Asphalt mixture uniformity evaluation method based on CT scanning technology

Abstract

The invention discloses an asphalt mixture uniformity evaluation method based on a CT scanning technology, and the method comprises the following specific steps: S1, obtaining an asphalt mixture cylinder test piece, and scanning the asphalt mixture cylinder test piece by using CT scanning equipment; s2, constructing a three-dimensional visual model of the test piece by using the slice image; s3, carrying out image processing on the three-dimensional visualization model, and carrying out component identification; s4, establishing an XYZ coordinate system by taking the geometric center of the cylindrical test piece as an original point and the height direction as a Z axis; s5, dividing the cylindrical test piece into N equal cubic grids, and defining the minimum unit size; s6, dividing the aggregate in the asphalt mixture into multiple grades of aggregate according to the equivalent diameter of the aggregate; and S7, calculating a uniformity index K. According to the invention, based on CT and three-dimensional image division, the uniformity of the whole recycled asphalt mixture test piece is evaluated, and the defects of the integrity of the split mixture and large calculation error are overcome.

Description

Technical Field

[0001] The present invention belongs to the technical field of asphalt mixtures, and in particular relates to an asphalt mixture uniformity evaluation method based on CT scanning technology. Background Art

[0002] At present, most studies on the internal structural uniformity of asphalt mixtures are based on CT scanning slices. The internal structural image information is obtained through CT scanning or cutting and shooting. Then, the single slice is divided into fan-shaped, square and other area regions. The coefficient of variation of the aggregate area between different regions is used as an indicator to quantitatively evaluate the uniformity of the asphalt mixture. There are the following limitations: (1) There are many CT slices of the specimen. In order to evaluate the uniformity of the specimen, it is necessary to calculate the cross-sectional evaluation index of many slices, which is a complicated process; (2) There is a certain error in the calculation of each cross-sectional uniformity index, which ultimately results in a large cumulative error. In recycled asphalt mixtures, the agglomeration phenomenon caused by the uneven distribution of new and old asphalt mortars is the main factor causing the uneven distribution of new and old materials in recycled asphalt mixtures. In view of this, it is necessary to improve the traditional asphalt mixture uniformity evaluation method. Summary of the Invention

[0003] In order to address the deficiencies of the prior art, the purpose of the present invention is to provide an asphalt mixture uniformity evaluation method based on CT scanning technology. Based on CT and three-dimensional image segmentation, the uniformity of the entire recycled asphalt mixture specimen is evaluated, overcoming the shortcomings of fragmenting the integrity of the mixture and large calculation errors.

[0004] In order to achieve the above objectives, the present invention adopts the following technical solutions: A method for evaluating the uniformity of asphalt mixture based on CT scanning technology, the specific steps are as follows: S1. Obtain an asphalt mixture cylindrical specimen, scan it using a CT scanner, and obtain a slice image of the specimen; S2. constructing a three-dimensional visual model of the specimen using the slice image obtained in step S1; S3. Perform image processing on the three-dimensional visualization model to identify components, remove asphalt mortar and voids, and retain aggregate particles; S4. Establish an XYZ coordinate system with the geometric center of the cylindrical specimen as the origin and the height direction as the Z axis; S5. Divide the cylindrical specimen into N equal cubic grids. To ensure that a single grid can accommodate at least one largest aggregate and two smallest aggregates, define the minimum unit size. The calculation formula is as follows: (1); in is the minimum unit size; is the maximum aggregate size in the specimen; is the minimum aggregate size in the specimen; 、 、 is the number of grids in each dimension, rounded down; R is the specimen radius; H is the specimen height; N is the total number of grids. For boundary grids (boundary grids are determined by the ratio of aggregate to grid), only the part of the grid with a specimen volume coverage of more than 50% is counted; S6. Divide the aggregates in the asphalt mixture into multiple grades based on the equivalent diameter of the aggregates; S7. For the i-th aggregate, calculate the volume proportion of the aggregate in each grid, count the volume proportions in all grids, calculate the uniformity coefficient, and finally weight the uniformity index K of each aggregate. The calculation formula is as follows (2): (2); in, is the uniformity coefficient of each aggregate; N is the total number of grids; is the bth grid Volume proportion of aggregate; For all grids Volume proportion of aggregate; is a very small constant (such as 1e-6, to avoid the denominator being zero); It is the ratio of the volume of the i-th grade aggregate to the total aggregate volume; the smaller the K value is, the better the uniformity of the asphalt mixture is; conversely, the worse the uniformity of the asphalt mixture is.

