Field of the invention

The present invention relates in a first aspect to a method for identification of a presence of an anomaly in a specific wind turbine blade of a rotor of a wind turbine. In its second aspect the present invention relates to a system for identification of an anomaly in a specific wind turbine blade of a rotor of a wind turbine. In a third aspect the present invention relates to a computer program product which is adapted to perform the method according to the first aspect of the present invention. In its fourth aspect, the present invention relates to a wind turbine comprising the system according to the second aspect of the present invention, or comprising a data storage comprising a computer program product according to the third aspect of the present invention. In its fifth aspect the present invention relates to the use of a system according to the second aspect of the present invention or of a computer program product according to third aspect of the present invention, or of a wind turbine according to the fourth aspect of the present invention, for identification of an anomaly in a specific wind turbine blade of a rotor of a wind turbine.

Background of the invention In the last decades a tremendous number of wind turbines has been built and set up in many countries.

Wind turbines provide for an environmentally friendly source of energy relying on moving air masses which in turn originates from meteorological phenomena, such as meteorological low pressure and high pressure regions, causing wind to blow across the surface of the ground or of the surface of the sea.

When wind is blowing across an area comprising a wind turbine the wind will exert a force on each of the wing shaped turbine blades of the wind turbine which will result in a rotational movement of the rotor of the wind turbine. The rotor is coupled to a generator for transforming a rotational movement of the rotor into to an induced electric power which will be distributed to the power grid.

The wind turbine blades will take up kinetic energy from the wind according to various physical factors and convert this kinetic energy into torque in the wind turbine rotor.

According to fluid dynamics a wind blowing over ground or over sea water will have a wind speed of zero (0) at the ground or at the sea level. Above ground or sea level, the wind gradient becomes increasingly more profound with increased wind speeds at increased heights. Hence, the blowing wind contributes to kinetic energy to an area of a wind turbine blade depending on the height of that blade. Another factor influencing the take-up of kinetic energy from the wind by the wind turbine blade is the so-called shadowing effect of the tower. As the rotor carrying the wind turbine blade is ratably mounted on the top of a wind turbine tower and as the wind turbine tower has a not insignificant lateral extension the tower will exert a shadowing effect on the strength of the wind meaning that the take-up of kinetic energy from the wind by a wind turbine blade is reduced in the area immediately in front of the wind turbine tower due to lower wind speed at this position

For economic reasons it is paramount to the feasibility of a wind turbine or of a whole wind turbine park that the power yield of each wind turbine to the extent possible continuously is optimum, However, although wind turbine blades are designed with the view of having structural integrity and strength it may happen that a specific wind turbine blade of a rotor of a wind turbine, either slowly over time or even suddenly, will encounter structural defects which may have the consequence that that specific turbine blade may not provide an optimum

performance, which in turn leads to a non-optimum contribution of power of that specific wind turbine blade and hence also of that wind turbine.

In order to monitor the performance of a wind turbine various systems are known in the art. These systems all relate to monitoring the performance of the rotor itself including all the blades of the rotor and hence does not provide for detecting or identifying a defect in a single, individual wind turbine blade. Accordingly, in these prior art systems, when a non-optimum performance of an assembly of wind turbine blades has been detected, the wind turbine must be brought to a hold.

Subsequently, each wind turbine blade must be carefully inspected visually in order to see whether any defects may be detected in this way. In case it is not possible to detect any faults this way, it may be necessary to dismantle each wind turbine blade and perform further inspections of the blades on ground, optionally involving thorough physical measurements and tests of the wind turbine blades.

It is obvious that in such a situation the down time of the wind turbine amounts to a loss in power production. To this must be added expenses in performing the inspection of the blades and optionally dismantling the blades and expenses in relation to the further testing.

Another problem of the prior art monitoring systems arises in relation to safety. As the prior art systems monitors the integrity of the whole assembly, each turbine blade contributes to a third the power measured in case the rotor comprises three blades (and half the power in case the rotor comprises two turbine blades; or a fourth of the power in case the rotor comprises four turbine blades; and so forth). Accordingly, a defect in a single turbine blade, may contribute to the power loss of the whole rotor only a small amount which may not be fully detected by the prior art monitoring system, monitoring only the performance of the whole rotor and not each individual turbine blades themselves.

Hence, a defect caused by a partly break down of the integrity of a single wind turbine blade, may only contribute to a small variation in the performance of the whole rotor which may not immediately be detected by the prior art monitoring system. However, such a partly break down of the integrity of a single wind turbine blade may still pose a potential severe hazard in that the turbine blade subsequently and suddenly may encounter a full break down which results in disintegration of the blade. Such disintegration may therefore represent a severe hazard to people on the ground due to falling wind turbine elements.

The same hazards will be present in case an ice built-up has occurred in respect of one or more of the wind turbine blades. Such ice built-up may suddenly result in release of lumps of ice which due to the angular velocity of the wind turbine blade may be swung to the ground at a velocity of the order hundreds of kilometer per hour. Accordingly, the prior art monitoring systems for monitoring the integrity of a wind turbine blade assembly do not provide for optimum surveillance of the integrity of an assembly of wind turbine blades of a wind turbine and does not fully prevent disintegration hazards in case a wind turbine blade suffers a partly break down of its integrity or in case of ice built-up. It is an objective of the present invention to alleviate, or even eliminate the above described disadvantages of the prior art monitoring systems.

Brief description of the invention

These objects are achieved according to the present invention in its first, second, third, fourth, and fifth aspects.

Accordingly, in a first aspect, the present invention relates to a method for identification of the presence of an anomaly in a specific wind turbine blade of a rotor of a wind turbine comprising a number N of wind turbine blades, said number N being an integer greater than one; said method comprises: i) while operating the wind turbine, performing physical measurements of one or more physical parameters of the wind turbine, or parts thereof, during a predetermined period of time T or during a predetermined magnitude of rotation R of the rotor; said one or more measured physical parameters being related to a performance yield of the wind turbine; ii) logging the azimuth angle of one or more of said wind turbine blades while performing the measurements in step i); iii) on the basis of step ii) correlating the azimuth position of one or more of said wind turbine blades to the measured physical parameter(s) measured in step i); iv) based on the correlation performed in step iii) identifying the presence of an asymmetric and/or unequal performance yield of one or more wind turbine blades amongst the number N of wind turbine blades; v) on the basis of the identification performed in step iv) determining which wind turbine blade, if any, exhibiting an anomaly; or alternatively, on the basis of the identification performed in step iv) determining that no wind turbine blades are exhibiting an anomaly.

