Numerical simulation method and system for chloride ion erosion of recycled concrete based on recursive region segmentation and five-phase meso model

Abstract

This invention discloses a method for constructing a microscopic model of recycled concrete and simulating chloride ion erosion based on recursive region segmentation and random convex polygons. The method includes: initializing model parameters; generating a first layer of aggregate and its intermediate zone (ITZ) using a subtractive placement strategy based on recursive region segmentation; subtracting the aggregate and its ITZ from the effective placement area using Boolean subtraction to update the effective placement area; generating a second layer of aggregate candidates within the remaining area; performing adhesion screening by calculating the Boolean intersection area of ​​the second layer aggregate with the first layer aggregate; merging all the first layer aggregates and shifting them outwards to generate a new ITZ; constructing a five-phase microscopic geometric model and exporting it as a standard file; importing the geometric model into simulation software; assigning different diffusion coefficients to the five phases; and performing transient simulation of chloride ion erosion. This invention significantly improves aggregate placement efficiency through recursive region segmentation and Boolean operations, realistically reflecting the multi-layered structure and interconnected interface areas of recycled concrete, and achieving seamless integration of modeling and simulation.

Description

Technical Field

[0001] This invention relates to the field of numerical simulation technology for building materials, and in particular to a numerical simulation method and system for chloride ion erosion of recycled concrete based on recursive region segmentation and a five-phase microscopic model. Background Technology

[0002] Recycled concrete refers to concrete made by crushing, cleaning, and grading waste concrete blocks, then mixing them with aggregates in a certain proportion to partially or completely replace natural aggregates. As a green and environmentally friendly building material, recycled concrete has significant advantages in saving resources and reducing construction waste, and its research and application have become a key focus in the current construction industry.

[0003] The performance of recycled concrete largely depends on its microstructural characteristics. At the microscale, recycled concrete can be considered a multiphase composite material composed of natural aggregates, attached old mortar, fresh mortar, and different interfacial transition zones (ITZs). The morphology, gradation, and spatial distribution of the aggregates, as well as the thickness and connectivity of the interfacial transition zones, have a decisive impact on the mechanical and durability properties of recycled concrete. Therefore, establishing a numerical model that can realistically reflect the complex microstructure of recycled concrete is an important means to deeply study the evolution of its macroscopic properties.

[0004] However, existing methods for microstructure modeling of recycled concrete have the following technical drawbacks: First, aggregate morphology is distorted. Traditional aggregate modeling methods often use simplified shapes such as circles, ellipses, or regular polygons to simulate aggregates, which cannot accurately reflect the irregular angular shapes formed after recycled aggregates are crushed. This morphological simplification leads to distortion of stress distribution in subsequent mechanical analysis and deviations in diffusion paths in chloride ion transport simulations, affecting the accuracy of performance predictions.

[0005] Second, aggregate placement efficiency is low. Existing technologies generally employ a random placement-overlap detection method, which involves randomly generating aggregate positions within the entire model area and then detecting whether they overlap with already generated aggregates. As the number of aggregates and the volume fraction increase, the collision detection failure rate rises sharply, the algorithm converges slowly, and it may even be difficult to generate high-density aggregate models, severely limiting modeling efficiency.

[0006] Third, the interface transition zone is poorly treated. The interface transition zone is the weakest link in concrete, with high porosity and initial microcracks, making it a key channel for the rapid penetration of harmful media such as chloride ions. However, existing technologies often ignore or simplify the interface transition zone, treating it as a single-layer unit or a constant-thickness ring. This makes it impossible to accurately construct an interconnected and complex ITZ network, and thus difficult to realistically simulate the impact of interface effects on the durability of recycled concrete.

[0007] Fourth, the structural hierarchy is simplified. Recycled concrete has a multi-layered composite structure of natural aggregate, old mortar, and new mortar, including multiple interfaces such as the old interface transition zone between aggregate and old mortar, and the new interface transition zone between old mortar and new mortar. Existing technologies mostly focus on simulating single-layer aggregates, lacking effective modeling of the multi-layered structure of recycled aggregates and their interactions, which makes it difficult to meet the research needs of the complex multiphase structure of recycled concrete.

[0008] Fifth, the modeling and simulation processes are disconnected. In existing technologies, there is a lack of standardized data interfaces between geometric modeling and subsequent numerical simulation software. The generated geometric models require a lot of manual repair and conversion before they can be used for simulation analysis. The low degree of automation in the process limits the promotion of the models in engineering applications and performance analysis.

[0009] In summary, there is an urgent need to develop a method for efficiently generating random, multi-layered microstructural models of recycled concrete that include complete interface transition zones, and to achieve seamless integration with numerical simulation, so as to improve the accuracy of performance prediction of recycled concrete and the practicality of its engineering applications. Summary of the Invention

[0010] In view of this, the purpose of this invention is to provide a numerical simulation method and system for chloride ion erosion of recycled concrete based on recursive region segmentation and a five-phase microscopic model. This method adopts a subtractive placement strategy based on recursive region segmentation to solve the problem of low placement efficiency of high-density aggregates; and solves the problem of distortion in the modeling of multi-layer structures of recycled concrete by generating double-layer aggregates and Boolean intersection adhesion detection.

[0011] To achieve the above objectives, the present invention provides the following technical solution: The numerical simulation method for chloride ion attack in recycled concrete based on recursive region segmentation and a five-phase microscopic model provided by this invention includes the following steps: S1: Initialize model parameters, including model size, target aggregate volume fraction, aggregate particle size range, ITZ thickness, and recycled aggregate replacement rate. S2: The first layer of aggregate and its interface transition zone (ITZ) are generated using a subtractive placement strategy based on recursive region segmentation. Specifically, this includes: S21: Define the entire model area as the initial effective delivery area; S22: Generate a random convex polygon within the effective delivery area as the first layer of aggregate; S23: Translate each side of the first layer of aggregate outward by the thickness of the interface transition zone ITZ to generate a first layer interface transition zone ITZ polygon that encloses the aggregate. S24: Perform a Boolean union operation on the first layer of aggregate and its first layer interface transition zone ITZ polygon to obtain the occupied area, and subtract the occupied area from the current effective delivery area by Boolean, and update the remaining space to at least one new effective delivery sub-region that is not connected to each other. S25: Repeat steps S22 to S24 until the cumulative volume fraction of the first layer of aggregate reaches the target value calculated from the target aggregate volume fraction and the recycled aggregate replacement rate, or the effective placement area is empty. S3: Generate a second layer of aggregate, allowing it to adhere to or fill the gaps between the first layer of aggregate, specifically including: S31: Generate a random convex polygon with a particle size smaller than the first layer of aggregate within the current remaining effective delivery area, as a candidate for the second layer of aggregate; S32: Calculate the Boolean intersection area of ​​the second layer aggregate candidate and the union area of ​​all first layer aggregates. If the ratio of the intersection area to the area of ​​the second layer aggregate candidate itself meets the preset attachment threshold range, then the second layer aggregate candidate is accepted and the second layer aggregate is regarded as the natural aggregate phase. S33: Repeat steps S31 to S32 until the cumulative volume fraction of the second layer of aggregate reaches the target value; S4: Construct a five-phase micro-geometric model. Perform a Boolean union operation on all the first-layer aggregates to obtain the old mortar phase. Translate the boundary of the old mortar phase outwards by the thickness of the interface transition zone (ITZ) to generate a new interface transition zone (ITZ) phase that encapsulates all the old mortar phases. This results in the construction of a five-phase micro-geometric model that includes the natural aggregate phase, the old interface transition zone (ITZ) phase, the old mortar phase, the new interface transition zone (ITZ) phase, and the new mortar phase. S5: Export standard geometry file, export the boundary data of the five-phase micro-geometric model as a standard geometry file; S6: Perform a numerical simulation of chloride ion erosion. Import the standard geometric file into the finite element simulation software, assign different chloride ion diffusion coefficients to the five phases respectively, set erosion boundary conditions, perform a transient numerical simulation of chloride ion erosion, and obtain the chloride ion concentration distribution results.

