METHOD AND SYSTEM FOR PARAMETERIZATION OF A CONTROLLER FOR A WIND ENERGY INSTALLATION AND/OR OPERATION OF A WIND ENERGY INSTALLATION

Abstract

A method of parameterizing a controller of a first wind energy installation wherein the controller sets a manipulated variable of the wind energy installation as a function of an input variable. An artificial intelligence determines at least one value of a parameter of the controller for at least one state/degree of being iced up of the wind energy installation based on a power curve, load, and/or downstream flow of the wind energy installation predicted with a mathematical model of the wind energy installation for at least one state/degree of being iced up, and/or determines at least one value of a parameter of the controller for at least one state/degree of being iced up of the wind energy installation, based on at least one determined state/degree of being iced up and a power, load, and/or downstream flow of the wind energy installation and/or at least one second wind energy installation.

Claims

1-9. (canceled)

10. A method of parameterizing a controller of a first wind energy installation, wherein the controller is configured to set a manipulated variable of the first wind energy installation based on an input variable, the method comprising at least one of: determining with an artificial intelligence at least one value of a parameter of the controller for at least one state/degree of the first wind energy installation being iced up on the basis of at least one of a power curve, a load, or a downstream flow of the first wind energy installation predicted with a mathematical model of the wind energy installation for at least one state/degree of being iced up; or determining with the artificial intelligence at least one value of a parameter of the controller for at least one state/degree of the first wind energy installation being iced up, on the basis of at least one determined state/degree of the first wind energy installation being iced up, and at least one of a power, a load, or a downstream flow of at least one of the first wind energy installation or at least one second wind energy installation that is determined for the state/degree of being iced up.

11. The method of claim 10, wherein at least one of: the method is performed in at least one of a multi-stage or adaptive manner; determining the at least one value of a parameter with artificial intelligence comprises adapting or adjusting the at least one value with the with artificial intelligence; the mathematical model used to predict the power curve, load, or downstream flow is for the same state/degree of being iced up for which the at least one value of the parameter is being determined; the at least one determined state/degree of being iced up is the same state/degree of being iced up for which the at least one value of the parameter is being determined; the at least one second wind energy installation is a wind energy installation of the same type as the first wind energy installation; the at least one of a power, a load, or a downstream flow is determined by a measurement.

12. The method of claim 10, wherein the state/degree of being iced up is at least one of: determined for a time interval of at most 5 minutes; determined with the aid of at least one of: at least one wind measuring device, or at least one sensor; or determined on the basis of at least one of a determined power of the wind energy installation, at least one determined temperature, or at least one determined humidity.

13. The method of claim 12, wherein at least one of: the at least one wind measuring device comprises at least one wind energy installation-side wind measuring device; or the at least one sensor comprises at least one sensor arranged on a rotor blade.

14. The method of claim 10, wherein at least one of: the input variable is dependent upon at least one of a wind speed, a rotational speed, an electrical power, or a mechanical power of the wind energy installation; or at least one of a pitch angle, a heating of at least one blade of a rotor of the first wind energy installation, a wind tracking feature of the rotor, or a braking torque of a generator of the first wind energy installation is set based on the manipulated variable.

15. The method of claim 10, wherein at least one of: the parameter is selected from a set of possible parameters of the controller; at least one of an adjustable starting value or a permissible range of values of the parameter is specified to the artificial intelligence; or the artificial intelligence determines a sensitivity, with respect to various components of the parameter, of at least one of the power, the load or the downstream flow.

16. The method of claim 10, wherein the artificial intelligence determines the parameter value with the aid of machine learning.

17. The method of claim 16, wherein the machine learning comprises reinforcement learning.

18. A method of operating a wind energy installation, the method comprising: adjusting the manipulated variable of the wind energy installation with a controller based on the input variable in response to a state/degree of being iced up being determined; wherein the controller has been parameterized according to the method of claim 10 for the state/degree of the wind energy installation being iced up.

19. The method of claim 18, wherein the state/degree of being iced up is at least one of: determined for a time interval of at most 5 minutes; determined with the aid of at least one of: at least one wind measuring device, or at least one sensor; or determined on the basis of at least one of a determined power of the wind energy installation, at least one determined temperature, or at least one determined humidity.

20. The method of claim 19, wherein at least one of: the at least one wind measuring device comprises at least one wind energy installation-side wind measuring device; or the at least one sensor comprises at least one sensor arranged on a rotor blade.

