Method for predicting chloride diffusion in cement-based materials based on multiscale pore images

Abstract

The application discloses a cement-based material chloride ion diffusion prediction method based on multi-scale pore images, and comprises the following steps: obtaining two-dimensional sequence images of a cement-based material sample under two different scales through a focused ion beam / scanning electron microscope method and an X-ray computed tomography method; pre-processing the two-dimensional sequence images; using Avizo software to perform three-dimensional modeling on the two-dimensional sequence images to generate two three-dimensional pore models; separating the three-dimensional pore models to generate separated pore models and connected pore models; quantitatively analyzing to obtain pore structure characteristic parameters of the cement-based material sample; and using a lattice Boltzmann method to simulate chloride ion diffusion. The method is based on the multi-scale characteristics of the microstructure of the cement-based material, three-dimensional pore images under two scales are established through the XCT and FIB-SEM methods, multi-scale pore structure characteristic parameters of the cement-based material are characterized, and the pore characteristics of the cement-based material are truly reflected.

Description

Technical Field

[0001] This invention belongs to the field of cement-based material durability research, and specifically relates to a method for predicting chloride ion diffusion in cement-based materials. Background Technology

[0002] Concrete, a cement-based material, is widely used in the construction industry and plays a vital role in the national economy. The degradation of reinforced concrete structural performance due to steel corrosion is one of the most significant threats to structural durability, especially in marine engineering projects. Therefore, for reinforced concrete structures in marine environments, the chloride ion diffusion coefficient is commonly used as an important durability indicator characterizing the concrete's resistance to chloride ion corrosion. Currently, traditional testing methods for confirming the chloride ion diffusion coefficient include the natural diffusion method, the electrical conductivity method, and the electrically accelerated chloride ion migration method. However, traditional testing methods have drawbacks such as being time-consuming, subject to complex influencing factors, and having large errors.

[0003] With the deepening research on the durability of cement-based materials, pore structure has been recognized as a crucial microstructural feature, considered the most direct factor influencing chloride ion transport. Many empirical formulas and models have been applied to permeation simulations, including the Carman-Kozeny model and the Katz-Thompson model. While these empirical formulas offer convenience, they cannot analyze the interaction mechanisms of pore structures at different components and scales. Therefore, numerical methods for predicting the chloride ion diffusion coefficient of cement-based materials based on pore structure are becoming a popular research direction.

[0004] However, cement-based materials exhibit a wide range of pore sizes and complex pore structures, spanning macroscopic dimensions (>10). -1 m), detailed observation (10 -4 ~10 -1 m), microscopic (10 -8 ~10 -4 The pore structure is typically characterized by three scales: porosity, pore size distribution, pore specific surface area, characteristic pore size, pore connectivity, and pore tortuosity. Therefore, accurate experimental characterization of the pore structure of cement-based materials and simulation of fluid and ion transport based on real pore structures are pressing issues for predicting the chloride ion diffusion coefficient of cement-based materials at the microscopic scale.

[0005] Therefore, there is an urgent need for a method that can better simulate the transport phenomenon of chloride ions in pores, under the premise of more accurate experimental characterization of the pore structure of cement-based materials, so as to predict the chloride ion diffusion performance of cement-based materials. Summary of the Invention

[0006] Based on the above-mentioned shortcomings and deficiencies in the existing technology, the present invention provides a method for predicting the chloride ion diffusion properties of cement-based materials based on multi-scale pore structure. This method can accurately characterize the micron and nanoscale pore structure of cement-based materials and more accurately simulate the transport phenomenon of chloride ions in the pores, thereby achieving the goal of accurately predicting the chloride ion diffusion performance of cement-based materials.

[0007] To achieve the above-mentioned objectives, the present invention adopts the following technical solution:

[0008] A method for predicting chloride ion diffusion in cement-based materials based on multi-scale pore images, comprising:

[0009] S1. Two-dimensional sequential images of cement-based material samples with micron and nanometer precision were obtained by X-ray computed tomography (XCT) and focused ion beam / scanning electron microscopy (FIB / SEM), respectively.

[0010] S2. Preprocess the two-dimensional image sequence;

[0011] S3. Avizo software was used to generate two 3D porosity models from the preprocessed 2D sequence images.

[0012] S4. Perform pore segmentation on the pore 3D model to generate separate pore models and connected pore models;

[0013] S5. Use the separated pore model and the connected pore model to perform quantitative analysis and obtain the pore structure characteristic parameters of the cement-based material samples;

[0014] S6. Using the lattice Boltzmann method, the diffusion of chloride ions is simulated based on two three-dimensional pore models and pore structure characteristic parameters.

