Impeller unbalance monitoring method based on blade three-point identification visual technology

Abstract

The invention discloses an impeller unbalance monitoring method based on a blade three-point identification visual technology. The impeller unbalance monitoring method comprises the steps that three identification points are marked on each blade; in the working process of the wind turbine generator, the cabin clearance monitoring camera is used for shooting images of the blades in real time; identifying the pixel coordinates of the identification points in the image through an image technology, and converting the pixel coordinates into world coordinates; fitting a blade track according to the world coordinates; and identifying the impeller imbalance condition according to the blade motion trail. According to the method, image data collected by a cabin clearance monitoring camera (located at the bottom of a cabin and close to a hub shooting blade) is utilized, pixel coordinates of marking points and internal and external parameters of the camera are recognized through an image technology, the positions of the marking points in a world coordinate system are converted, and the position of the marking points is determined according to the difference of tracks of the marking points of the three blades. And identifying the impeller imbalance condition of the wind generating set.

Description

Technical Field

[0001] The present invention relates to the technical field of wind turbine impeller imbalance detection, and in particular to an impeller imbalance monitoring method based on blade three-point identification visual technology. Background Art

[0002] In recent years, with the global emphasis on renewable energy and the continuous advancement of wind power technology, wind turbine capacity has increased, and rotor diameters have become larger. This has placed higher demands on wind turbine performance, reliability, and safety. Wind turbine rotor imbalance has gradually become a key operational indicator of concern. Wind turbine rotor imbalance seriously impacts the unit's operating efficiency and safety. Specifically, it can lead to increased vibration in the transmission system, which in turn increases the vibration amplitude of the tower and nacelle. This not only reduces equipment stability but also increases wear on blades and transmission components, shortening equipment life. In extreme cases, rotor imbalance can even cause blade breakage or damage to transmission components, leading to serious accidents.

[0003] With increasing attention paid to impeller imbalance, related monitoring methods have become increasingly sophisticated. Common methods include vibration monitoring, load monitoring, headroom monitoring, and power statistics monitoring. For example, patent publication number CN117989071A discloses a method for identifying impeller imbalance by monitoring the frequency anomaly of double the rotational speed in nacelle vibration. Patent publication number CN117968553A discloses a method for identifying impeller imbalance by using a camera inside the blades to identify blade deformation and calculate the root load difference. Patent publication number CN118293024A discloses a method for identifying impeller imbalance by calculating the difference in blade headroom values ​​using a camera installed at the bottom of the nacelle. Patent publication number CN116412082A discloses a method for identifying impeller imbalance by identifying the load difference during impeller rotation using generator axial displacement data. Patent publication number CN105221353B discloses a method for identifying impeller imbalance by using statistics of single-phase current, speed, and power of the rotor and stator of a doubly-fed generator set.

[0004] These methods all indirectly calculate the impeller imbalance value by using the blade status, which often results in limited accuracy and lag. Therefore, how to directly and effectively obtain the impeller imbalance value is a technical problem that needs to be solved urgently. Summary of the Invention

[0005] In order to address the shortcomings of the existing technology, the present invention provides an impeller imbalance monitoring method based on blade three-point identification visual technology. It uses image data collected by a cabin clearance monitoring camera (located at the bottom of the cabin close to the hub to shoot blades), identifies the pixel coordinates of the marked points and the camera's intrinsic and extrinsic parameters through image technology, converts the position of the marked points in the world coordinate system, and identifies the impeller imbalance of the wind turbine generator set based on the differences in the trajectories of the three-blade marked points.

[0006] The embodiment of the present invention provides the following solutions: An embodiment of the present invention provides an impeller imbalance monitoring method based on blade three-point identification visual technology, the method comprising: Step 1: Mark three identification points on each leaf; Step 2: When the wind turbine is operating, use the cabin clearance monitoring camera to capture images of the blades in real time; Step 3: Identify the pixel coordinates of the identified points in the image through image technology and convert the pixel coordinates into world coordinates; Step 4: Fit the blade trajectory according to the world coordinates; Step 5: Identify the impeller imbalance based on the blade motion trajectory.

[0007] In an optional embodiment, the cabin clearance monitoring camera described in step 2 is located at the bottom of the cabin near the wheel hub.

[0008] In an optional embodiment, step one uses blade navigation paint to mark identification points.

[0009] In an optional embodiment, among the three identification points, one is close to the blade tip and is the blade tip marking point; and two are close to the blade root and are the blade root marking points.