In a second aspect, the present invention relates to a system for identification of an anomaly in a specific wind turbine blade of a rotor of a wind turbine comprising a number N of wind turbine blade, said number N being an integer greater than one; said system comprises:

-one or more probes for making physical measurements of one or more physical parameters of a wind turbine, or parts of the a wind turbine; said one or more physical parameters being related to a power yield of the wind turbine; - a control unit configured to receive data input associated with said one or more physical parameters;

-wherein said control unit comprises a data storage, wherein said data storage comprises a computer program configured to perform calculations associated with the method according to the first aspect of the present invention. In a third aspect, the present invention relates to computer program product, when running on a computing device, being adapted to perform the method according to the first aspect of the present invention.

In its fourth aspect, the present invention relates to a wind turbine comprising the system according to the second aspect of the present invention, or comprising a data storage comprising a computer program product according to the third aspect of the present invention.

In a fifth aspect, the present invention relates to the use of a system according to the second aspect of the present invention or of a computer program product according to the third aspect or of a wind turbine according to the fourth aspect of the present invention for identification of an anomaly in a specific wind turbine blade of a rotor of a wind turbine; or for determining a reduced rotor performance of the rotor itself.

The present invention in its various aspects provides for determining the presence of an anomaly in a specific wind turbine blade of a wind turbine, thereby allowing detection of defects of a specific wind turbine blade with the view to alleviating or eliminating such defect for improving performance economy and safety in the operation of a wind turbine. Brief description of the figures

Fig. 1 is a front view illustrating the concept of azimuth angles of blades of a wind turbine.

Fig. 2 is a 3D diagram illustrating the effect of wind gradient or wind shear as a function of height and of the shadowing effect of a wind turbine tower.

Fig. 3 is a diagram illustrating the torque generated by three healthy wind turbine blades of a rotor as a function of azimuth angle of each turbine blade.

Fig. 4 is a diagram showing two graphs; one depicting a measured power of a wind turbine rotor depending on the azimuth angle of a specific rotor blade as depicted in the second graph. Fig. 5 is a polar diagram illustrating the power produced by a wind turbine rotor depending on the rotor position.

Fig. 6 is a diagram illustrating the torque generated by three individual wind turbine blades of a rotor as a function of azimuth angle of each turbine blade, wherein one blade exhibits an anomaly. Fig. 7 is a polar diagram illustrating the torque generated by three individual wind turbine blades of a rotor and illustrating the total torque generated, as a function of azimuth angle of each turbine blade, wherein one blade exhibits an anomaly.

Fig. 8 is a plan view illustrating the principle of applying a grid of cells overlaying the swept area of the rotor plane. Fig. 9 is a polar diagram showing the power available at various azimuth angles of the rotor plane at a given wind speed at hub height.

Fig. 10 is a plan view illustrating the power available at various azimuth angles of the rotor plane at a given wind speed at hub height.

Fig. 11 is a plan view illustrating an example of the power available in the three different 120° sectors defining the rotor plane.

Fig. 12 a plan view illustrating the average wind speeds available at various azimuth angles of the rotor plane at a given wind speed at hub height of 9 m/s. Fig. 13 is a diagram illustrating the stacked power generated by three wind turbine blades of a rotor as a function of azimuth angle of each turbine blade, wherein one blade is exhibiting a loss of efficiency of 50%.

Detailed description of the invention

The first aspect of the present invention

According to the first aspect the present invention relates to a method for identification of the presence of an anomaly in a specific wind turbine blade of a rotor of a wind turbine comprising a number N of wind turbine blades, said number N being an integer greater than one; said method comprises: i) while operating the wind turbine, performing physical measurements of one or more physical parameters of the wind turbine, or parts thereof, during a predetermined period of time T or during a predetermined magnitude of rotation R of the rotor; said one or more measured physical parameters being related to a performance yield of the wind turbine; ii) logging the azimuth angle of one or more of said wind turbine blades while performing the measurements in step i); iii) on the basis of step ii) correlating the azimuth position of one or more of said wind turbine blades to the measured physical parameter(s) measured in step i); iv) based on the correlation performed in step iii) identifying the presence of an asymmetric and/or unequal performance yield of one or more wind turbine blades amongst the number N of wind turbine blades; v) on the basis of the identification performed in step iv) determining which wind turbine blade, if any, exhibiting an anomaly; or alternatively, on the basis of the identification performed in step iv) determining that no wind turbine blades are exhibiting an anomaly.

The method according to the first aspect is based on the principle of utilizing information available from analyzing wind shear and tower effect and the 3P effect of a wind turbine operating in a wind field for determining the presence of an anomaly of one or more wind turbine blades of a wind turbine.

In the present description and the appended claims the following definitions shall be adhered to. Wind shear relates to the phenomenon that the wind speed varies with height above ground level. As a result, a wind turbine blade is at the highest part of its cycle it will experience a greater wind speed than that of one at the lowest part of its cycle.

Tower shadow refers to the phenomenon that the wind slows down in front of the tower of a wind turbine. 3P effect refers to the torque oscillations due to wind shear and tower shadow during a 360 cycle of the rotor.

The theoretical basis for these considerations are provided by the equations is provided by the equations (1), (2) and (3) below:

(1) P = ½ p A v ³ - power P available in an area A of the rotor plane being

subjected to a wind of speed v in an air mass having the density p.

The effect of wind shear is given as:

(2) v(z) =V _H (z/H) ^a ; wherein v(z) is the wind speed at height z above ground; VH is the wind speed at a known height H; and wherein a is an empirical roughness parameter relating to the terrain at the location of the wind turbine.

The shadowing effect of the tower on the wind speed can be calculated from the following formula:

(3) Vtower = Vo a ² (y ² - x ² )/(x ² +y ² ) ² ; wherein Vo corresponds to the wind speed at a given height as determined by equation (2) above; a being the tower radius at that given height; y being the lateral distance from the blade to the tower midline; x being the distance from the blade midline to the tower midline. On a practical level the method of the first aspect of the present invention can be performed by measuring a performance yield, such as power at various azimuth angles of the blades and logging or monitoring the position of each blade during the measurements, Subsequently, based on the effect of the wind shear and the tower shadow and also based on the 3P effect it will be possible to determine whether each blade is taking up an equal amount of power and on this basis it is possible to assess whether one or more (and which one or more) is/are exhibiting an anomaly, if any.

In the present description and in the appended claims the term "anomaly of a wind turbine blade" shall be construed to mean a wind turbine blade exhibiting a reduced efficiency, or efficiency coefficient, compared to the other blades of the rotor of rotor blades.

In one embodiment of the method of the first aspect of the present invention, the identification of the presence of an anomaly in a specific wind turbine blade is based on a qualitative determination of an anomaly, or is based on a quantitative determination of an anomaly.

In the most simple embodiment, the method of the first aspect allows the determination of the presence of an anomaly in respect of one or more of the blades in terms of a qualitative assessment.

The physical parameter(s) being measured and which relate(s) to a performance yield of the wind turbine preferably relate(s) to a performance yield of the rotor of the wind turbine.

In a more advanced embodiment, the method of the first aspect allows the determination of the presence of an anomaly in respect of one or more of the blades in terms of a quantitative assessment. This embodiment allows defining a threshold of acceptable anomalies.

