Control of a wind turbine comprising multi-axial accelerometers

Abstract

The invention relates to control of a wind turbine comprising a plurality of multi-axial accelerometers mounted at different positions in the nacelle and/or in a top portion of the tower. The position and orientation of each accelerometer as mounted is obtained, accelerations in at least two different directions by each accelerometer are measured during operation of the wind turbine. From a number of predetermined mode shapes for the movement of the wind turbine is then determined an absolute position of at least one of the accelerometers during operation of the wind turbine based on the measured accelerations, the mount position and orientation of each accelerometer and the pre-determined mode shapes. Hereby a more precise absolute position during operation is obtained which can be used in the controlling of the turbine.

Claims

1. A method of controlling a wind turbine, the wind turbine comprising a tower supporting a nacelle and a rotor with a number of pitch-adjustable rotor blades, the wind turbine further comprising a control system for changing the pitch of the rotor blades and/or the generator torque, and a plurality of multi-axial accelerometers mounted at different positions in the nacelle and/or in a top portion of the tower, each accelerometer being mounted in a defined orientation, the method comprising: obtaining the position and orientation of each accelerometer as mounted; measuring accelerations in at least two different directions by each accelerometer during operation of the wind turbine; obtaining a number of pre-determined mode shapes for the movement of the wind turbine based at least on dimensions of the tower and a weight of the nacelle and the rotor as supported by the tower, the number of pre-determined mode shapes including at least one natural mode shape in a plane corresponding to fore-aft movements and at least one natural mode shape in a plane corresponding to side-side movements; determining an absolute position of at least one of the accelerometers during operation of the wind turbine based on the measured accelerations, the mount position and orientation of each accelerometer and the pre-determined mode shapes; determining a control parameter of the wind turbine as a function of the determined absolute position; and controlling the wind turbine according to the control parameter.

2. The method of controlling according to claim 1 further comprising determining a velocity of at least one of the accelerometers during operation of the wind turbine based on the measured accelerations, the position and orientation of each accelerometer as mounted, and the mode shapes.

3. The method of controlling according to claim 1, wherein the number of pre-determined mode shapes of the wind turbine is determined from the natural mode shapes of a beam fixed in one end and with a point mass at an opposite free end.

4. The method of controlling according to claim 1, wherein the number of pre-determined mode shapes comprises at least the two first natural mode shapes in a plane corresponding to fore-aft movements of the wind turbine.

5. The method of controlling according to claim 1, wherein the number of mode shapes comprises at least the two first natural mode shapes in a plane corresponding to side-side movements of the wind turbine.

6. The method of controlling according to claim 1, further comprising estimating a thrust force acting on the rotor blades by the wind based on the determined absolute position of the at least one of the accelerometers during operation of the wind turbine.

7. The method of controlling according to claim 5 further comprising estimating a wind speed based on the estimated thrust and parameters including a rotational speed of the rotor blades, a pitch angle of each of the rotor blades, and an air density.

8. The method of controlling according to claim 1, wherein the absolute position of the accelerometer(s) is determined by means of a Kalman filtering.

9. The method of controlling according to claim 1, wherein the multi-axial accelerometers are each mounted with two axes of measurement arranged in an essentially horizontal plane.

10. The method of controlling according to claim 1, wherein at least two of the plurality of the multi-axial accelerometers are mounted such that the axes of measurement of the multi-axial accelerometers are oriented in the same directions.

11. The method of controlling according to claim 1, wherein at least two of the plurality of the multi-axial accelerometers are mounted such that the axes of measurement of the multi-axial accelerometers are oriented differently.

12. The method of controlling according to claim 1, wherein a first accelerometer is mounted at a first position in a first vertical plane parallel to an axis of rotation of the rotor blades of the wind turbine, and a second accelerometer is mounted at a second position in a second vertical plane parallel to the axis of rotation, the first plane being different from the second plane.

13. The method of controlling according to claim 1, wherein a first accelerometer is mounted at a first position at a first plane perpendicular to an axis of rotation of the rotor blades of the wind turbine, and a second accelerometer is mounted at a second position in a second plane perpendicular to the axis of rotation, the first plane being different from the second plane.

14. The method of controlling according to claim 1, where the control parameter comprises a pitch parameter of one or more of the rotor blades and the controlling of the wind turbine comprises pitching one or more of the blades according to the pitch parameter.

15. The method of controlling according to claim 1, where the control parameter comprises a torque parameter and the controlling of the wind turbine comprises adjusting the torque of a wind turbine generator according to the torque parameter.