[0005] Preferably, in formula (2):

[0006] is the volume of the i-th aggregate in the b-th grid; It is the volume of a single voxel, determined by the CT resolution; is the number of voxels of the i-th aggregate in the b-th grid. The position coordinates of all aggregates are derived from AVIZO software, and the number of voxels in each grid is counted using MATLAB software.

[0007] Preferably, in the aforementioned step S1, the radius of the cylindrical specimen is ≥50 mm, and the height is ≥50 mm; the CT tomography scanning direction is the height direction of the specimen, and the spacing is ≤1.5 mm.

[0008] Preferably, in the aforementioned step S2, the three-dimensional visualization model of the specimen is constructed using AVIZO software.

[0009] Preferably, in the aforementioned step S3, the specific steps of performing image processing on the three-dimensional visualization model are as follows: (1) Using AVIZO software to read and convert grayscale images of asphalt mixture CT scans; (2) Using grayscale conversion to adjust the contrast and brightness of grayscale images and using median filtering to remove granular noise in asphalt mixture digital images to achieve image enhancement; (3) Use interactive threshold segmentation and slice editing to separate, check and correct aggregates in 5 to 10 consecutive fault images, which serve as material training models for deep learning; (4) Use the trained deep learning model to pre-identify the components of the remaining tomographic images to enhance the image contrast between the components; (5) Based on the bimodal method, the gray value boundary of the components is determined, and the binary images of the aggregate, asphalt mortar, and void components are separated from the image using the gray threshold; (6) The adhered particles are separated by morphological opening operation.

[0010] Preferably, in the aforementioned step S6, the equivalent diameter of the aggregate is the diameter of a sphere having the same volume as the aggregate.

[0011] Preferably, in the aforementioned step S6, the aggregates in the asphalt mixture are divided into four grades according to the equivalent diameter of the aggregates: 2.36 mm to 4.75 mm, 4.75 mm to 9.5 mm, 9.5 mm to 13.2 mm and above 13.2 mm.

[0012] The benefits of the present invention lie in that: based on CT and three-dimensional image segmentation, the present invention systematically evaluates the uniformity of the entire recycled asphalt mixture specimen. Through three-dimensional full-domain analysis, dynamic grid division and hierarchical weighted evaluation mechanism, it systematically solves the core problems of dimensional limitations, fragmentation integrity and error accumulation in traditional two-dimensional evaluation methods, while ensuring the accuracy of calculation results and improving calculation efficiency. Its technical advantages have been fully verified in the examples of new materials (SMA-13, AC-13) and recycled materials (conventional RAP, finely separated RAP). BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is the gradation curve of the asphalt mixture in Example 1; Figure 2 This is the operation step of performing image processing on the three-dimensional visualization model in Example 1; Figure 3 Schematic diagram of each particle size in Example 1; Figure 4 is the K value of the three asphalt mixtures in Example 1; Figure 5 is the gradation curve of the asphalt mixture in Example 2; Figure 6is the K value of the three asphalt mixtures in Example 2. DETAILED DESCRIPTION

