An ultrasonic automatic recognition method for composite material defects considering the detection probability

Abstract

The present invention relates to the field of nondestructive testing technology, and particularly to an ultrasonic automatic recognition method for composite material defects considering the detection probability. The method includes the steps of: obtaining ultrasonic C-scan information; setting an automatic recognition threshold G for defects u ; setting an automatic evaluation threshold for defects; automatically recognizing defects; visually marking defects; and evaluating the robustness of automatic defect recognition. It is not necessary to separately prepare an ultrasonic detection comparison test block for composite materials, and it can better evaluate the effectiveness and correctness of the automatic evaluation result of the ultrasonic C-scan detection result of a large composite material structure. Furthermore, it significantly improves the accuracy and reliability of the evaluation of the ultrasonic C-scan detection result, and significantly increases the visualization degree of the evaluation of the ultrasonic C-scan detection result.

Description

Technical Field

[0001] The present invention relates to the technical field of non-destructive testing, and particularly to an ultrasonic automatic recognition method for composite material defects considering the detection probability. Background Art

[0002] Large composite material structures are currently widely used in the industrial field and belong to very important composite material components, with very high requirements for quality, cost, safety, and performance. In order to ensure the quality of such key composite material structures, ultrasonic C-scanning is usually used for 100% non-destructive testing, and then, inspection technicians judge whether there are defects exceeding the design requirements based on the ultrasonic C-scanning images. For this purpose, it is necessary for inspection technicians to observe and judge the ultrasonic C-scanning images one by one, and finally give the inspection results.

[0003] Currently, when performing ultrasonic C-scanning detection of composite materials, manual evaluation of ultrasonic C-scanning images is usually adopted to judge the detection results. Its obvious deficiencies are: (1) Since the amount of ultrasonic C-scanning data of large composite material structures is very large, and the physical display size of the computer screen is very limited, only compressed display can be adopted, which is likely to cause partial small defects to be missed in display and visually missed in judgment; (2) Due to the large amount of ultrasonic C-scanning data, it is easy to have teaching fatigue and drawing evaluation fatigue in defect evaluation, and thus it is easy to cause defect missed judgment; (3) The judgment results of ultrasonic C-scanning are marked and recorded manually, which is easy to make mistakes; (4) Long-term visual judgment of detection results on the computer screen is likely to cause visual fatigue, and thus cause misjudgment and missed detection; (5) It is easily affected by the comprehensive technology and experience factors of the detection result judgment personnel; (6) The judgment efficiency of the detection results is low. As an improvement, an automatic discrimination method is also introduced to evaluate the detection results of ultrasonic C-scanning, but its main deficiencies are: (1) The verification of the robustness of the automatic discrimination method is not considered; (2) The mutual influence relationship between the ultrasonic C-scanning images and the detected object, its materials, processes, and structural behaviors is not considered, which significantly affects the applicability and robustness of the automatic discrimination method. Summary of the Invention

[0004] (1) Technical Problems to be Solved

[0005] An embodiment of the present invention provides an ultrasonic automatic recognition method for composite material defects considering the detection probability, which solves the technical problems of low detection efficiency and poor detection reliability in ultrasonic C-scanning detection of large composite material structures.

[0006] (2) Technical Solutions

[0007] An embodiment of the present invention proposes an ultrasonic automatic recognition method for composite material defects considering the detection probability, including the steps of: obtaining ultrasonic C-scanning information; setting the automatic defect recognition threshold G u; Set the automatic defect evaluation threshold; automatically identify defects; visually identify defects; evaluate the robustness of automatic defect identification.

[0008] Further, the obtaining of the ultrasonic C-scan information includes the steps of: for the formed ultrasonic C-scan image format, read and store it in the computer memory I nm (x, y, z, c), where x, y, and z respectively correspond to the position coordinates in the ultrasonic C-scan image, and satisfy: x = k x x o (1), y = k y y o (2), z = k z z o (3), where (x o , y o , z o ) respectively represent the position coordinates on the surface of the composite material part to be detected, and (k x , k y , k z ) respectively represent the conversion coefficients between the position coordinates on the surface of the composite material part to be detected and the position coordinates in the ultrasonic C-scan image; c is the color value of the ultrasonic C-scan image corresponding to (x, y, z), and satisfies: c = k u A u (4), k u is the conversion coefficient between the ultrasonic signal and the color value in the image, and A u is the ultrasonic signal used for ultrasonic C-scan imaging; for the ultrasonic C-scan result in the original data format of detection, convert the ultrasonic C-scan original data into the ultrasonic C-scan image format according to equations (1) to (4), and read it into the computer memory I nm (x, y, z, c).