16. A control system for a wind turbine comprising a tower supporting a nacelle and a rotor with a number of pitch-adjustable rotor blades, the wind turbine further comprising a plurality of multi-axial accelerometers mounted at different positions in the nacelle and/or in a top portion of the tower, each accelerometer being mounted in a defined orientation, the control system being configured to perform the steps of: obtaining the position and orientation of each of a plurality of multi-axial accelerometers as mounted; receiving data of the acceleration in at least two different directions as measured by each of the accelerometers during operation of the wind turbine; obtaining a number of pre-determined mode shapes for the movement of the wind turbine based at least on dimensions of the tower and a weight of the nacelle and rotor as supported by the tower, the number of pre-determined mode shapes including at least one natural mode shape in a plane corresponding to fore-aft movements and at least one natural mode shape in a plane corresponding to side-side movements; determining an absolute position of at least one of the accelerometers during operation of the wind turbine based on the measured accelerations, the mount position and orientation of each accelerometer and the pre-determined mode shapes; determining a control parameter of the wind turbine as a function of the determined absolute position; and controlling the wind turbine according to the control parameter.

17. A wind turbine comprising a tower supporting a nacelle and a rotor with a number of pitch-adjustable rotor blades, the wind turbine further comprising a plurality of multi-axial accelerometers mounted at different positions in the nacelle and/or in a top portion of the tower, each accelerometer being mounted in a defined orientation, and a control system according to claim 16.

18. A computer program configured when executed for controlling a processor to perform the method according to claim 1.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) In the following different embodiments of the invention will be described with reference to the drawings, wherein:

(2) FIG. 1 shows a wind turbine equipped with a plurality of multi-axial accelerometers and during a fore-aft movement,

(3) FIG. 2 shows a wind turbine equipped with a plurality of multi-axial accelerometers and during a side-side movement,

(4) FIG. 3 is a sketch of a wind turbine modeled as a beam with a point mass at its free end,

(5) FIGS. 4 and 5 illustrate the first mode shapes in the fore-aft and side-side direction, respectively,

(6) FIG. 6 is a flow chart illustrating an embodiment of the invention,

(7) FIG. 7 illustrates a flow diagram showing relationship of parameters relating to the tower top position(s) and velocitie(s), and

(8) FIG. 8 shows an extended Kalman filter for estimated states.

DESCRIPTION OF EMBODIMENTS

(9) FIGS. 1 and 2 illustrate a wind turbine 100 equipped with a plurality of multi-axial accelerometers 101. Here two three-axial accelerometers 101 are mounted at different positions in the nacelle 102. The multi-axial accelerometers are indicated by their measuring axes and how these are oriented when the wind turbine 100 is in its un-deformed and upright position 104 and as oscillating 105 (the movements and deformation of the wind turbine are exaggerated for clarity). The accelerometers 101 measure components of both the actual accelerations from the tower top movements as well as from the gravity (illustrated with the arrow 106), decomposed in the three axes depending on the pose of the accelerometer. Pose of each accelerometer is a combination of the a priori known geometry of the installation in the nacelle/hub (i.e. the position and orientation of the accelerometer as mounted) and how the accelerometer is moved (translation, rotation, yawing) due to the tower top movement. This is illustrated for movements and oscillations in the fore-aft direction, with two accelerometers depicted as three-axis coordinate systems, in FIG. 1, and for movements and oscillations in the side-side direction (seen from the back of the nacelle) in FIG. 2. In FIG. 2, the nacelle 102 is simply indicated by a box. For better capturing the movement of the wind turbine, the accelerometers are mounted at different positions and preferably not directly above each other, preferably one behind the other, and preferably at different sideways positions (when seen as in FIG. 2 from the back of the nacelle). Hereby the components from the gravity in the measured acceleration signals are different on each accelerometer.

(10) FIG. 3 illustrates a sketch of a wind turbine. The tower 301 of the wind turbine 100 can be modeled as a beam which is fixed 302 in one end while a force or a torque is applied at the free end. On top of the beam (the free end) the nacelle 102 and the rotor can be modeled as a point mass. The movement of the tower sections can be described using the natural mode shapes of such a beam for example established by means of finite-element modeling and dynamic analyses.

(11) FIG. 4 illustrates the first three mode shapes 400 in the fore-aft direction where in particular the first mode 401 is dominating, driven by variations in the thrust force. In the figure is shown the modal displacement, 410, along the (normalised) tower height 420. The second mode, 402, reflects the mode with maximal displacement approximately halfway up the tower.