In one embodiment of the method of the first aspect of the present invention, the identification of the presence of an anomaly in a specific wind turbine blade is based on a quantitative determination of an anomaly; wherein said quantitative determination of an anomaly is expressed in terms of an efficiency coefficient of one or more of the N wind turbine blades; wherein said efficiency coefficient of a specific wind turbine blade of the N wind turbine blades represents the efficiency of said specific wind turbine blade to take up power from the wind.

An efficiency coefficient assigned to one or more of the N wind turbine blades may express the ability of a specific wind turbine blade to take up power from the blowing wind. In one embodiment such an efficiency coefficient may express the percentage of the available power taken up by a specific wind turbine blade. Such a parameter provides a very convenient expression of the efficiency of a specific wind turbine blade.

In one embodiment of the method of the first aspect of the present invention, the efficiency coefficient of a specific wind turbine blade is determined in respect of each of said N wind turbine blades.

In this mode of performing the method a quantitative assessment of an anomaly is easily determined by comparing the magnitude of efficiency coefficients associated with each of said N wind turbine blades. In one embodiment of the method of the first aspect of the present invention the

measurement(s) performed in step (i) and/or the logging performed in step (ii) is/are performed at a sampling rate of 1 ms - 5 second, such as 5 ms - 4 seconds, for example 10 ms - 3 seconds, e.g. 50 ms - 2 seconds, such as 100 ms - 1 second, such as 200 ms - 800 ms or 400 - 600 ms. In one embodiment of the method of the first aspect of the present invention, the method is a method for identification of a sudden appearing anomaly in said specific wind turbine blade; said method involved continuously or repeatedly performing the steps i) - v).

Being able to determine a sudden appearing anomaly in a specific wind turbine blade may provide enhanced safety in the operation of the wind turbine in that the wind turbine may be shut down immediately in such a situation, thereby avoiding debris falling at high speed in case a wind turbine blade is about to disintegrate.

In one embodiment of the method of the first aspect of the present invention, said method is continuously or repeatedly performed over a time span of one week or more, such as two weeks or more, such as over a time span of four weeks or more, e.g. over a time span of two months or more, such as over a time span of four months or more, for example over a time span of six months or more; such as over a time span of 9 months or more; or over a time span of one year or more.

In one embodiment of the method of the first aspect of the present invention, said method is performed repeatedly at time intervals of 2 hours or less, such as 1 hour or less, for example 30 min or less, such as 15 min or less, for example 5 min or less, such as 1 min or less, such as 30 seconds or less; or 15 seconds or less.

Continuously or repeatedly performing the method of the first aspect of the invention provides the possibility of immediately detecting any defects of a specific wind turbine blade. In one embodiment of the method of the first aspect of the present invention, said magnitude of rotation R during which the physical measurements in step i) are performed is 5° or more, such as 10° or more, for example 15° or more, such as 20° or more, for example 30° or more, such as 45° or more, e.g. 60° or more, such as 90° or more, for example 120° or more, such as 240° or more, for example 360° or more. These magnitudes of rotation R provides sufficient data acquisition on the one hand and yet may still provide that a too large wind speed variation detrimentally affects the result of the method on the other hand.

In one embodiment of the method of the first aspect of the present invention, said magnitude of rotation R during which the physical measurements in step i) are performed is a number of full rotations of the rotor, such as 1 full rotation of the rotor or more; or 2 full rotations of the rotor or more, for example 5 full rotations of the rotor or more, e.g. 10 full rotations of the rotor or more, such as 25 full rotations of the rotor or more, for example 50 full rotations of the rotor or more, such as 100 full rotations of the rotor or more, for example 500 full rotations of the rotor or more; or 1000 full rotations of the rotor or more. In one embodiment of the method of the first aspect of the present invention the number N of wind turbine blades of the wind turbine being 2, 3, 4, 5 or 6.

These numbers of blades of the rotor of a wind turbine are most common.

In one embodiment of the method of the first aspect of the present invention, said one or more measured physical parameters and/or said performance yield being a function of one or more of wind speed of said wind turbine blade, and azimuth angle of said wind turbine blade, and air temperature and air pressure.

In one embodiment of the method of the first aspect of the present invention, said method is being performed repeatedly during said predetermined magnitude of rotation R of the rotor in respect of a number of rotations of said rotor, and wherein the data obtained are pooled into two or more groups representing data measured under similar conditions; and wherein one or more of the steps iii) - v) are being performed in respect of averages of data associated with each group of the pool of groups.

In one embodiment of the method of the first aspect of the present invention, said similar conditions relate to similar wind speed and/or similar wind gradient above ground and/or similar wind direction and/or similar air pressure and/or similar air temperature.

Pooling together data corresponding to data measured under similar conditions into two or more groups representing data measured under similar conditions may allow eliminating errors due to variations in wind speeds during measurements.

In one embodiment of the method of the first aspect of the present invention, said method further comprising measuring wind speeds in the vicinity of the tower of the wind turbine corresponding to one or more heights thereof.

Knowing the variation in wind speeds during performing the measurements of the physical parameters may provide improved accuracy in the determination of an anomaly of one or more wind turbine blades. In one embodiment of the method of the first aspect of the present invention, said one or more measured physical parameters measured in step i) relate to torque of the rotor axle; or relate to measured effect as monitored by the control system of the wind turbine; or relate to current and/or voltage induced in the generator; or relate to twist angle of the main shaft of the rotor; or relate to measured surface strain of the main shaft of the rotor. These measured physical parameters provide a good starting point for assessment of the presence of an anomaly.

In one embodiment of the method of the first aspect of the present invention, said one or more measured physical parameters measured in step i) relate to power as calculated as the product of current and voltage as measured on the generator. Although such a combined measurement and calculation in an academic sense does not qualify as being a pure measurement, such combination of measurements and calculation shall nevertheless in the present description and in the appended claims be construed as being a "measurement". In one embodiment of the method of the first aspect of the present invention, said method relates to a case in which no anomaly is detected in respect of any of the wind turbine blades, wherein the performance yield is compared with performance yield determined previously under similar conditions in order to determine a slowly progression of a reduced performance yield of the wind turbine.

Being able to detect a slowly progression of a reduced performance yield of the wind turbine allows determination of certain types of defects of the wind turbine blades.

In one embodiment of the method of the first aspect of the present invention, said step iii) is performed by identifying a number M of segments of azimuth angles in respect of which said measured physical parameter of the wind turbine, as provided in step i), are either at a maximum, or at a minimum; and wherein step iv) is performed by comparing, in respect of each of said M segments of azimuth angles, said measured physical parameter of the wind turbine; and wherein step v) is performed on the basis of said comparison of said measured physical parameter of the wind turbine in respect of each of said M segments of azimuth angles.

