hereinafter referred to as monitoring device 6 for short, comprises at least one vibration sensor 61, hereinafter referred to as sensor 61 for short.

In the present case, a sensor 61 is arranged in each of the rotor blades 41 shown. Each sensor 61 is connected to an evaluation unit 63 via a sensor line 62. The type of connection is shown in FIG. 1, purely by way of example. As a rule, there is a connection between the sensors 61 and the evaluation unit 63 via sensor lines running in the rotor blade 41 as far as the spinner 43, from where a usually wireless transmission to the evaluation unit 63 takes place. In alternative configurations, the sensors 61 can be coupled to energy harvesting units, see above that they draw energy, for example, from the rotation of the rotor 4 and transmit data directly from the rotor blade 41 via radio to the evaluation unit 63. Energy supply of the sensors 61 via optical fibers within the rotor blades 41, as well as optical data transmission from the sensors 61 to the evaluation unit 63 or at least to a radio relay station in the spinner 43, is also conceivable.

The sensors 61 are vibration sensors that detect a vibration of the Ro torblatts 41. The sensors 61 can be acceleration, expansion or rotation rate sensors. An oscillation is then recorded as a change in a measured acceleration value, a measured speed or a measured expansion. The arrangement of the sensors 61 within the rotor blade 51 can be such that vibrations in the pivoting direction (“edge”) and / or in the flapping direction (“flap”) and / or in the torsion direction of the respective rotor blade 41 are detected.

Purely by way of example, the two sensors 61 shown in FIG. 1 are arranged approximately in a lower third of the rotor blade 41. The sensors 61 can, however, also be arranged at other positions in the rotor blade 41. In addition, it is possible to arrange several sensors 61 in each rotor blade 41, which are evaluated jointly or independently of one another.

It is also possible to detect vibrations of the rotor blades 41 on other components of the wind energy installation 1, on which corresponding vibration sensors are then arranged. For example, sensors can be arranged in the hub 42 and / or along the drive train 5, with vibrations of the rotor blade 41, which show up in these sensors, based on, for example, their frequency range of inherent vibrations on the drive train 5, for example due to gear meshes in the transmission 42, can be distinguished.

Fig. 2 shows in a schematic representation initially possible oscillation states 7 of a rotor blade, for example one of the rotor blades 41 according to FIG. 1. Shown is an oscillation amplitude on the vertical axis of the diagram as a function of a position along the rotor blade on the horizontal axis.

Each of the curves 71-74 represents an instantaneous deflection which is characteristic of the respective oscillation state 7. The position "0" on the horizontal axis corresponds to the position of the blade root and the position “max” on the horizontal axis corresponds to the position of the blade tip.

In Fig. 2 four oscillation states 7 are shown, a basic state in curve 71, a first harmonic in curve 72, which is characterized by a vibration node along the extension of the rotor blade, a second harmonic in curve 73, which is characterized by two vibration nodes and one third harmonic in curve 74, which is characterized by three oscillation nodes along the rotor blade. In the following, the fundamental oscillation according to curve 71 is referred to as the first natural frequency state and the first, second and third harmonic as the second, third and fourth natural frequency state.

In Fig. 2, transverse vibrations, that is, vibrations in the pivoting direction or direction of impact of the rotor blade are shown. A comparable picture also emerges for torsional vibrations, i.e. rotations of the rotor blade about its longitudinal axis.

When the monitoring device 6 is in operation, the time-dependent oscillation deflection derived from its measurement signals is recorded for a specific period of time for each of the sensors 61.

An amplitude spectrum is then preferably determined from the oscillation recorded in the time domain. The transformation into the frequency range, that is to say the representation as a spectrum, can take place, for example, by means of a fast Fourier transformation (FFT) or a wavelet transformation. Alternatively, instead of a transformation into the frequency domain, natural frequency states can also be determined in the time domain by appropriate filtering or by stochastic methods, for example via the so-called “Stochastic Subspace Identification” (SSI).

3 shows a spectrum transformed, for example via FFT, from the vibration recordings in the time domain into the frequency domain in a spectral curve 75. The vertical axis shows the amplitude of the vibration as a function of the frequency plotted on the horizontal axis.

