DE102014208681B4 - Method for monitoring a condition of a rotor blade moving in a fluid - Google Patents

Abstract

Method for monitoring a state of a rotor blade (1) moving in a fluid, comprising the following steps: a) Detection of vibrations generated by structure-borne sound waves in the rotor blade (1) by means of at least one measuring device attached to the rotor blade (1) or integrated into the rotor blade (1) ( 2), b) determining the structure-borne noise level over a predetermined frequency range from the recorded vibrations, c) determining a flow velocity v of the fluid relative to the rotor blade (1) from the structure-borne noise level, and d) determining a parameter representing the state of the rotor blade (1) from the determined flow velocity v, where the parameter is either d1) a strain d and the following relationship is used to determine the strain: where c is a constant, or d2) that a drag force K is determined as a parameter and the following relationship is used to determine the drag force K is used:, whereC = the resistance number, ρ = the density of the fluid, A = the effective cross section or the aerodynamic end face of the rotor blade.

Description

The invention relates to a method for monitoring a state of a rotor blade moving in a fluid.

In the prior art, wind turbines are generally known. Such wind turbines have a rotor with several rotor blades. During periodic rotation, the rotor blades are excited to vibrate at low frequencies by the so-called “tower shadow effect”. If the low-frequency vibrations are in the range of a resonance frequency, the rotor blade can be overloaded and thus undesirably damaged even at relatively low wind speeds. In addition, aperiodic loads on the rotor blade due to gusts can also cause vibrations, which load the rotor asymmetrically and thus damage the bearings and / or gears. Similar problems occur not only with rotors of wind power plants, but also with rotors of helicopters, turbines of water power plants, ship propellers and the like.

From the US 2009/0169378 A1 a wind turbine is known with sensors embedded in the rotor blades, which are designed, for example, as microphones or vibration sensors. By evaluating sensor signals, it can be determined whether there is a flow separation on a rotor blade. The wind turbine has a tilt angle adjustment that can be controlled on the basis of the sensor signals.

The DE 695 31 315 T2 discloses a method for real-time monitoring of fatigue and stress corrosion cracks in helicopter rotor heads. Using a piezoelectric PVDF sound emission energy converter on the component to be monitored, non-repeating, high-frequency sound emission events in a frequency range of 1 to 10 MHz are measured and the resulting data are transmitted to a signal processor. From the occurrence of a noise emission event, it can be concluded that the monitored component has been damaged.

The US 2011/0135475 A1 discloses a method for monitoring a condition of a rotor blade of a wind turbine. A vibration of the rotor blade is detected by means of at least one sensor. The sensor is attached to a hub, a rotor shaft or other components of a drive train of the rotor. Depending on the measured vibrations of the rotor, in the event of damage to the rotor blade, the wind turbine can be put out of operation by changing the pitch angle accordingly.

With the known methods it is possible to detect damage to a component. Occasionally, however, it cannot be prevented that damage and thus breakage of the component occur.

The object of the invention is to eliminate the disadvantages of the prior art. In particular, a method that can be carried out as simply and inexpensively as possible is to be specified, with which a condition of a rotor blade can be monitored with improved accuracy. According to a further aim of the method, load states are to be recognized even before damage occurs, which is likely to lead to damage.

This object is achieved by the features of claim 1. Appropriate refinements of the invention result from the features of claims 2 to 14.

