Three-blade deformation comparison impeller imbalance identification method based on tower bottom video

Abstract

The invention discloses a tower bottom video-based three-blade deformation comparison impeller imbalance identification method, which comprises the following steps of: deploying a camera at the tower bottom of a wind turbine generator, and shooting real-time working videos of three blades; impeller azimuth angle information is obtained in real time; in each period of rotation of the impeller, when the three blades respectively reach specific impeller azimuth angles, blade contour and key point extraction is performed on the video frame, and blade deformation differences of the three blades are compared, so that the unbalance degree of the impeller is judged, and an unbalance diagnosis result of the impeller is output. According to the three-blade deformation comparison impeller unbalance identification method based on the tower bottom video provided by the invention, a camera deployed at the tower bottom can be utilized to shoot the blades of the wind turbine generator, and impeller unbalance monitoring is economically, efficiently and accurately realized by identifying the position contour of the blades and extracting key points.

Description

Technical Field

[0001] The invention relates to the technical field of wind turbine state monitoring, and in particular to a method for identifying impeller imbalance by comparing three-blade deformations based on tower bottom video. Background Art

[0002] In the field of wind power generation, wind turbines are core equipment, and their stable operation is crucial to power generation efficiency and the sustainability of energy supply. As a key component of wind turbines to capture wind energy, impellers are prone to imbalance due to various factors during long-term high-speed rotation, such as blade wear, corrosion, icing, and manufacturing and installation errors. Impeller imbalance not only causes severe vibration and abnormal noise in the unit, but also accelerates the wear of bearings, gears and other components. In severe cases, it may even cause the unit to fail and shut down, greatly affecting power generation efficiency and increasing maintenance costs.

[0003] With the increasing attention paid to the problem of impeller imbalance, relevant monitoring methods have become increasingly mature, and the more common ones include vibration monitoring methods, load monitoring methods, power statistics monitoring methods, etc. These solutions often require special equipment, which is costly, difficult to deploy, and increases the cost of subsequent maintenance.

[0004] In order to reduce the monitoring cost, there is a plan to place multiple sets of cameras at a certain distance from the tower bottom, shoot the blades obliquely upward, and monitor the blade conditions in combination with the monitored clearance data (public patent CN113565697A). However, the camera deployment position of this solution needs to be far away from the unit, usually more than 1.5 times the distance from the blade to ensure processing accuracy. Since wind turbines are often located in mountainous areas in order to obtain higher wind speeds, it is difficult to obtain a good test orientation in the initial deployment. In addition, after the unit adjusts its yaw with the wind direction, the shooting equipment also needs to adjust its deployment position with the windward side, and there are also high requirements for the terrain, which makes this solution difficult to adapt.

[0005] Therefore, how to provide a cost-effective impeller imbalance monitoring method that can be deployed in any terrain and in any scenario is a technical problem that needs to be solved urgently. Summary of the invention

[0006] In order to address the deficiencies in the prior art, the present invention provides a method for identifying impeller imbalance by comparing three-blade deformations based on tower bottom video. A camera deployed at the bottom of the tower can be used to shoot the blades of a wind turbine. By identifying the blade position profile and extracting key points, impeller imbalance monitoring can be achieved economically, efficiently and accurately.

[0007] The embodiment of the present invention provides the following solution:

[0008] The embodiment of the present invention provides a method for identifying impeller imbalance based on three-blade deformation comparison of tower bottom video, the method comprising:

[0009] Step 1: Deploy a camera at the bottom of the wind turbine tower to capture videos of the real-time operation of the three blades.

[0010] Step 2: Obtain the information of the impeller azimuth angle in real time.

[0011] Step 3: In each rotation cycle of the impeller, when the three blades reach specific impeller azimuth angles respectively, extract the blade contours and key points from the video frames, compare the blade deformation differences of the three blades, so as to judge the degree of impeller imbalance and output the impeller imbalance diagnosis result.

[0012] In an optional embodiment, the camera described in Step 1, after position adjustment, includes the contours of the 3 blades in the captured image when the impeller azimuth angle is at 270°.

[0013] In an optional embodiment, the camera described in Step 1 communicates with a processing device, and the processing device communicates with the nacelle data acquisition unit of the wind turbine in real time.