[0009] Further, the ultrasonic C-scan image format includes BMP, TIFF, and JPEG bitmap formats.

[0010] Further, the setting of the automatic defect identification threshold G u includes: according to the material, process, and structural characteristics of the composite material part to be detected and the characteristics of its ultrasonic C-scan image, determine the calculation formula for the automatic defect identification threshold as: where Q ij (x ij , y ij , c ij ) is the i-th row and j-th column of the selected ultrasonic C-scan feature image area, x ij , y ij , c ijThey are the corresponding image coordinates and image color values respectively. The size of the ultrasonic C-scan feature image area is m×n, which is selected according to the read ultrasonic C-scan image features, and the defect-free area of the composite material part to be detected corresponding to the ultrasonic C-scan feature image area is determined through ultrasonic C-scan detection tests.

[0011] Further, the setting of the defect automatic evaluation threshold includes: setting the defect automatic evaluation threshold according to the quality acceptance requirements of the composite material part to be detected

[0012] Further, the is the area or diameter or length, and the unit is millimeters.

[0013] Further, the automatic identification of defects includes the steps of: reading the defect automatic identification image area from I nm (x,y,z,c) Recognition threshold comparison: If there is no image in which the color value c is greater than G u , it is judged whether the recognition is completed. If the recognition is not completed, then i = i + 1, and continue to read the defect automatic identification image area from I nm (x,y,z,c) If the recognition is completed, end the recognition; If there is an image in which the color value c is greater than G u , then calculate the area of the image in which the color value c is greater than G u in and its length L, width W and equivalent diameter D, and the corresponding position coordinates (x,y), and save them in the array , judge whether the recognition is completed. If the recognition is not completed, then i = i + 1, and continue to read the defect automatic identification image area from I nm (x,y,z,c) If the recognition is completed, end the recognition.

[0014] Further, the visual identification of defects includes: according to the recorded results in i carry out defect F

[0015] Further, the evaluation of the robustness of automatic defect recognition includes the steps of: selecting a standard sample for evaluating the ultrasonic C-scan detection effect of the composite material, where the selection of the standard sample is the same as that of the composite material part to be detected, and the defect distribution and quantity in the standard sample meet the detection probability requirements of the ultrasonic C-scan; using the same ultrasonic C-scan detection conditions, detecting the standard sample 3 times respectively, and saving the detection results each time; automatically recognizing the defects in the 3D ultrasonic automatic C-scan results of the evaluated part, and obtaining the number of defects N automatically recognized by the ultrasonic C-scan for 3 times d , and the formula for calculating the robustness is:

[0016] (III) Beneficial effects

[0017] In summary, the present invention takes into account that the amount of ultrasonic C-scan data of large composite structures is very large, while the physical display size of the computer screen is very limited. The compressed display method is prone to missing the display of some small defects and visual misjudgment; it overcomes the deficiencies of easy teaching fatigue and drawing evaluation fatigue during manual evaluation, and will not cause missed defect judgment; the ultrasonic C-scan judgment results are automatically marked and recorded, and it is not easy to make mistakes; there will be no misjudgment and missed detection caused by viewing angle fatigue; it is not affected by the comprehensive technology and experience factors of the detection result evaluators; the evaluation efficiency of the detection results is very high; it considers the verification of the robustness of the automatic discrimination method; it considers the mutual influence relationship between the ultrasonic C-scan image and the detected object, its material, process, and structural behavior, significantly improving the applicability and robustness of the defect automatic discrimination method. Furthermore, it significantly improves the accuracy and reliability of the evaluation of the ultrasonic C-scan detection results of large composite structures, and significantly increases the visualization degree of the evaluation of the ultrasonic C-scan detection results. Description of the drawings

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required to be used in the embodiments of the present invention. Obviously, the following described drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0019] Figure 1 It is a schematic flow chart of an ultrasonic automatic defect recognition method for composite materials considering the detection probability according to an embodiment of the present invention. Detailed implementation manners

[0020] The following further describes in detail the embodiments of the present invention in conjunction with the accompanying drawings and embodiments. The detailed description and drawings of the following embodiments are used to exemplarily illustrate the principles of the present invention, but cannot be used to limit the scope of the present invention, that is, the present invention is not limited to the described embodiments, and covers any modifications, substitutions, and improvements of parts, components, and connection methods without departing from the spirit of the present invention.