1. a) Erfassung von durch Körperschallwellen im Rotorblatt erzeugte Schwingungen mittels zumindest einer am Rotorblatt angebrachten oder in das Rotorblatt integrierten Messeinrichtung,
2. b) Bestimmen des Körperschallpegels über einem vorgegebenen Frequenzbereich aus den erfassten Schwingungen,
3. c) Ermittlung einer Strömungsgeschwindigkeit v des Fluids relativ zum Rotorblatt aus dem Körperschallpegel, und
4. d) Bestimmung eines den Zustand des Rotorblatts wiedergebenden Parameters aus der ermittelten Strömungsgeschwindigkeit v, wobei

der Parameter entweder

• d1) eine Dehnung d ist und zur Bestimmung der Dehnung die folgende Beziehung verwendet wird: $$d=c\cdot v^2$$
  
  , wobei c eine Konstante ist, oder
• d2) dass als Parameter eine Strömungswiderstandskraft K bestimmt wird und zur Bestimmung der Strömungswiderstandskraft K die folgende Beziehung verwendet wird: $$K=\frac{1}{2}\cdot C_w\cdot \rho \cdot v^2\cdot A$$
  
  , wobei

  C_w =

  die Widerstandszahl,

  ρ=

  die Dichte des Fluids,

  A

  = der effektive Querschnitt bzw. die aerodynamische Stirnfläche des Rotorblatts

  ist.

According to the invention, a method for monitoring a state of a rotor blade moving in a fluid is proposed with the following steps:

1. a) Detection of vibrations generated by structure-borne sound waves in the rotor blade by means of at least one measuring device attached to the rotor blade or integrated into the rotor blade,
2. b) Determination of the structure-borne noise level over a specified frequency range from the recorded vibrations,
3. c) determining a flow velocity v of the fluid relative to the rotor blade from the structure-borne sound level, and
4. d) Determination of a parameter representing the state of the rotor blade from the determined flow velocity v, wherein

the parameter either

• d1) is an elongation d and the following relationship is used to determine the elongation: $$d=c\cdot {\mathrm{<figure-callout\; id="2"\; label="v"\; filenames="DE102014208681B4\_0013.png,DE102014208681B4\_0014.png"\; state="\{\{state\}\}">v</figure-callout>}}^2$$
  
  , where c is a constant, or
• d2) that a drag force K is determined as a parameter and the following relationship is used to determine the drag force K: $$K=\frac{1}{2}\cdot {\mathrm{C.}}_w\cdot \rho \cdot {\mathrm{<figure-callout\; id="2"\; label="v"\; filenames="DE102014208681B4\_0013.png,DE102014208681B4\_0014.png"\; state="\{\{state\}\}">v</figure-callout>}}^2\cdot \mathrm{A.}$$
  
  , in which

  C _w =

  the resistance number,

  ρ =

  the density of the fluid,

  A.

  = the effective cross-section or the aerodynamic frontal area of the rotor blade

  is.

The invention is based on the knowledge that the flow around a rotor blade with a fluid causes an acoustic excitation of the fluid as well as an acoustic excitation of the rotor blade. The structure-borne sound formed in the rotor blade is attenuated to a comparatively high degree compared to the sound formed in the fluid. According to the invention, it is proposed to measure vibrations generated in the rotor blade by structure-borne sound waves by means of at least one measuring device attached to the rotor blade or integrated into the rotor blade, ie. H. to detect a change in the amplitude of the oscillations over time.

In the context of the present invention, the term “fluid” is understood to mean a liquid, in particular water, or a gas, in particular air.

The invention is also based on the knowledge that the structure-borne sound level increases with increasing flow velocity v of the fluid surrounding the rotor blade in a known frequency range or at specific frequencies. In particular, an average frequency of a maximum structure-borne sound level increases with increasing flow velocity v. As a result, it is possible to determine the flow velocity v of the fluid relative to the rotor blade from the structure-borne sound level. It is also possible in a particularly advantageous manner to determine a parameter reproducing the state of the rotor blade from the determined flow velocity v.

The method according to the invention can be carried out comparatively simply and inexpensively. It allows the condition of a rotor blade, in particular a rotor blade moving in air, to be monitored with high accuracy. The method allows in particular the detection of load states which are likely to lead to damage to the rotor blade. Measures can thus be taken at an early stage to prevent damage to the rotor blade.

