Methods, apparatus, equipment and readable storage media for determining the damage level of blade cracks

Abstract

This invention provides a method, apparatus, device, and readable storage medium for determining the damage level of blade cracks. The method includes: acquiring the centroid coordinates, width, and image of the blade; performing crack detection on the blade image to obtain the total number of cracks and the minimum bounding box coordinates of each crack; cropping all crack images from the blade image based on the minimum bounding box coordinates of each crack, and obtaining crack feature information for each crack based on each crack image; obtaining a blade crack feature index value based on the centroid coordinates, width, total number of cracks, minimum bounding box coordinates of each crack, and crack feature information of each crack; and determining the blade crack damage level based on the total number of cracks and the blade crack feature index value. This invention, with equipment as the primary driver, automatically determines the blade crack damage level, reducing manual intervention and improving the efficiency of blade flaw detection testing.

Description

Technical Field

[0001] This invention relates to the field of detection technology, and in particular to a method, apparatus, equipment, and readable storage medium for determining the level of blade crack damage. Background Technology

[0002] Turbocharger turbine blades are constantly operating under high temperature, high speed, and overload conditions, making them prone to cracking. When these cracks penetrate, they can easily cause the blades to break. Therefore, to ensure the stable operation of the turbocharger, it is necessary to perform flaw detection tests on the turbine blades to determine the extent of damage.

[0003] Currently, when conducting flaw detection tests on turbine blades, ultrasonic testing of the rotor must be performed manually, which is not only labor-intensive but also has a long testing cycle. Summary of the Invention

[0004] The main objective of this invention is to provide a method, apparatus, device, and readable storage medium for determining the level of blade crack damage, aiming to solve the technical problem of time-consuming and labor-intensive manual flaw detection testing of turbine blades in the prior art.

[0005] In a first aspect, the present invention provides a method for determining the level of blade crack damage, the method comprising:

[0006] Obtain the centroid coordinates of the blade, the blade width, and the blade image;

[0007] Crack detection is performed on the blade image to obtain the total number of cracks and the coordinates of the minimum bounding rectangle of each crack;

[0008] All crack images are cropped from the blade image based on the minimum bounding rectangle coordinates of each crack, and crack feature information of each crack is obtained based on each crack image.

[0009] The blade crack characteristic index value is obtained based on the blade centroid coordinates, blade width, total number of cracks, minimum bounding rectangle coordinates of each crack, and crack characteristic information of each crack.

[0010] The blade crack damage level is determined based on the total number of cracks and the characteristic index value of the blade crack.

[0011] Optionally, the steps of obtaining the blade centroid coordinates, blade width, and blade image include:

[0012] Leaf detection is performed on the original image to obtain the leaf boundary coordinates and leaf centroid coordinates. The original image is an image captured of the leaf.

[0013] The coordinates of the minimum bounding rectangle of the blade and the blade width are obtained from the blade boundary coordinates.

[0014] The leaf image is cropped from the original image based on the coordinates of the minimum bounding rectangle of the leaf.

[0015] Optionally, the step of providing the crack feature information includes:

[0016] Maximum crack width, crack fracture surface type, and crack orientation.

[0017] Optionally, the step of obtaining the blade crack feature index value based on the blade centroid coordinates, blade width, total number of cracks, minimum bounding rectangle coordinates of each crack, and crack feature information of each crack includes:

[0018] Based on the maximum crack width of each crack and the blade width, the characteristic index value of the blade crack width is obtained.

[0019] Based on the crack fracture surface type of each crack, the characteristic index value of the blade crack fracture surface is obtained.

[0020] Based on the crack direction of each crack, the characteristic index value of the blade crack direction is obtained;

[0021] The centroid coordinates of each crack are determined based on the coordinates of the minimum bounding rectangle of each crack.

[0022] Based on the centroid coordinates of each crack, the characteristic index value of the blade crack location is obtained;

[0023] Based on the centroid coordinates of the blade and the centroid coordinates of each crack, the discrete characteristic index value of the blade crack is obtained.

