REPAIR SYSTEM AND METHOD FOR PERFORMING AT LEAST ONE REPAIR TASK IN A WIND TURBINE

Abstract

The repair system for performing at least one repair task in at least one wind turbine, in particular the nacelle, includes:

a controllable robotic repair device installed in the wind turbine, in particular the nacelle,

an operating device for remotely operating the robotic repair device, which comprises a control device and virtual reality equipment providing a virtual reality environment of the interior of the wind turbine, in particular the nacelle, for at least one human service user provided remote to the wind turbine,

wherein the control device of the operating device is adapted to remotely control the robotic repair device according to user input commands provided via the virtual reality equipment.

Claims

1. A repair system for performing at least one repair task in at least one wind turbine, the repair system comprising: a controllable robotic repair device installed in the wind turbine, and comprising at least one environment sensor, an operating device for remotely operating the robotic repair device, which comprises a control device and virtual reality equipment providing a virtual reality environment of the interior of the wind turbine, for at least one human service user provided remote to the wind turbine, wherein the control device of the operating device is adapted to remotely control the robotic repair device, according to user input commands provided via the virtual reality equipment, to determine, based on status information received from a wind turbine controller and/or sensor data of the at least one environment sensor, and/or receive repair task information describing a current repair task, and to, in the virtual reality environment, highlight objects of the interior, relevant for the repair task according to the repair task information and/or fade objects not relevant for the repair task according to the repair task information.

2. The repair system according to claim 1, wherein the virtual reality equipment comprises virtual reality glasses, wherein the control device is adapted to generate the virtual reality environment for the virtual reality glasses based on sensor data from the at least one sensor of the robotic repair device, and/or in that the virtual reality equipment further comprises virtual reality gloves, wherein the control device is adapted to derive at least one user input command from movement data of the virtual reality gloves.

3. The repair system according to claim 1, wherein the control device is adapted to include virtual help and/or status information objects into the virtual reality environment according to the repair task information and/or the status information.

4. The repair system according to claim 1, wherein the robotic repair device comprises at least one manipulator, comprising at least one tool and/or adapted to handle a tool provided in the interior of a nacelle.

5. The repair system according to claim 4, wherein, if the robotic repair device also comprises at least one environment sensor, the at least one environment sensor and the at least one manipulator are arranged in a human eye-to-arm relationship.

6. The repair system according to claim 4, wherein the robotic repair device further comprises a positioning assembly for positioning the manipulator.

7. The repair system according to claim 6, wherein the positioning assembly comprises a first guiding device or first guide rail, and a second guiding device or second guide rail, wherein the second guiding device is guided in the first guiding device in a first direction and the manipulator is guided in the second guiding device in a second direction, which is perpendicular to the first direction.

8. The repair system according to claim 7, wherein the first direction corresponds to a longitudinal direction of the nacelle and/or the direction of the rotational axis of the wind turbine rotating assembly, and/or the first guiding device is mounted to a ceiling of the nacelle.

9. The repair system according to claim 6, wherein the positioning range of the positioning assembly, in the first guiding device, spans at least partly a rotor bearing and/or a gearbox and/or a power generation assembly, or a generator, in the nacelle.

10. A method for performing at least one repair task in at least one wind turbine, using the repair system, according to claim 1, the repair system comprising a controllable robotic repair device installed in the wind turbine, an operating device for remotely operating the robotic repair device, which comprises a control device and virtual reality equipment providing a virtual reality environment of the interior of the wind turbine, for at least one human service user provided remote to the wind turbine, wherein the robotic repair device remotely controlled according to user input commands provided via the virtual reality equipment, wherein repair task information describing a current repair task is determined based on status information received from a wind turbine controller and/or sensor data of at least one environment sensor of the robotic repair device and/or received, and wherein, in the virtual reality environment, objects of the interior, relevant for the repair task according to the repair task information are highlighted and/or fade objects not relevant for the repair task according to the repair task information are faded.

11. The method according to claim 10, wherein protocol information describing each repair task is stored, in the control device, the protocol information comprising a repaired component of the wind turbine, used tools and/or processing steps to perform the repair according to user input commands.

12. The method according to claim 11, wherein stored protocol information for multiple repair tasks is used to train an artificial intelligence automatic control algorithm, wherein the trained automatic control algorithm is used for fully automatic performance of at least one future repair task by, remotely, controlling the robotic repair device.

13. The method according to claim 12, wherein the automatic control algorithm comprises a neural network and/or training is performed as, deep, reinforcement learning.