[0012] Furthermore, the generation of random convex polygons in step S22 includes: The angles of each vertex are randomly generated in polar coordinates. and radius , where the radius A random perturbation factor is introduced to ensure the irregularity of the aggregate shape; the polar coordinates are converted to Cartesian coordinates to obtain the initial vertex set; the Andrew algorithm is used to calculate the convex hull of the initial vertex set to obtain the vertex sequence of the convex polygon.

[0013] Furthermore, step S23, which involves shifting the thickness of the interface transition zone ITZ outward from each edge of the first layer of aggregate, includes: The aggregate is translated a distance t along the outward normal direction of each side. The translated sides are extended to both ends, and the coordinates of the intersection points of two adjacent extension lines are obtained. The intersection points are connected in sequence to form a closed interface transition zone ITZ polygon, and the miter connection method is used to maintain the sharpness of the geometric features.

[0014] Furthermore, the preset adhesion threshold range in step S32 is 5% to 15%. When the ratio of the intersection area to the area of ​​the second layer aggregate candidate itself is less than the lower limit of the adhesion threshold range, the aggregate is determined to meet the adhesion filling conditions.

[0015] Furthermore, step S4 is followed by: Boolean operations are performed on the five-phase micro-geometric model to delete the intersecting parts between each phase, ensuring that the geometric boundaries of each phase are clear and do not overlap.

[0016] Furthermore, the standard geometry file mentioned in step S5 is a DXF format file.

[0017] Furthermore, step S6 involves assigning different chloride ion diffusion coefficients to the five phases, specifically including: The diffusion coefficient of the new mortar phase is set as the reference value D0; The diffusion coefficient of the ITZ phase in the new interface transition region is set to k1×D0, where k1 is the first multiple coefficient; The diffusion coefficient of the first layer of aggregate, i.e., the old mortar phase, is set as k2×D0, where k2 is the second multiple coefficient; The diffusion coefficient of the first-layer interface transition region ITZ, i.e. the old interface transition region ITZ phase, is set as k3×D0, where k3 is the third multiple coefficient. The diffusion coefficient of the second layer of aggregate, i.e., the natural aggregate phase, is set to approximately 0. Where k1, k2, and k3 are real numbers greater than 1, and k3 is greater than both k1 and k2.

[0018] Furthermore, the setting of erosion boundary conditions in step S6 includes: One boundary of the model is set as a chloride ion concentration boundary to simulate a chloride ion erosion source; the other boundaries of the model are set as no flux boundaries to simulate unidirectional erosion conditions.

[0019] Furthermore, prior to performing the transient numerical simulation of chloride ion corrosion in step S6, the following steps are also included: The five-phase micro-geometric model is meshed, and the interface transition zone (ITZ) is locally refined to ensure that at least 2-3 layers of mesh elements are distributed within the thickness of the interface transition zone (ITZ).

[0020] The present invention provides a system for constructing a microscopic model of recycled concrete and a numerical simulation system for chloride ion erosion based on recursive region segmentation and random convex polygons, comprising: The parameter initialization module is used to initialize model parameters, including model size, target aggregate volume fraction, aggregate particle size range, interface transition zone (ITZ) thickness, and recycled aggregate replacement rate. The first-layer aggregate generation module is used to generate the first-layer aggregate and its interface transition zone (ITZ) using a subtractive placement strategy based on recursive region segmentation. Specifically, it includes: Define the entire model area as the initial effective deployment area; A random convex polygon is generated within the effective delivery area as the first layer of aggregate; The thickness of the interface transition zone ITZ is translated outward from each side of the first layer of aggregate to generate a first layer interface transition zone ITZ polygon that encloses the aggregate. Perform a Boolean union operation on the first layer of aggregate and its first layer interface transition zone (ITZ) polygon to obtain the occupied area, and subtract the occupied area from the current effective delivery area in Boolean order, and update the remaining space with at least one new, unconnected effective delivery sub-area. Repeat the above operation until the cumulative volume fraction of the first layer of aggregate reaches the target value calculated from the target aggregate volume fraction and the recycled aggregate replacement rate, or the effective placement area is empty; The second-layer aggregate generation module is used to generate a second layer of aggregate, which is then attached to or fills the gaps in the first layer of aggregate. Specifically, it includes: Generate a random convex polygon with a particle size smaller than the first layer of aggregate within the current remaining effective delivery area, as a candidate for the second layer of aggregate; Calculate the Boolean intersection area of ​​the second aggregate candidate and the union area of ​​all first aggregates. If the ratio of the intersection area to the area of ​​the second aggregate candidate itself meets a preset adhesion threshold range, then the second aggregate candidate is accepted and the second aggregate is regarded as the natural aggregate phase. Repeat the above operation until the cumulative volume fraction of the second layer of aggregate reaches the target value; The five-phase model construction module is used to perform Boolean union operation on all the first-layer aggregates to obtain the old mortar phase, and to translate the boundary of the old mortar phase outward by the thickness of the interface transition zone ITZ to generate a new interface transition zone ITZ phase that encapsulates all the old mortar phases, thereby constructing a five-phase micro-geometric model containing the natural aggregate phase, the old interface transition zone ITZ phase, the old mortar phase, the new interface transition zone ITZ phase, and the new mortar phase. The data export module is used to export the boundary data of the five-phase microscopic geometric model as a standard geometric file; The numerical simulation module is used to import the standard geometric file into the finite element simulation software, assign different chloride ion diffusion coefficients to the five phases respectively, set erosion boundary conditions, perform transient numerical simulation of chloride ion erosion, and obtain chloride ion concentration distribution results.