[0024] Optionally, the step of obtaining the blade crack width characteristic index value based on the maximum crack width of each crack and the blade width includes:

[0025] The maximum widths of all cracks are summed to obtain the sum value;

[0026] Calculate the ratio of the sum to the blade width, and use the ratio as the characteristic index value of the blade crack width.

[0027] Optionally, the step of obtaining the characteristic index value of the blade crack fracture surface based on the crack fracture surface type of each crack includes:

[0028] Based on the fracture surface type of each crack, obtain the fracture surface type label value corresponding to each crack;

[0029] The fracture surface type label values ​​of all cracks are summed, and the sum is used as the characteristic index value of the blade crack fracture surface.

[0030] Optionally, the step of determining the blade crack damage level based on the total number of cracks and the blade crack characteristic index value includes:

[0031] The blade crack damage coefficient is obtained by weighting the total number of cracks, the characteristic index value of blade crack width, the characteristic index value of blade crack fracture surface, the characteristic index value of blade crack direction, the characteristic index value of blade crack location, and the characteristic index value of blade crack discreteness.

[0032] The blade crack damage level is determined based on the blade crack damage coefficient.

[0033] Secondly, the present invention also provides a blade crack damage level determination device, the blade crack damage level determination device comprising:

[0034] The acquisition module is used to acquire the centroid coordinates of the blade, the blade width, and the blade image.

[0035] The detection module is used to detect cracks in the blade image to obtain the total number of cracks and the coordinates of the minimum bounding rectangle of each crack.

[0036] The feature extraction module is used to crop all crack images from the blade image based on the coordinates of the minimum bounding rectangle of each crack, and to obtain the crack feature information of each crack based on each crack image.

[0037] The calculation module is used to obtain the blade crack feature index value based on the blade centroid coordinates, blade width, total number of cracks, minimum bounding rectangle coordinates of each crack, and crack feature information of each crack.

[0038] The rating module is used to determine the blade crack damage level based on the total number of cracks and the blade crack characteristic index value.

[0039] Thirdly, the present invention also provides a blade crack damage level determination device, the blade crack damage level determination device including a processor, a memory, and a blade crack damage level determination program stored in the memory and executable by the processor, wherein when the blade crack damage level determination program is executed by the processor, the steps of the blade crack damage level determination method as described above are implemented.

[0040] Fourthly, the present invention also provides a readable storage medium storing a blade crack damage level determination program, wherein when the blade crack damage level determination program is executed by a processor, the steps of the blade crack damage level determination method as described above are implemented.

[0041] In this invention, the centroid coordinates, width, and image of the blade are acquired. Crack detection is performed on the blade image to obtain the total number of cracks and the minimum bounding box coordinates of each crack. Based on the minimum bounding box coordinates of each crack, all crack images are cropped from the blade image, and crack feature information for each crack is obtained. Based on the centroid coordinates, width, total number of cracks, minimum bounding box coordinates of each crack, and crack feature information, a blade crack feature index value is obtained. The blade crack damage level is determined based on the total number of cracks and the blade crack feature index value. This invention automatically determines the blade crack damage level using equipment, reducing manual intervention and improving the efficiency of blade flaw detection testing. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the hardware structure of the blade crack damage level determination device involved in the embodiment of the present invention;

[0043] Figure 2 This is a flowchart illustrating an embodiment of the blade crack damage level determination method of the present invention;

[0044] Figure 3 This is a schematic diagram of the functional modules of an embodiment of the blade crack damage level determination device of the present invention.

[0045] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0046] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0047] In a first aspect, embodiments of the present invention provide a blade crack damage level determination device, which can be a personal computer (PC), laptop computer, server or other device with data processing capabilities.

