A method and system for predicting the rotational speed of a small vertical-axis wind turbine.

Abstract

This invention belongs to the field of wind turbine technology and provides a method and system for predicting the rotational speed of a small vertical-axis wind turbine. The method includes: acquiring an operating video of the small wind turbine; performing cropping and resolution reduction preprocessing on the operating video; decomposing and numbering the preprocessed operating video to obtain numbered static images; extracting feature colors from the static images; calculating the feature color area of ​​each static image and plotting a curve corresponding to the feature color area according to the static image number; obtaining the predicted rotational speed value of the small wind turbine based on the number of frames between adjacent peaks in the curve. Based on feature color recognition, this method achieves the goal of predicting the rotational speed of the small wind turbine, solving the problem of requiring specialized equipment and personnel for rotational speed detection, and replacing the traditional method of measuring and calibrating rotational speed by installing sensors.

Description

Technical Field

[0001] This invention belongs to the field of wind turbine technology, and particularly relates to a method and system for predicting the rotational speed of a small vertical-axis wind turbine. Background Technology

[0002] Small wind turbine generators have the advantages of small footprint and flexible layout, making them very suitable for integrated smart energy projects and multi-energy complementary microgrids, virtual power plants, and other scenarios. They can serve as an effective supplement to large wind turbine generators and improve the utilization rate of wind resources.

[0003] Currently, the measurement of the rotational speed of small wind turbines mostly relies on various speed sensors installed inside the unit. The unit's rotational speed is obtained by analyzing and processing the signals such as sound, light, voltage, and current collected by the sensors. The rotational speed is then displayed in the central control system, and the unit's operating status is controlled based on the rotational speed.

[0004] Unlike centralized wind farms that use large wind turbines, small wind turbines are flexible in their layout and can be placed in factories, hospitals, schools, parks, and scenic areas as needed. These scenarios often lack professional wind turbine maintenance personnel, making it impossible to understand the turbine speed by checking the control system. Furthermore, when the control system malfunctions, there is a lack of professional equipment and personnel to promptly detect the turbine speed. At the same time, because small wind turbines are manufactured and installed with considerations for compact structure and aesthetic appearance, it is also difficult to measure and calibrate the speed by adding sensors. Summary of the Invention

[0005] To address the aforementioned problems, this invention proposes a method and system for predicting the rotational speed of a small vertical-axis wind turbine. Based on rotational videos of small wind turbines captured by mobile devices such as smartphones, tablets, and miniature cameras, this invention predicts the turbine's rotational speed using characteristic color recognition. This enriches the methods for measuring the rotational speed of small wind turbines, providing a convenient and quick way to calculate the turbine's rotational speed for users of small wind turbine sets who lack professional operation and maintenance knowledge. Furthermore, it provides a basis for remote technical services and support for small wind turbine sets.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solution:

[0007] In a first aspect, the present invention provides a method for predicting the rotational speed of a small vertical-axis wind turbine, comprising:

[0008] Obtain a video of a small wind turbine in operation;

[0009] The video of the operation is preprocessed and broken down into still images frame by frame;

[0010] Number the static images in chronological order;

[0011] Feature color extraction is performed on static images to convert them into binary images containing only black and white.

[0012] Calculate the area of ​​the characteristic color in each static image based on the binarized image;

[0013] Plot a curve showing the relationship between the area of ​​the characteristic color in a static image as the vertical axis and the time sequence number as the horizontal axis;

[0014] The predicted rotational speed of a small wind turbine is obtained based on the number of frames between adjacent peaks in the corresponding curve.

[0015] Furthermore, after acquiring the operating video of the small wind turbine, the key parameters of the operating video are determined; the key parameters include the encoding format, frame rate, and resolution.

[0016] Furthermore, the rotating part of the small wind turbine is used as the foreground, separating it from the environmental background in the operation video.

[0017] Furthermore, based on the frame rate of the running video, the video file is decomposed into static images frame by frame, and the static images are numbered sequentially according to time order.