[0015] In a preferred embodiment, step S4 includes the following steps:

[0016] S41. Use the overflow method to obtain the pore segmentation threshold of the pore 3D model;

[0017] S42. Use the pore segmentation threshold to segment the pores to obtain a two-dimensional pore slice image and a three-dimensional pore structure image.

[0018] As a further preferred embodiment, step S5 includes the following steps:

[0019] S51. Perform quantitative analysis on the two-dimensional slice diagram and three-dimensional structural diagram of the pores, and calculate the porosity;

[0020] S52. Use the 3D structure diagram to calculate the percentage of connected pore volume to the total pore volume, and generate a 3D rendering of connected pores.

[0021] S53. Divide a single hole in the 3D rendering of the connected pores into multiple interconnected small holes to generate a pore network model.

[0022] S54. Calculate the pore structure characteristic parameters of the pore network model.

[0023] As a further preferred embodiment, the pore structure characteristic parameters include:

[0024] Porosity, pore size distribution, pore shape, pore orientation, connectivity, and tortuosity are all considered in this study.

[0025] In a preferred embodiment, step S6 includes the following steps:

[0026] S61. Convert the two pore 3D models into a digital sample matrix;

[0027] S62. Use pore structure characteristic parameters to define the initial conditions of each grid point in the digital sample matrix;

[0028] S63. Use pore structure characteristic parameters to define the boundary conditions of each grid point in the digital sample matrix;

[0029] S64. Simulate the collision and migration process of particles in the digital sample matrix according to the discrete equation, and update the chloride ion concentration at each node until the steady-state diffusion condition is reached.

[0030] S65. Macroscopic parameters of each grid point in the statistical numerical sample matrix.

[0031] As a preferred implementation, the lattice Boltzmann method in step S6 uses the D3Q7 model.

[0032] As a preferred embodiment, for two-dimensional sequence images acquired by the focused ion beam / scanning electron microscopy method, step S2 uses the following preprocessing steps:

[0033] S21. Correct image resolution;

[0034] S22. Correct the relative position of the image;

[0035] S23, Image noise reduction.

[0036] As a preferred embodiment, for two-dimensional sequence images acquired by X-ray computed tomography, step S2 uses the following preprocessing steps:

[0037] S24. Convert to an 8-bit grayscale image;

[0038] S25. Use a median filter for noise reduction.

[0039] As a preferred embodiment, the focused ion beam / scanning electron microscopy method uses a 5×5×3mm sheet-like cement-based sample;

[0040] The X-ray computed tomography method uses a cylindrical cement-based sample with a diameter of 2 mm and a height of 2 mm.

[0041] Compared with the prior art, the beneficial effects of this invention are:

[0042] The method for predicting chloride ion diffusion in cement-based materials of this invention establishes three-dimensional pore images at two scales: micron and nanometer, using both X-ray computed tomography and focused ion beam / scanning electron microscopy. Therefore, the method of this invention can accurately characterize the pore structure parameters of cement-based materials at multiple scales, truly reflecting the pore characteristics of cement-based materials. Furthermore, it combines the advantages of the lattice Boltzmann method, such as easy description of fluid interactions, easy setting of complex boundaries, and ease of parallel computing, to simulate the diffusion performance of chloride ions within the pores, calculate the chloride ion diffusion coefficient, and then use the chloride ion diffusion coefficient to help predict the resistance of cement-based materials to chloride ion erosion. Attached Figure Description

[0043] Figure 1 The flowchart shown is a method for predicting chloride ion diffusion in cement-based materials according to the present invention.

[0044] Figure 2 The image shown is a two-dimensional cross-sectional view of the pores in an embodiment of the present invention;

[0045] Figure 3 The diagram shown is a three-dimensional pore structure diagram of an embodiment of the present invention;

[0046] Figure 4 The image shown is a three-dimensional rendering of the connecting pores in an embodiment of the present invention;

[0047] Figure 5 The image shows a pore network model according to an embodiment of the present invention;

[0048] Figure 6 The figure shows the concentration distribution of chloride ions diffused in different directions according to an embodiment of the present invention. Detailed Implementation

[0049] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the figures in the embodiments of this application.