[0048] Reference Figure 1 , Figure 1This is a schematic diagram of the hardware structure of the blade crack damage level determination device involved in the embodiment of the present invention. In this embodiment, the blade crack damage level determination device may include a processor 1001 (e.g., a Central Processing Unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to realize communication between these components; the user interface 1003 may include a display screen or an input unit such as a keyboard; the network interface 1004 may optionally include a standard wired interface or a wireless interface (e.g., Wireless Fidelity, Wi-Fi interface); the memory 1005 may be high-speed random access memory (RAM) or stable memory (non-volatile memory), such as a disk storage device; the memory 1005 may also optionally be a storage device independent of the aforementioned processor 1001. Those skilled in the art will understand that… Figure 1 The hardware structure shown does not constitute a limitation of the invention and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0049] Continue to refer to Figure 1 , Figure 1 The memory 1005, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a blade crack damage level determination program. The processor 1001 can call the blade crack damage level determination program stored in the memory 1005 and execute the blade crack damage level determination method provided in this embodiment of the invention.

[0050] Secondly, embodiments of the present invention provide a method for determining the level of blade crack damage.

[0051] In one embodiment, reference is made to Figure 2 , Figure 2 This is a flowchart illustrating an embodiment of the blade crack damage level determination method of the present invention. Figure 2 As shown, the methods for determining the damage level of blade cracks include:

[0052] Step S10: Obtain the centroid coordinates of the blade, the blade width, and the blade image;

[0053] In this embodiment, the leaf is first photographed, and then image processing is performed on the photographed image to obtain the centroid coordinates, width, and image of the leaf in the image.

[0054] Further, in one embodiment, step S10 includes:

[0055] Step S101: Perform leaf detection on the original image to obtain the leaf boundary coordinates and leaf centroid coordinates. The original image is an image obtained by taking pictures of the leaf.

[0056] In this embodiment, the leaf to be detected is first photographed to obtain the original image. Then, leaf detection is performed on the original image to obtain the leaf boundary coordinates and the leaf centroid coordinates. Specifically, the leaf boundary can be delineated in the original image using a leaf segmentation model to obtain the leaf boundary coordinates and the leaf centroid coordinates. xy = (x0, y0). The blade segmentation model is obtained through training. The network structure of the blade segmentation model can be Unet++. The blade segmentation model is used to identify the position of the blade from the image.

[0057] Step S102: Obtain the coordinates of the minimum bounding rectangle of the blade and the blade width based on the blade boundary coordinates;

[0058] In this embodiment, the coordinates of the minimum bounding rectangle of the blade and the blade width W0 can be obtained from the blade boundary coordinates.

[0059] Step S103: Crop the leaf image from the original image according to the coordinates of the minimum bounding rectangle of the leaf.

[0060] In this embodiment, the leaf image defined by the minimum bounding rectangle coordinates of the leaf can be cropped from the original image.

[0061] Step S20: Crack detection is performed on the blade image to obtain the total number of cracks and the minimum bounding rectangle coordinates of each crack;

[0062] In this embodiment, a pre-trained blade crack detection model can be used to detect cracks in the blade image, obtaining the total number of cracks and the minimum bounding box coordinates of each crack. The network structure of the blade crack detection model can be YOLOv3, and the model is used to determine the minimum bounding box coordinates of each crack in the blade image.

[0063] Step S30: Based on the coordinates of the minimum bounding rectangle of each crack, crop all crack images from the blade image, and obtain crack feature information of each crack based on each crack image;

[0064] In this embodiment, all crack images are cropped from the blade image based on the minimum bounding rectangle coordinates of each crack, and then image detection is performed on each crack image to obtain the crack feature information of each crack.

[0065] Furthermore, in one embodiment, the step of providing the crack feature information includes:

[0066] Maximum crack width, crack fracture surface type, and crack orientation.

[0067] In this embodiment, image analysis is performed on each crack image using Halcon software to obtain the maximum crack width for each crack. It is easy to understand that one crack image represents one crack.

[0068] Each crack image is input into a pre-trained crack fracture surface type recognition model to obtain the crack fracture surface type for each crack. VGG16 is preferentially selected and trained based on sample images to obtain the crack fracture surface type recognition model. The labels include smooth and uneven, indicating whether the obtained crack fracture surface type is smooth or uneven.