[0021] The beneficial effects of this invention are as follows: This invention provides a numerical simulation method and system for chloride ion erosion in recycled concrete based on recursive region segmentation and a five-phase microscopic model. The method includes: initializing model parameters; generating a first layer of aggregate and its intermediate zone (ITZ) using a subtractive placement strategy based on recursive region segmentation; subtracting the aggregate and its ITZ from the effective placement area using Boolean subtraction to update the effective placement area; generating a second layer of aggregate candidates within the remaining area; performing adhesion screening by calculating the Boolean intersection area of ​​the second layer aggregate with the first layer aggregate; merging all first layer aggregates and shifting them outwards to generate a new ITZ; constructing a five-phase microscopic geometric model and exporting it as a standard file; importing the geometric model into simulation software; assigning different diffusion coefficients to the five phases; and performing transient simulation of chloride ion erosion. This invention significantly improves aggregate placement efficiency through recursive region segmentation and Boolean operations, realistically reflecting the multi-layered structure and interconnected interface areas of recycled concrete, and achieving seamless integration of modeling and simulation. The method provided by this invention is based on the construction of a microstructure model of recycled concrete using recursive region segmentation and random convex polygons, and simulation of chloride ion erosion using numerical simulation. The five-phase model allows for a realistic microstructure simulation of recycled concrete. This is achieved through a three-layer material system of natural aggregate-old mortar-new mortar and a three-layer interface system of old ITZ-new ITZ (natural aggregate side)-new ITZ (old mortar side), thus forming a complete representation of the realistic microstructure. The five phases of this invention, from the inside out, are natural aggregate, old interface transition zone, old mortar, new interface transition zone, and new mortar, thus constituting a five-phase structure in a material sense. This method has the following beneficial effects: First, it significantly improves modeling efficiency and model density. This invention employs a subtractive placement strategy based on recursive region segmentation, subtracting aggregates and their interface transition zones from the effective placement area using Boolean subtraction. This ensures that subsequent aggregate placement only occurs within the remaining effective geometric space, fundamentally eliminating the possibility of invalid attempts in already filled areas. Compared to the traditional trial-and-error method of random placement and overlap detection, this invention greatly improves the convergence speed of the algorithm in the later placement stages, enabling rapid generation of high-volume-fraction aggregate models and effectively solving the technical problem of low placement efficiency for high-density aggregates.

[0022] Secondly, this invention realistically reproduces the multi-layered composite structure of recycled concrete. Through a two-layer aggregate generation strategy, the first layer of aggregate simulates the old mortar in recycled aggregate, while the second layer simulates natural aggregate. Boolean intersection operations are used for adhesion screening, ensuring that the second layer of aggregate accurately fills the gaps between the first layer or adheres to its surface. This layered strategy more realistically simulates the complex microstructure of natural aggregate-old mortar-new mortar in recycled concrete, laying a solid geometric foundation for the accurate prediction of subsequent mechanical and durability properties.

[0023] Third, it accurately constructs a network of interconnected interface transition zones. This invention not only generates an interface transition zone of equal thickness for each aggregate independently, but also generates a new interface transition zone encompassing all old mortar areas by performing a Boolean union of all the first-layer aggregates and then translating it outwards. This hierarchical ITZ construction method can automatically generate complex interconnected interface zone networks, accurately simulating the spatial distribution characteristics of multiple interfaces in recycled concrete. This is crucial for studying the rapid penetration channels of harmful media such as chloride ions along interfaces, significantly improving the accuracy of durability simulation.

[0024] Fourth, it automates the entire modeling and simulation process. This invention breaks down the data barriers between modeling software and finite element simulation software by exporting the boundary data of the five-phase microstructure geometric model as a standard DXF format file. Users can directly obtain a high-fidelity simulation model with clear material partitioning, eliminating tedious manual repair and conversion work. This greatly improves the workflow efficiency from microstructure research to macroscopic performance prediction, providing a convenient and efficient technical tool for the material design and performance evaluation of recycled concrete.

[0025] The above and other objects, advantages, and features of the present invention will be more fully set forth and demonstrated through the following detailed description of specific embodiments in conjunction with the accompanying drawings. Those skilled in the art, upon referring to the following detailed description and the accompanying drawings, will be able to better understand and realize the above advantages of the present invention. Other objects, features, and advantages of the present invention will become clearer after being described in detail in the detailed description section in conjunction with the accompanying drawings. Attached Figure Description

[0026] To make the objectives, technical solutions, and beneficial effects of this invention clearer, the following drawings are provided for illustration.

[0027] Figure 1 Flowchart for generating random convex polygon aggregates for recycled concrete; Figure 2 A schematic diagram of a single recycled aggregate; Figure 3 A numerical model diagram of the five-phase microstructure of recycled concrete generated by the program; Figure 4The overall flow of the program for creating a random convex polygon aggregate model for recycled concrete; Figure 5 Boundary conditions for the five-phase microstructure numerical model of recycled concrete; Figure 6 This is a schematic diagram of the model mesh generation; Figure 7 This is a diagram showing the chloride ion concentration distribution in the microscopic numerical model of recycled concrete. Detailed Implementation

[0028] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments described are not intended to limit the present invention. Example 1

[0029] like Figure 1 As shown in the figure, the method for constructing a five-phase microscopic model of concrete and simulating chloride ion erosion based on recursive region segmentation and random convex polygons provided in this embodiment has the following specific steps: Step 1: Initialize Model Parameters Set the basic parameters of the concrete model: Model dimensions: Length X, Width Y; Target aggregate volume fraction r; Aggregate particle size range: minimum particle size Dmin, maximum particle size Dmax; Interface transition zone thickness t; Step 2: Convex Polygon Aggregate Generation Algorithm: 1. First, define the parameter space: the range of the number of vertices satisfies The size is controlled by the diameter of the circumscribed circle. Center point coordinates Located in the effective delivery area Inside.

[0030] 2. The steps for generating polar coordinate vertices are as follows: the angle of each vertex. It follows a uniform distribution, where, After sorting these angles, an ordered sequence is obtained. .radius The generation requires controlling the degree of vertex dispersion, expressed as: ; in, Represents a uniformly random number between 0 and 1; Represents the vertex number or index; in a polygon aggregate, it represents the k-th vertex currently being calculated. This represents the total number of vertices in the aggregate. It determines the number of sides of the random polygon; generally, the larger n is, the closer the aggregate shape is to a circle. This represents the nth polar angle after sorting. In generation algorithms, it is usually first... A series of angles are randomly generated and sorted from smallest to largest to form an ordered sequence. This is to ensure that the lines connecting the generated aggregate vertices do not self-intersect; This indicates the maximum particle size (or equivalent diameter) of the aggregate. It is a core parameter controlling aggregate size, determining the maximum span that the aggregate occupies in space. therefore Next, the polar coordinates are converted to Cartesian coordinates to obtain the coordinates of each vertex. and The calculation formula is: , .

[0031] This represents the polar angle corresponding to the k-th vertex. These angles are usually randomly generated and arranged in ascending order to ensure the convexity of the polygon. 3. Centering is performed to ensure that the generated convex polygon is located in a specified position.