This embodiment of the method of the first aspect is suitable for determining, on a qualitative basis, the presence of an anomaly of a wind turbine blade.

In one embodiment of this embodiment said number of segments M being equal to the number of wind turbine blades N; and/or wherein said segments of azimuth angles being of equal magnitude.

This embodiment will allow easy assessment, on a qualitative basis, the presence of an anomaly of a wind turbine blade.

In one embodiment of the method of the first aspect of the present invention, said method comprises: defining in respect of each blade of the rotor, N complementary segments Si, S ₂ , . . . SN of the rotor plane, wherein step i) is being performed by allowing each rotor blade to consecutively sweep one such segment Si, S¾... SN of the rotor plane; and wherein step iv) is being performed by: iva) defining, in respect of the area swept by the rotor, a grid comprising a number of cells; ivb) calculating, in respect of each cell of the grid defined in step iva), at the relevant wind speed, the power available for uptake by the rotor when sweeping said cell; ivc) calculating, in respect of each segment Si, S ₂ , ... SN associated with each blade, the power available for uptake by the blade sweeping each said segment as a sum of powers available in each cell of each said segment; ivd) on the basis of step iii), in respect of each specific wind turbine blade, calculating the contribution from each said segment Si, S¾ ... SN to the total physical parameter measured, during the sweeping of each segment by said specific rotor blade; ive) expressing a number N of equations, each equating the contribution from each said segment Si, S¾... SN obtained in step ivd) to a sum of terms, each term expressing a product of an efficiency coefficient of each blade times the power available in the specific segment Si, S¾ .. . SN, swept by said blade; ivf) solving said N equations in order to determine in respect of each blade, that specific blade's efficiency coefficient; wherein step v) is performed by comparing the N efficiency coefficients obtained in step ivf).

This embodiment of the method of the first aspect of the present invention, allow easy assessment, on a quantitative basis, of the presence of an anomaly of a wind turbine blade.

In step ivb) instead of calculating, in respect of each cell of the grid defined in step iva), the power available for uptake by the rotor when sweeping said cell, it may as an alternative be sufficient to calculate only the power available for uptake by the rotor in respect of those cell actually being about to be swept by a wind turbine blade.

In one embodiment of the method of the first aspect of the present invention, said the measurement is performed in a full 360° rotation of the rotor blade.

Performing the method by allowing the blades to sweep the rotor plane by a full 360° rotation provides improved accuracy in the determination of the presence of an anomaly and the magnitude thereof.

In one embodiment of the method of the first aspect of the present invention, each said segment Si, S¾... SN corresponds to a span of azimuth angles of 360 N. In one embodiment of the method of the first aspect of the present invention, each said segment Si, S ₂ , . . . SN corresponds to a span of azimuth angles of 1 - 180°, such as 2 - 135°, such as 3 - 120°, for example 4 - 90°, such as 5 - 60°, e.g. 10 - 50°, such as 15 - 45° or 20 - 30°.

These embodiments simplify the mathematical calculations to be performed in the method. In one embodiment of the method of the first aspect of the present invention, said step ivf) is being performed by solving N equations having N unknowns.

In one embodiment of the method of the first aspect of the present invention, said step ivd) is being performed as set out above in respect of the method stated as being appropriate for a qualitative assessment of an anomaly. In one embodiment of the method of the first aspect of the present invention, said grid defined in step iva) is having a cell size corresponding to 1 m ² or less, such as 0.5 m ² or less, for example 0.25 m ² or less, such as 0.1 m ² or less, such as 0.05 m ² or less, for example 0.01 m ² or less, such as 0.005 m ² or less, for example 0.001 m ² or less, such as 0.0005 m ² or less 0.0001 m ² or less. These cell areas provide a sufficient accurate basis for the calculations to be performed.

In one embodiment of the method of the first aspect of the present invention the method of applying an imaginary grid as disclosed above is generalized. In this generalized definition of the method it is a requirement that in respect of each blade of the rotor, N complementary segments Si, S¾ . . . SN of the rotor plane are defined and wherein step i) is being performed by allowing each rotor blade to consecutively sweep one such segment Si, S¾ . . . SN of the rotor plane; and wherein step iv) is being performed by: iva) calculating, in respect of each segment Si, S¾ . . . SN associated with each blade, the power available for uptake by the blade sweeping each said segment; ivb ) on the basis of step iii), in respect of each specific wind turbine blade, calculating the contribution from each said segment Si, S¾ . . . SN to the total physical parameter measured, during the sweeping of each segment by said specific rotor blade; ivc) expressing a number N of equations, each equating the contribution from each said segment Si, S¾ . . . SN obtained in step ivb) to a sum of terms, each term expressing a product of an efficiency coefficient of each blade times the power available in the specific segment S i, S ₂ , . . . SN, swept by said blade; ivd) solving said N equations in order to determine in respect of each blade, that specific blade' s efficiency coefficient; wherein step v) is performed by comparing the N efficiency coefficients obtained in step ivd).

This embodiment of the method according to the first aspect of the present invention is a more generalized form of the quantitative method outlined above. This method allow determination of the power availvable for uptake by the wind turbine blades by any calculation method.

In one embodiment of this method the measurement is performed in a full 360° rotation of the rotor blade.

In one embodiment of this method each said segment Si, S¾ . . . SN corresponds to a span of azimuth angles of 1 - 180°, such as 2 - 135°, such as 3 - 120°, for example 4 - 90°, such as 5 - 60°, e.g. 10 - 50°, such as 15 - 45° or 20 - 30°.

In one embodiment of this method the each segment Si, S¾ . . . SN corresponds to a span of azimuth angles of 360 N.

In one embodiment of this method step ivd) is being performed by solving N equations having N unknowns.

In one embodiment of this method the step ivb) is being performed by performing the following steps: performing step iii) by identifying a number M of segments of azimuth angles in respect of which said measured physical parameter of the wind turbine, as provided in step i), are either at a maximum, or at a minimum; and performing step iv) by comparing, in respect of each of said M segments of azimuth angles, said measured physical parameter of the wind turbine; and performing step v) on the basis of said comparison of said measured physical parameter of the wind turbine in respect of each of said M segments of azimuth angles.

In one embodiment of this method the power available for uptake by the blade sweeping each said segment as set out in step iva) is being performed by integration. Obtaing the power available for uptake by a wind turbine blade in a given segment by use of integration is a more generalized form of calculation, compared to the summation involving cells of the segment as set out above.

In one embodiment of this embodiment the power available for uptake by the blade sweeping each said segment is defined as:

r A area

wherein: Psx is denoting the total power available for uptake in a specific Segment X;

P(r,A)/area is the power density of the wind available at a point Q located at a radius r and at an azimuth angle A of Segment X.

In one embodiment of this method P(r,A) is calculated from the formula P(r,A) = ½ p A _r v ³ ; wherein v is wind speed of the air having the density p, at an area A _r at the position (r,A) in the specific segment.