[0018] Furthermore, after determining the feature color, the feature color is represented using the HSV model, and a range threshold is set; the HSV color value of each pixel in the static image is calculated; it is determined whether each pixel is within the set range of the feature color, and pixels within the set range are retained; pixels outside the set range are removed.

[0019] Furthermore, by using the area of ​​the characteristic color in the static image as the vertical axis and the video frame number after time numbering as the horizontal axis, the corresponding curve is obtained.

[0020] Furthermore, the rotational speed of a small wind turbine is calculated by converting the time interval between two adjacent peaks in the characteristic curve.

[0021] Secondly, the present invention also provides a vertical axis small wind turbine speed prediction system, comprising:

[0022] The data acquisition module is configured to acquire video footage of the operation of a small wind turbine.

[0023] The key parameter extraction module is configured to: determine the key parameters of the running video;

[0024] The preprocessing module is configured to perform cropping preprocessing and resolution reduction preprocessing on the running video;

[0025] The numbering module is configured to decompose and number the pre-processed operation video to obtain a numbered static image.

[0026] The conversion module is configured to: extract feature colors from the static image and convert the static image into a binary image containing only black and white;

[0027] The area calculation module is configured to: calculate the feature color area of ​​each static image and draw the corresponding curve of the feature color area according to the static image number;

[0028] The prediction module is configured to obtain the predicted rotational speed of a small wind turbine based on the number of frames between adjacent peaks in the curve.

[0029] Thirdly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the vertical axis small wind turbine speed prediction method described in the first aspect.

[0030] Fourthly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the vertical axis small wind turbine speed prediction method described in the first aspect.

[0031] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0032] This invention achieves the goal of predicting the rotational speed of small wind turbines by capturing videos of their operation, calculating the characteristic color area of ​​each static image, and plotting the corresponding curve of the characteristic color area according to the static image number. Based on the number of frames between adjacent peaks in the curve, the invention obtains the predicted rotational speed of the small wind turbine. This solves the problem of requiring specialized equipment and personnel for rotational speed detection and replaces the traditional method of measuring and calibrating rotational speed by adding sensors. Attached Figure Description

[0033] The accompanying drawings, which form part of this embodiment, are used to provide a further understanding of this embodiment. The illustrative embodiments and their descriptions are used to explain this embodiment and do not constitute an improper limitation of this embodiment.

[0034] Figure 1 This is a flowchart of Embodiment 1 of the present invention;

[0035] Figure 2 This is a graph showing the characteristic colors corresponding to the number of frames in Embodiment 1 of the present invention. Detailed Implementation

[0036] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0037] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of this application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0038] Small wind turbine generators are generally defined as wind turbine generators with a power output of 10 kilowatts or less.

[0039] Example 1:

[0040] This embodiment provides a method for predicting the rotational speed of a small vertical-axis wind turbine, including:

[0041] You can capture videos of small wind turbines in operation; these can be filmed directly using mobile devices such as smartphones and tablets.

[0042] Extract key parameters from the operation video;

[0043] The video is preprocessed and broken down frame by frame into still images;

[0044] Number the static images in chronological order;

[0045] Feature color extraction is performed on static images, and by retaining only the feature color portion, the static image is converted into a binary image containing only black and white.

[0046] Calculate the area of ​​the original feature color in each static image based on the binarized image;

[0047] Plot a curve showing the relationship between the area of ​​the characteristic color in the static image and the frame number of the static image on the x-axis.

[0048] The predicted rotational speed of a small wind turbine is obtained based on the number of frames between adjacent peaks in the curve.

[0049] Specifically, this method utilizes the operational video of a vertically mounted small wind turbine to predict its rotational speed by analyzing the changing patterns of characteristic color areas within the video footage. The characteristic colors to be identified can originate from the turbine blades themselves, which possess distinct color features, or from rotational speed identification markers added to the blades for ease of inspection. Specifically, during the operation of a vertically mounted small wind turbine, the turbine blades rotate periodically around the main axis. The angle between a specific position of the blade and the line connecting the main axis and the observation point also changes periodically. Therefore, in a video captured from a fixed position, the area of ​​the characteristic color on the blade surface changes with the angle between the blade and the line connecting the main axis and the observation point. This embodiment leverages this pattern through video image processing and calculation to predict the rotational speed of the small wind turbine, solving the problem of requiring specialized equipment and personnel for speed detection and replacing the traditional method of measuring and calibrating the speed using sensors.