21. A system for parameterizing a controller of a first wind energy installation, wherein the controller is configured to set a manipulated variable of the first wind energy installation as a function of an input variable, the system comprising at least one of: an artificial intelligence configured to determine at least one value of a parameter of the controller for at least one state/degree of the first wind energy installation being iced up on the basis of at least one of a power curve, a load, or a downstream flow of the first wind energy installation predicted with a mathematical model of the wind energy installation for at least one state/degree of being iced up; or an artificial intelligence configured to determine at least one value of a parameter of the controller for at least one state/degree of the first wind energy installation being iced up, on the basis of at least one determined state/degree of the first wind energy installation being iced up, and at least one of a power, a load, or a downstream flow of at least one of the first wind energy installation or at least one second wind energy installation that is determined for the state/degree of being iced up.

22. The system of claim 21, further comprising the controller that is being parameterized.

23. A computer program product for parameterizing a controller of a first wind energy installation, wherein the controller is configured to set a manipulated variable of the first wind energy installation as a function of an input variable, the computer program product comprising program code stored in a non-transitory, machine-readable storage medium, the program code configured to, when executed by a computer, cause the computer to: determine with an artificial intelligence at least one value of a parameter of the controller for at least one state/degree of the first wind energy installation being iced up on the basis of at least one of a power curve, a load, or a downstream flow of the first wind energy installation predicted with a mathematical model of the wind energy installation for at least one state/degree of being iced up; or determine with the artificial intelligence at least one value of a parameter of the controller for at least one state/degree of the first wind energy installation being iced up, on the basis of at least one determined state/degree of the first wind energy installation being iced up, and at least one of a power, a load, or a downstream flow of at least one of the first wind energy installation or at least one second wind energy installation that is determined for the state/degree of being iced up.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate exemplary embodiments of the invention and, together with a general description of the invention given above, and the detailed description given below, serve to explain the principles of the invention.

[0058] FIG. 1 shows a system for operating a first wind energy installation 10 and/or for parameterizing a controller of this wind energy installation in accordance with an embodiment of the present invention;

[0059] FIG. 2 shows a method of operating the first wind energy installation or of parameterizing the controller in accordance with an embodiment of the present invention; and

[0060] FIG. 3 shows characteristic curves of the controller for different states/degrees of being iced up, as determined by an artificial intelligence.

DETAILED DESCRIPTION

[0061] FIG. 1 shows a system for operating a first wind energy installation 10 and/or for parameterizing a controller 2 of this wind energy installation in accordance with an embodiment of the present invention, and FIG. 2 shows a corresponding method.

[0062] In a manner known per se, the first wind energy installation comprises a nacelle 11, which is mounted on a tower 12 so as to be rotatable and which comprises a rotor with adjustable blades 13, which rotor is coupled to a generator 14.

[0063] The controller 2 of the wind energy installation adjusts, on the basis of a measured generator power and/or a wind speed measured by means of two wind measuring devices 15, 15′ which are fixed with respect to the wind energy installation, a braking torque of the generator, a wind tracking feature of the nacelle about a yaw axis which, in FIG. 1, is vertical, and/or collectively or individually pitch angles β of the rotor blades, for example by means of a threshold control and/or a proportional control, an integral control and/or a differential control, in one embodiment a cascaded threshold control and/or a cascaded proportional control, a cascaded integral control and/or a cascaded differential control, or a different type of control.

[0064] For this purpose, according to one embodiment of the present invention, the controller is parameterized depending on the determined state/degree of being iced up of the wind energy installation, and/or specific to the state/degree of being iced up, for example, for different degrees of being iced up ICE.sub.0, ICE.sub.1 and ICE.sub.2, in each case, different (parameter) values for amplification coefficients, threshold values or the like are set in a manner which is specific to the degree of being iced up.

[0065] For this purpose, in a first step S10 (cf. FIG. 2), an artificial intelligence 30, in the example embodiment a computer with software for reinforced (machine) learning, is given initial values and/or permissible value ranges for the components of the multi-dimensional parameter by means of an interface 31, for example an input menu or the like.

[0066] In a second step S20, it is determined by means of the artificial intelligence and with the aid of a mathematical model 10 of the wind energy installation, which mathematical model 10, for parameter values of the co-modeled controller specified by the artificial intelligence, specified virtual wind speed values v and specified virtual states/degrees of being iced up, simulates or predicts a respective generated electrical power P of the modeled wind energy installation, how strong, within their permissible value range, the influence of the various components of the parameter, i. e. for example individual amplification coefficients or the like, is on the power. In step S20, the artificial intelligence then determines (in each case) a (multi-dimensional parameter) value that optimizes the power for the respective state/degree of being iced up. In this process, a load on the wind energy installation, in particular on its rotor blades 13, and/or an avoidance of a stall, can also be taken into account.