In one embodiment of this method the wind speed, v(z) of the air at a given height z is defined as: v(z) =V _H (z/H) ^a ; wherein VH is the wind speed at a known height H; and wherein a is an empirical roughness parameter relating to the terrain at the location of the wind turbine; and wherein the wind speed, when taking into account the shadowing effect of the tower, is defined in the following formula: Vtower = Vo a ² (y ² - x ² )/(x ² +y ² ) ² ; wherein Vo corresponds the wind speed at a given height; a being the tower radius at that given height; y being the lateral distance from the blade to the tower midline; x being the distance from the blade midline to the tower midline.

From the two latter equations one may calculate the wind speed at a given height and at a given lateral distance from the vertical midline of the tower.

It should be noted, though, that it must be borne in mind that for obvious reasons there is no shadowing effect above the top of the tower.

Having formulated expressions for the wind speed at various positions in the area swept by the rotor of the wind turbine, one can find the corresponding expression for the power density and perform the integral set out in the equation above.

Although the wind speed equations above are defined in a Cartesian coordinate reference frame, a person skilled in the art will be able to translate such formulas into the polar coordinate system applied in respect of the integration equation above.

It should also be noted that when performing the integration of the power density throughout the area of a specific segment, one may chose not to integrate over all radii of that specific segment. This is because a wind turbine blade at small radii does not take up power from the wind.

Only at radii in respect of which the wind turbine blade is exhibiting a "wing profile", viz. at radii corresponding to positions at larger radii than the "root of the blade", power is taken up by the wind blowing at positions of such radii.

The extent of the root of a wind turbine blade varies from manufacturer to manufacturer and from model to model.

When performing the integration of the power density in a segment one may integrate from a radius of the segment corresponding to 1 - 10 %, such as 2 - 9 %, for example 3 - 8 %, such as 4 - 7 % or 5 - 6 % of the radial extension of the blade, and up to the total radial extension of the blade.

In one embodiment of the method of the first aspect of the present invention, said method further comprising transmission of an alert signal to an external unit, in case an anomaly is detected in respect of one or more of the wind turbine blades, and in case the anomaly is exceeding a predetermined value.

Proving an alert signal allows for improved safety in that the wind turbine may immediately be shut down and inspected and repaired in case an anomaly is detected. In one embodiment of the method of the first aspect of the present invention, said method further comprising shutting down the wind turbine in case an anomaly is detected in respect of one or more of the wind turbine blades and in case said anomaly is exceeding a predetermined value.

In one embodiment of the method of the first aspect of the present invention the shutting down of the wind turbine is being performed automatically.

Such automatically shutting down provides enhanced safety and a minimum of human labor hours when surveilling the performance of a wind turbine.

In one embodiment of the method of the first aspect of the present invention, said method further comprising inspecting said one or more wind turbine blades exhibiting an anomaly and optionally repairing or alleviating the defects of one or more of said one or more blades or replacing one or more of said one or more blades.

The second aspect of the present invention

According to the second aspect the present invention relates to a system for identification of the presence of an anomaly in a specific wind turbine blade of a rotor of a wind turbine comprising a number N of wind turbine blade, said number N being an integer greater than one; said system comprises:

-one or more probes for making physical measurements of one or more physical parameters of a wind turbine, or parts of the a wind turbine during operation thereof; said one or more physical parameters being related to a performance yield of the wind turbine;

- a control unit configured to receive data input associated with said one or more measured physical parameters; -wherein said control unit comprises a data storage, wherein said data storage comprises a computer program configured to perform the steps associated with the method according to the first aspect of the present invention.

In one embodiment the system of the second aspect of the present invention furthermore comprising an alert unit configured to transmit an alert signal to an external unit, in case an anomaly is detected in respect of one or more of the wind turbine blades, and in case the anomaly is exceeding a predetermined value.

In one embodiment the system of the second aspect of the present invention furthermore comprising a stop mechanism for automatically shutting down a wind turbine in case said control system detects the presence of an anomaly in one or more wind turbine blades of said wind turbine.

Such an alert unit and/or stop mechanism provide enhanced safety and a minimum of human labor hours when surveilling the performance of a wind turbine.

The one or more probes of the system for making physical measurements of one or more physical parameters of a wind turbine preferably relate(s) to probes for measurement of torque of the rotor axle; or to measurement of effect as monitored by the control system of the wind turbine; or relate to measurement of current and/or voltage induced in the generator; or relate to measurement of twist angle of the main shaft of the rotor; or relate to measurement of surface strain of the main shaft of the rotor; or relate to measurement of power as calculated as the product of current and voltage as measured on the generator.

The third aspect of the present invention

According to the third aspect the present invention relates to a computer program product, when running on a computing device, being adapted to perform the method according to the first aspect of the present invention.

The fourth aspect of the present invention According to the fourth aspect the present invention relates to a wind turbine comprising a system according to the second aspect of the present invention, or comprising a data storage comprising a computer program product according to the third aspect of the present invention.

The fifth aspect of the present invention

According to the fifth aspect the present invention relates to a use of a system according to the second aspect of the present invention, or of a computer program product according to the third aspect of the present invention, or of a wind turbine according to the fourth aspect of the present invention for identification the presence of an anomaly in a specific wind turbine blade of a rotor of a wind turbine; or for determining a reduced performance yield of the rotor itself.

In one embodiment of the use of the fifth aspect of the present said anomaly and/or said reduced performance yield is caused by one or more of the following causes: reduced structural integrity of one or more blades; dirt or surface defects, such as coating defects or presence of algae or insects on the surface of one or more blades; ice build-up on one or more blades.

In the present description and the appended claims the following definitions shall be adhered to.

Wind shear relates to the phenomenon that the wind speed varies with height above ground level. As a result, a wind turbine blade is at the highest part of its cycle it will experience a greater wind speed than that of one at the lowest part of its cycle.

Tower shadow refers to the phenomenon that the wind slows down in front of the tower of a wind turbine.

3P effect refers to the torque oscillations due to wind shear and tower shadow during a 360 cycle of the rotor.

Referring now to the drawings for illustrating the present invention, Fig. 1 illustrates the concept of azimuth angle of a wind turbine blade. Fig. 1 shows a wind turbine 100 comprising a wind turbine tower 2 and rotor 4 comprising wind turbine blades B l, B2 and B3. The three blades B l, B2 and B3 define a rotational plane 8. A center line 6 aligned with the direction of gravity extends through the center of the tower 2. The azimuth angle A in respect of a specific wind turbine blade, B 1 is defined as the angle between an upper portion of the center line 6 and a cord section of that specific wind turbine blade B l, when looking into the rotational plane in the wind direction. In fig. 1, the azimuth angle A of blade B 1 is approximately 20°. The azimuth angle, A of blade B2 is approximately 140°, and the azimuth angle, A of blade B3 is approximately 260°.