[0010] In an optional embodiment, the two blade root marking points adopt the same color, and the one blade tip marking point adopts a color different from that of the blade root marking point.

[0011] In an optional embodiment, step three converts the image pixel coordinates into image physical coordinates and camera coordinates in sequence, and finally converts them into world coordinates.

[0012] In an optional embodiment, step three includes the following process: S3.1. Identify the pixel positions of three marker points on the leaf; S3.2. Derivation of the impeller azimuth angle based on the pixel coordinates of the two blade root identification points; S3.3. Derive the three-dimensional coordinates of the two blade root identification points in the world coordinate system based on the impeller azimuth angle; S3.4. Derivation of the depth matrix Z based on the world coordinates and pixel coordinates of the two leaf root identification points and the camera internal and external parameters C ; S3.5, according to the depth matrix Z C The three-dimensional coordinates of the world coordinate system are derived from the pixel coordinates of the point close to the leaf tip.

[0013] In an optional embodiment, before step S3.1, it is first determined whether the image contains leaves, and if so, the process proceeds to step S3.1.

[0014] In an optional embodiment, step five includes the following process: S5.1. When the data of the three blades are complete, the discreteness of the motion trajectories of the three blades in the most recent cycle is traversed to calculate the imbalance degree; S5.2. Determine whether there is an upward trend based on the historical impeller imbalance, and output a prompt when it exceeds the alarm threshold or fault threshold.

[0015] In an optional embodiment, before step S5.1, it is first determined whether the image contains leaves. If not, the process proceeds to step S5.1.

[0016] The beneficial effects of the present invention based on its technical solution are: (1) The present invention provides a method for obtaining the impeller imbalance directly by detecting images, without indirectly calculating it through other indicators of the blades, and the numerical value is accurate; (2) The effectiveness of the calculation process of this method can be proved by mathematical deduction. By introducing the depth matrix and combining the impeller direction angle and the internal and external parameters of the camera, the coordinates of the identification point in the two-dimensional image can be used to calculate the three-dimensional world coordinates, and then the blade motion trajectory in the three-dimensional space can be fitted, and the impeller imbalance can be directly calculated. The calculation method is direct and effective. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 It is a schematic diagram of the process of the present invention.

[0019] Figure 2 Schematic diagram of the location of the marking points.

[0020] Figure 3 Schematic diagram of image data.

[0021] Figure 4 Schematic diagram of the coordinate system transformation principle.

[0022] Figure 5 Schematic diagram of the imaging principle of the marker point.

[0023] Among them: 1-tower, 2-blade, 3-blade tip identification point, 4-blade root identification point, 5-camera optical center, 6-imaging plane. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field fall within the scope of protection of the embodiments of the present invention.

[0025] Reference Figure 1 The embodiment of the present invention provides an impeller imbalance monitoring method based on blade three-point identification visual technology, the method comprising: Step 1: Mark three points on each leaf. Figure 2 Among the three marking points, one is close to the blade tip, which is the blade tip marking point 3; and two are close to the blade root, which are the blade root marking point 4.

[0026] The marking points can be painted with blade beacon paint. Blade beacon paint is a kind of marking and anti-corrosion paint specially used for large outdoor structures such as wind turbine blades. The main components are acrylic resin, silicone resin, pigment, filler, UV absorber, additives, etc., which are ground together. These ingredients together give the beacon paint excellent weather resistance, heat resistance, cold resistance, aging resistance, UV resistance and good anti-corrosion properties. Blade beacon paint is widely used for marking, decoration and anti-corrosion of wind turbine blades, outdoor chimney walls, steel structures, overhead pipelines and buildings. It can clearly mark the location and status of these structures, improving safety and recognition. At the same time, its excellent anti-corrosion performance can also extend the service life of these structures.

[0027] Choose a color that's easy to identify. The two markers at the base of the blade should be a different color than the marker near the tip. For example, if the tip of the blade is painted red, it's recommended to choose green for the two markers at the base and blue for the marker near the tip. The markers should be selected based on the size of the blade. A single marker that's 50cm x 50cm or larger is ideal.

[0028] Step 2: During the operation of the wind turbine, a cabin clearance monitoring camera located at the bottom of the cabin near the hub is used to capture images of the blades in real time.

[0029] Some parameters of the cabin clearance monitoring camera are shown in the following table:

[0030] Shooting effect Figure 3 As shown, the tower 1, the blade 2, and the blade root identification point 4 and the blade tip identification point 3 on the blade can be clearly seen in the picture.