In this representation, natural frequency states can be identified in a simple manner as maxima of the spectral curve 75. The assignment of the maxima to the different natural frequency states is possible through the increasing frequency.

The described vibration measurement and generation of a spectrum is repeated at regular time intervals. If a mass distribution on the rotor blade 41 changes, for example due to ice formation, the natural frequency states determined from the spectrum according to FIG. 3 change. These are characterized by their frequency and an assigned maximum amplitude.

It turns out that the natural frequency states change in different ways, depending on how the increase in mass of the rotor blade is locally distributed.

This is shown by way of example in FIG. 4 using a model calculation. Fig. 4 shows in a diagram on the vertical axis, the maximum amplitude de of two natural frequency states, namely the second natural frequency and the third natural frequency state, during the growth of ice (or more generally an additional mass), which is presented on the horizontal axis . The units on the horizontal axis are percentages that indicate an increase of 0-100% of the additional mass.

The maximum amplitudes of the natural frequency states are normalized in such a way that they have the value 1 if the rotor blade is in its original state, i.e. no additional mass has yet grown (i.e. at the value 0% on the horizontal axis).

4 shows the results 8 of two different model calculations. The first model calculation shows the development of the second and third natural frequency state when the ice accumulation occurs evenly over the entire length of the blade. The correspondingly calculated ratios se of the frequency to the ice-free state are shown in the diagram by diamonds 81 for the second natural frequency state and circles 82 for the third natural frequency state.

In a second scenario, the ice does not grow evenly over the entire length of the blade, but is concentrated in an outer area of the rotor blade. For this scenario, the ratio of the frequency to the ice-free state of the second oscillation state is given by squares 83 and the Ratios of the frequency to the ice-free state of the third oscillation state represented by triangles 84 in the diagram.

4 shows that in both cases, with increasing additional mass, the natural frequency states shift to lower frequencies. The shift is more pronounced for concentrated growth of the ice at the tip of the blade than for uniform growth.

It is particularly relevant for the method according to the application that the ratio of the frequency to the ice-free state of the second and third eigenfrequency state behaves differently in the two scenarios. If the ice layer grows evenly, the changes in the maximum amplitude for the second and third natural frequency state are essentially uniform over the entire calculated area, whereas if the additional mass is located at the tip of the blade, the amplitude of the second oscillation state decreases more significantly as the additional mass increases than that of the third natural frequency state.

From this different behavior, by comparing the changes in the natural frequency states, either with regard to their frequency but also with regard to their amplitude, with the model calculations, conclusions can be drawn about the mass distribution of the additional mass on the rotor blade.

In a method according to the application, the changes in the natural frequency states can accordingly be compared with previously made model calculations for mass distribution or with in-situ model calculations for mass distribution and a mass distribution that results from the increase in the additional mass can be derived. An exemplary embodiment of a method according to the application is explained in more detail below with reference to FIG. 5 using a flow chart. The method can be carried out, for example, in connection with the wind energy installation 1 shown in FIG. 1. The method is explained by way of example with reference to the wind energy installation 1 according to FIG. 1.

In a first step S1, first natural frequency states of the rotor blades 41 of the rotor 4 of the wind energy installation 1 are determined in an ice-free state of the rotor 4. Natural frequency states are those oscillation states in which the rotor blades 41 oscillate after being excited to oscillate. How is explained in connection with FIG. 3, natural frequency states appear at maxima of an oscillation spectrum. The first natural frequency states can accordingly be determined in step S1 by measuring and recording a time curve of vibrations, for example by one or more sensors 61. A vibration spectrum is calculated from the time curve by transforming it into the frequency range. The natural frequency states then result from the position (and possibly fleas) of the maxima in the spectrum.

As an alternative to a measurement on the wind energy installation 1, the first natural frequency states can be determined in a comparable manner using measurements that are already carried out by the manufacturer on test stands as part of quality controls. It is also possible to calculate the first eigenfrequency states on the basis of model data of the rotor blades 41.

The natural frequency states of the rotor blades 41 in the ice-free state represent reference measurements with a first mass distribution of the rotor blades 41.

In a second step S2, vibrations of the rotor blades 41 are measured at a second, unknown and possibly deviating from the first mass distribution and, as described above, second natural frequency states are determined therefrom in a step S3.