(12) FIG. 5 illustrates the movements of the tower in the side-side direction, by the first three mode shapes 500 for side-side movements, together with the couplings to the twist of the tower. Variations in the wind turbine generator torque primarily drive the 2.sup.nd side-side mode, 502.

(13) By assuming that the tower only vibrates according to a finite number of mode shapes (as described above and shown in FIGS. 4 and 5), equations are setup relating the tower top position(s) and velocitie(s) to the pose of each of the accelerometers and again relating this to the acceleration components as measured. Such equations can be described on the following form:
x.sub.k=f(x.sub.k1, u.sub.k, k)+w.sub.k1
y.sub.k=h(x.sub.k, u.sub.k, k)
{tilde over (y)}.sub.k=y.sub.k+v.sub.k

(14) Which is further illustrated in FIG. 7, where: k denotes a discrete point in time (with k1 being the immediate past time point). u.sub.k is a vector of inputs (here it can be thrust force and generator torque). x.sub.k is a vector of the actual states (e.g. the pose of a node in the tower model). y.sub.k is a vector of the actual process outputs (e.g. the actual acceleration components). {tilde over (y)}.sub.k is a vector of the measured process outputs (e.g. the measured acceleration components). w.sub.k and v.sub.k are process and output noise respectively. They are assumed to be zero mean Gaussian. f(.) and h(.) are generic non-linear functions relating the past state, current input, and current time to the next state and current output respectively.

(15) In an embodiment, the tower is described by its first fore-aft mode, q.sub.1, and its first side-side mode, q.sub.2. Each of the modes vibrates according to the equation of motion, i.e.,
{umlaut over (q)}.sub.1m.sub.1+{dot over (q)}.sub.1c.sub.1+q.sub.1k.sub.1=F.sub.1
{umlaut over (q)}.sub.2m.sub.2+{dot over (q)}.sub.2c.sub.2+q.sub.2k.sub.2=F.sub.2

(16) where m, k, c and F are the modal mass, modal stiffness, modal damping, and the modal force, respectively. In this case,

(17) $x=\left[\begin{array}{c}{q}_1\\ q_2\end{array}\right]$$u=\left[\begin{array}{c}{F}_1\\ F_2\end{array}\right]$

(18) f(.) is given by the two equations of motion above. In case of a single 3-axis accelerometer measuring Acc.sub.1, Acc.sub.2, and Acc.sub.3,

(19) $y=\left[\begin{array}{c}{\mathrm{Acc}}_1\\ {\mathrm{Acc}}_2\\ {\mathrm{Acc}}_3\end{array}\right]$

(20) and h(.) is the function that translates the two modal coordinates, q.sub.1and q.sub.2 to the three measured components of the acceleration. h(.) is given by the geometry, i.e., how the accelerometer is located relative to the node for which the equations of motion are derived.

(21) Given the inputs, measured outputs and assumptions on the model and on the process and output noise, the purpose of an Extended Kalman Filter (EKF) is to estimate unmeasured states and the actual process outputs. This is shown below where the estimated states are {circumflex over (x)}.sub.k and .sub.k are the estimated measured outputs, as schematically shown in FIG. 8.

(22) FIG. 6 is a flow chart illustrating an embodiment of the control method as described in the above and according to the invention. In a first step 601, the accelerations Acc.sub.1, Acc.sub.2, . . . , Acc.sub.n are measured by the multi-axial accelerometers. By means of Kalman filtering, 602, the unmeasured states {circumflex over (x)}.sub.k and the actual measured process outputs .sub.k are estimated. Hereby, the absolute position and velocity of at least one of the accelerometers is determined, 603, and parameters such as the tower top displacement, tower top velocity, thrust force acting on the rotor can be determined.

(23) The wind speed can then be estimated based on the determined thrust by:

(24) $F_T=\frac{1}{2}.Math..Math.A.Math.V^2.Math.C_T$$C_T=f\left(V,,\right)$$V=\sqrt{\frac{2.Math.F_T}{.Math.A.Math.f\left(V,,\right)}}$

(25) where:

(26) F.sub.T=Estimated or Measured Thrust

(27) C.sub.T=Thrust Coefficient

(28) P=Air Density

(29) A=Rotor Area

(30) V=Estimated Wind Speed

(31) =Pitch Angle

(32) =Rotor rotational speed

(33) While preferred embodiments of the invention have been described, it should be understood that the invention is not so limited and modifications may be made without departing from the invention. The scope of the invention is defined by the appended claims, and all devices that come within the meaning of the claims, either literally or by equivalence, are intended to be embraced therein.