[0050] The method for image processing and analysis of video footage of a small wind turbine in this embodiment, and for predicting its rotational speed, mainly includes the following steps:

[0051] S1. Obtain key parameters of the running video.

[0052] After obtaining the video of the small wind turbine in operation, the key information such as video encoding format, frame rate (FPS), and resolution is extracted first in order to perform subsequent analysis and processing of the video; among them, the frame rate (FPS) can be represented by the symbol f.

[0053] S2. Preprocessing of running video.

[0054] Based on the acquired operational video information, the video is preprocessed. Specifically, to improve calculation speed, the video size can be appropriately cropped and the video resolution reduced. For operational video files with complex environmental backgrounds, motion target locking can be used to separate the rotating part of the small wind turbine as the foreground from the environmental background in the operational video, and then the preprocessed video file is saved.

[0055] S3, Video frame-by-frame decomposition.

[0056] The video file is broken down into still images frame by frame according to the video frame rate, and the images are numbered sequentially according to time. For example, still image 1 represents the first frame of the video.

[0057] S4. Static Image Analysis.

[0058] S4.1 Extract feature colors;

[0059] Before analyzing static images, it is necessary to extract the characteristic colors of the target area of ​​the small wind turbine, as well as the characteristic colors or marker colors of the blade surface. The characteristic colors are represented using an HSV model, and their HSV values ​​can be quickly obtained using image processing software.

[0060] S4.2, Set the feature color threshold;

[0061] In the HSV color representation model, H represents hue, S represents saturation, and V represents value. Therefore, by simply changing the H value, a color that conforms to human perception habits can be obtained. To avoid deviations caused by changes in lighting and equipment acquisition quality during video recording, certain deviation thresholds ΔH, ΔS, and ΔV can be set during the feature color recognition process. For example, if the HSV values ​​of the feature colors are H0, S0, and V0, the upper limit for recognizing feature colors can be set to H0+ΔH, S0+ΔS, and V0+ΔV, and the lower limit for recognizing feature colors can be set to H0-ΔH, S0-ΔS, and V0-ΔV.

[0062] S4.3, Feature Color Recognition;

[0063] The HSV color value H of each pixel in a static image n S n and V n Perform calculations to determine whether the pixel is within the defined range of the feature color. If the following conditions are met simultaneously:

[0064] H0-ΔH≤H n ≤H0+Δ H

[0065] S0-ΔS≤S n ≤S0+ΔS

[0066] V0-ΔV≤V n ≤V0+ΔV

[0067] If a pixel is within the set range of the feature color, then the pixel is retained and replaced with white. This is represented as (0, 0, 1) using the HSV model and (255, 255, 255) using the RGB model.

[0068] If the pixel is not within the set range of the feature color, the pixel is not retained and its color is replaced with black, which is represented as (0, 0, 0) using the HSV model and (0, 0, 0) using the RGB model.

[0069] Through the above calculations, the original static image can be converted into a binary image containing only black and white, and the static image retains only the regions with characteristic colors from the original image, which are represented by white.

[0070] By calculating the number of white pixels in the binarized image, the area An of the characteristic color region in the original static image can be obtained.

[0071] S4.4. Draw the feature color area change curve.

[0072] By calculating the characteristic color area of ​​each still image, the sequence of characteristic color areas (A1, A2, ..., An) in each frame of the video over time can be obtained. For example... Figure 2 As shown, by using the area of ​​the characteristic color in the static image as the vertical axis and the video frame number as the horizontal axis (1, 2, ..., n), the curve corresponding to the video frame number and the area of ​​the characteristic color in that frame can be obtained.