[0032] First, calculate the centroid of the initial vertex set, whose coordinates are: .

[0033] Then, by translation, all vertices are moved toward the target center. The coordinates of the translated vertex are: .

[0034] and This represents the initial coordinates of the k-th vertex in Cartesian coordinates during the initial generation phase. These coordinates are obtained through the polar coordinate transformation formula... , Calculated; Represents the vertex number or index; in a polygon aggregate, it represents the k-th vertex currently being calculated. This represents the total number of vertices in the aggregate. It determines the number of sides of the random polygon; generally, the larger n is, the closer the aggregate shape is to a circle. 4. The convex hull calculation uses Andrew's algorithm, which consists of two steps: lower convex hull and upper convex hull. First, the vertices are sorted in ascending order of their x-coordinates, and then each vertex is processed sequentially. The condition for constructing the lower convex hull is: when there are at least two vertices in the stack, if the cross product of the current vertex and the top two vertices of the stack is less than or equal to zero, that is... If the vertex is not found, the top vertex needs to be popped from the stack. After processing, the current vertex is pushed back onto the stack. Similarly, the upper convex hull is processed in descending order of x-coordinate, and the same cross product condition check is repeated. Finally, the results of the lower and upper convex hulls are merged to obtain the complete convex hull vertex sequence. .

[0035] This represents the m-th vertex in the stack. During algorithm execution, it typically refers to the top vertex of the stack on the currently determined convex hull boundary. This represents the k-th vertex currently being processed. The algorithm iterates through the sorted set of points sequentially, judging each vertex one by one. Should it be included within the convex hull boundary? This represents the original set of points, i.e., the set of all randomly generated vertices, and the convex hull function. The input object; 5. Validity verification ensures that the generated convex polygon meets all constraints. First, it's necessary to check if the area is non-degenerate, using the calculation formula: , in, , The area A is required to be greater than a set threshold. .

[0036] Represents the area of ​​the polygon; and Represents the x and y coordinates of the k-th vertex of the polygon; and This represents the x-coordinate and y-coordinate of the first vertex of the polygon. x is mentioned at the end of the formula. n+1 =x1 is used to connect the ends of the polygon, completing the closure calculation; The second step is to check the dimensional constraints, calculate the width and height of the circumscribed rectangle, and ensure the minimum side length. Not smaller than the specified minimum size ; The formula for calculating the width is as follows: ; Height calculation formula: ; This represents the width of the bounding rectangle; This represents the maximum x-coordinate among all vertices of the polygon; This represents the minimum x-coordinate among all vertices of the polygon; Indicates the height of the bounding rectangle; This represents the maximum value of the y-coordinate among all vertices of the polygon; This represents the minimum value of the y-coordinate among all vertices of the polygon; Step 3: Calculate the cumulative volume fraction V of the generated aggregate according to the following formula: V = V + Vi; Where Vi represents the volume fraction of the random aggregate of the i-th convex polygon; Determine if the cumulative volume fraction V of the generated aggregate satisfies the following relationship: V≥r*X*Y; if not, return to continue generating random convex polygon aggregates; if yes, save the particle size di of all generated convex polygon aggregates and the point and line data of the vertices of the convex polygon aggregates. Where r represents the target volume fraction of the convex polygon aggregate to be generated, X represents the length of the concrete numerical model to be constructed, and Y represents the width of the concrete numerical model to be constructed. Step 4: Each newly generated aggregate needs to undergo rigorous conflict detection, i.e., checking whether it overlaps with the existing buffer zone around each existing aggregate. Circumscribed circle quick screening: If the center distance between the circumscribed circles of two aggregates meets the following conditions... If the initial screening is successful, then the precise intersection determination is achieved by detecting vertex inclusion using the ray casting method and solving the edge intersection parametric equations. (non-parallel condition).

[0037] and These represent the coordinates of the circumcenters of the i-th and j-th aggregates, respectively. and These represent the circumradii of the i-th and j-th aggregates, respectively. This indicates the thickness of the interface transition zone, used to ensure sufficient slurry gaps between aggregates; and Represents the coordinates of the two endpoints of the first line segment (edge) to be measured; and Represents the coordinates of the two endpoints of the second line segment (side) to be tested, which is used to determine whether the two sides are parallel or intersecting by using the determinant; Step 5: Building the Interface Transition Zone Generate a uniform thickness transition zone: translate a distance along the outer normal direction of each edge of the aggregate. Connecting the translated vertices forms a closed polygon; merging and connecting adjacent transition areas generates the interface transition area. Step 6: Target volume fraction of the second layer of aggregate ,in This is the proportion coefficient for the second layer.

[0038] The particle size range is adjusted to simulate the secondary distribution of small-sized aggregates.

[0039] Overlap area threshold: This ensures that the new aggregate is fully encapsulated within the first layer of structure.

[0040] Step 7: Second layer aggregate generation algorithm: The vertex perturbation rule of the first layer is used, but a particle size scaling factor is introduced. This produces aggregates with smaller particle sizes: ,in, ; Vertex coordinate calculation: , ; in, Enhance angular perturbation to simulate irregular shapes.

[0041] and These represent the aggregate particle sizes generated in the first and second layers, respectively. This indicates the initial random particle size of the first layer of aggregate. This indicates the minimum particle size limit required for the first layer of aggregate; This indicates the maximum particle size limit required for the first layer of aggregate; Indicates the geometric center coordinates of the second layer of aggregate; This represents the coordinates of the k-th vertex of the second layer of polygonal aggregate; This represents a random angle perturbation term, used to increase the irregularity of aggregate shape; This represents the original distribution angle of the k-th vertex; This represents the generation radius of the k-th vertex of the second layer of aggregate; Convex hull construction optimization: An improved Andrew monotonic chain algorithm is used to exclude non-convex vertices and generate strictly convex polygons. ; With fixed buffer thickness Expand the polygon to generate a transition area. ; Step 8: To ensure the second layer of aggregate adheres tightly to the first layer, new aggregate is required. The overlap area with the base region meets the requirements. If the condition is not met, the currently generated aggregate is discarded; if it is met, the aggregate is trimmed to the overlapping area. .

[0042] This represents the valid aggregate polygon. It is the result of intersecting the newly generated aggregate with the base region. This indicates the merged base area (usually referring to the area of ​​the first layer of old mortar). Indicates a valid interface transition area; This represents the initial interface transition area; Step 9: To avoid aggregate overlap in the second layer, the new aggregate needs to be inspected. and its transition zone Is it consistent with all the second layer aggregates that have already been generated? A conflict occurs. For any i, the following must be satisfied: and .

[0043] This condition is achieved using a polygon intersection area calculation function, with the core criterion being: in, The floating-point error threshold is usually taken as... .