Another advantage of the method according to the invention is that the structure-borne noise can be measured at any point on the rotor blade by means of the measuring device. For example, it is possible to record the structure-borne noise in the area of a root of the rotor blade. From the detected structure-borne sound, it is possible, for example, to infer an average strain or deflection of the rotor blade at a position remote from the measuring device.

wobei c eine Konstante ist. Die Konstante c kann beispielsweise mittels der folgenden Beziehung ermittelt werden: $$c=\frac{1}{4}\cdot C_w\cdot \rho \cdot A\cdot \frac{\left(z\cdot y\right)}{\left(I_z\cdot E\right)}$$

wobei

c =

die effektive Entfernung der punktuellen Ersatzkraft mit dem Betrag K' zum Ort der Dehnungsmessung,

C_w =

die Widerstandszahl,

ρ =

die Dichte des Fluids,

A =

der effektive Querschnitt bzw. die aerodynamische Stirnfläche des Rotorblatts,

z =

die effektive Entfernung der punktuellen Ersatzkraft zum Ort der Dehnungsmessung,

y =

der Abstand der gedehnten Faser zur neutralen Faser,

I_Z =

das axiale Flächenträgheitsmoment,

E =

der E-Modul,

ist.According to a first embodiment of the invention, the parameter is an elongation d and the following relationship is used to determine the elongation: $$d=c\cdot {\mathrm{<figure-callout\; id="2"\; label="v"\; filenames="DE102014208681B4\_0013.png,DE102014208681B4\_0014.png"\; state="\{\{state\}\}">v</figure-callout>}}^2$$

where c is a constant. For example, the constant c can be determined using the following relationship: $$c=\frac{1}{\mathrm{4th}}\cdot {\mathrm{C.}}_w\cdot \rho \cdot \mathrm{A.}\cdot \frac{\left(z\cdot y\right)}{\left({\mathrm{I.}}_z\cdot \mathrm{E.}\right)}$$

in which

c =

the effective distance of the punctual substitute force with the amount K 'to the location of the strain measurement,

C _w =

the resistance number,

ρ =

the density of the fluid,

A =

the effective cross-section or the aerodynamic face of the rotor blade,

z =

the effective distance of the punctual substitute force to the location of the strain measurement,

y =

the distance between the stretched fiber and the neutral fiber,

I _Z =

the axial geometrical moment of inertia,

E =

the modulus of elasticity,

is.

wobei

C_w =

die Widerstandszahl,

ρ =

die Dichte des Fluids,

A =

der effektive Querschnitt bzw. die aerodynamische Stirnfläche des Rotorblatts

ist.According to a second alternative embodiment of the invention, a flow resistance force K is used as the parameter and the following relationship is used to determine the flow resistance force K: $$K=\frac{1}{2}\cdot {\mathrm{C.}}_w\cdot \rho \cdot {\mathrm{<figure-callout\; id="2"\; label="v"\; filenames="DE102014208681B4\_0013.png,DE102014208681B4\_0014.png"\; state="\{\{state\}\}">v</figure-callout>}}^2\cdot \mathrm{A.}$$

in which

C _w =

the resistance number,

ρ =

the density of the fluid,

A =

the effective cross-section or the aerodynamic face of the rotor blade

is.

According to a further advantageous embodiment of the invention, the measuring device is attached in the area of a root of the rotor blade. The signals or data generated by the measuring device are expediently transmitted wirelessly to a data processing device. In this case, the data processing device is located on a housing component, which is fixed relative to the rotor, of a device accommodating the rotor. This enables the method to be carried out relatively easily.

A strain measuring device and / or an acceleration measuring device and / or a compression measuring device is expediently used as the measuring device. Such measuring devices are generally known. In the case of a strain measuring device, a voltage proportional thereto is generated, for example, by means of a piezoelectric element as a function of strain. Suitable piezoelectric transducer elements are generally known in the prior art. For example, they can be made using PZT fibers or PZT films. - The measuring device advantageously comprises a signal amplifier, signal filter, signal converter and / or a radio device, e.g. B. a transponder, for the transmission of data or signals.