[0014] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0015] Example 1: Asphalt mixture uniformity evaluation method based on CT scanning technology, the specific steps are as follows: (1) Three types of Marshall specimens were made by indoor molding, namely, new SMA-13, conventional crushed and screened RAP recycled SMA-13 ​​mixture, and finely separated RAP recycled mixture. The specimen size was Φ101.6 mm × 63.5 mm. The gradation curve of asphalt mixture is shown in Figure 2. Figure 1 As shown. CT scanning equipment was used to scan it with a scanning accuracy of 0.1 mm, and a total of 635 slice images along the height direction of the specimen were obtained; (2) Use the slice image obtained in step A to construct a 3D visualization model of the specimen using AVIZO software; perform image processing on the 3D visualization model to identify components, remove asphalt mortar and voids, and retain aggregate particles; perform image processing on the 3D visualization model. The operation steps are shown in Figure 2 ; (3) Using AVIZO software to read and convert grayscale of asphalt mixture CT scan images; then using grayscale transformation to adjust the contrast and brightness of the grayscale image and using median filtering to remove granular noise in the asphalt mixture digital image to achieve image enhancement; (4) Use interactive threshold segmentation and slice editing to separate and check the aggregates in 5 to 10 consecutive fault images, and use them as material training models for deep learning; use the trained deep learning model to pre-identify the components of the remaining fault images to enhance the image contrast between the components; (5) Based on the bimodal method, the grayscale value boundary of the components is determined, and the binary images of the aggregate, asphalt mortar, and voids are separated from the image using the grayscale threshold. Secondly, the adhering particles are disconnected through morphological opening operations such as opening pre-segmentation and watershed segmentation. (6) An XYZ coordinate system is established with the geometric center of the cylindrical specimen as the origin and the height direction as the Z axis. The cylindrical specimen is divided into N equal cubic grids. To ensure that a single grid can accommodate at least one largest aggregate and two smallest aggregates, the minimum unit size is defined, and the calculation formula is as shown in Equation (1): (1); in is the minimum unit size; is the maximum aggregate size in the specimen; is the minimum aggregate size in the specimen; 、 、 is the number of grids in each dimension, rounded down; R is the specimen radius; H is the specimen height; N is the total number of grids. For boundary grids, only the grids with a specimen volume coverage exceeding 50% are counted; (7) According to the equivalent diameter of the aggregate, the aggregate in the asphalt mixture is divided into four grades: 2.36 mm~4.75 mm, 4.75~9.5 mm, 9.5~13.2 mm and above 13.2 mm. The schematic diagram of each grade of particle size is shown as follows: Figure 3 As shown; (8) For the i-th grade of aggregate, calculate the volume proportion of the aggregate in each grid, count the volume proportions in all grids, calculate the uniformity coefficient, and finally calculate the weighted comprehensive uniformity index K of each grade of aggregate. The calculation formula is as follows: (2); Where, is the uniformity coefficient of each aggregate; N is the total number of grids; is the number of the bth grid in the bth grid. Volume proportion of aggregate; is the volume proportion of the first-tier aggregate in all grids; is a very small constant (such as 1e-6, to avoid the denominator being zero); It is the ratio of the volume of the i-th grade aggregate to the total aggregate volume; the smaller the K value is, the better the uniformity of the asphalt mixture is; conversely, the worse the uniformity of the asphalt mixture is.

[0016] See also Figure 4 The K values ​​of different types of SMA-13 ​​asphalt mixtures ranged from 0.06 to 0.13. The K values ​​of both the finely separated recycled mixture and the conventionally crushed and screened recycled mixture were higher than those of the new mixture, indicating that the incorporation of RAP reduced the uniformity of the recycled mixture. The K value of the finely separated recycled mixture was 26.1% lower than that of the conventionally crushed and screened recycled mixture, indicating that the finely separated recycled mixture had better uniformity. This is because the fine separation technology reduces RAP agglomeration, which facilitates the falling of RAP surface particles during mixing, resulting in a better dispersion uniformity in the corresponding recycled mixture.

[0017] Example 2: The steps are the same as those in Example 1, except that the Marshall specimens are new AC-13, conventional crushed and screened RAP recycled AC-13 mixture, and refined separated RAP recycled mixture. Figure 5 As shown in the figure, the K values ​​of the three asphalt mixtures are as follows: Figure 6 shown.

[0018] See also Figure 6The K value of the new material is the lowest, indicating that its dispersion uniformity is the best. The K value of the finely separated recycled mixture is 30.4% lower than that of the conventional crushed and screened recycled mixture, which is consistent with the law shown in Example 1.

[0019] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the above embodiments do not limit the present invention in any form, and any technical solutions obtained by equivalent replacement or equivalent transformation fall within the scope of protection of the present invention.