Description

BRIEF DESCRIPTION

[0039] Some of the embodiments will be described in detail, with reference to the following figures, wherein like designations denote like members, wherein:

[0040] FIG. 1 shows a principle drawing of a repair system according to the invention implemented for a wind turbine;

[0041] FIG. 2 shows a view of a robotic repair device installed in the nacelle of the wind turbine; and

[0042] FIG. 3 shows a flowchart describing an embodiment of a method according to the invention.

DETAILED DESCRIPTION

[0043] FIG. 1 shows a repair system 1 implemented for a wind turbine 2. It is, however, noted already at this point, that in exemplary embodiments, the repair system 1 is provided for multiple wind turbines 2. Additionally, the repair system may comprise multiple sets of virtual reality equipment 3 such that multiple repair tasks may be performed simultaneously by multiple human service users, in particular service technicians.

[0044] As can be seen, the wind turbine 2 comprises a tower 4 on which a nacelle 5 is rotatably mounted. A rotating assembly 6 comprising three wind turbine blades 7 and a rotor hub 8 is mounted on the nacelle 5. In the nacelle 5, a rotor bearing 9, a gear box 10, and a power generation assembly 11 are housed, as well as a controller 12 of the wind turbine 2. The power generation assembly 11, in this embodiment, comprises a generator 13 and at least one converter 14.

[0045] To be able to perform repair tasks in the nacelle 5 of the wind turbine 2, the repair system 1 comprises a robotic repair device 15 installed inside the nacelle 5, in this case comprising at least one manipulator 16 only indicated in FIG. 1 and a positioning assembly 17 for positioning the manipulator 16 inside the nacelle 5.

[0046] The positioning assembly 17 comprises a first guiding device 18, in this case a guard rail extending in a first direction, which is the longitudinal direction of the nacelle 5 and thus the direction of the rotating axis of the rotating assembly 6. The guard rail spans at least part of the rotor bearing 9, the gear box 10, the generator 13 and, at least on part, the at least one converter 14 such that a second guiding device 19 guided in the first guiding device 18 can be positioned above each of those components, since, in this case, the positioning assembly 17 is mounted to the ceiling 20 of the nacelle 5. The second guiding device 19 is also a guide rail extending in a second horizontal direction perpendicular to the first direction, thus perpendicular to the longitudinal direction of the nacelle 5. The manipulator 16 is coupled to the second guiding device 19 by a height adjustment device 21, in this case a telescope arm which is perpendicular to the first and second guiding devices. In this manner, the manipulator 16 can be positioned anywhere in the three-dimensional space of the nacelle 5 allowed by the positioning ranges of the components of the positioning assembly 17, that is, the first guiding device 18, the second guiding device 19 and the height adjustment device 21. Each of these components of the positioning assembly 17 comprises at least one controllable actuator, as does the manipulator 16, which may be or comprise a robotic arm or a grabbler. The robotic repair device 15 further comprises at least one environment sensor 22, in this case at least one camera, in particular a 3D-camera, mounted on the manipulator 16 or the height adjustment device 21. It is noted that the controller 12 of the wind turbine 2 might also be connected to further sensors and/or control units in the wind turbine 2, which are not shown for simplicity reasons, which provide status information or, for example comprising alarm information, regarding the wind turbine 2, in particular the components and sub-components housed in the nacelle 5.

[0047] The repair system 1 further comprises a control device 23. The control device 23 is adapted to receive sensor data of the at least one environment sensor 22 and to control the robotic repair device 15 by sending corresponding control signals using a wireless communication link 24. The control device 23 also provides a virtual reality environment of the interior of the nacelle 5 for the already mentioned at least one set of virtual reality equipment 3 and is adapted to derive user input commands from user data received from the virtual reality equipment 3. In the embodiment of FIG. 1, each set of virtual reality equipment 3 comprises virtual reality glasses 25 as well as virtual reality gloves 26. As in principle known from the state of the art, the virtual reality glasses 25 comprise at least one user monitoring sensor, wherein the control device 23 is adapted to adapt a representation of the virtual reality environment based on sensor data from the user monitoring sensor and to derive user input commands from sensor data of the user monitoring sensor. For example, interaction with objects of the virtual reality environment may be implemented by fixating objects with the gaze of the human service user and performing predefined blinking movements and/or eye movements by the user. Furthermore, of course, user input commands can also be derived from movement data of the virtual reality gloves 26.