Fig. 2 illustrates the variation of wind speed at different locations in relation to the wind turbine tower. Fig. 2 shows a "cut" cylinder "coloured" with different shades. In fig. 2 one should imagine the location of rotor axis to be in the axial center of the cylinder with the rotor plane located at the top of the cylinder and being perpendicular to the axial axis of the cylinder.

The actual wind speed encountered at various positions in the vicinity of the tower is affected by two factors. One is the wind shear or wind gradient. The other effect affecting the actual wind speed is the shadowing effect of the wind turbine tower. Fig. 2 shows that the actual wind speed generally increases with height (wind shear). Fig. 2 also shows that in a vertical area in front of the wind turbine tower, the actual wind speed is reduced due to the shadowing effect of the wind turbine tower.

Fig. 3 illustrates the variation of torque generated by the three wind turbine blades B l, B2 and B3 of a rotor of a wind turbine as a function of azimuth angle of the specific blade. The Y-axis represents torque. The X-axis represents the azimuth angle of each blade in tens of degrees.

Referring to the curve representing the blade B l, it can be seen that the torque generated by blade B l peaks at an azimuth angle of 0°. It then slowly decreases until an azimuth angle of approximately 155° is being reached (corresponding to a position in close vicinity to the wind turbine tower), whereafter the decrease becomes more profound. The torque generated by blade B l has it lowest value at approximately 175 - 185° (corresponding to a downward pointing direction of the blade B l). After this lowest point, the torque generated by blade B l increases rapidly until an azimuth angle of approximately 205° is being reached, whereafter the torque increases more slowly. The torque patterns in respect of the three blades B l, B2 and B3 is approximate identical, only displaced by a 120° displacement. The wind turbine providing the data of fig. 3 seems to comprise three rotor blades being equally healthy, i.e. having equal efficiency coefficients.

This is in contrast to the wind turbine rotor providing the data of fig. 6.

In fig. 6, the torque patterns in respect of the three blades B l, B2 and B3 corresponds to the shadowing effect and the wind shear effect as explained in respect of fig. 2 above.

Fig 6 illustrates the variation of torque generated by the three wind turbine blades B l, B2 and B3 of a rotor of a wind turbine as a function of azimuth angle of the specific blade.

The Y-axis represents torque as measured in NM. The X-axis represents the azimuth angle in tens of degrees. Referring to the curve representing the blade B 1, it can be seen that the torque generated by blade B l peaks at an azimuth angle of 0°. It then decreases until an azimuth angle of approximately 180° (corresponding to a downward pointing direction of the blade B l). After this lowest point, the torque generated by blade B 1 increases The torque patterns in respect of the three blades B 1 and B2 are approximate identical. However, although the torque in respect of the turbine blade B3 follows the same pattern as in respect of B 1 and B2 (albeit lacking behind B l with a displacement of approximately 120°), the torque in respect of the turbine blade B3 also exhibit a lower magnitude, reduced approximately 10% in relation the torque generated by B 1 and B2. This reduced torque measured in respect of B3 may indicative of a structural anomaly in the blade B3. Fig. 7 is a polar diagram illustrating the theoretical calculated torque (NM) of a single rotation for each of the turbine blades of the rotor used for illustration in fig. 6.

Also illustrated in Fig. 7 is the sum of the calculated power output of a single rotation each of the turbine blades, thus representing the torque of the rotor as a function of azimuth angle.

Fig. 7 is to be compared with fig. 5, and is showing the measured power from the generator. Fig. 7 illustrates the contribution of each blade of the rotor to the combined torque on the main shaft.

Fig. 4 and Fig. 5 are disclosed thoroughly in relation to example 1 below. Fig. 8 illustrates a NEC Micon NM 72 used in the examples below. Fig. 8 illustrates the concept of overlay ed the rotor plane with an imaginary grid of cells. The cells each having a cell size of 1.0 x 1.0 m. The imaginary grid is used for calculating the power available in respect of an area being swept by a wind turbine blade, as explained in more detail in example 2 below.

Fig. 12 illustrates the average wind speed encountered by the wind turbine blades in segments corresponding to a span of azimuth angles of 10°.

In fig. 12, wind shear and tower shadow have been taken into account.

The method of performing these measurements is more thoroughly described in example 2 below.

Fig. 10 illustrates the power available for uptake by the wind turbine blades in segments corresponding to a span of azimuth angles of 10°.

These available powers have been calculated for each cell in each specific segment and by summarizing the power available in each cell associated with a specific segment. The information relating to wind speeds in fig. 12 forms the basis for the calculation of the available powers depicted in fig. 10.

The method of performing these measurements is more thoroughly described in example 2 below.

Fig 9 is a polar representation of the information depicted in fig. 10. Fig. 11 illustrates an example of the power available for uptake by the wind turbine blades in segments corresponding to a span of azimuth angles of 120°. The available powers obtained this way have been calculated for each cell in each specific 120° segment and by summarizing the power available in each cell associated with a specific segment.

Fig. 13 illustrates a simulated stacked power yield measurement of three blades during 36 10° segments (corresponding to a full 360° rotation of the rotor) in respect of a rotor exhibiting a 50 % loss of efficiency coefficient in respect of one of the blades. The lower curve corresponds to the power yield measurement of blade 1 as a function of azimuth angle of blade 1. The middle curve corresponds to the sum of the power yield measurement of blade 1 and blade 2 as a function of azimuth angle. The upper curve corresponds to the sum of the power yield measurement of blade 1 and blade 2 and blade 3 as a function of azimuth angle.

It can be seen that blade 2 shows a reduced power uptake from the wind available.

Examples

The following examples illustrate the methods according to the present invention.

Example 1

This example illustrates the method according to the first aspect of the present invention.

Experimental part

This example relate to measurements performed on a NEG Micon NM72 wind turbine located at Overgaard, Randers, Denmark. The wind turbine is wind turbine No. 18 of a cluster of 20 wind turbines in the Overgaard wind turbine park. The NM72 wind turbine has three blades, 72 m diameter rotor and Nominal Power 1,500 kW. The swept area of the rotors is 4072 m ² . The hub height is 63.5 m. Rotational speed is 17.3 min ^"1 .

The wind turbine is supplied with probes for measuring current and voltage; respectively induced at its generator.

Also, an iSpin anemometer was used for measuring the wind speed on the rotor as well as the position of each blade.

The iSpin device records data from three combined anemometers and accelerometers placed at the spinner, each anemometer measures correlated data about wind speed and position of the blade.

The wind turbine was allowed to run for a period of time of 30 days at naturally occurring wind speeds as measured by the anemometer. During the experiment the current and the voltage was logged by the iSpin data logger, recording 10 minute average data as well as high speed data with a sample rate of 10 Hz.

Additionally, during the experiment the azimuth angle of each of the wind turbine blades was also logged using the iSpin data logger where three accelerometers are placed on the hub between the blades. The position was recorded with at sample rate of 10 Hz.