[0069] Each crack image is input into a pre-trained crack orientation recognition model to obtain the crack orientation of each crack. ResNet50 is preferred, and the model is trained based on sample images to obtain the crack orientation recognition model. The labels include vertical, horizontal, diagonal, and divergent, meaning the obtained crack orientation is vertical, horizontal, diagonal, or divergent.

[0070] Step S40: Based on the blade centroid coordinates, blade width, total number of cracks, minimum bounding rectangle coordinates of each crack, and crack feature information of each crack, obtain the blade crack feature index value.

[0071] In this embodiment, the blade crack characteristic index value is obtained based on the blade centroid coordinates, blade width, total number of cracks, minimum bounding rectangle coordinates of each crack, and crack characteristic information of each crack, according to a preset calculation method.

[0072] Further, in one embodiment, step S40 includes:

[0073] Step S401: Based on the maximum crack width of each crack and the blade width, obtain the blade crack width characteristic index value.

[0074] In this embodiment, assuming there are 5 cracks with maximum crack widths W1, W2, W3, W4, and W5, and the blade width is W0, W1, W2, W3, W4, W5, and W0 can be calculated according to a preset calculation method to obtain the blade crack width characteristic index value. The preset calculation method is determined according to actual needs and is not limited here.

[0075] Further, in one embodiment, step S401 includes:

[0076] The maximum widths of all cracks are summed to obtain a sum value; the ratio of the sum value to the blade width is calculated, and the ratio is used as the characteristic index value of the blade crack width.

[0077] In this embodiment, W1, W2, W3, W4, and W5 are first summed, i.e., W1 + W2 + W3 + W4 + W5, to obtain the sum S1. Then, the ratio of S1 to W0 is calculated, i.e., S1 / W0, and the calculated ratio is used as the blade crack width characteristic index value label. W .

[0078] Step S402: Based on the crack fracture surface type of each crack, obtain the characteristic index value of the blade crack fracture surface.

[0079] In this embodiment, it is assumed that there are 5 cracks, and the crack fracture surface types of the 5 cracks are smooth, smooth, uneven, smooth, and uneven, respectively. Based on the correspondence between crack fracture surface types and fracture surface type label values, the fracture surface type label value corresponding to each crack is determined. Then, the 5 fracture surface type label values ​​corresponding to the 5 cracks are calculated according to a preset calculation method to obtain the characteristic index value of the blade crack fracture surface. The preset calculation method is determined according to actual needs and is not limited here.

[0080] Further, in one embodiment, step S402 includes:

[0081] Based on the fracture surface type of each crack, obtain the fracture surface type label value corresponding to each crack; sum the fracture surface type label values ​​of all cracks, and use the sum as the characteristic index value of the blade crack fracture surface.

[0082] In this embodiment, if the fracture surface types of the five cracks are smooth, smooth, uneven, smooth, and uneven, respectively, the correspondence between the fracture surface type and the fracture surface type label value is shown in Table 1:

[0083] Crack fracture surface type Fracture surface type label value smooth 0 uneven 1

[0084] Table 1

[0085] The fracture surface type label values ​​corresponding to the 5 cracks are 0, 0, 1, 0, and 1, respectively. The sum of these values ​​(0 + 0 + 1 + 0 + 1) is calculated to obtain the sum S2. S2 is then used as the label value representing the characteristic index of the blade crack fracture surface. D .

[0086] It should be noted that the above correspondence between crack fracture surface types and fracture surface type label values ​​is only illustrative and does not constitute a limitation on the correspondence between crack fracture surface types and fracture surface type label values.

[0087] Step S403: Based on the crack direction of each crack, obtain the characteristic index value of the blade crack direction.

[0088] In this embodiment, it is assumed that there are 5 cracks, and the crack directions of the 5 cracks are vertical, horizontal, oblique, oblique, and divergent, respectively. The correspondence between crack direction and direction label value is shown in Table 2:

[0089] Crack direction Towards Label Value Vertical 0 Horizontal 1 oblique 2 divergent 3

[0090] Table 2

[0091] The orientation label values ​​corresponding to the 5 cracks are 0, 1, 2, 2, and 3, respectively. The sum of these values ​​(0 + 1 + 2 + 2 + 3) is obtained as S3, and S3 is used as the label value representing the crack orientation of the blade. Z .