[0031] Step 3: Use image technology to identify the pixel coordinates of the identified points in the image and convert the pixel coordinates into world coordinates. Figure 4 As shown in the figure, the principle is to convert the image pixel coordinates into image physical coordinates and camera coordinates in sequence, and finally into world coordinates. World coordinate system (3D): A reference coordinate system (X W , Y W , Z W ).

[0032] Camera coordinate system (3D): The bridge connecting the world coordinate system and the image coordinate system. Generally, the optical axis of the camera is taken as the z axis (X C , Y C , Z C ).

[0033] Image physical coordinate system (2D): Introduced based on the projection relationship to facilitate further deriving pixel coordinates. The unit is millimeters, and the coordinate origin is the intersection position (x, y) of the camera optical axis and the image physical coordinate system.

[0034] Image pixel coordinate system (2D): The information actually read from the camera, the discretization of the image physical coordinates, in pixels, with the coordinate origin as point (u, v).

[0035] The conversion of the world coordinate system to the pixel coordinate system point, that is, the projection of the three-dimensional point to the two-dimensional plane, is: (1) Where: s represents the resolution electronic magnification of the pixel coordinate system; f Represents the focal length of the camera; dX and dY are the length and width of a pixel, that is, the element size; u0 and v0 are the horizontal and vertical coordinates of the image center; fx=f / dX and fy=f / dY are called the normalized focal length on the x-axis and y-axis, respectively, that is, the number of pixels per unit length; T is the translation matrix, which translates the origin of the world coordinate system to the optical center of the camera; R is the rotation matrix, which rotates the direction of the world coordinate system after the translation matrix transformation to the camera coordinate system.

[0036] The above parameters can all be determined by camera parameters.

[0037] Since 3D has one more dimension than 2D, 2D image coordinates cannot be directly converted to 3D world coordinates. However, when converting from 3D to 2D, the calibration plate directly sets the distance from the calibration plate to the camera to 0. When converting from 2D image coordinates to 3D world coordinates, one of the dimensions can be set to a known quantity. That is, the depth matrix Zc can be derived and obtained. When Zc is known, the pixel to world coordinate conversion can be performed. It can be proved that: (2) Among them, k1 represents the camera external parameter , k2 represents the perspective projection matrix , k3 represents the transformation matrix from the image coordinate system to the pixel coordinate system After camera calibration, the camera’s intrinsic parameters, including k2 and k3, can be directly obtained. The camera’s extrinsic parameters, i.e., k1, can be obtained according to the camera’s installation position.

[0038] Reference Figure 5 , points A, B, and C are the 3D coordinate positions of the three annotation points of a single leaf in the camera coordinate system, P A 、P B 、P C is a point in the pixel coordinate system on the imaging plane 6. The propagation direction of the light passing through the optical center 5 of the camera will not change. Since the two marked points on the blade root are close to the blade root, the blade root has high rigidity and the elastic deformation is small. Therefore, points A and B basically move in a circle around the hub center. The camera position and the hub center position are known. The motion trajectories of points A and B in the picture are fixed and can be seen through the P A and P B The impeller rotation angle ω is derived from the pixel coordinate system position, and then the world coordinate system coordinates and camera coordinate system coordinates of points A and B are derived based on the impeller rotation angle ω. Point C is located in the middle of the blade, which is biased towards the tip (not at the tip where the blade is prone to deflection). Since the middle section of the blade has a higher strength along the blade direction, it is not easy to produce compression or tension deformation, nor is it easy to produce deflection deformation. Therefore, the ABC triangle only produces Figure 3 The angle rotation of ABC and ABC1 is the main reason why the pixel coordinates of point C in the image change. Therefore, the problem becomes: A 、P B 、P C , world coordinates A and B, and the process of deriving the world coordinate C is easy to find based on the principle of coordinate transformation.

[0039] Based on the above principles, this step specifically includes the following processes: S3.1. Identify the pixel positions of three marker points on the leaf; S3.2. Derivation of the impeller azimuth angle based on the pixel coordinates of the two blade root identification points; S3.3. Derive the three-dimensional coordinates of the two blade root identification points in the world coordinate system based on the impeller azimuth angle; S3.4. Derivation of the depth matrix Z based on the world coordinates and pixel coordinates of the two leaf root identification points and the camera internal and external parameters C ; S3.5, according to the depth matrix Z C The three-dimensional coordinates of the world coordinate system are derived from the pixel coordinates of the point close to the leaf tip.