Claims

1. A method for evaluating the uniformity of asphalt mixture based on CT scanning technology, characterized in that: The specific steps are as follows: S1. Obtain an asphalt mixture cylindrical specimen, scan it using a CT scanner, and obtain a slice image of the specimen; S2. constructing a three-dimensional visual model of the specimen using the slice image obtained in step S1; S3. Perform image processing on the three-dimensional visualization model to identify components, remove asphalt mortar and voids, and retain aggregate particles; S4. Establish an XYZ coordinate system with the geometric center of the cylindrical specimen as the origin and the height direction as the Z axis; S5. Divide the cylindrical specimen into N equal cubic grids. To ensure that a single grid can accommodate at least one largest aggregate and two smallest aggregates, define the minimum unit size. The calculation formula is as follows: （1）； in is the minimum unit size; is the maximum aggregate size in the specimen; is the minimum aggregate size in the specimen; 、 、 is the number of grids in each dimension, rounded down; R is the specimen radius; H is the specimen height; N is the total number of grids. For boundary grids, only the grids with a specimen volume coverage exceeding 50% are counted; S6. Divide the aggregates in the asphalt mixture into multiple grades based on the equivalent diameter of the aggregates; S7. For the i-th aggregate, calculate the volume proportion of the aggregate in each grid, count the volume proportions in all grids, calculate the uniformity coefficient, and finally weight the uniformity index K of each aggregate. The calculation formula is as follows (2): （2）； in, is the uniformity coefficient of each aggregate; N is the total number of grids; is the bth grid Volume proportion of aggregate; is the volume proportion of the first-tier aggregate in all grids; is a minimum constant; It is the ratio of the volume of the i-th grade aggregate to the total aggregate volume; the smaller the K value is, the better the uniformity of the asphalt mixture is; conversely, the worse the uniformity of the asphalt mixture is.

2. The asphalt mixture uniformity evaluation method based on CT scanning technology according to claim 1 is characterized in that: In the step S7, ； in, is the volume of the i-th aggregate in the b-th grid; It is the volume of a single voxel, determined by the CT resolution; is the number of voxels of the i-th aggregate in the b-th grid. The position coordinates of all aggregates are derived from AVIZO software, and the number of voxels in each grid is counted using MATLAB software.

3. The asphalt mixture uniformity evaluation method based on CT scanning technology according to claim 1 is characterized in that: In step S1, the radius of the cylindrical specimen is ≥50 mm, and the height is ≥50 mm; the CT tomography scanning direction is the height direction of the specimen, and the spacing is ≤1.5 mm.

4. The asphalt mixture uniformity evaluation method based on CT scanning technology according to claim 1 is characterized in that: In step S2, the three-dimensional visualization model of the specimen is constructed using AVIZO software.

5. The asphalt mixture uniformity evaluation method based on CT scanning technology according to claim 1 is characterized in that: In step S3, the specific steps of performing image processing on the three-dimensional visualization model are as follows: (1) Using AVIZO software to read and convert grayscale images of asphalt mixture CT scans; (2) Using grayscale conversion to adjust the contrast and brightness of grayscale images and using median filtering to remove granular noise in asphalt mixture digital images to achieve image enhancement; (3) Use interactive threshold segmentation and slice editing to separate, check and correct aggregates in 5 to 10 consecutive fault images, which serve as material training models for deep learning; (4) Use the trained deep learning model to pre-identify the components of the remaining tomographic images to enhance the image contrast between the components; (5) Based on the bimodal method, the gray value boundary of the components is determined, and the binary images of the aggregate, asphalt mortar, and void components are separated from the image using the gray threshold; (6) The adhered particles are separated by morphological opening operation.

6. The asphalt mixture uniformity evaluation method based on CT scanning technology according to claim 1 is characterized in that: In step S6, the equivalent diameter of the aggregate is the diameter of a sphere with the same volume as the aggregate.

7. The asphalt mixture uniformity evaluation method based on CT scanning technology according to claim 1 is characterized in that: In step S6, the aggregates in the asphalt mixture are divided into four grades according to the equivalent diameter of the aggregates: 2.36 mm to 4.75 mm, 4.75 mm to 9.5 mm, 9.5 mm to 13.2 mm, and above 13.2 mm.

Images