[0050] The following description provides several embodiments of this application. Different embodiments can be substituted or combined. Therefore, this application can also be considered to include all possible combinations of the same and / or different embodiments described. Thus, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then this application should also be considered to include embodiments containing one or more other possible combinations of A, B, C, and D, even if such embodiments are not explicitly described in the following text.

[0051] The following description provides examples and does not limit the scope, applicability, or examples set forth in the claims. Changes may be made to the function and arrangement of the described elements without departing from the scope of this application. Various processes or components may be appropriately omitted, substituted, or added to the examples. For example, the described methods may be performed in a different order than described, and various steps may be added, omitted, or combined. Furthermore, features described with respect to some examples may be combined into other examples.

[0052] This application provides a method for predicting chloride ion diffusion in cement-based materials, the flowchart of which is shown below. Figure 1 As shown, it includes the following steps:

[0053] S1. Two-dimensional sequential images of cement-based material samples with micron and nanometer precision were obtained by X-ray computed tomography (XCT) and focused ion beam / scanning electron microscopy (FIB / SEM), respectively.

[0054] S2. Preprocess the two-dimensional image sequence;

[0055] S3. Avizo software was used to generate two 3D porosity models from the preprocessed 2D sequence images.

[0056] S4. Perform pore segmentation on the pore 3D model to generate separate pore models and connected pore models;

[0057] S5. Use the separated pore model and the connected pore model to perform quantitative analysis and obtain the pore structure characteristic parameters of the cement-based material samples;

[0058] S6. Using the lattice Boltzmann method, the diffusion of chloride ions is simulated based on two three-dimensional pore models and pore structure characteristic parameters.

[0059] In this embodiment, step S1 uses different cement-based samples depending on the imaging method and scale. Specifically, for the focused ion beam / scanning electron microscopy (FIB-SEM) method, step S1 uses the following method to prepare the cement-based sample and acquire it as a sequence of images.

[0060] Cement paste sheets with a size of approximately 5×5×3mm were used for hydration termination and vacuum drying, followed by sample pretreatment such as pre-grinding, vacuum impregnation, grinding and polishing, and vacuum drying.

[0061] A field emission scanning electron microscope (FSE) with a focused ion beam system, model Crossbeam 540, manufactured by Carl Zeiss AG, Germany, was used. The argon-ion polished and gold-sprayed sample was fixed on the sample stage, perpendicular to the electron beam. The electron beam was turned on, and a target area of ​​11 × 10 μm was selected in SE mode, avoiding areas with microcracks. A positioning mark was made on the selected target area, and platinum was deposited on its upper surface.

[0062] The target area is etched with an ion beam to the front, left and right sides, with a depth of 20 μm and a width of 30 μm. Then, the imaging surface is precisely cut with a small current ion beam until the surface is flat.

[0063] The ion beam was used to cut the surface every 10 nm, and the fresh cut surfaces were imaged. The parameters were set as follows: gallium ion beam operating voltage of 30 kV and operating current of 700 pA, electron beam operating voltage of 1.5 kV and operating current of 500 pA. The cutting was repeated 1000 times, resulting in 1048 two-dimensional sequential images with a resolution of 5 nm / voxel.

[0064] For the X-ray computed tomography (XCT) method, step S1 uses the following method to prepare a cement-based sample and acquire it as a sequence of images.

[0065] A cylindrical cement-based sample with a diameter of 2 mm and a height of 2 mm was used. The target area of ​​the sample was photographed with a lens of lower magnification, and then further determined with a lens of higher magnification.

[0066] After setting appropriate X-ray voltage, current, and focal spot size, and adjusting the position parameters, the sample was ensured to be completely scanned. The instrument used a 4032×2688 pixel CCD detector, with a distance of 9.735 mm between the X-ray source and the sample, and 146.813 mm between the X-ray source and the detector; the voltage was 80 kV, the current was 280 μA, and the exposure time was 1800 ms. The reconstructed image size was 4032×4032 pixels, with a resolution of 0.60 μm / voxel. After selecting appropriate image capture parameters, X-ray computed tomography (XCT) data acquisition began, resulting in a two-dimensional sequence of images.

[0067] In this embodiment, step S2 also involves different preprocessing procedures based on the two-dimensional sequence images obtained by the two methods of focused ion beam / scanning electron microscopy (FIB-SEM) and X-ray computed tomography (XCT).

[0068] Specifically, for two-dimensional sequence images acquired by the focused ion beam / scanning electron microscopy (FIB-SEM) method, step S2 uses the following preprocessing steps:

[0069] S21. Correct image resolution;

[0070] S22. Correct the relative position of the image;

[0071] S23, Image noise reduction.