The above steps a) to d) are expediently carried out repeatedly with a predetermined clock frequency. The clock frequency can be selected so that the parameter is determined in real time.

According to a further embodiment of the method, the structure-borne sound level is averaged over the frequency range and the averaged values of the structure-borne sound level are used to determine the flow velocity v. This enables a particularly simple and rapid determination of the parameter representing the state of the rotor blade.

According to a further particularly advantageous embodiment of the invention, a structure-borne sound level spectrum is generated by frequency-resolved detection of the structure-borne sound level. Natural frequencies can be determined from the structure-borne sound level spectrum, and the modulus of elasticity or the mass and / or a mass distribution can be determined from the natural frequencies. Furthermore, loads or load collectives can be determined from the structure-borne noise level. Apart from this, damage in the rotor blade can be determined from burst signals identified in the structure-borne sound level spectrum. Using at least two measuring devices attached to the rotor blade at a distance from one another or integrated into the rotor blade, it is also possible to determine a location of damage in the rotor blade from the structure-borne sound levels determined thereby.

The structure-borne sound level spectrum can be determined in a known manner, for example using a signal analyzer with the aid of the fast Fourier or wavelet transformation, or according to the principle of the superposition receiver.

According to a further advantageous embodiment, a temperature of the rotor blade is measured by means of a temperature measuring device attached to the rotor blade, and the measured temperature is used for calibration and / or for correction when determining the parameter. The temperature measuring device can be part of the measuring device for detecting the vibrations generated by the structure-borne sound waves in the rotor blade. This can further increase the accuracy of the parameter used to describe the state of the rotor.

Furthermore, the flow velocity v of the fluid relative to the rotor blade is advantageously determined from the structure-borne noise level on the basis of calibration values previously determined for the rotor blade. The calibration values can relate, for example, to the modulus of elasticity, the constant c and its temperature and / or location dependency.

According to a further embodiment of the method, a frequency of 0 to 500 kHz, preferably 0.1 Hz to 10 kHz, particularly preferably 10 Hz to 100 Hz, is used as the frequency range. The choice of the respective frequency range depends on the geometry of the rotor blade. A suitable limitation of the frequency range can further increase the accuracy of the method.

Using suitable filters, in addition to the structure-borne sound frequencies typically lying in the frequency range from 0 to 1 kHz, higher-frequency frequency ranges above 1 kHz can also be observed. In the higher-frequency frequency ranges, burst signals occur which are caused by structural breaks or micro-cracks. In addition to the parameters mentioned above, any damage that may occur on the rotor blade can thus be detected at an early stage.

• 1 schematisch einen Versuchsaufbau,
• 2 die Dehnung über der Zeit für verschiedene Windgeschwindigkeiten,
• 3 die Dehnung über der Windgeschwindigkeit,
• 4 der Körperschallpegel über der Frequenz für die unterschiedlichen Windgeschwindigkeiten und
• 5 den mittleren Körperschallpegel über der Windgeschwindigkeit für ausgewählte Frequenzbereiche.

The method according to the invention is explained in more detail below with reference to the drawings. Show it:

• 1 schematically an experimental setup,
• 2 the elongation over time for different wind speeds,
• 3 the elongation over the wind speed,
• 4th the structure-borne noise level over the frequency for the different wind speeds and
• 5 the mean structure-borne noise level over the wind speed for selected frequency ranges.

1 shows an experimental set-up with which the method was carried out by way of example. On a rotor blade 1 which can be made of glass fiber reinforced or carbon fiber reinforced plastic, for example, is a strain gauge 2 appropriate. The rotor blade 1 is by means of a holding device 3 on a carrier 4th mounted, which can be arranged for example in a wind tunnel. With the reference number 5 is a cup anemometer with which the horizontal wind speed can be measured. The experimental set-up also includes that with the reference number 6th designated wind direction transmitter for measuring the horizontal wind direction. One generally with the reference number 7th designated data processing device is removed from the rotor blade 1 on the carrier 4th assembled.

