CN-121993362-A - Blade pair jointing compliance detection method, device and readable storage medium

CN121993362ACN 121993362 ACN121993362 ACN 121993362ACN-121993362-A

Abstract

The application provides a blade jointing regularity detection method, equipment and a readable storage medium, wherein a blade and a hub are jointed through a plurality of connecting pieces, an electronic device acquires an original image for presenting the jointing state of the blade and the hub, the original image is preprocessed to obtain an enhanced image, and a first mark of a first color and a second mark of a second color in the enhanced image are subjected to semantic segmentation to execute the blade jointing regularity detection. By adopting the scheme, the target tool and the precise assembly hole are not required to be installed, the color mark of the original process step is utilized, the blade butt joint compliance is detected based on the deep learning of monocular vision, the purpose of improving the operability is realized, and the digital level of management and the transparency of information can be improved. Moreover, the universality is not required to be calibrated, the original working flow is adapted, the additional process and the production cost are not generated, and the hardware cost is greatly reduced.

Inventors

WANG RUI
CHEN PENG
LIU YANG
HUANG DIE
Ni Weilun
PAN YALU
Muzhaparmaitinin

Assignees

远景能源有限公司

Dates

Publication Date: 20260508
Application Date: 20251217

Claims (10)

1.A method of blade pair compliance detection, comprising: Acquiring an original image for presenting a butt joint state of a blade and a hub, wherein the blade and the hub are in butt joint through a plurality of connecting pieces; Preprocessing the original image to obtain an enhanced image, wherein a first mark of a first color and a second mark of a second color are highlighted in the enhanced image, the first mark is used for indicating a first target position, the second mark is used for indicating a second target position, the first target position is a position marked by the first color in a plurality of first positions on the blade, the second target position is a position marked by the second color in a plurality of second positions on the hub, and the connecting piece is inserted between the first position and the corresponding second position so as to enable the blade and the hub to be in butt joint; Semantic segmentation is performed on the first identification of the first color and the second identification of the second color in the enhanced image to perform leaf-engagement compliance detection.
2. The method of claim 1, wherein semantically segmenting the first identification of the first color and the second identification of the second color in the enhanced image to perform leaf-pair compliance detection comprises: determining a first polygonal contour line surrounding the first mark and a second polygonal contour line surrounding the second mark according to the enhanced image; Determining the number of target pairs according to the first polygonal contour and the second polygonal contour, wherein one target pair comprises a pair of the first polygonal contour and the second polygonal contour; and when the number is equal to a preset number, performing blade-to-joint compliance detection according to the first polygonal contour and the second polygonal contour.
3. The method of claim 2, wherein performing a blade pair compliance detection from the first polygonal contour and the second polygonal contour when the number is equal to a preset number comprises: When the number is equal to a preset number, a first rectangle is determined according to the first polygonal contour line, a second rectangle is determined according to the second polygonal contour line, the first rectangle is the largest inscribed rectangle of the first polygonal contour line, and the second rectangle is the largest inscribed rectangle of the second polygonal contour line; and performing blade-to-joint compliance detection according to the first rectangle and the second rectangle.
4. A method according to claim 3, wherein said performing blade-to-mating compliance detection from said first rectangle and said second rectangle comprises: Determining the vertical distance between the central point of the first rectangle and a fitting line, wherein the fitting line is a middle branching line in the horizontal direction of the second rectangle, the second mark is in a strip shape, and the extending direction is in the horizontal direction; Determining a ratio of the vertical distance to a long side of the first rectangle, the long side being a longer side of the length and width of the first rectangle; and when the ratio is larger than a preset ratio, determining that the blade is not in fit.
5. A method according to claim 3, wherein said determining a first rectangle from said first polygonal contour comprises: Determining a mask matrix according to the first polygonal contour line and the second polygonal contour line, wherein the mask matrix is used for indicating mask values of pixels in the enhanced image, the mask value of the pixels in the enhanced image, which are positioned in the first polygonal contour line or the second polygonal contour line, is 1, and the mask value of the pixels, which are positioned outside the first polygonal contour line and the second polygonal contour line, is 0; rotating a first region surrounded by the first polygon contour line in the mask matrix by the rotation angle according to each rotation angle in the angle set to obtain a plurality of rotated first regions; The first rectangle is determined from the plurality of rotated first regions.
6. The method of claim 5, wherein the determining the first rectangle from the plurality of rotated first regions comprises: for each rotated first region, determining the position coordinates of each pixel in the first region; determining a rotated first region from the mask matrix according to the position coordinates of each pixel in the first region, and setting the mask value of each pixel in the rotated first region to be 1; Determining a plurality of histograms according to a mask matrix containing the rotated first region, wherein the histograms in the plurality of histograms are in one-to-one correspondence with rows in the mask matrix, and the histograms are used for indicating the number of pixels with mask values of 1 continuously appearing in the rows corresponding to the histograms; Determining the maximum inscribed rectangle corresponding to the current rotation angle according to the plurality of histograms to obtain the maximum inscribed rectangle corresponding to each first region; And determining the first rectangle according to the maximum inscribed rectangle corresponding to each first region.
7. The method according to any one of claim 2 to 6, wherein, And when the number is not equal to the preset number, displaying target images and outputting prompt information, wherein the target images are used for presenting correct images of the butt joint state of the blade and the hub, and the prompt information is used for prompting a user to shoot the original images again.
8. The method according to any one of claims 1 to 6, wherein after performing semantic segmentation on the first identifier of the first color and the second identifier of the second color in the enhanced image to perform blade-to-joint compliance detection, further comprises: sending a detection result obtained by detecting the bonding compliance of the execution blade to a server; And displaying an early warning signal when the detection result indicates that the blade joint is not in compliance.
9. An electronic device comprising a processor, a memory and a computer program stored on the memory and executable on the processor, wherein execution of the computer program by the processor causes the electronic device to implement the method of any one of claims 1 to 8.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the method according to any one of claims 1 to 8.