[0072] The specific implementation of the above steps is as follows:

[0073] S21. Correcting image resolution: When the focused ion beam / scanning electron microscope (FIB-SEM) is used to take pictures, the electron beam is not perpendicular to the observation plane, but at a 54° angle. Therefore, the following formula is used to correct the resolution of the original image in the y direction.

[0074]

[0075] Among them, R y R′ represents the resolution in the y-direction of the original image. y This represents the true spatial resolution in the y-direction.

[0076] S22. Correcting the relative position of images: The least squares method is used to compare the pixels of the two images to check the similarity between the two slices until the highest similarity is obtained, that is, to make the best position adjustment between the two images.

[0077] S23. Image Denoising: In the Avizo software, the cropped cubic region data volume undergoes three denoising processes: Fast Fourier Transform (FFT) Filter, Non-Local Means Filter, and Median Filter. These three filtering processes effectively eliminate noise. The denoised image then requires brightness adjustment, which is performed automatically using the Background Detection Correction command in Avizo.

[0078] Specifically, for two-dimensional sequence images acquired by X-ray computed tomography (XCT), step S2 uses the following preprocessing steps:

[0079] S24. Convert to an 8-bit grayscale image;

[0080] S25. Use a median filter for noise reduction.

[0081] The specific implementation of the above steps is as follows:

[0082] S24. The original image of X-ray computed tomography (XCT) is generally a 16-bit grayscale image. In order to segment it after obtaining the pore threshold, the 16-bit grayscale image is converted to 8-bit using the convert image type in Avizo.

[0083] S25. Use the median filter in Avizo to reduce noise in the converted image.

[0084] This embodiment also provides a specific implementation of steps S3-S5, wherein step S3 uses the 3D modeling software Avizo to import the 2D sequence image after preprocessing in step S2 and generate a pore 3D model.

[0085] Step S4 includes the following steps:

[0086] S41. Use the overflow method to obtain the pore segmentation threshold of the pore 3D model;

[0087] S42. Use the pore segmentation threshold to segment the pores to obtain a two-dimensional pore slice image and a three-dimensional pore structure image.

[0088] Specifically, step S41, based on the image grayscale after noise reduction in step S2, uses the "overflow method" to obtain the pore segmentation threshold, and then uses the pore segmentation threshold to segment the 3D pore model, resulting in the following: Figure 2 The two-dimensional slice diagram of the pores shown and as follows Figure 3 The diagram shows the three-dimensional structure of the pores. Among them, Figure 2 The image on the left is a two-dimensional section of pores obtained using the focused ion beam / scanning electron microscopy (FIB-SEM) method. Figure 2 The image on the right is a two-dimensional slice of the pore obtained by X-ray computed tomography (XCT). Figure 3 The image on the left shows a three-dimensional structure of the pores obtained using the focused ion beam / scanning electron microscopy (FIB-SEM) method. Figure 3 The image on the right is a three-dimensional structure diagram of the pores obtained by X-ray computed tomography (XCT).

[0089] Step S5 includes the following steps:

[0090] S51. Perform quantitative analysis on the two-dimensional slice diagram and three-dimensional structural diagram of the pores, and calculate the porosity;

[0091] S52. Use the 3D structure diagram to calculate the percentage of connected pore volume to the total pore volume, and generate a 3D rendering of connected pores.

[0092] S53. Divide a single hole in the 3D rendering of the connected pores into multiple interconnected small holes to generate a pore network model.

[0093] S54. Calculate the pore structure characteristic parameters of the pore network model.

[0094] Specifically, step S51 performs quantitative analysis on the two-dimensional slice image and three-dimensional structure image of the pores. The porosity of the two-dimensional slice image and three-dimensional structure image is obtained by using Avizo's Volume Fraction tool. The ratio of the number of voxel points representing pores to the total number of voxel points of the entire cube is used as the porosity.

[0095] Step S52 uses Avizo's Axis Connectivity tool, setting the direction to the z-axis, to obtain the connectivity of the pore structure in the z-direction, i.e., the percentage of connected pore volume to the total pore volume, as shown below. Figure 4 The three-dimensional rendering of the connected pores shown.

[0096] S53 uses the Separate Objects tool in Avizo to divide a single large hole into multiple connected small holes in the data volume of connected holes. Then, it uses the Generate Pore Network Model in Avizo to establish a model for the newly generated data volume. Figure 5 The pore network model shown is a simplification of the complex pore structure of cement-based materials. It simplifies the irregular pore structure into pores and throats, where pores represent larger spaces and throats represent relatively narrow spaces.