[0092] It should be noted that the above correspondence between crack orientation and orientation label values ​​is for illustrative purposes only and does not constitute a limitation on the correspondence between crack orientation and orientation label values. The above calculation method for the orientation label values ​​corresponding to all cracks is only for illustrative purposes; the specific calculation method can be designed according to actual needs.

[0093] Step S404: Determine the centroid coordinates of each crack based on the coordinates of the minimum bounding rectangle of each crack;

[0094] In this embodiment, the minimum bounding rectangle coordinates of each crack are denoted as (x... i0 ,y i0 ,x i1 ,y i1 If the centroid coordinates of each crack are...

[0095] Step S405: Obtain the characteristic index value of the blade crack location based on the centroid coordinates of each crack;

[0096] In this embodiment, the ordinate of the centroid coordinate of each crack is obtained based on the centroid coordinates of each crack. Assuming there are 5 cracks, and the ordinates of the centroids of the 5 cracks are L1, L2, L3, L4, and L5, then the maximum value among L1, L2, L3, L4, and L5 is used as the label value, which is the characteristic index value of the crack location on the blade. L .

[0097] Step S406: Based on the centroid coordinates of the blade and the centroid coordinates of each crack, obtain the discrete characteristic index value of the blade crack.

[0098] In this embodiment, the centroid coordinate of the blade is 0. xy= (x0, y0), then the distance between the centroid of each crack and the centroid of the blade is:

[0099]

[0100] Suppose there are 5 cracks, and the distances between the centroid of the 5 cracks and the centroid of the blade are r1, r2, r3, r4, and r5, respectively. Then, perform specific operations on r1, r2, r3, r4, and r5, such as calculating the standard deviation, calculating the average, etc., and use the calculated values ​​as the discrete characteristic index values ​​(labels) of the blade cracks. r .

[0101] Step S50: Determine the blade crack damage level based on the total number of cracks and the blade crack characteristic index value.

[0102] In this embodiment, the total number of cracks and the characteristic index value of blade cracks are calculated in a specific way to obtain the blade crack damage coefficient, and then the blade crack damage level is determined according to the magnitude of the blade crack damage coefficient.

[0103] In this embodiment, the blade centroid coordinates, blade width, and blade image are acquired. Crack detection is performed on the blade image to obtain the total number of cracks and the minimum bounding box coordinates of each crack. Based on the minimum bounding box coordinates of each crack, all crack images are cropped from the blade image, and crack feature information for each crack is obtained based on each crack image. Based on the blade centroid coordinates, blade width, total number of cracks, minimum bounding box coordinates of each crack, and crack feature information for each crack, a blade crack feature index value is obtained. Based on the total number of cracks and the blade crack feature index value, the blade crack damage level is determined. This embodiment, with equipment as the primary driver, automatically determines the blade crack damage level, reducing manual intervention and improving the efficiency of blade flaw detection testing.

[0104] Further, in one embodiment, step S50 includes:

[0105] The blade crack damage coefficient is obtained by weighting the total number of cracks, the blade crack width characteristic index value, the blade crack fracture surface characteristic index value, the blade crack direction characteristic index value, the blade crack location characteristic index value, and the blade crack discrete characteristic index value; the blade crack damage level is determined based on the blade crack damage coefficient.

[0106] In this embodiment, the total number of cracks m and the characteristic index value label of blade crack width are used. W Blade crack fracture surface characteristic index value label D Blade crack orientation characteristic index value label Z Blade crack location characteristic index value label Land the label of discrete characteristic index value of blade crack r By performing a weighted calculation, the blade crack damage coefficient A is obtained, namely:

[0107] A = λ1·m + λ2·label w +λ3·label D +λ4·label Z +λ5·label L +λ6·label r

[0108] Wherein, λ1, λ2, λ3, λ4, λ5, and λ6 are weight values, obtained by training machine learning models such as decision trees and random forests.