[0044] , This represents the existing i-th second-layer aggregate and its transition zone, used to check whether there is a conflict between the old and new aggregates; and This represents a general polygon designation, which in the overlaps(A,B) function represents any two geometric shapes that need to be checked for overlap. Data output and storage organization: Output data structure: 1. Aggregate geometry data is stored in cell array XY; 2. The transition region data is stored in the cell array fitz; 3. Generate standard DXF format geometry files. Example 2

[0045] This embodiment demonstrates a method for constructing a concrete microstructure model and numerically simulating chloride ion attack based on co-simulation using MATLAB and COMSOL Multiphysics. The method first uses MATLAB scripts to generate a complex two-dimensional geometric model containing a first layer of aggregate (simulating old mortar) and a second layer of aggregate (simulating natural aggregate) and their corresponding interface transition zone (ITZ). The geometric data is then imported into COMSOL for chloride ion diffusion analysis. Figure 1 As shown, Figure 1 The flowchart for generating random convex polygonal aggregates in recycled concrete is shown. This embodiment provides a method for generating polygonal random aggregate particles, including the following steps: I. Construction of the micro-geometric model of concrete (MATLAB part) Step S1: Parameter initialization and region definition Define the basic dimensions and control parameters of the model in MATLAB: Computational domain size: Set the model area to a square of 100mm × 100mm.

[0046] Aggregate control parameters: Set the target area ratio of the first layer of aggregate to 30%.

[0047] Interface Transition Zone (ITZ) Parameters: Set the ITZ thickness to 0.05mm and specify the connection method as 'miter' when the polygon expands outward to maintain the sharpness of the geometric features.

[0048] Initial available regions: Define the initial effective deployment regions available_regions as the entire 99mm×99mm internal area.

[0049] Step S2: Generation and addition of the first layer of aggregate (old mortar / large aggregate) The first layer of aggregate is generated using a subtractive addition strategy: The subtraction-based placement strategy provided in this embodiment defines an initial effective area. After each aggregate and its interface transition area are generated, its occupied area is subtracted from the current effective area using a Boolean subtraction operation, dynamically updating it to several unconnected sub-regions. Subsequent aggregates are only generated and placed within these remaining sub-regions, fundamentally eliminating invalid attempts and significantly improving the placement efficiency of high-density aggregates. The specific process is as follows: 1. Randomly select a region: Randomly select a polygonal region from available_regions.

[0050] In the process of generating random convex polygon aggregates, when available_regions consists of multiple discontinuous sub-regions, the specific implementation method should preferentially adopt an area-weighted random selection strategy, that is, first calculate the area A of each sub-region. i A random number is generated that falls within the total area, and a target sub-region is selected probabilistically based on the area proportion. This strategy ensures that large areas have a higher probability of obtaining new aggregate, thus making the global distribution more uniform and effectively avoiding excessive accumulation of aggregate in small fragment areas or sparse distribution in open areas. After selecting the region, the algorithm calculates the target sub-region based on the maximum particle size d. max exist Randomly generate vertex radius r within the range k polar angles arranged in ascending order Through formula and Calculate the initial vertex coordinates. To ensure the aggregate is precisely positioned within its predetermined location within the sub-region, a centering process is required, i.e., the centroid of the initial vertex set is first calculated. Then, using the translation formula Move it toward the target center (x) c ,y c If multiple deliveries fail within a selected sub-area, consider skipping that area or fine-tuning the aggregate size to ensure overall delivery efficiency and isotropic physical properties.

[0051] 2. Generate a random convex polygon: Generate a random convex polygon new_poly within the selected area, with a size between [5mm, 20mm] and 5-9 vertices.

[0052] 3. Construct ITZ: Using the polybuffer function, expand new_poly outward in a 'miter' manner to generate a ring-shaped region buffer_poly with a width of 0.05mm (as the first layer ITZ).

[0053] 4. Integrity check and update: Construct the full region: full_region = union(new_poly, buffer_poly).

[0054] Check if the full_region overlaps with the already deployed area. If there is no overlap, save the aggregate and its ITZ.

[0055] Key step: Subtract the full region from the available regions (subtract operation), breaking the remaining space into multiple small sub-regions for subsequent placement, thereby achieving close packing of aggregates.

[0056] 5. Repeat the above process until the target area percentage is reached or there are no more areas available for deployment.

[0057] In the random aggregate delivery system, the determination of areas with no delivery points employs a dual standard of static geometric filtering and dynamic trial constraints to ensure that the program can safely exit the loop when space is saturated or geometrically constrained. The static determination is based on an area threshold: after each Boolean subtraction operation to update available_regions, the system immediately verifies the area S of each sub-region. sub If it is less than the preset minimum aggregate area threshold (usually based on the minimum particle size d), min =5 (calculated), then the sub-region is determined to be an invalid fragment that cannot accommodate aggregate and is removed. Dynamic determination is based on the success rate of attempts: for residual spaces that are too long and narrow or have difficulty in placement due to ITZ conflicts, if the system cannot generate a full_region that meets the non-overlapping condition in N consecutive random attempts, then it is determined that there is no available physical placement space under the current geometric topology. Therefore, the loop will automatically terminate when the available_regions list is empty due to static filtering, the global consecutive attempt count reaches the upper limit, or the generated cumulative volume fraction V reaches the target ratio r·X·Y.

[0058] Step S3: Generation of the new interface transition area (first layer overall wrapping) To simulate the interface between the first layer of aggregate (such as old mortar blocks used as recycled aggregate) and the new mortar matrix: 1. Perform a Boolean union operation on all the first-layer aggregates to form a single geometric object.

[0059] 2. Perform a polybuffer operation (0.05mm thickness, 'miter' connection) on the merged geometry object again to generate a new interface transition area that wraps around all the first layer of aggregate.

[0060] 3. Save the transition area data for subsequent export.

[0061] Step S4: Filling and adding the second layer of aggregate (natural aggregate) Fill the voids in the first layer of aggregate with a second, smaller layer of aggregate (as the final natural aggregate phase): 1. Target setting: Set the target area of ​​the second layer of aggregate to 50% of the area already generated in the first layer.

[0062] 2. Random generation and filtering: Randomly generate smaller convex polygons (size range dynamically adjusted).

[0063] The size d of the second layer of aggregate (2) The random allocation mechanism, which adopts proportional scaling, is based on the initial characteristic particle size d of the first layer of aggregate. (1) In practice, a grain size scaling factor is introduced. Satisfying the formula ,in It follows a uniform distribution of U(0.5,0.8). This adjustment mechanism ensures that the size of the secondary natural aggregate always maintains physical coordination with the size of its host old mortar block, avoiding shearing failure caused by the natural aggregate exceeding the boundary of the old mortar, and simulating the randomness of the thickness of the old mortar coating layer in real recycled aggregate.

[0064] 3. Boolean intersection detection: Calculate the intersection of the newly generated aggregate with the first layer of aggregate region.

[0065] If the intersection area is insufficient, it means that most of the aggregate is located outside the first layer of aggregate and should be retained; otherwise, it should be discarded.

[0066] This logic simulates the situation where natural aggregate fills the gaps between old aggregate or is partially embedded.