[0097] Based on the established pore network model, S54 obtains pore structure characteristic parameters such as porosity, pore diameter distribution, pore shape, pore orientation, connectivity, and tortuosity.

[0098] This embodiment also provides a specific implementation of step S6. In this embodiment, step S6 uses the D3Q7 model in the lattice Boltzmann model to simulate the chloride ion diffusion performance. The evolution equation of the particle distribution function in the D3Q7 model is as follows:

[0099]

[0100] In the formula: i represents the discretized diffusion direction, with orientation [0,1,2,3,4,5,6], indicating 7 different directions; e i f is the discretized diffusion rate of chloride ions along the i-th direction; the left side of the equation is the migration term, and the right side is the collision term. i (x,t) is the distribution function of the particle at position x in the i-th direction at time t; f i (x+e i Δt, t+Δt) is the distribution function of the particle at position x in the i-th direction at time t+Δt; τ is the relaxation time; f i eq(x,t) is the equilibrium distribution function of the particle at the same position at time t; Δt is the time increment.

[0101] The chloride ion concentration C(x,t) and the distribution function satisfy Chloride ion diffusion is considered a pure diffusion problem; therefore, the seven discrete directions (i = 0, 1, 2, 3, 4, 5, 6) can satisfy the accuracy requirements. The equilibrium distribution function is:

[0102]

[0103] Relaxation time τ and chloride ion diffusion coefficient D of pore fluid p The following relationship must be satisfied:

[0104]

[0105] Chloride ion diffusion flux J e Satisfy the following formula:

[0106]

[0107] The effective diffusion coefficient of chloride ions, De, satisfies the following equation:

[0108]

[0109] Where L and ΔC are the length of the channel and the chloride ion concentration difference between the inlet and outlet, respectively.

[0110] In one specific implementation of step S6, the steps are as follows:

[0111] S61. Convert the two pore 3D models into a digital sample matrix;

[0112] S62. Use pore structure characteristic parameters to define the initial conditions of each grid point in the digital sample matrix;

[0113] S63. Use pore structure characteristic parameters to define the boundary conditions of each grid point in the digital sample matrix;

[0114] S64. Simulate the collision and migration process of particles in the digital sample matrix according to the discrete equation, and update the chloride ion concentration at each node until the steady-state diffusion condition is reached.

[0115] S65. Macroscopic parameters of each grid point in the statistical numerical sample matrix.

[0116] Specifically, step S61 uses the imread function in Matlab to convert the separate pore model and the connected pore model into a digital sample matrix;

[0117] Step S62: Based on the relation Convert Dp (i.e., the chloride ion diffusion coefficient of the pore liquid, which is a known quantity) into relaxation time; based on the initial concentration value and the equation Make the initial particle distribution function f of all nodes i (x,t) is equal to the initial equilibrium distribution function f i eq (x,t), i.e., f i (x,t)=f i eq (x,t) is used to define the initial conditions for each grid point in the digital sample matrix.

[0118] Step S63 calculates the boundary where the normal diffusion flux is zero, and the relationship is as follows:

[0119] Where α and β are perpendicular to the boundary and opposite directions, and β is the direction facing the interface; the boundary between the inlet and outlet is represented by the following formula: Where C0 represents the chloride ion concentration at the inlet and outlet. The boundary conditions for each grid point in the digital sample matrix are defined using the above relationship.

[0120] Step S64 implements the particle collision and migration process based on discrete equations, f i (x,t) will move along the lattice direction and reach an adjacent node; this process is called migration. When particles from different directions reach the same lattice node, they will collide with each other, and the original particles moving in each direction will be redistributed; this process is usually called collision. According to the evolution equation of the particle distribution function... By rearranging terms, f i Move (x,t) to the right side of the equation and let According to f i (x+e i Δt, t+Δt)=f i * (x,t), distribution function f i * (x,t) is transferred to x+e i The adjacent nodes at Δt become the new distribution function.

[0121] After each collision and migration calculation is completed, according to the equation Update the concentration C(x,t) at each node to obtain the equilibrium distribution function at time step t+Δt;

[0122] Repeat the above collision, migration, and concentration update process until steady-state diffusion conditions are reached, according to the formula. To obtain the chloride ion diffusion flux J during stable transport across the entire pore structure e According to the formula Calculate the effective diffusion coefficient D of the pore structuree .