[0014] In an optional embodiment, the processing device uses the SCADA (Supervisory Control And Data Acquisition) system to collect the impeller azimuth angle information in real time.

[0015] In an optional embodiment, Step 3 specifically includes the following processes:

[0016] S3.1: When the impeller azimuth angle reaches the first preset angle, use the image segmentation method to identify the contour of the first blade in the video frame, and use the key point recognition method to identify the deformation key points of the first blade according to the contour.

[0017] S3.2: When the impeller azimuth angle reaches the second preset angle, use the image segmentation method to identify the contour of the second blade in the video frame, and use the key point recognition method to identify the deformation key points of the second blade according to the contour; the second preset angle is the first preset angle + 120°.

[0018] S3.3: When the impeller azimuth angle reaches the third preset angle, use the image segmentation method to identify the contour of the third blade in the video frame, and use the key point recognition method to identify the deformation key points of the third blade according to the contour; the third preset angle is the second preset angle + 120°.

[0019] S3.4: Calculate the deformation differences between the three blades using the deformation key points. If the deformation differences between the three blades exceed the threshold, it is judged as a single-cycle anomaly, and a single-cycle anomaly diagnosis report is output.

[0020] S3.5. Repeat steps S3.1 to S3.4 for a preset number of n times, calculate the average value of the deformation differences of each blade for n times, and if it exceeds the threshold, it is determined that the impeller is actually unbalanced, and an impeller actual unbalance report is output.

[0021] In an alternative embodiment, the first preset angle is set to 270°.

[0022] In an alternative embodiment, n is not less than 20.

[0023] In an alternative embodiment, the blade deformation difference described in step three is calculated through the following process:

[0024] a1. Extract the upper contour line l of the blade from the video frame 1 and the lower contour line l 2 ;

[0025] a2. Divide the image frame into N equal parts along the x-axis. The intersection points of the upper contour line l 1 and the lower contour line l 2 with the i-th equal division line are respectively denoted as a i and b i ;

[0026] a3. Take the midpoint of a i and b i as the i-th key point s i , and the i-th key point of the j-th blade is denoted as s i,j ;

[0027] a4. The blade deformation difference dis of the three blades is calculated by the following formula:

[0028]

[0029] where dis() is the two-point distance formula, and ω i is the deformation weight of the i-th key point, which is determined and preset by the distance from this point to the blade root.

[0030] In an alternative embodiment, the dis() adopts the Euclidean distance formula:

[0031]

[0032] where x c and y c are the abscissa and ordinate of point c respectively, and x d and y d are the abscissa and ordinate of point d respectively. The beneficial effects of the present invention based on its technical solution are as follows:

[0033] (1) The impeller imbalance identification method based on tower bottom video and three-blade deformation comparison provided by the present invention only requires one camera to be deployed close to the bottom of the tower barrel, and can complete the monitoring of the impeller imbalance state. The processing and calculation process is efficient and accurate, not restricted by the terrain where the tower barrel is located, has a wide application scenario, and has low deployment and maintenance costs;

[0034] (2) The impeller imbalance identification method based on tower bottom video and three-blade deformation comparison provided by the present invention can use the impeller azimuth angle data collected by the built-in device of the wind turbine itself, and trigger the extraction of blade contours and key points immediately when the blade rotates to a special angle. This angle is the most prominent position of the blade contour, which is beneficial to improving the calculation accuracy, ensuring the impeller imbalance identification effect, and can greatly improve the operation safety of the wind turbine. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present specification or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present specification. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0036] Figure 1 It is a schematic flow chart of the impeller imbalance identification method based on tower bottom video and three-blade deformation comparison provided by the present invention.

[0037] Figure 2 It is a schematic diagram of camera deployment.

[0038] Figure 3 It is a schematic diagram of a video frame.

[0039] Figure 4 It is a schematic diagram of the extraction of the three-blade contours.

[0040] Figure 5 It is a schematic diagram of key point extraction.

[0041] Figure 6 It is a schematic diagram of imbalance degree calculation.

[0042] In the figure, 1 - tower barrel, 2 - camera, 3 - blade, 4 - portable laptop, 5 - SCADA system. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art belong to the scope protected by the embodiments of the present invention.