In a subsequent step S4, the differences in the natural frequency states are determined, in particular any frequency shifts that have occurred in the natural frequency states are determined. The second mass distribution is then derived from these in a step S5. The amount and position of any ice accumulation can be calculated from the difference between the second and the first mass distribution.

Steps S2 to S5 are preferably repeated at certain time intervals in order to ensure (quasi-) continuous monitoring of the rotor blades 41 with regard to ice or dirt accumulation.

To determine the second mass distribution, empirical or methodical knowledge about the effect of additional masses on the eigenfrequency states can be used, which can be derived, for example, from curves as shown in FIG. 4. In particular, the effect of access set masses to a ratio of frequencies of the natural frequency states of different natural frequencies are taken into account.

If a mathematical model is available for the vibration behavior of the rotor blades 41, a model calculation can also be used in which a distribution of additional masses is varied and natural frequency states derived from the model are compared with the measured ones. The additional mass distribution is varied until a “best fit” is achieved. If the method is used over a longer period of time, within which aging of the rotor blades 41 becomes noticeable in the first natural frequency states, it makes sense to repeat step S1 and thus to keep a current reference of the natural frequency states in an ice-free state.

Reference number

1 wind turbine

2 tower

3 gondolas

4 rotor

41 sheets

42 hub

43 weirdos

5 drive train

51 rotor shaft

52 transmission

53 gear shaft

54 clutch

55 generator

6 Monitoring device

61 sensor

62 sensor cable

63 Evaluation unit

7 State of vibration 71-74 curve

75 Spectral curve

8 Result of the model calculation

81 diamond (2nd natural frequency state, uniform ice accumulation)

82 circle (3rd natural frequency state, uniform ice accumulation)

83 square (2nd natural frequency state, local ice accumulation)

84 triangle (3rd natural frequency state, local ice accumulation)

S1-S5 step

Claims

Expectations

1. A method for determining a change in a mass distribution of a rotor blade (41) of a wind turbine (1) with the following steps:

- Determination of first natural frequency states of the rotor blade (41) of the wind energy installation (1) with a first mass distribution of the rotor blade (41);

- Detecting vibrations of the rotor blade (41) with a second mass distribution, which may differ from the first mass distribution under;

- Determination of second natural frequency states of the rotor blade (41) from the detected vibrations;

- Determining changes between the first and second eigenfrequency states; and

- Deriving a distribution of an additional mass which corresponds to the difference between the second and the first mass distribution.

2. The method according to claim 1, wherein the vibrations of the rotor blade (41) are determined in the first mass distribution and the first eigenfre frequency states of the rotor blade (41) are determined from the detected vibrations.

3. The method according to claim 1 or 2, in which the vibrations are detected using measurement signals from sensors (63), time-dependent vibration deflection derived from the measurement signals being recorded for a specific period of time.

4. The method as claimed in claim 3, in which an amplitude spectrum is determined from the oscillation deflections recorded in the time domain by transformation into the frequency range.

5. The method according to claim 4, in which natural frequency states are determined on the basis of maxima of the amplitude spectrum.

6. The method according to claim 1, wherein the first natural frequency states are calculated using model data of the rotor blade (41).

7. The method according to any one of claims 1 to 6, wherein the natural frequency states are characterized by a frequency and / or a vibration amplitude GE.

8. The method according to any one of claims 1 to 7, wherein the second mass distribution is determined based on known dependencies of the frequencies of the egg gene frequency states of mass distributions.

9. The method according to claim 8, wherein the second mass distribution is determined on the basis of known dependencies of the ratio of the frequencies of different natural frequency states of mass distributions.

10. The method according to any one of claims 1 to 9, in which the second mass distribution is determined by determining natural frequency states of assumed mass distributions using mathematical models of the rotor blade (41) and comparing them with the measured second natural frequency states.

11. Method according to Claim 10, in which the mathematical models take into account the first natural frequency states.

12. Device for determining a change in a mass distribution of a rotor blade (41) of a wind turbine (1), having at least one sensor (61) for recording a vibration of the rotor blade (41) and an evaluation unit (63), the device for performing a Ver Fahrens according to one of claims 1 to 11 is set up.