[0109] If A ≤ the first threshold, the blade crack damage level is determined to be mild; if the first threshold < A ≤ the second threshold, the blade crack damage level is determined to be moderate; if the second threshold < A, the blade crack damage level is determined to be severe. This completes the blade crack damage level determination.

[0110] Thirdly, embodiments of the present invention also provide a device for determining the level of blade crack damage.

[0111] In one embodiment, reference is made to Figure 3 , Figure 3 This is a functional module diagram of an embodiment of the blade crack damage level determination device of the present invention. Figure 3 As shown, the blade crack damage level determination device includes:

[0112] The acquisition module 10 is used to acquire the centroid coordinates of the blade, the blade width, and the blade image;

[0113] Detection module 20 is used to detect cracks in the blade image to obtain the total number of cracks and the minimum bounding rectangle coordinates of each crack.

[0114] The feature extraction module 30 is used to crop all crack images from the blade image based on the coordinates of the minimum bounding rectangle of each crack, and to obtain crack feature information of each crack based on each crack image.

[0115] The calculation module 40 is used to obtain the blade crack feature index value based on the blade centroid coordinates, blade width, total number of cracks, minimum bounding rectangle coordinates of each crack, and crack feature information of each crack.

[0116] The rating module 50 is used to determine the blade crack damage level based on the total number of cracks and the blade crack characteristic index value.

[0117] Furthermore, in one embodiment, the acquisition module 10 is used for:

[0118] Leaf detection is performed on the original image to obtain the leaf boundary coordinates and leaf centroid coordinates. The original image is an image captured of the leaf.

[0119] The coordinates of the minimum bounding rectangle of the blade and the blade width are obtained from the blade boundary coordinates.

[0120] The leaf image is cropped from the original image based on the coordinates of the minimum bounding rectangle of the leaf.

[0121] Furthermore, in one embodiment, the step of obtaining crack feature information includes:

[0122] Maximum crack width, crack fracture surface type, and crack orientation.

[0123] Furthermore, in one embodiment, the calculation module 40 is used for:

[0124] Based on the maximum crack width of each crack and the blade width, the characteristic index value of the blade crack width is obtained.

[0125] Based on the crack fracture surface type of each crack, the characteristic index value of the blade crack fracture surface is obtained.

[0126] Based on the crack direction of each crack, the characteristic index value of the blade crack direction is obtained;

[0127] The centroid coordinates of each crack are determined based on the coordinates of the minimum bounding rectangle of each crack.

[0128] Based on the centroid coordinates of each crack, the characteristic index value of the blade crack location is obtained;

[0129] Based on the centroid coordinates of the blade and the centroid coordinates of each crack, the discrete characteristic index value of the blade crack is obtained.

[0130] Furthermore, in one embodiment, the calculation module 40 is used for:

[0131] The maximum widths of all cracks are summed to obtain the sum value;

[0132] Calculate the ratio of the sum to the blade width, and use the ratio as the characteristic index value of the blade crack width.

[0133] Furthermore, in one embodiment, the calculation module 40 is used for:

[0134] Based on the fracture surface type of each crack, obtain the fracture surface type label value corresponding to each crack;

[0135] The fracture surface type label values ​​of all cracks are summed, and the sum is used as the characteristic index value of the blade crack fracture surface.

[0136] Furthermore, in one embodiment, the classification module 50 is used for:

[0137] The blade crack damage coefficient is obtained by weighting the total number of cracks, the characteristic index value of blade crack width, the characteristic index value of blade crack fracture surface, the characteristic index value of blade crack direction, the characteristic index value of blade crack location, and the characteristic index value of blade crack discreteness.

[0138] The blade crack damage level is determined based on the blade crack damage coefficient.

[0139] The functions of each module in the blade crack damage level determination device correspond to the steps in the blade crack damage level determination method embodiment, and their functions and implementation processes will not be described in detail here.