[0073] S5. Characteristic curve analysis.

[0074] During the rotation of the wind turbine, the characteristic color area exhibits periodic changes in the video image. Therefore, by calculating the frame interval fn between two adjacent peaks of the curve, the time interval t1 between two consecutive peaks of the characteristic color area in the video can be calculated, with the time unit being seconds.

[0075] If two adjacent peaks in the curve are (i1, A) i1 (i2, A) and (i2, A) i2 ),

[0076] fn1 = i2 - i1,

[0077] t1=fn1*(1 / f)=fn1 / f

[0078] Where f is the video frame rate (FPS), which is the number of static frames per second in the video.

[0079] S6. Fan speed calculation.

[0080] The period at which the characteristic color of a small wind turbine appears with the largest area in the video frame is the reciprocal of the two adjacent peaks in the curve; the rotational speed of the small wind turbine can be calculated from the number of times the characteristic color area appears with the largest area in the video frame during one rotation of the small wind turbine. Specifically:

[0081] After obtaining the time interval t1 between two consecutive peak values ​​of the feature color area in the video, the corresponding rotational speed N1 of the wind turbine can be calculated. The unit of rotational speed can be revolutions per second.

[0082] For units where only one blade contains the characteristic color:

[0083] N1 = 1 / t1 = f / fn1

[0084] For a unit with k blades containing a characteristic color, k can be set to an integer greater than 1:

[0085] N1=1 / (t1*k)=(1 / t1) / k=(f / fn1) / k=(k*f) / fn1

[0086] The average rotational speed of the wind turbine during video recording can be obtained by taking the arithmetic mean of the rotational speeds corresponding to the peak values ​​of each pair of adjacent feature colors.

[0087] N ave =(∑ n i＝1 N i ) / n

[0088] The rotational speed prediction method in this embodiment enriches the prediction methods for the rotational speed of vertical-axis small wind turbine generators, improves the convenience of rotational speed calculation, and reduces the reliance of users on wind turbine expertise and testing equipment. Users can calculate the rotational speed of a small wind turbine simply by recording a video. When users lack the necessary analytical and calculation capabilities, they can send the video to maintenance personnel for analysis and calculation, providing a convenient and quick way to obtain the rotational speed calculation results. This saves time and personnel / equipment costs associated with the operation and inspection of small wind turbines, encouraging users to increase the frequency and number of operational status checks. It also provides a basis for remote technical support and services from maintenance personnel, facilitating the safe and stable operation of small wind turbine generators and the early detection of potential faults. Furthermore, the increased convenience of daily inspections promotes the application and promotion of small wind turbine generators, improving the utilization of renewable resources.

[0089] Example 2:

[0090] This embodiment provides a vertical axis small wind turbine speed prediction system, including:

[0091] The acquisition module is configured to acquire video footage of the operation of a small wind turbine.

[0092] The preprocessing module is configured to preprocess the running video and decompose it frame by frame into still images.

[0093] The numbering module is configured to number static images in chronological order.

[0094] The conversion module is configured to: extract feature colors from static images and convert the static images into binary images containing only black and white;

[0095] The area calculation module is configured to calculate the area of ​​the feature colors in each static image based on the binarized image;

[0096] The curve construction module is configured to plot a curve corresponding to the frame number and the area of ​​the feature color in the static image as the vertical axis and the time sequence number as the horizontal axis.

[0097] The prediction module is configured to obtain the predicted rotational speed of a small wind turbine based on the number of frames between adjacent peaks in the corresponding curve.

[0098] The working method of the system is the same as that of the vertical axis small wind turbine speed prediction method in Example 1, and will not be repeated here.

[0099] Example 3:

[0100] This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the vertical axis small wind turbine speed prediction method described in Embodiment 1.

[0101] Example 4:

[0102] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the steps of the vertical axis small wind turbine speed prediction method described in Embodiment 1.