[0067] 4. Conflict detection: Ensure that the newly generated second layer of aggregate and its ITZ do not overlap with the existing second layer of aggregate.

[0068] 5. Save data: Save the second layer of aggregate that meets the conditions and its ITZ to the corresponding list.

[0069] like Figure 2 As shown, Figure 2 This is a schematic diagram of a single recycled aggregate. Figure 2 This visually demonstrates the complex multiphase microstructure of a single recycled aggregate in the numerical model, clearly illustrating the hierarchical nesting relationships between the various physical phases. The core region (dark area) represents the natural aggregate generated by the second-layer delivery algorithm, simulating the original coarse aggregate within the recycled aggregate. The thin layer immediately adjacent to the natural aggregate is the old interface transition zone, reflecting the weak interface between the coarse aggregate and old mortar in the original concrete. The tan area surrounding the old interface represents the attached old mortar, i.e., the geometric core generated by the first layer of aggregate. The outermost boundary, outlined by cyan lines, is the new interface transition zone surrounding the entire recycled aggregate, generated using polybuffer operations; the outermost light blue area represents the new mortar.

[0070] like Figure 3 As shown, Figure 3 This is a numerical model diagram of a five-phase microstructure of recycled concrete generated by the program. (Example:) Figure 3 As shown, the five-phase microscopic numerical model of recycled concrete generated by this program realistically reproduces the complex internal geometry of recycled concrete through a parametric random placement algorithm. In the figure, a light blue background represents continuous new mortar, and a cyan frame generated using polybuffer operations and wrapped around the outermost layer of recycled aggregate represents the new interface transition zone. The main body of the model consists of light brown old mortar blocks generated and tightly packed using a subtractive placement strategy, with black natural aggregate nested inside as the original core, and the two connected by a dark gray old interface transition zone. In addition, the model also includes some independent natural aggregates directly placed into the matrix to meet gradation requirements. At the algorithmic level, the model utilizes an improved Andrew algorithm to ensure the rigorous convex polygonal shape of the aggregates, and through precise Boolean intersection detection and overlap area determination, it guarantees the logical rigor and physical compatibility of each physical phase in two-dimensional space, providing a high-fidelity geometric platform for further research on numerical simulation of chloride ion erosion in recycled concrete.

[0071] Step S5: Export DXF geometry file Using a custom function, `export_polygons_to_dxf`, the four types of generated geometric objects are exported as separate DXF files for COMSOL to recognize: old_aggregate_phase.dxf: The first layer of aggregate entities (such as old mortar / recycled aggregate core).

[0072] old_transition_zone.dxf: The interface transition zone corresponding to the first layer of aggregate.

[0073] old_mortar_phase.dxf: The second layer of filler aggregate (natural aggregate in the new mortar).

[0074] new_transition_zone.dxf: A new interface transition zone that wraps around the entire first layer of aggregate.

[0075] like Figure 4 As shown, Figure 4 The overall flowchart of the program for creating a random convex polygon aggregate model for recycled concrete.

[0076] Step S51: Start and set the aggregate volume fraction, particle size range, and interface transition zone thickness; Step S52: Generate the first aggregate cycle from [DMIN, DMAX]; Step S53: Randomly select a region to generate convex polygon aggregate; Step S54: Perform conflict detection and determine whether a conflict exists. If no conflict exists, return to step S52; otherwise, proceed to the next step. Step S55: Record aggregate and update aggregate area; Step S56: Generate the first layer of interface transition area (new interface transition area phase); Step S57: The second layer of aggregate and the interface transition zone are generated cyclically; Step S58: Perform a conflict detection. If no, return to step S57; if yes, proceed to the next step. Step S59: Preserve overlapping areas Step S510: Determine whether the overlap area requirement is met. If not, return to step S57; if yes, end.

[0077] II. Numerical Simulation of Chloride Ion Erosion (COMSOL Section) Step S6: Simulation Environment Setup and Geometry Import 1. Launch COMSOLMultiphysics, select the two-dimensional space dimension, and add the rare matter transfer (tds) physics interface.

[0078] 2. In the geometry node, import the four DXF files generated in step S5 in sequence.

[0079] 3. Construct a FormUnion to ensure mesh continuity between regions, forming a complex multiphase geometry containing the following regions: Natural aggregate phase; Old interface transition phase (OldITZPhase); Old Mortar Phase New ITZ Phase; New Mortar Phase (i.e., the remaining part after subtracting the above two phases from the background area).

[0080] Step S7: Setting Material Properties and Transfer Parameters Based on the transport characteristics of chloride ions in different media, the diffusion coefficient is defined as follows: New mortar phase: Serving as the reference transport medium for the model, its diffusion coefficient is determined by the mix proportions. Set as the reference diffusion coefficient Dnew_mortar = D0 (typically ranging from 1e-11 to 1e-12m). 2 (on the order of magnitude of / s).

[0081] New interface transition zone phase: Due to the large local water-cement ratio and the presence of wall effect, it has a high porosity. It is set to be k1 times the diffusion coefficient of the new mortar, i.e., Dnew_itz = k1 × D0.

[0082] Old mortar phase: This phase has undergone degradation during early service and fine crack damage from crushing and processing, resulting in a lower density than freshly mixed mortar. It is set to k² times the diffusion coefficient of fresh mortar, i.e., Dold_mortar = k² × D0.

[0083] The old interface transition zone phase, being the weakest link in the entire multi-interface system, not only has high original porosity but is also extremely prone to interfacial delamination and fine crack accumulation during the crushing process, serving as a rapid short-circuit channel for chloride ion penetration. It is set to be k3 times the diffusion coefficient of the new mortar, i.e., Dold_itz = k3 × D0.

[0084] Natural aggregate phase: set as impermeable, diffusion coefficient .

[0085] To ensure the accuracy of the numerical simulation, the values ​​of k1, k2, and k3 should be defined based on the microscopic porosity characteristics and structural damage degree of each physical phase. The physical meaning of the new interface transition zone coefficient k1 mainly stems from the wall effect on the aggregate surface, which leads to a local increase in water-cement ratio and porosity. Its recommended value range is usually between 1.3 and 16.2 of the new mortar phase, used to quantify its accelerated transport effect relative to the new mortar matrix. The old mortar phase coefficient k2 measures the carbonization degradation of the attached mortar due to early service and the accumulation of microcracks introduced by later mechanical crushing. Its diffusion coefficient is usually set to 0.2 to 5 times that of the new mortar phase, and its value reflects the quality of the recycled aggregate source and its degree of damage. The old interface transition zone coefficient k3 is the most critical degradation parameter in the five-phase system. It combines the original high porosity interface with the significant interface peeling damage caused by secondary processing. Its diffusion capacity is set to the highest level, and its value range is generally between 1.3 and 16.2 for the old mortar phase. As a fast short-circuit channel for chloride ion penetration, it is shown in the concentration cloud map as chloride ions preferentially bypassing the dense aggregate core along this path.