Description

Blade pair jointing compliance detection method, device and readable storage medium Technical Field The embodiment of the application relates to the technical field of intelligent operation and maintenance of industrial equipment, in particular to a method and equipment for detecting the butt joint compliance of blades and a readable storage medium. Background With the continuous development of society, the demand for energy is increasing, and wind energy is becoming a clean renewable energy source and is increasingly valued in all countries of the world. The wind generating set is a main device for generating electric energy by utilizing wind energy, such as a wind generating set, a fan and the like. The wind generating set comprises a wind wheel system, a transmission system and the like, wherein the wind wheel system is a core for capturing wind energy and comprises blades and a hub, and the hub is a key component of a main shaft for connecting the blades and the transmission system. Blade butt joint compliance detection is a key link when installing blades of a fan in hoisting butt joint. In the traditional scheme, a special sensor and a sucker device are arranged on the blade, signals are collected by the sensor, and the blade jointing compliance is detected according to the signals collected by the sensor. However, the blade pair compliance detection method is complex in operation steps and high in cost. Disclosure of Invention The embodiment of the application provides a blade pair jointing regularity detection method, a device and a readable storage medium, which are used for carrying out blade pair jointing regularity detection based on monocular vision deep learning and achieving the purpose of reducing the complexity and cost of blade pair jointing regularity detection. In a first aspect, an embodiment of the present application provides a method for detecting blade pair compliance, including: Acquiring an original image for presenting a butt joint state of a blade and a hub, wherein the blade and the hub are in butt joint through a plurality of connecting pieces; Preprocessing the original image to obtain an enhanced image, wherein a first mark of a first color and a second mark of a second color are highlighted in the enhanced image, the first mark is used for indicating a first target position, the second mark is used for indicating a second target position, the first target position is a position marked by the first color in a plurality of first positions on the blade, the second target position is a position marked by the second color in a plurality of second positions on the hub, and the connecting piece is inserted between the first position and the corresponding second position so as to enable the blade and the hub to be in butt joint; Semantic segmentation is performed on the first identification of the first color and the second identification of the second color in the enhanced image to perform leaf-engagement compliance detection. In a second aspect, an embodiment of the present application provides a blade pair compliance detection apparatus, including: The acquisition module is used for acquiring an original image for presenting the butt joint state of the blade and the hub, and the blade and the hub are in butt joint through a plurality of connecting pieces; A preprocessing module, configured to preprocess the original image to obtain an enhanced image, where a first identifier of a first color and a second identifier of a second color are highlighted in the enhanced image, where the first identifier is used to indicate a first target position, and the second identifier is used to indicate a second target position, where the first target position is a position identified by the first color in a plurality of first positions on the blade, and the second target position is a position identified by the second color in a plurality of second positions on the hub, and the first position and the corresponding second position are interposed between the connecting piece so that the blade and the hub are in butt joint; and the processing module is used for carrying out semantic segmentation on the first mark of the first color and the second mark of the second color in the enhanced image so as to execute blade-to-joint compliance detection. In a third aspect, an embodiment of the present application provides an electronic device, including a processor, a memory, and a computer program stored on the memory and executable on the processor, where the processor executes the computer program to cause the electronic device to implement the method as described above in the first aspect or the various possible implementations of the first aspect. In a fourth aspect, embodiments of the present application provide a computer readable storage medium having stored therein computer instructions which, when executed by a processor, are adapted to carry out the method according to the first aspect or the va