[0021] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The following will describe the present application in detail with reference to the accompanying drawings and embodiments.

[0022] Please refer to Figure 1 , an ultrasonic automatic recognition method for composite material defects considering the detection probability is proposed in the embodiments of the present invention, including the steps of: obtaining ultrasonic C-scan information; setting an automatic defect recognition threshold G u ; setting an automatic defect evaluation threshold; automatically recognizing defects; visually marking defects; evaluating the robustness of automatic defect recognition. Considering that the amount of ultrasonic C-scan data of large composite material structures is very large and the physical display size of the computer screen is very limited, the compressed display method is adopted, which is prone to the deficiencies of missing display of some small defects and visual misjudgment; it overcomes the deficiencies of easy teaching fatigue and drawing evaluation fatigue during manual evaluation, and will not cause misjudgment of defects; the ultrasonic C-scan evaluation results are automatically marked and recorded, and it is not easy to make mistakes; there will be no misjudgment and missed detection caused by viewing fatigue; it is not affected by the comprehensive technology and experience factors of the detection result evaluators; the evaluation efficiency of the detection results is very high; the verification of the robustness of the automatic discrimination method is considered; the mutual influence relationship between the ultrasonic C-scan image and the detected object, its materials, processes, and structural behaviors is considered, significantly improving the applicability and robustness of the defect automatic discrimination method. Furthermore, the accuracy and reliability of the ultrasonic C-scan detection results evaluation of large composite material structures are significantly improved, and the visualization degree of the ultrasonic C-scan detection results evaluation is significantly increased.

[0023] In some embodiments, the step of obtaining ultrasonic C-scan information includes: for the formed ultrasonic C-scan image format, reading and storing it in the computer memory I nm (x,y,z,c), where x, y, z respectively correspond to the position coordinates in the ultrasonic C-scan image, and satisfy: x = k x x o (1), y = k y y o (2), z = k z z o (3), where (x o , y o , z o ) respectively represent the position coordinates on the surface of the detected composite material part, (k x, k y , k z ) represent the conversion coefficients between the position coordinates on the surface of the detected composite material part and the position coordinates in the ultrasonic C-scan image; c is the color value of the ultrasonic C-scan image corresponding to (x, y, z), and satisfies: c = k u A u (4), k u is the conversion coefficient between the ultrasonic signal and the color value in the image, and A u is the ultrasonic signal used for ultrasonic C-scan imaging; for the ultrasonic C-scan result in the format of the original detection data, convert the original ultrasonic C-scan data into the ultrasonic C-scan image format according to formulas (1) to (4), and read it into the computer memory I nm (x, y, z, c). Further, the ultrasonic C-scan image format includes BMP, TIFF, and JPEG bitmap formats.

[0024] In some embodiments, the setting of the automatic defect recognition threshold G u includes: according to the material, process, and structural characteristics of the detected composite material part and the characteristics of its ultrasonic C-scan image, determine the calculation formula for the automatic defect recognition threshold as: where Q ij (x ij , y ij , c ij ) is the i-th row and j-th column of the selected ultrasonic C-scan feature image area, and x ij , y ij , c ij are the corresponding image coordinates and image color values respectively. The size of the ultrasonic C-scan feature image area is m×n, which is selected according to the read ultrasonic C-scan image characteristics, and the defect-free area of the detected composite material part corresponding to the ultrasonic C-scan feature image area is determined through ultrasonic C-scan detection tests.