[0044] This embodiment provides a method for identifying impeller imbalance by comparing the deformation of three blades based on the bottom video of the tower. Refer to Figure 1 , the method includes:

[0045] Step 1: Refer to Figure 2 , deploy a camera 2 at the bottom of the tower barrel 1 of the wind turbine to capture the real-time working video of the three blades 3. The camera tooling can be a strong magnetic adsorption type. When on-site personnel need to test the impeller imbalance of the unit during stable operation of the fan (the wind direction is relatively stable, there is no yaw action within 10 minutes, and the rotational speed is greater than 5 rpm), deploy the camera tooling in the direction of the nacelle at the bottom of the tower. Debug the camera. When at the impeller azimuth angle of 270°, it is optimal to include the contours of the 3 blades in the captured image, as shown in Figure 3 .

[0046] Connect a portable laptop 4 through a network cable. The portable laptop communicates with the SCADA system 5 built into the wind turbine.

[0047] Step 2: Obtain the impeller azimuth angle information in real time through the SCADA system.

[0048] Step 3: In each cycle of impeller rotation, when the three blades reach specific impeller azimuth angles respectively, extract the blade contours and key points from the video frames, compare the blade deformation differences of the three blades, so as to judge the degree of impeller imbalance and output the impeller imbalance diagnosis result. Specifically, it includes the following processes:

[0049] S3.1: When the impeller azimuth angle reaches the first preset angle, use the image segmentation method to identify the contour of the first blade in the video frame, and use the key point recognition method to identify the deformation key points of the first blade according to the contour; in this embodiment, the first preset angle is set to 270°;

[0050] S3.2: When the impeller azimuth angle reaches the second preset angle, use the image segmentation method to identify the contour of the second blade in the video frame, and use the key point recognition method to identify the deformation key points of the second blade according to the contour; the second preset angle is the first preset angle + 120°;

[0051] S3.3: When the impeller azimuth angle reaches the third preset angle, use the image segmentation method to identify the contour of the third blade in the video frame, and use the key point recognition method to identify the deformation key points of the third blade according to the contour; the third preset angle is the second preset angle + 120°;

[0052] S3.4: Calculate the deformation differences between the three blades using the deformation key points. If the deformation differences between the three blades exceed the threshold, it is judged as a single-cycle anomaly, and a single-cycle anomaly diagnosis report is output;

[0053] S3.5. Repeat steps S3.1 to S3.4 for 20 times, calculate the average value of the deformation differences of each blade for n times. If it exceeds the threshold, it is determined that the impeller is actually unbalanced, and an impeller actual unbalance report is output.

[0054] The blade deformation difference described in Step 3 is calculated through the following process:

[0055] a1. Extract the upper contour line l of the blade from the video frame 1 and the lower contour line l 2 . The contour lines extracted from the first blade, the second blade, and the third blade can be distinguished by different colors during visualization, as shown in Figure 4 (a), (b), and (c) of

[0056] a2. Refer to Figure 5 , divide the image frame into N equal parts along the x-axis. The intersection points of the upper contour line l 1 and the lower contour line l 2 with the i-th equal division line are respectively denoted as a i and b i ;

[0057] a3. Take the midpoint of a i and b i as the i-th key point s i . The i-th key point of the j-th blade is denoted as s i,j ;

[0058] a4. The blade deformation difference dis of the three blades is calculated through the following formula:

[0059]

[0060] where dis() is the two-point distance formula, and ω i is the deformation weight of the i-th key point, which is determined and preset by the distance between this point and the leaf root. As shown in Figure 6 (a) and (b) of

[0061] are the overlapping effects of the contour lines and key points of the three blades respectively. It can be seen that it is normal for the tip of the leaf to have a greater deviation than the root of the leaf. Therefore, the deformation weight near the leaf root should be set larger than that far from the leaf root.

[0062]

[0063] where x c and y c are the horizontal and vertical coordinates of point c respectively, and x d and y d are the horizontal and vertical coordinates of point d respectively.

[0064] A method for identifying impeller imbalance by comparing the deformation of three blades based on the bottom tower video provided by the present invention can use a camera deployed at the bottom of the tower to photograph the blades of a wind turbine, and realize the monitoring of impeller imbalance economically, efficiently and accurately by identifying the blade position contour and extracting key points.

[0065] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0066] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (modules, systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a machine for realizing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 a device for the functions specified in one block or multiple blocks.