[0086] Step S8: Boundary Conditions and Mesh Generation 1. Boundary conditions: Left boundary: set as concentration boundary to simulate chloride ion erosion source, concentration c=c0.

[0087] Right, top, and bottom boundaries: set as flux-free boundaries to simulate unidirectional erosion conditions, such as... Figure 5 As shown. Figure 5 This figure shows the boundary condition configuration for a five-phase numerical model of recycled concrete, designed to simulate the unidirectional erosion and diffusion process of chloride ions in a multiphase complex medium. The boundary on the left side of the figure is indicated by a blue arrow. - The symbol indicates the concentration boundary (c=c0), simulating the initial erosion surface of a concrete specimen exposed to an external high-concentration chloride ion source. Conversely, the right, upper, and lower boundaries of the model are marked with solid blue lines and are set as no-flux boundaries, simulating controlled unidirectional erosion conditions and forcing chloride ions to diffuse only horizontally. This boundary setting simplifies the complex diffusion problem into a one-dimensional erosion model, allowing for precise observation and quantification of the specific effects of five physical phases (new mortar phase, new interface transition zone phase, old mortar phase, old interface transition zone phase, and natural aggregate phase) with different diffusion coefficients on the erosion depth. In particular, it allows for analysis of how the shortcut effect of the interface transition zone and the barrier effect of aggregates alter the chloride ion transport path.

[0088] 2. Grid generation: A free triangle mesh is used.

[0089] For the ITZ region, an extremely fine mesh size is set to ensure that at least 2-3 mesh cells are distributed within the thin layer to accurately capture local concentration gradients. For example... Figure 6 As shown, Figure 6 This diagram illustrates the finite element mesh generation results for the microscopic numerical model of recycled concrete. The model is discretized using free triangular elements, and by setting differentiated mesh densities for different phases, both computational accuracy and numerical efficiency are balanced. Within the larger areas of the new mortar matrix and aggregate, the mesh size is relatively uniform; however, in the geometrically complex interface regions, the mesh exhibits a clear adaptive refinement characteristic. This high-fidelity mesh generation provides a solid discretization foundation for subsequent chloride ion diffusion simulations, ensuring the continuity and convergence of the physical field at the multiphase medium interface. For the extremely thin interface transition zone (ITZ), the model implements an ultra-fine mesh strategy: Quantization standard: Within the ITZ region, the maximum size of the mesh element l... e It is strictly limited to less than half the ITZ thickness t (i.e. Distribution Requirements: This quantization relationship ensures that at least two layers of triangular mesh elements are distributed within a thin layer width of only 0.05 mm. Simulation Significance: This accuracy setting aims to accurately capture the drastically changing local concentration gradients within the ITZ, preventing numerical oscillations or distortions in the erosion front simulation caused by overly coarse meshes, thereby realistically reproducing the physical effects of the interface region as a fast track for chloride ion erosion.

[0090] Step S9: Solving and Post-processing 1. Solver settings: Select the transient solver, set the time step (e.g., 1 day) and the total simulation time (e.g., 300 days).

[0091] 2. Results Analysis: Concentration cloud map: Generates a two-dimensional chloride ion concentration distribution cloud map, such as... Figure 7 As shown, Figure 7This contour plot visually illustrates the two-dimensional chloride ion concentration distribution in a micro-numerical model of recycled concrete at a specific erosion time. The color gradient represents the dynamic evolution of chloride ion concentration diffusion from the high-concentration source (red area) on the left boundary to the low-concentration zone (blue area) on the right. Due to the low diffusion coefficient of the natural aggregate phase, the aggregate area outlined in black clearly exhibits a significant barrier effect, causing its core concentration evolution rate to lag far behind that of the surrounding mortar matrix. Simultaneously, constrained by the complex geometric distribution of the new / old mortar and the highly diffusive new / old interface transition zone (ITZ), the erosion front exhibits obvious non-uniform fluctuation characteristics. Sectional concentration curves: Chloride ion concentration versus time curves were plotted by taking segments at different depths. Flux analysis: The total chloride ion flux through the specimen was calculated to evaluate the impact of different aggregate distributions on the chloride ion penetration resistance of concrete.

[0092] The above-described embodiments are merely preferred embodiments provided to fully illustrate the present invention, and the scope of protection of the present invention is not limited thereto. Equivalent substitutions or modifications made by those skilled in the art based on the present invention are all within the scope of protection of the present invention. The scope of protection of the present invention is defined by the claims.

Claims

1. A numerical simulation method for chloride ion erosion of recycled concrete based on recursive domain segmentation and a five-phase microscopic model, characterized in that, Includes the following steps: S1: Initialize model parameters, including model size, target aggregate volume fraction, aggregate particle size range, ITZ thickness, and recycled aggregate replacement rate. S2: The first layer of aggregate and its interface transition zone (ITZ) are generated using a subtractive placement strategy based on recursive region segmentation. Specifically, this includes: S21: Define the entire model area as the initial effective delivery area; S22: Generate a random convex polygon within the effective delivery area as the first layer of aggregate; S23: Translate each side of the first layer of aggregate outward by the thickness of the interface transition zone ITZ to generate a first layer interface transition zone ITZ polygon that encloses the aggregate. S24: Perform a Boolean union operation on the first layer of aggregate and its first layer interface transition zone ITZ polygon to obtain the occupied area, and subtract the occupied area from the current effective delivery area by Boolean, and update the remaining space to at least one new effective delivery sub-region that is not connected to each other. S25: Repeat steps S22 to S24 until the cumulative volume fraction of the first layer of aggregate reaches the target value calculated from the target aggregate volume fraction and the recycled aggregate replacement rate, or the effective placement area is empty. S3: Generate a second layer of aggregate, allowing it to adhere to or fill the gaps between the first layer of aggregate, specifically including: S31: Generate a random convex polygon with a particle size smaller than the first layer of aggregate within the current remaining effective delivery area, as a candidate for the second layer of aggregate; S32: Calculate the Boolean intersection area of ​​the second layer aggregate candidate and the union area of ​​all first layer aggregates. If the ratio of the intersection area to the area of ​​the second layer aggregate candidate itself meets the preset attachment threshold range, then the second layer aggregate candidate is accepted and the second layer aggregate is regarded as the natural aggregate phase. S33: Repeat steps S31 to S32 until the cumulative volume fraction of the second layer of aggregate reaches the target value; S4: Construct a five-phase micro-geometric model. Perform a Boolean union operation on all the first-layer aggregates to obtain the old mortar phase. Translate the boundary of the old mortar phase outwards by the thickness of the interface transition zone (ITZ) to generate a new interface transition zone (ITZ) phase that encapsulates all the old mortar phases. This results in the construction of a five-phase micro-geometric model that includes the natural aggregate phase, the old interface transition zone (ITZ) phase, the old mortar phase, the new interface transition zone (ITZ) phase, and the new mortar phase. S5: Export standard geometry file, export the boundary data of the five-phase micro-geometric model as a standard geometry file; S6: Perform a numerical simulation of chloride ion erosion. Import the standard geometric file into the finite element simulation software, assign different chloride ion diffusion coefficients to the five phases respectively, set erosion boundary conditions, perform a transient numerical simulation of chloride ion erosion, and obtain the chloride ion concentration distribution results.