[0025] In some embodiments, the setting of the automatic defect evaluation threshold includes: setting the automatic defect evaluation threshold according to the quality acceptance requirements of the detected composite material part Further, the is the area or diameter or length, and the unit of

[0026] In some embodiments, the automatic recognition of defects includes the steps of: reading the automatic defect recognition image area from I nm (x, y, z, c) Recognition threshold comparison: If there is no image in which the color value c is greater than G u , then determine whether the recognition is completed. If the recognition is not completed, then i = i + 1, and continue to read from I nmRead the defect automatic recognition image area from (x, y, z, c) If the recognition is completed, end the recognition; if there is an image with color value c greater than G u in it, then calculate the area of the image with color value c greater than G u in it its length L, width W and equivalent diameter D, and the corresponding position coordinates (x, y), and save them in the array , determine whether the recognition is completed. If not, then i = i + 1, and continue to read the defect automatic recognition image area from I nm Read the defect automatic recognition image area from (x, y, z, c) If the recognition is completed, end the recognition.

[0027] In some embodiments, visually identifying the defects includes: according to the recorded results in it, perform defect F i identification in the ultrasonic C-scan image; according to the recorded results in it, make a text list according to Table 1

[0028] Table 1 Defect Visualization List Format

[0029]

[0030] In some embodiments, the evaluation of the robustness of automatic defect recognition includes the steps of: selecting a standard sample for evaluating the ultrasonic C-scan detection effect of the composite material. The selection of the standard sample is the same as the composite material part to be detected, and the defect distribution and quantity in the standard sample meet the detection probability requirements of the ultrasonic C-scan; using the same ultrasonic C-scan detection conditions, perform 3 detections on the standard sample respectively, and save the results of each detection; perform automatic defect recognition on the 3 ultrasonic three-dimensional automatic C-scan results of the evaluated part, and obtain the number of defects N d automatically recognized by the ultrasonic C-scan in 3 times, and calculate the robustness formula as:

[0031] (1) When γ ≥ 98%, the robustness of ultrasonic C-scan defect automatic recognition is rated as very robust, defined as grade A;

[0032] (2) When 95% > γ ≥ 98%, the robustness of ultrasonic C-scan defect automatic recognition is rated as robust, defined as grade B;

[0033] (3) When 90% ≤ γ < 95%, the robustness of ultrasonic C-scan defect automatic recognition is rated as less robust, defined as grade D;

[0034] (4) When γ < 90%, the robustness of ultrasonic C-scan defect automatic recognition is rated as not robust, defined as grade E;

[0035] Among them, grade A indicates that the automatic defect recognition result is the most correct, grade E indicates that the automatic defect recognition result is the least correct, and grades B, C, and D are in between grade A and grade E in sequence.

[0036] When it is verified to reach grade A, it can be used; when it reaches grade B, it has practicality; if other grades are obtained, optimization is required.

[0037] Example:

[0038] Ultrasonic C-scan data of large composite wall panels with sizes ranging from 3200mm×1600mm to 6000mm×1600mm were respectively selected. Automatic defect recognition was carried out for multiple cases. Combining the automatic defect evaluation results, the evaluation standard sample with N s = 93 was selected. After carrying out the robustness evaluation, for two types of defects, the number of automatically recognized defects was both M d = 279, and the robustness was both γ = 100%, belonging to grade A. The test results show that the proposed method can efficiently and accurately evaluate the ultrasonic C-scan detection results of composites, with very high evaluation efficiency, high correctness of evaluation results, extremely low labor intensity, very low cost, and very short evaluation cycle. Furthermore, it is more conducive to improving the accuracy, reliability, and detection efficiency of ultrasonic C-scan detection results.

[0039] It should be clear that each embodiment in this specification is described in a progressive manner. The same or similar parts among the embodiments can be referred to each other, and the key points of each embodiment are the differences from other embodiments. The present invention is not limited to the specific steps and structures described above and shown in the figures. And, for the sake of brevity, the detailed description of known method technologies is omitted here.