[0040] Before step S3.1, it can be determined whether the image contains leaves. If so, the process proceeds to step S3.1.

[0041] Step 4: Fit the blade trajectory according to the world coordinate system. For each rotation of the blade, we can obtain the trajectory point set blade1, blade2, and blade3 of the world coordinate system marker point C when the three blades pass through the tower bottom. By fitting the three-dimensional curve, we can obtain the fitting curve when the three blades pass through the tower.

[0042] Step 5: Identify the impeller imbalance based on the blade motion trajectory, including the following process: S5.1. When the data of the three blades are complete, the discreteness of the motion trajectories of the three blades in the most recent cycle is traversed to calculate the imbalance degree; S5.2. Based on the historical impeller imbalance, determine whether there is an upward trend. When it exceeds the alarm threshold or fault threshold, perform corresponding high-precision and fault reporting and output the impeller imbalance diagnosis results.

[0043] Before step S5.1, it can be determined whether the image contains leaves. If not, then step S5.1 is entered.

[0044] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0045] The present invention is described with reference to flowcharts and / or block diagrams of methods, apparatus (modules, systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0046] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0047] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0048] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0049] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A method for monitoring impeller imbalance based on blade three-point identification visual technology, characterized in that: The method comprises: Step 1: Mark three identification points on each leaf; Step 2: When the wind turbine is operating, use the cabin clearance monitoring camera to capture images of the blades in real time; Step 3: Identify the pixel coordinates of the identified points in the image through image technology and convert the pixel coordinates into world coordinates; Step 4: Fit the blade trajectory according to the world coordinates; Step 5: Identify the impeller imbalance based on the blade motion trajectory.

2. The impeller imbalance monitoring method based on blade three-point identification visual technology according to claim 1 is characterized by: The cabin clearance monitoring camera described in step 2 is located at the bottom of the cabin near the wheel hub.

3. The impeller imbalance monitoring method based on blade three-point identification visual technology according to claim 1 is characterized by: Step 1: Use blade navigation paint to mark the identification points.

4. The impeller imbalance monitoring method based on blade three-point identification visual technology according to claim 1 is characterized in that: Among the three marking points, one is close to the blade tip and is the blade tip marking point; and two are close to the blade root and are the blade root marking points.

5. The impeller imbalance monitoring method based on blade three-point identification visual technology according to claim 4 is characterized in that: The two blade root marking points adopt one color, and the one blade tip marking point adopts a color different from that of the blade root marking point.

6. The impeller imbalance monitoring method based on blade three-point identification visual technology according to claim 1 is characterized in that: Step 3 converts the image pixel coordinates into image physical coordinates and camera coordinates in turn, and finally into world coordinates.

7. The impeller imbalance monitoring method based on blade three-point identification visual technology according to claim 5 or 6 is characterized in that: Step 3 includes the following process: S3.

1. Identify the pixel positions of three marker points on the leaf; S3.

2. Derivation of the impeller azimuth angle based on the pixel coordinates of the two blade root identification points; S3.

3. Derive the three-dimensional coordinates of the two blade root identification points in the world coordinate system based on the impeller azimuth angle; S3.

4. Derivation of the depth matrix Z based on the world coordinates and pixel coordinates of the two leaf root identification points and the internal and external parameters of the camera C ; S3.5, according to the depth matrix Z C The three-dimensional coordinates of the world coordinate system are derived from the pixel coordinates of the point close to the leaf tip.

8. The impeller imbalance monitoring method based on blade three-point identification visual technology according to claim 7 is characterized in that: Before step S3.1, it is first determined whether the image contains leaves. If so, the process proceeds to step S3.

1.

9. The impeller imbalance monitoring method based on blade three-point identification visual technology according to claim 1 is characterized in that: Step 5 includes the following process: S5.

1. When the data of the three blades are complete, the discreteness of the motion trajectories of the three blades in the most recent cycle is traversed to calculate the imbalance degree; S5.

2. Determine whether there is an upward trend based on the historical impeller imbalance, and output a prompt when it exceeds the alarm threshold or fault threshold.

10. The impeller imbalance monitoring method based on blade three-point identification visual technology according to claim 9 is characterized in that: Before step S5.1, it is first determined whether the image contains leaves. If not, the process proceeds to step S5.1.

Images