2. The numerical simulation method for chloride ion erosion of recycled concrete based on recursive region segmentation and a five-phase microscopic model according to claim 1, characterized in that, The step S22 of generating a random convex polygon includes: The angles of each vertex are randomly generated in polar coordinates. and radius , where the radius A random perturbation factor is introduced to ensure the irregularity of the aggregate shape; the polar coordinates are converted to Cartesian coordinates to obtain the initial vertex set; the Andrew algorithm is used to calculate the convex hull of the initial vertex set to obtain the vertex sequence of the convex polygon.

3. The numerical simulation method for chloride ion erosion of recycled concrete based on recursive region segmentation and a five-phase microscopic model according to claim 1, characterized in that, Step S23, which involves shifting the thickness of the interface transition zone (ITZ) outward from each edge of the first layer of aggregate, includes: The aggregate is translated a distance t along the outward normal direction of each side. The translated sides are extended to both ends, and the coordinates of the intersection points of two adjacent extension lines are obtained. The intersection points are connected in sequence to form a closed interface transition zone ITZ polygon, and the miter connection method is used to maintain the sharpness of the geometric features.

4. The numerical simulation method for chloride ion erosion of recycled concrete based on recursive region segmentation and a five-phase microscopic model according to claim 1, characterized in that, The preset adhesion threshold range in step S32 is 5% to 15%. When the ratio of the intersection area to the area of ​​the second layer aggregate candidate is less than the lower limit of the adhesion threshold range, the aggregate is determined to meet the adhesion filling conditions.

5. The numerical simulation method for chloride ion erosion of recycled concrete based on recursive region segmentation and a five-phase microscopic model according to claim 1, characterized in that, Step S4 is followed by: Boolean operations are performed on the five-phase micro-geometric model to delete the intersecting parts between each phase, ensuring that the geometric boundaries of each phase are clear and do not overlap.

6. The numerical simulation method for chloride ion erosion of recycled concrete based on recursive region segmentation and a five-phase microscopic model according to claim 1, characterized in that, The standard geometry file mentioned in step S5 is a DXF format file.

7. The numerical simulation method for chloride ion erosion of recycled concrete based on recursive region segmentation and a five-phase microscopic model according to claim 1, characterized in that, Step S6 describes assigning different chloride ion diffusion coefficients to the five phases, specifically including: The diffusion coefficient of the new mortar phase is set as the reference value D0; The diffusion coefficient of the ITZ phase in the new interface transition region is set to k1×D0, where k1 is the first multiple coefficient; The diffusion coefficient of the first layer of aggregate, i.e., the old mortar phase, is set as k2×D0, where k2 is the second multiple coefficient; The diffusion coefficient of the first-layer interface transition region ITZ, i.e. the old interface transition region ITZ phase, is set as k3×D0, where k3 is the third multiple coefficient. The diffusion coefficient of the second layer of aggregate, i.e., the natural aggregate phase, is set to approximately 0. Where k1, k2, and k3 are real numbers greater than 1, and k3 is greater than both k1 and k2.

8. The numerical simulation method for chloride ion erosion of recycled concrete based on recursive region segmentation and a five-phase microscopic model according to claim 1, characterized in that, Setting the erosion boundary conditions in step S6 includes: One boundary of the model is set as a chloride ion concentration boundary to simulate a chloride ion erosion source; the other boundaries of the model are set as no flux boundaries to simulate unidirectional erosion conditions.

9. The numerical simulation method for chloride ion erosion of recycled concrete based on recursive region segmentation and a five-phase microscopic model according to claim 1, characterized in that, Before performing the transient numerical simulation of chloride ion corrosion as described in step S6, the following steps are also included: The five-phase micro-geometric model is meshed, and the interface transition zone (ITZ) is locally refined to ensure that at least 2-3 layers of mesh elements are distributed within the thickness of the interface transition zone (ITZ).

10. A numerical simulation system for chloride ion attack in recycled concrete based on recursive domain segmentation and a five-phase microscopic model, characterized in that, include: The parameter initialization module is used to initialize model parameters, including model size, target aggregate volume fraction, aggregate particle size range, interface transition zone (ITZ) thickness, and recycled aggregate replacement rate. The first-layer aggregate generation module is used to generate the first-layer aggregate and its interface transition zone (ITZ) using a subtractive placement strategy based on recursive region segmentation. Specifically, it includes: Define the entire model area as the initial effective deployment area; A random convex polygon is generated within the effective delivery area as the first layer of aggregate; The thickness of the interface transition zone ITZ is translated outward from each side of the first layer of aggregate to generate a first layer interface transition zone ITZ polygon that encloses the aggregate. Perform a Boolean union operation on the first layer of aggregate and its first layer interface transition zone (ITZ) polygon to obtain the occupied area, and subtract the occupied area from the current effective delivery area in Boolean order, and update the remaining space with at least one new, unconnected effective delivery sub-area. Repeat the above operation until the cumulative volume fraction of the first layer of aggregate reaches the target value calculated from the target aggregate volume fraction and the recycled aggregate replacement rate, or the effective placement area is empty; The second-layer aggregate generation module is used to generate a second layer of aggregate, which is then attached to or fills the gaps in the first layer of aggregate. Specifically, it includes: Generate a random convex polygon with a particle size smaller than the first layer of aggregate within the current remaining effective delivery area, as a candidate for the second layer of aggregate; Calculate the Boolean intersection area of ​​the second aggregate candidate and the union area of ​​all first aggregates. If the ratio of the intersection area to the area of ​​the second aggregate candidate itself meets a preset adhesion threshold range, then the second aggregate candidate is accepted and the second aggregate is regarded as the natural aggregate phase. Repeat the above operation until the cumulative volume fraction of the second layer of aggregate reaches the target value; The five-phase model construction module is used to perform Boolean union operation on all the first-layer aggregates to obtain the old mortar phase, and to translate the boundary of the old mortar phase outward by the thickness of the interface transition zone ITZ to generate a new interface transition zone ITZ phase that encapsulates all the old mortar phases, thereby constructing a five-phase micro-geometric model containing the natural aggregate phase, the old interface transition zone ITZ phase, the old mortar phase, the new interface transition zone ITZ phase, and the new mortar phase. The data export module is used to export the boundary data of the five-phase microscopic geometric model as a standard geometric file; The numerical simulation module is used to import the standard geometric file into the finite element simulation software, assign different chloride ion diffusion coefficients to the five phases respectively, set erosion boundary conditions, perform transient numerical simulation of chloride ion erosion, and obtain chloride ion concentration distribution results.

Images