Finally, during the experiment the wind speed as measured by the anemometer was also logged from the iSpin device with a sample rate of 10 Hz.

After 30 days the experimental part of the example was concluded.

Theoretical part

The immediate power produced by the wind turbine at a given time was calculated as the product of the voltage measured and the current measured.

In fig. 4 is illustrated the variation of the power of the wind turbine as calculated as the product of measured voltage and current induced in the generator, and as a function of azimuth angle of the rotor.

Fig. 4 shows two graphs. The upper graph represents the actual power output in kW calculated from the voltage and current measured on the Generator. The Y-axis is in kW. The X-axis is time.

The lower graph illustrates the rotor position shifting between a +180° position and a -180° position. The Y-axis is Angle measured in degrees, the X-axis is time. Hence, a total of 18 cycles of shifting between a +180° position and a -180° position in respect of a specific wind turbine blade is shown in fig. 4.

The upper curve corresponds to the effect of a slow variation of measured power associated with a slow variation in wind speed which leads to a relative long wave length curve upon which is added a "ripple" corresponding to a fast variation of measured power which can be attributed to a fast variation caused by different power uptake by the rotor depending on the configuration of the rotor in terms of azimuth angle of a specific wind turbine blade (the 3p effect). It is noted that there is a relatively large variation in the measured power output within each 360° cycle and that there seems to be a relation between the azimuth angle of the rotor and the generated power.

Fig. 5 illustrates a polar diagram showing, in respect of one 360° cycle of the rotor, the power as determined by multiplication of the induced voltage and the current as measured on the generator.

The measured power ranges from approximately 600 kW to approximately 1150 kW.

Because a wind turbine rotor provides maximum power in a configuration where the three blades is having an azimuth angle of 0°, 120°, 240° respectively, the three power peaks shown in the curve in Fig. 5 correspond to such situations where the blade is less affected by "wind shear". Likewise the three power lows shown in fig. 5 correspond to a configuration of the rotor, in which the three blades are having an azimuth angle of 60°, 180° and 300°, respectively ( where the blade is placed in front of the tower and thus is affected by the tower shadowing effect). It is noted that the three power peaks and the three power lows are slightly offset in relation to the exact theoretical azimuth angles of 0°, 60°, 120°, 180°, 240° and 300°, respectively. This offset may be attributable to slow data transmissions in the measuring and data processing devices.

The circumference of the curve in the polar diagram of fig 5 is having the shape of a trefoil leaf.

The power diagram of fig. 5 reveals that the maximum power corresponding to each of the three "foils" are approximately 1150 kW, HOOkW and 1000 kW, respectively. Further, the power diagram of fig. 5 reveals that the trefoil curve is not symmetrical. Such asymmetry is indicative of a qualitative determination of an anomaly of a wind turbine blade of a rotor of wind turbine blades.

These differences in maximum power measured and asymmetry can be attributed to rotor having three blades which are not equally "healthy", i.e. having different efficiency coefficients.

An unhealthy turbine blade not providing optimum power will reveal itself most profoundly when that blade is located in the segment correspond to an azimuth angle of 300° - 60°. Therefore, by comparing the curve in fig. 5 with the logged azimuth angle of the turbine blades, it is found that the Blade 1 is the blade which is in a top position (azimuth angle of 0°) at the point where the maximum power is 1150 kW. Blade 2 is the blade which is in a top position (azimuth angle of 0°) at the point where the maximum power is 1100 kW, and Blade 3 is the blade which is in a top position (azimuth angle of 0°) at the point where the maximum power is 1000 kW.

Accordingly, in this way blade 3 is diagnosed as being an unhealthy wind turbine blade, i.e. a blade having a reduced efficiency coefficient, which may need further inspection.

Based on asymmetry in measured 3p effects in each sector of the rotor of a wind turbine, a health factor or efficiency coefficient can be calculated for each of the three blades in the rotor and thus monitor the state of each blade in the rotor.

This calculation is illustrated below in example 2.

Example 2 This example illustrates the method according to the first aspect of the present invention for determining the presence of an anomaly in a wind turbine blade. This example is based on simulated measurements of power of the wind turbine of example 1 and having an anomaly in one of its blades.

The example simulates a situation of running the wind turbine at a wind speed of 9 m/s at hub height, in a terrain classified as "rough terrain with trees and buildings" and at an air density of 1.225 kg/m ³ .

In order to perform a more thorough analysis relating to the magnitude of an anomaly of one unhealthy wind turbine blade of a rotor of a wind turbine the following method was applied:

First, an imaginary square grid of square cells is overlayed the rotor plane in such a way that the center of the grid coincides with the center of the rotor plane.

The overlaying of such a grid onto the rotor plane of the NEC Micon NM 72 is illustrated in fig. 8. For sake of simplicity, the grid illustrated in fig. A is made up of cells having a cell size of 1.0 x 1.0 m.

However, for more accuracy in the calculations performed in this example, the actual cell size was only 0.1 x 0.1 m. Accordingly, in the present example a 72 x 72 m grid or matrix is imagined having a cell size of 0.1 m x 0.1 m. Hence, the grid or matrix comprises 518,400 cells (72 x 10 x 72 x 10).

The swept rotor area is divided into three sectors, viz. Sector 1 comprising the area having azimuth angles of [0;120°[; Sector 2 comprising the area having azimuth angles of [120;240°[; and Sector 3 comprising the area having azimuth angles of [240;360°[. The power available for uptake by a rotor of rotor blades in a rotor plane sweeping an area A and at wind speed v and air density p (with no compensation for shadow effect and shear) is given as:

(1) P = ½ p A v ³ .

Accordingly, the power available for uptake by a rotor in the NEC Micon NM 72 at the given conditions is:

P = ½ p A v ³ = ½ x 1.225 kg/m ³ x 4072 m ³ x (9 m/s) ³ = 1818198.9 W.

The effect of wind shear is given as:

(2) v(z) =V _H (z/H) ^a ; wherein v(z) is the wind speed at height z above ground; VH is the wind speed at a known height H; and wherein a is an empirical roughness parameter relating to the terrain at the location of the wind turbine. In this example, the roughness parameter, a is the parameter related to "rough terrain with trees and buildings". In the present example this value of a is 0.25.

The shadowing effect of the tower on the wind speed can be calculated from the following formula:

(3) Vtower = Vo a ² (y ² - x ² )/(x ² +y ² ) ² ; wherein Vo corresponds the wind speed at a given height as determined by equation (2); a being the tower radius at that given height; y being the lateral distance from the blade to the tower midline; x being the distance from the blade midline to the tower midline.