[0067] In a third step S30, the controller of the first wind energy installation 10 and controllers of further, second wind energy installations 50-52 of the same type are parameterized with the (parameter) values found in this way.

[0068] In a fourth step S40, the controllers of these wind energy installations 10, 50-52 are further re-parameterized during operation in a manner analogous to the step S20 described above, with the aid of the same or one or more further artificial intelligences, in the example embodiment with the aid of the computer or one or more further computers with software for reinforced (machine) learning.

[0069] In this context, for optimizing the controller of the first wind energy installation 10, the artificial intelligence uses data from the second wind energy installations 50-52 as a reference, so that a kind of swarm intelligence can be used in an advantageous manner.

[0070] During operation of the wind energy installations 10, 50-52, their respective current state/degree of being iced up is determined for a short time interval of about 0.5 to 2 minutes by means of a comparison of the wind speeds measured by the unheated cup anemometer 15 and those measured by the heated ultrasonic anemometer 15′.

[0071] On the basis of the state/degree of being iced up determined in this way, in step S40, the controller is then re-parameterized, specific to the state/degree of being iced up, with the (parameter) value (so far) determined for this purpose, or a respective (parameter) value determined for this state/degree of being iced up of the wind energy installation is set if this state/degree of being iced up is determined. Accordingly, at the same time, an in situ control is carried out with the (parameter) value specific to the state/degree of being iced up determined so far, and, at the same time, this is (further) optimized on the basis of the power determined in the process.

[0072] If, in a variant—for example by means of temperature sensors and/or load sensors in the individual rotor blades—their individual ice load is determined, components of the (parameter) value which determine the individual blade pitch angles of the rotor blades can be individually adapted to the respective state/degree of being iced up of the rotor blades and, purely by way of example, a more heavily iced-up blade can be pitched more strongly in partial load operation in order to take account of its deteriorated aerodynamics and, in particular, in order to avoid a stall.

[0073] FIG. 3 shows, for illustration purposes, characteristic curves of the controller 2 for different states/degrees of being iced up ICE.sub.0, ICE.sub.1 and ICE.sub.2, respectively, which characteristic curves are determined by the artificial intelligence by means of reinforced learning, whereby the degree of being iced up increases from ICE.sub.0 to ICE.sub.1 and from ICE.sub.1 to ICE.sub.2, respectively. The (parameter) values which are specific to the state/degree of being iced up and which are determined for these states/degrees of being iced up determine the shape and the position of the characteristic curves, by means of which the controller individually sets the blade angles β of the rotor blades as a function of the tip-speed ratio λ and thus as a function of the rotational speed and the wind speed.

[0074] Although example embodiments have been explained in the preceding description, it is to be noted that a variety of variations are possible.

[0075] It is also to be noted that the example embodiments are merely examples which are not intended to limit the scope of protection, the applications and the structure in any way. Rather, the preceding description provides the person skilled in the art with a guideline for the implementation of at least one example embodiment, whereby various modifications, in particular with regard to the function and the arrangement of the components described, can be made without departing from the scope of protection as it results from the claims and combinations of features equivalent to these.

[0076] While the present invention has been illustrated by a description of various embodiments, and while these embodiments have been described in considerable detail, it is not intended to restrict or in any way limit the scope of the appended claims to such de-tail. The various features shown and described herein may be used alone or in any combination. Additional advantages and modifications will readily appear to those skilled in the art. The invention in its broader aspects is therefore not limited to the specific details, representative apparatus and method, and illustrative example shown and described. Accordingly, departures may be made from such details without departing from the spirit and scope of the general inventive concept.

LIST OF REFERENCE SIGNS

[0077] 10 first wind energy installation [0078] 10′ model [0079] 11 nacelle [0080] 12 tower [0081] 13 rotor (blade) [0082] 14 generator [0083] 15; 15′ wind measuring device [0084] 2 controller [0085] 30 computer with software for reinforced learning (AI) [0086] 31 interface [0087] 50, 51, 52 second wind energy installation [0088] P electric power [0089] v averaged wind speed [0090] β blade pitch angle [0091] λ tip-speed ratio