[0048] The control device 23 may be configured to determine repair task information concerning a current repair task by itself, in particular by evaluating status information received from the controller 12 and/or sensor data received from the at least one environment sensor 22. Status information from the controller 12 may, of course, also comprise alarm information if a fault condition, in particular a malfunction, has been detected by the controller 12. Using such information, the control device 23 may derive a repair task to be performed and compiles respective repair task information. Alternatively or additionally, repair task information may also be received by the control device 23, for example from the controller 12 or a higher-level device.

[0049] If it was determined that the wind turbine 2 needs repair by performing a corresponding repair task, a human service user, in particular a service technician, is contacted and may use the virtual reality equipment 3 to act in the virtual reality environment 23 generated by the control device 23 and resembling the interior of the nacelle 5. The virtual reality environment depicts, using the virtual reality glasses 25, the current state of the interior of the nacelle 5, in particular based on the sensor data of the camera as environment sensor 22 and/or status information from the controller 12. However, the virtual reality environment is enriched with additional information or help by the control device 23 based on the repair task information. In particular, objects relevant for the repair task may be highlighted, for example by representing them in signal colors instead of the natural colors in the virtual reality environment or increasing their brightness. On the other hand, objects not relevant for the repair task may be faded out, for example represented blurred and/or in grey scale colors. Further, the control device 23 adds virtual help information objects and virtual status information objects, for example showing the temperature of certain components and/or sub-components and providing hints on performing the repair task, for example by arrow objects. In particular regarding status information objects, of course, also status information received from the controller 12 may be used directly.

[0050] While the service technician performs the repair task by generating corresponding user input commands resulting in control signals for the robotic repair device 15, protocol data describing the execution of the repair task is collected. It is noted that, for performing the repair task, tools 27 and/or spare parts 28 located in the nacelle 5, in particular as part of the repair system 1, may be used. Tools 27 may be integrated into the manipulator 16 or picked up and handled by the manipulator 16. Protocol information may, in particular, comprise the component or sub-component to be repaired, tools used and processing steps to perform the repair task.

[0051] FIG. 2 shows an exemplary concrete embodiment of the robotic repair device 15 as a schematical drawing. In this embodiment, both the first guiding device 18 as well as the second guiding device 19 of the positioning assembly 17 are realized as guide rails 29 in perpendicular first and second directions 30, 31. The height adjustment device 21 is mounted to a sled of the guide rail 29 of the second guiding device 19, such that the, in this case two, manipulators 16, which may be grabblers or robotic arms, may also be adjusted in a height direction 32. In this case, the environment sensor 22 implemented as a camera is positioned above the manipulators 16 in a human eye-to-arm relationship. This allows for intuitive input of user input commands by the service technician and simplifies conversion of user input commands to respective control signals.

[0052] It is noted that, in this embodiment, the positioning assembly 17 further comprises a gimbal assembly 33 having a corresponding actuator, such that the environment sensor 22 and/or the manipulator 16 may be brought into different orientations, in particular according to user monitoring data and/or movement data from the virtual reality equipment 3. Of course, additionally or alternatively, each manipulator 16 may also comprise at least one joint.

[0053] FIG. 3 finally shows a flowchart of an exemplary embodiment of a method according to embodiments of the invention. In a step S1, repair task information concerning at least one repair task in at least one wind turbine 2 is determined and/or received, as explained above. In a step S2, a human service user, in this case a service technician, performs the repair task based on the repair task information and uses the repair system 1, in particular the virtual reality equipment 3. The execution of this repair task, performed in step S2, is protocolled in a parallel step S3. Protocol information, comprising at least the component repaired, the tools used and the process steps performed/initiated by the service technician in step S2, is stored. All repair task information determined and/received in step S1 may be included into the protocol information data set of a repair task. Since, as indicated by arrow 34, multiple repair tasks are performed over time, a database 35 containing a large number of protocol information data sets is generated.

[0054] These protocol information data sets can, once sufficient data has been acquired, in particular for a certain class of repair tasks, be used, in a step S4, to train an artificial intelligence automatic control algorithm to completely automatically perform repair tasks, in particular of the class of repair tasks to which the protocol information used as training data is related. Once the automatic control algorithm has been sufficiently trained in step S4, such that it can robustly perform repair tasks, at least of the repair task class, it may be employed in the control device 23 to automatically perform at least some of the repair tasks.

[0055] Although the present invention has been disclosed in the form of embodiments and variations thereon, it will be understood that numerous additional modifications and variations could be made thereto without departing from the scope of the invention.

[0056] For the sake of clarity, it is to be understood that the use of “a” or “an” throughout this application does not exclude a plurality, and “comprising” does not exclude other steps or elements.