[0067] These computer program instructions can 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 generate a manufactured article including an instruction device, and the instruction device realizes the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 a device for the functions specified in one block or multiple blocks.

[0068] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in one Figure 1 one flow or multiple flows and / or blocks Figure 1 a device for the functions specified in one block or multiple blocks.

[0069] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made to these embodiments by those skilled in the art once they learn the basic inventive concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications that fall within the scope of the present invention.

[0070] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A method for identifying impeller imbalance based on three-blade deformation comparison of tower bottom video, characterized in that: The method comprises: Step 1: Deploy a camera at the bottom of the wind turbine tower to capture the real-time working video of the three blades; Step 2: Obtain the impeller azimuth information in real time; Step 3: In each cycle of impeller rotation, when the three blades reach specific impeller azimuth angles respectively, the blade contours and key points of the video frames are extracted, and the blade deformation differences of the three blades are compared to determine the degree of impeller imbalance and output the impeller imbalance diagnosis results.

2. The method for identifying impeller imbalance based on three-blade deformation comparison of tower bottom video according to claim 1 is characterized in that: The camera described in step 1 is adjusted in position so that the outlines of the three blades are included in the shooting picture when the impeller azimuth angle is 270°.

3. The method for identifying impeller imbalance based on three-blade deformation comparison of tower bottom video according to claim 1 is characterized in that: The camera described in step 1 communicates with the processing device, and the processing device communicates with the cabin data acquisition unit of the wind turbine generator set in real time.

4. The method for identifying impeller imbalance based on three-blade deformation comparison of tower bottom video according to claim 3 is characterized in that: The processing equipment uses the SCADA system to collect the impeller azimuth information in real time.

5. The method for identifying impeller imbalance based on three-blade deformation comparison of tower bottom video according to claim 1 is characterized in that: Step 3 Specific The process includes: S3.

1. When the impeller azimuth angle reaches a first preset angle, an image segmentation method is used to identify the contour of the first blade in the video frame, and a key point recognition method is used to identify the deformation key points of the first blade according to the contour; S3.2, when the impeller azimuth angle reaches a second preset angle, using an image segmentation method to identify the contour of the second blade in the video frame, and using a key point recognition method to identify the deformation key points of the second blade according to the contour; The second preset angle is the first preset angle + 120°; S3.3, when the impeller azimuth angle reaches a third preset angle, using an image segmentation method to identify the contour of the third blade in the video frame, and using a key point recognition method to identify the deformation key points of the third blade according to the contour; The third preset angle is the second preset angle + 120°; S3.

4. Calculate the deformation difference between the three blades using the deformation key points. If the deformation difference between the three blades exceeds a threshold, it is judged as a single-cycle anomaly and a single-cycle anomaly diagnosis report is output; S3.5, repeat steps S3.1 to S3.4 for a preset number of n times, calculate the average value of the deformation difference of each blade n times, if it exceeds the threshold, it is judged that the impeller is actually unbalanced, and output the impeller actual imbalance report.

6. The method for identifying impeller imbalance based on three-blade deformation comparison of tower bottom video according to claim 1 is characterized in that: The first preset angle is set to 270°.

7. The method for identifying impeller imbalance based on three-blade deformation comparison of tower bottom video according to claim 1 is characterized in that: n is not less than 20.

8. The method for identifying impeller imbalance based on three-blade deformation comparison of tower bottom video according to claim 1 is characterized in that: The blade deformation difference described in step 3 is calculated by the following process: a1, extract the upper contour line l1 and the lower contour line l2 of the leaf from the video frame; a2, divide the image into N equal parts along the x-axis, and record the intersection points of the upper contour line l1 and the lower contour line l2 with the i-th dividing line as a i and b i ; a3, take a i and b i The midpoint of is the i-th key point s i , the i-th key point of the j-th leaf is recorded as s i,j ; a4. The blade deformation difference dis of the three blades is calculated by the following formula: Among them, dis() is the distance formula between two points, ω i is the deformation weight of the i-th key point, which is determined and preset by the distance between the point and the leaf root.

9. The method for identifying impeller imbalance based on three-blade deformation comparison of tower bottom video according to claim 8 is characterized in that: The dis() described uses the Euclidean distance formula: Among them, x c and c are the horizontal and vertical coordinates of point c, x d and d are the horizontal and vertical coordinates of point d respectively.

Images