[0140] Fourthly, embodiments of the present invention also provide a readable storage medium.

[0141] The present invention stores a blade crack damage level determination program on a readable storage medium, wherein when the blade crack damage level determination program is executed by a processor, the steps of the blade crack damage level determination method described above are implemented.

[0142] The method implemented when the blade crack damage level determination procedure is executed can be referred to in various embodiments of the blade crack damage level determination method of the present invention, and will not be repeated here.

[0143] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0144] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0145] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device to execute the methods described in the various embodiments of the present invention.

[0146] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A method of determining a blade crack damage level, characterized by, The blade crack damage grade determination method comprises: obtaining blade centroid coordinates, blade width and blade image; carrying out crack detection on the blade image to obtain total number of cracks and minimum circumscribed rectangle box coordinates of each crack; based on the minimum circumscribed rectangle box coordinates of each crack, all crack images are cropped from the blade image, and based on each crack image, crack feature information of each crack is obtained, the crack feature information comprising: crack maximum width, crack fracture surface type and crack direction; total number of cracks, blade crack width feature index value, blade crack fracture surface feature index value, blade crack direction feature index value, blade crack position feature index value and blade crack dispersion feature index value are calculated by weighting, to obtain blade crack damage coefficient; and the blade crack damage grade is determined according to the blade crack damage coefficient. The step of obtaining blade centroid coordinates, blade width and blade image comprises:

2. The blade crack damage level determination method according to Claim 1, characterized by, carrying out blade detection on the original image to obtain blade boundary coordinates and blade centroid coordinates, the original image being an image obtained by photographing the blade; obtaining blade minimum circumscribed rectangle box coordinates and blade width according to the blade boundary coordinates; cropping the blade image from the original image according to the blade minimum circumscribed rectangle box coordinates. The blade crack damage grade determination device comprises:

3. A device for determining the damage level of blade cracks, characterized in that, an acquisition module configured to obtain blade centroid coordinates, blade width and blade image; a detection module configured to carry out crack detection on the blade image to obtain total number of cracks and minimum circumscribed rectangle box coordinates of each crack; a feature extraction module configured to crop all crack images from the blade image based on the minimum circumscribed rectangle box coordinates of each crack, and obtain crack feature information of each crack based on each crack image, the crack feature information comprising: crack maximum width, crack fracture surface type and crack direction. ​ The computing module is configured to accumulate and sum up the maximum crack widths of all the cracks to obtain a sum value, calculate a ratio of the sum value to the blade width, and take the ratio as a blade crack width characteristic index value; obtain a crack fracture surface type label value corresponding to each crack according to the crack fracture surface type of each crack, accumulate and sum up the crack fracture surface type label values of all the cracks to obtain a sum value, and take the sum value as a blade crack fracture surface characteristic index value; obtain a blade crack orientation characteristic index value according to the crack orientation of each crack; determine the centroid coordinates of each crack according to the minimum circumscribed rectangular frame coordinates of each crack; obtain a blade crack position characteristic index value according to the centroid coordinates of each crack; and obtain a blade crack dispersion characteristic index value according to the blade centroid coordinates and the centroid coordinates of each crack. The grading module is configured to perform weighted calculation on the total number of cracks, the blade crack width characteristic index value, the blade crack fracture surface characteristic index value, the blade crack orientation characteristic index value, the blade crack position characteristic index value, and the blade crack dispersion characteristic index value to obtain a blade crack damage coefficient, and determine a blade crack damage grade according to the blade crack damage coefficient.

4. A blade crack damage level determination device, characterized in that, The blade crack damage grade determination device includes a processor, a memory, and a blade crack damage grade determination program stored on the memory and executable by the processor, wherein the blade crack damage grade determination program, when executed by the processor, implements the steps of the blade crack damage grade determination method of claim 1 or 2.

5. A readable storage medium, characterized by, The readable storage medium stores a blade crack damage grade determination program, wherein the blade crack damage grade determination program, when executed by the processor, implements the steps of the blade crack damage grade determination method of claim 1 or 2.

Images