Each cell of 0.1 x 0.1 m in the matrix or grid of the square of 72 x 72 m was assigned to belong to a specific Sector (Sector 1, 2 and 3, respectively) according to the following rules: i) In case the center of a specific cell is outside any sector, that specific cell is

discarded.

ii) In case the center of the cell is within a given sector, that specific cell is assigned to belong to that sector. By making use of equation (1) and equation (2) and equation (3) above, the power available for take up by the rotor in each of the 0.1 x 0.1 m cell of the 72 x 72 m matrix or grid can be calculated.

In fig. 10 the power available in each sector of 10 degrees azimuth angles is depicted as a sum of power available in each 0.1 x 0.1 m cell belonging to each such sector of 10 degrees azimuth angles. Fig. 9 displays a polar representation of the data of fig. 10.

In Fig. 11 the power available in each sector of 120 degrees azimuth angles is depicted as a sum of power available in each 0.1 x 0.1 m cell belonging to each such 120 degrees sector.

Fig. 11 reveals that the power available in the sector of azimuth angles of 0 - 120° is of the same magnitude as the power available in the sector of azimuth angles of 240 - 360 °, which makes perfect sense due to the symmetry of the wind turbine.

Fig. 12 illustrates the average wind speeds encountered in each sector of 10 degrees azimuth angles. These values have been calculated on the basis of equation (2) and equation (3) above.

In table 1 below is provided detailed information relating to the data leading to fig. 11. Sector No/Angular range Count Sum (W) Average Standard Variance deviation

Sector 1 - [0; 120°[ 135,729 670,454.1 4.939652 0.588406 0.346222

Sector 2 - [120; 240°[ 135,730 463,764.5 3.416817 0.454449 0.206524

Sector 3 - [240; 360°[ 135,729 670,454.1 4.939652 0.588406 0.346222

Sum 1.804.673

Table 1

Column 1 in table 1 above represents the three different sectors of the rotor plane, viz. Sector 1 (0 - 120 °), Sector 2 (120 - 240 °) and Sector 3 (240 - 360 °). The second column in table 1 represents the number of 0.1 x 0.1 m cells which fall into the given sector.

The third column in table 1 represents the effect available in a given sector, expressed as a sum of effects available for each 0.1 x 0.1 cell within the given sector.

The fourth column in table 1 represents the average effect available in one 0.1 x 0.1 m cell in a given sector.

Also included in table 1 is the effect available in all three sectors. This effect amounts to 1.804.673 W.

So far we have been dealing with the power available for uptake and conversion into torque of a wind turbine blade of a wind turbine rotor. The power actually extracted from the rotor of the wind turbine is defined as

(4) P = ½ p A v ³ Cp;

Cp being a power coefficient which is specific in respect of a specific wind turbine. For a NEC Micon NM 72, the power coefficient, Cp is 0.4444 at 9 m/s and at an air density of 1.225 kg/m ³ . During one full rotation of the rotor, each blade will sweep each sector exactly once. It follows that:

When blade A is in Sector 1, blade B will be in Sector 2 and blade C will be in Sector 3. When blade A is in Sector 3, blade B will be in Sector 1 and blade C will be in Sector 2. When blade A is in Sector 2, blade B will be in Sector 3 and blade C will be in Sector 1.

Accordingly, the following equations apply in this example:

(i) CpB _A S l + CpB _B S2 + CpBc S3 = 703810 W

(ii) CpB _A S3 + CpB _B S 1 + CpBc S2 = 643698 W

(iii) CpB _A S2 + CpB _B S3 + CpBc S I = 643741 W

The effects stated on the right hand side of these equations are the simulated effects measured in the various sectors.

S I being the possible power available in Sector 1 S2 being the possible power available in Sector 2

S3 being the possible power available in Sector 3; and

CpBA SI being blade A's contribution to the power uptake in sector 1 CpB _B S2 being blade B's contribution to the power uptake in sector 2 CpBc S3 being blade C's contribution to the power uptake in sector 3 CpBA S3 being blade A's contribution to the power uptake in sector 3 CpB _B SI being blade B's contribution to the power uptake in sector 1 CpBc S2 being blade C's contribution to the power uptake in sector 2 CpBA S2 being blade A's contribution to the power uptake in sector 2 C BB S3 being blade B's contribution to the power uptake in sector 3 CpBc SI being blade C's contribution to the power uptake in sector 1.

Using Cramer's rule, and solving the equations (i), (ii) and (iii) above, the terms CpBA, CpBB and CpBc can be determined to be:

CpB _A = 0.4646 CpB _B = 0.1739 CpBc = 0.4648.

This result accordingly reveals that blade B having a CpBB of 0.1739 has a reduced blade power of approximately 60%, compared to the CP of blade A and blade C, respectively.

Accordingly it can be concluded that blade B is not contributing its originally full power and therefore exhibits an anomaly. This blade accordingly needs further inspection in order to alleviate the cause of this anomaly.

This example is an example of a quantitative assessment of an anomaly of a wind turbine blade.

This example simulates measurements in a real life situation. In a real life situation, the right hand sides of the equations (i), (ii) and (iii) above, i.e. the measured power taken up by each blade, would have obtained from a curve corresponding to that on fig. 4 or fig. 5 and measured in respect of the specific turbine blade. In that case, the respective power uptakes in respect of a 120° rotation of each blade would be equal to:

Measured power uptake by a specific blade = (the sum of power measured in each sample during a 120° rotation) x (the duration between samples)/(time needed to take a 120° rotation). In this equation, the numerator represents energy taken up by a given blade during a 120° rotation; whereas the denominator represents the total sampling time. Accordingly, the measured power uptake by a specific blade found this way will be an average power uptake by a specific blade during a 120° rotation of the rotor.

In making such measurements, one must keep track of which blade is at which 120° sector during the measurement. Additionally, in making such measurements, one must make sure that the measurements are performed during a wind speed which corresponds to the wind speed at which the available power in each 0.1 x 0.1 m cell of the matrix or grid was calculated.

Alternatively, one may choose to calculate the available power in each 0.1 x 0.1 m cell of the matrix or grid in respect of an array of different wind speeds and perform the power measurements mentioned above at an arbitrary wind speed, however carefully selecting, in the calculations, the array of available power in each 0.1 x 0.1 m cell of the matrix or grid corresponding to the wind speed encountered during the measurements.

- End of examples -

In example 2 above the correlation made in equations (i), (ii) and (iii) relates to power.

However, generally according to the present invention such correlations may be made for other performance parameters, such as e.g. energy produced.

It should be understood that all features and achievements discussed above and in the appended claims in relation to one aspect of the present invention and embodiments thereof apply equally well to the other aspects of the present invention and embodiments thereof.

List of reference numerals

2 Wind turbine tower

4 Rotor

6 Vertical center line of wind turbine tower

8 Rotor plane of rotor

100 Wind turbine

B 1 Wind turbine blade No. 1

B2 Wind turbine blade No. 2

B3 Wind turbine blade No. 3

A Azimuth angle of a specific wind turbine blade B 1