• Mit dem Schalensternanemometer 5 wird eine herrschende Windgeschwindigkeit laufend gemessen und aufgezeichnet. Zeitgleich werden von der Dehnungsmesseinrichtung gemessene Dehnungen des Rotorblatts 1 erfasst. Die entsprechenden Signale werden mittels der Dehnungsmesseinrichtung digitalisiert. Die entsprechenden Daten werden per Funk an die Datenverarbeitungseinrichtung 7 übermittelt. Sie werden mittels der Datenverarbeitungseinrichtung 7 entsprechend dem erfindungsgemäßen Verfahren weiterverarbeitet.

The function of the experimental setup is as follows:

• With the cup anemometer 5 a prevailing wind speed is continuously measured and recorded. At the same time, the expansion of the rotor blade is measured by the expansion measuring device 1 detected. The corresponding signals are digitized by means of the strain measuring device. The corresponding data are sent by radio to the data processing device 7th transmitted. You are using the data processing device 7th processed according to the method according to the invention.

2 shows by means of the strain gauge 2 recorded strain values over time for various wind speeds. Such strain measurement values or data are obtained from the strain measuring device 2 to the data processing device 7th transmitted.

3 shows one of the measured values according to 2 determined mean elongation over wind speed. How out 3 As can be seen, the mean elongation is proportional to the square of the wind speed.

4th shows the structure-borne sound level versus frequency, ie a frequency spectrum of the structure-borne sound level for the various wind speeds. How out 4th can be seen, has the rotor blade 1 Resonance frequencies in the range of 5 Hz, 20 Hz and 49 Hz. The other peak at 280 Hz, on the other hand, is caused by wind noise.

5 shows the mean structure-borne sound level versus wind speed for selected frequency ranges of the structure-borne sound spectrum. The mean values of the structure-borne noise level are largely linear to the wind speed. Their slope is surprisingly independent of the selected frequency range over which the averaging was carried out.

From the in 5 The results shown can be derived from the structure-borne noise level on the wind speed and taking into account the in 3 The results shown on the elongation of the rotor blade 1 conclude.

• Ermittlung des Parameters d:
  • Eine Messung der Körperschallpegel in einem Frequenzbereich von 20 bis 300 Hz liefert bei einem Körperschallpegel von -61 dB eine dazu korrespondierende Windgeschwindigkeit v von 20 m/s (siehe 5). Die Kalibierkurve gemäß 3 wird unter Verwendung der allgemeinen Formel $$y=c\ast x^2$$
    
    gefittet. Es ergibt sich in diesem Fall beispielsweise c zu 0,15849 * 10^-6 s²/m^-2. Daraus ergibt sich der Parameter d aus der Formel $$d=c\ast v^2=\mathrm{0,15849}\ast {10}^{-6}{s}^2{m}^{-2}\ast {\left(20m/s\right)}^2=\mathrm{63,4}\mu m/m.$$
    
• Ermittlung des Parameters K:
  • Für eine Widerstandszahl Cw von etwa 1 (bei senkrechter Anströmung), einer Dichte von Luft von ρ = 1,2 kg/m³ und einem effektiven Querschnitt A von 0,1 m² ergibt sich der Parameter K, d. h. die Strömungswiderstandskraft zu $$K=\frac{1}{2}\ast 1\ast \mathrm{1,2}kg/m^3\ast \mathrm{0,1}{m}^2\ast {\left(20m/s\right)}^2=24N.$$
    

The parameters d and K can e.g. B. can be determined as follows:

• Determination of parameter d:
  • A measurement of the structure-borne noise level in a frequency range from 20 to 300 Hz provides a corresponding wind speed v of 20 m / s at a structure-borne noise level of -61 dB (see 5 ). The calibration curve according to 3 is made using the general formula $$y=c\ast x^2$$
    
    fitted. In this case, for example, c is 0.15849 * 10 ^-6 s ² / m ^-2 . This gives the parameter d from the formula $$d=c\ast v^2=0.15849\ast {10}^{-\mathrm{6th}}{s}^2{m}^{-2}\ast {\left(\mathrm{20th}m/s\right)}^2=63.4\mu m/m.$$
    
• Determination of the parameter K:
  • For a drag coefficient Cw of about 1 (with vertical flow), a density of air of ρ = 1.2 kg / m ³ and an effective cross-section A of 0.1 m ² , the parameter K results, ie the drag force to $$K=\frac{1}{2}\ast 1\ast 1.2kG/m^3\ast 0.1m^2\ast {\left(\mathrm{20th}m/s\right)}^2=24N.$$
    

Claims (14)

Method for monitoring a state of a rotor blade (1) moving in a fluid, comprising the following steps: a) Detection of vibrations generated by structure-borne sound waves in the rotor blade (1) by means of at least one measuring device attached to the rotor blade (1) or integrated into the rotor blade (1) ( 2), b) Determination of the structure-borne noise level over a predetermined frequency range from the recorded vibrations, c) Determination of a flow velocity v of the fluid relative to the rotor blade (1) from the structure-borne noise level, and d) Determination of a parameter reflecting the condition of the rotor blade (1) from the determined flow velocity v, where the parameter either d1) is a strain d and the following relationship is used to determine the strain: $$d=c\cdot v^2$$

, where c is a constant, or d2) that a drag force K is determined as a parameter and the following relationship is used to determine the drag force K: $$K=\frac{1}{2}\cdot {\mathrm{C.}}_w\cdot \rho \cdot v^2\cdot \mathrm{A.}$$

, where C _w = the drag coefficient, ρ = the density of the fluid, A = the effective cross-section or the aerodynamic face of the rotor blade. Method according to one of the preceding claims, wherein the measuring device (2) is attached in the region of a root of the rotor blade (1). Method according to one of the preceding claims, wherein the signals or data generated by the measuring device (2) are transmitted wirelessly to a data processing device (7). Method according to one of the preceding claims, wherein a strain measuring device (2) and / which a compression measuring device is used as the measuring device. Method according to one of the preceding claims, wherein steps a) to d) are repeated with a predetermined clock frequency. Method according to one of the preceding claims, wherein the structure-borne sound level is averaged over the frequency range, and the averaged values of the structure-borne sound level are used to determine the flow velocity v. Method according to one of the Claims 1 to 5 , whereby a structure-borne sound level spectrum is generated by frequency-resolved detection of the structure-borne sound level. Procedure according to Claim 6 , whereby natural frequencies are determined from the structure-borne sound level spectrum, and the modulus of elasticity and / or the mass and / or a mass distribution is / are determined from the natural frequencies. Procedure according to Claim 6 or 7th , whereby loads or load collectives are determined from the structure-borne noise level. Method according to one of the Claims 6 to 9 , damage in the rotor blade (1) being determined from burst signals identified in the structure-borne sound level spectrum. Method according to one of the preceding claims, wherein structure-borne sound levels are determined using at least two measuring devices attached to the rotor blade (1) at a distance from one another or integrated into the rotor blade (1), and a location of the damage in the rotor blade (1) is determined therefrom. Method according to one of the preceding claims, wherein a temperature of the rotor blade (1) is measured by means of a temperature measuring device attached to the rotor blade (1) and the measured temperature is used for calibration and / or for correction when determining the parameter. Method according to one of the preceding claims, wherein the determination of the flow velocity v of the fluid relative to the rotor blade (1) from the structure-borne sound level is carried out on the basis of calibration values previously determined for the rotor blade (1). Method according to one of the preceding claims, wherein the predetermined frequency range is 0 to 100 kHz, preferably 10 Hz to 1 kHz, particularly preferably 10 Hz to 500 Hz.

Images