[0123] Furthermore, after calculating the macroscopic parameters of each grid point in the statistical digital sample matrix, step S65 also outputs the calculation results and analyzes the concentration distribution of chloride ions diffused along different directions, as shown in the example below. Figure 6 The visualization shown is shown below.

[0124] This invention applies X-ray computed tomography (XCT) and focused ion beam scanning electron microscopy (FIB-SEM) combined with electronic image processing techniques and the lattice Boltzmann method to predict the chloride ion diffusion coefficient of cement-based materials. XCT is used to characterize large capillaries, pores, and cracks in cement-based materials, while FIB-SEM is used to characterize images at nanoscale precision. This creates three-dimensional pore images at two scales, more realistically reflecting the pore characteristics of cement-based materials and revealing more pore feature details without damaging the original sample, ensuring high repeatability. Furthermore, the lattice Boltzmann method leverages its advantages—ease of describing fluid interactions, ease of setting complex boundaries, and ease of parallel computation—to simulate the diffusion performance of chloride ions within pores, calculate the chloride ion diffusion coefficient, and then use this coefficient to help predict the resistance of cement-based materials to chloride ion erosion.

[0125] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Other embodiments of this disclosure will be readily apparent to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described herein. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A method for predicting chloride ion diffusion in a cement-based material based on multi-scale pore images, characterized by, The method comprises the following steps: S1, obtaining two-dimensional sequence images of micron level and nanometer precision of a cement-based material sample by X-ray computed tomography or focused ion beam scanning electron microscopy method respectively; S2, preprocessing the two-dimensional sequence images; S3, generating two kinds of pore three-dimensional models by using Avizo software on the preprocessed two-dimensional sequence images; S4, segmenting the pore three-dimensional models to generate separated pore models and connected pore models; S5, using the separated pore models and the connected pore models for quantitative analysis to obtain pore structure characteristic parameters of the cement-based material sample; S6, using lattice Boltzmann method to simulate chloride ion diffusion according to the two kinds of pore three-dimensional models and the pore structure characteristic parameters; The step S4 comprises the following steps: S41, obtaining a pore segmentation threshold of the pore three-dimensional model by using overflow method; S42, segmenting pores by using the pore segmentation threshold to obtain pore two-dimensional slice images and pore three-dimensional structure images; The step S5 comprises the following steps: S51, performing quantitative analysis on the pore two-dimensional slice images and the three-dimensional structure images to calculate porosity; S52, using the three-dimensional structure images to calculate the percentage of connected pore volume in total pore volume to generate a connected pore three-dimensional rendering image; S53, dividing a single hole in the connected pore three-dimensional rendering image into a plurality of connected small holes to generate a pore network model; S54, calculating pore structure characteristic parameters of the pore network model; The pore structure characteristic parameters comprise: porosity, connected distribution algorithm pore size distribution, pore shape, pore orientation, connectivity and tortuosity; The step S6 comprises the following steps: S61, converting the two kinds of pore three-dimensional models into a digital sample matrix; S62, defining initial conditions of each grid point in the digital sample matrix by using the pore structure characteristic parameters; S63, defining boundary conditions of each grid point in the digital sample matrix by using the pore structure characteristic parameters; S64, simulating collision and migration process of particles in the digital sample matrix according to a discrete equation, updating chloride ion concentration at each node until reaching a steady-state diffusion condition; S65, calculating macroscopic parameters of each grid point in the digital sample matrix; The lattice Boltzmann method of the step S6 uses a D3Q7 model.

2. The method of predicting chloride ion diffusion of a cement-based material based on multi-scale pore images according to claim 1, wherein, For the two-dimensional sequence images obtained by the focused ion beam scanning electron microscopy method, the step S2 uses the following steps for preprocessing: S21, correcting image resolution; S22, correcting image relative position; S23, image denoising.

3. The method of predicting chloride diffusion in cement-based materials based on multi-scale pore images according to claim 1, wherein, For the two-dimensional sequence images obtained by the X-ray computed tomography, the step S2 uses the following steps for preprocessing: S24, converting into an 8-bit gray scale image; S25, using a median filter for denoising.

4. The method of predicting chloride ion diffusion in cement-based materials based on multi-scale pore images according to claim 1, wherein, The focused ion beam scanning electron microscopy method uses a sheet-shaped cement-based sample with a size of 5*5*3 mm; The X-ray computed tomography method uses a cylindrical cement-based sample with a diameter of 2 mm and a height of 2 mm.

Images