[0040] The above description is only for the embodiments of the present application and does not limit the present application. For those skilled in the art, the present application can have various changes and modifications without departing from the scope of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. An ultrasonic automatic recognition method for composite material defects considering the detection probability, characterized in that, Including the steps: Obtain ultrasonic C-scan information; Set the automatic defect recognition threshold G u : According to the material, process, and structural characteristics of the composite material part to be detected and the characteristics of its ultrasonic C-scan image, the calculation formula for determining the automatic defect recognition threshold is as follows: where Q ij (x ij , y ij , c ij ) is the i-th row and j-th column of the selected ultrasonic C-scan feature image area, x ij , y ij , c ij are the corresponding image coordinates and image color values respectively. The size of the ultrasonic C-scan feature image area is m×n, which is selected according to the read ultrasonic C-scan image features, and the defect-free area of the composite material part to be detected corresponding to the ultrasonic C-scan feature image area is determined by ultrasonic C-scan detection tests; Set the automatic evaluation threshold for defects; Automatically identify defects; Visually identify defects; Automatically identify and evaluate the robustness of defects: Select a standard sample for evaluating the ultrasonic C-scan detection effect of composite materials. The selection of the standard sample is the same as that of the composite material parts to be detected, and the defect distribution and quantity in the standard sample meet the detection probability requirements of ultrasonic C-scan; Using the same ultrasonic C-scan detection conditions, detect the standard sample 3 times respectively, and save the detection results each time; Automatically identify the defects in the 3D ultrasonic automatic C-scan results of the evaluated parts, and obtain the number of defects N automatically identified by ultrasonic C-scan for 3 times d , and the formula for calculating the robustness is: , where N s = 93.

2. The ultrasonic automatic recognition method for composite material defects considering the detection probability according to claim 1, characterized in that The obtaining of ultrasonic C-scan information includes the steps: Read and store the formed ultrasonic C-scan image format into the computer memory I nm In (x, y, z, c), x, y, and z respectively correspond to the position coordinates in the ultrasonic C-scan image and satisfy: x = k x x o (1) y = k y y o (2) z = k z z o (3) Among them, (x o , y o , z o ) respectively represent the position coordinates on the surface of the composite material part to be detected, and (k x , k y , k z ) respectively represent the conversion coefficients between the position coordinates on the surface of the composite material part to be detected and the position coordinates in the ultrasonic C-scan image; c is the color value of the ultrasonic C-scan image corresponding to (x, y, z), and satisfies: c = k u A u (4) k u is the conversion coefficient of the ultrasonic signal to the color value in the image, and A u is the ultrasonic signal for ultrasonic C-scan imaging; For the ultrasonic C-scan result being the detected original data format, convert the ultrasonic C-scan original data into the ultrasonic C-scan image format according to Equations (1) to (4), and read it into the computer memory I nm (x, y, z, c).

3. An ultrasonic automatic recognition method for composite material defects considering the detection probability according to claim 2, characterized in that The ultrasonic C-scan image format includes BMP, TIFF, and JPEG bitmap formats.

4. The ultrasonic automatic recognition method for composite material defects considering the detection probability according to claim 1, characterized in that The setting of the automatic defect evaluation threshold includes: setting the automatic defect evaluation threshold according to the quality acceptance requirements of the composite material parts to be detected .

5. The ultrasonic automatic recognition method for composite material defects considering the detection probability according to claim 4, characterized in that The said is area or diameter or length, and the unit thereof is millimeter.

6. An ultrasonic automatic recognition method for composite material defects considering the detection probability according to claim 4 or 5, characterized in that Automatically identifying defects includes the steps: Read from I nm the defect automatic recognition image area in (x, y, z, c) ; Recognition threshold comparison: If there is no color value c greater than G in u the image, then determine whether the recognition is completed. If the recognition is not completed, then i = i + 1, and continue to read the defect automatic recognition image area from I nm (x, y, z, c) ; If the identification is completed, end the identification; If there is an image with a color value c greater than G u , calculate the area of the image with a color value c greater than G u in as well as its length L, width W and equivalent diameter D, and the corresponding position coordinates (x, y), and save them in the array . Determine whether the recognition is completed. If not, then i = i + 1, and continue to read the defect automatic recognition image area from I nm (x, y, z, c) ; If the identification is completed, end the identification.

7. An ultrasonic automatic recognition method for composite material defects considering the detection probability according to claim 6, characterized in that, Visual identification of defects includes: Based on the recorded results in i perform defect F identification in the ultrasonic C-scan image.

Images