[0103] The above description is merely a preferred embodiment of this practice and is not intended to limit the scope of this practice. Various modifications and variations can be made to this practice by those skilled in the art. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of this practice should be included within the protection scope of this practice.

Claims

1. A method for predicting the rotational speed of a small vertical-axis wind turbine, characterized in that, include: Obtain a video of a small wind turbine in operation; The video of the operation is preprocessed and broken down into still images frame by frame; Number the static images in chronological order; Feature color extraction is performed on static images to convert them into binary images containing only black and white. After determining the feature color, use the feature color. HSV The model is represented, and a range threshold is set; Calculate each pixel in the static image HSV Color value; determine whether each pixel is within the set range of the feature color, retain pixels within the set range; remove pixels outside the set range; calculate the feature color area in each static image based on the binarized image; The area of ​​the feature color region in the original static image is obtained by calculating the number of white pixels in the binarized image. Plot a curve showing the relationship between the area of ​​the characteristic color in a static image as the vertical axis and the time sequence number as the horizontal axis; Based on the number of frames between adjacent peaks in the corresponding curve, the predicted rotational speed of the small wind turbine is obtained. The period during which the characteristic color of a small wind turbine appears with the largest area in the video frame is the reciprocal of the two adjacent peaks in the curve; the rotational speed of the small wind turbine can be calculated based on the number of times the characteristic color area appears with the largest area in the video frame during one rotation of the small wind turbine.

2. The method for predicting the rotational speed of a small vertical-axis wind turbine as described in claim 1, characterized in that, After acquiring the operating video of the small wind turbine, the key parameters of the operating video are determined; the key parameters include the encoding format, frame rate, and resolution.

3. The method for predicting the rotational speed of a small vertical-axis wind turbine as described in claim 1, characterized in that, The rotating part of the small wind turbine is used as the foreground and separated from the environmental background in the operation video.

4. The method for predicting the rotational speed of a small vertical-axis wind turbine as described in claim 1, characterized in that, Based on the frame rate of the running video, the video file is decomposed into static images frame by frame, and the static images are numbered sequentially according to time.

5. The method for predicting the rotational speed of a small vertical-axis wind turbine as described in claim 1, characterized in that, Using the area of ​​the characteristic color in the static image as the ordinate and the video frame number after time numbering as the abscissa, the corresponding curve is obtained.

6. The method for predicting the rotational speed of a small vertical-axis wind turbine as described in claim 1, characterized in that, The rotational speed of a small wind turbine is calculated by converting the time interval between two adjacent peaks in the characteristic curve.

7. A vertical axis small wind turbine speed prediction system, characterized in that, include: The acquisition module is configured to acquire video footage of the operation of a small wind turbine. The preprocessing module is configured to preprocess the running video and decompose it frame by frame into still images. The numbering module is configured to number static images in chronological order. The conversion module is configured to: extract feature colors from static images and convert the static images into binary images containing only black and white; After determining the feature color, use the feature color. HSV The model is represented, and a range threshold is set; Calculate each pixel in the static image HSV Color value; determine whether each pixel is within the set range of the feature color, keep the pixels within the set range; remove the pixels outside the set range; The area calculation module is configured to calculate the area of ​​the feature colors in each static image based on the binarized image; The area of ​​the feature color region in the original static image is obtained by calculating the number of white pixels in the binarized image. The curve construction module is configured to plot a curve corresponding to the frame number and the area of ​​the feature color in the static image as the vertical axis and the time sequence number as the horizontal axis. The prediction module is configured to obtain the predicted rotational speed of a small wind turbine based on the number of frames between adjacent peaks in the corresponding curve. The period during which the characteristic color of a small wind turbine appears with the largest area in the video frame is the reciprocal of the two adjacent peaks in the curve; the rotational speed of the small wind turbine can be calculated based on the number of times the characteristic color area appears with the largest area in the video frame during one rotation of the small wind turbine.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the vertical axis small wind turbine speed prediction method as described in any one of claims 1-6.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the vertical axis small wind turbine speed prediction method as described in any one of claims 1-6.

Images