Cement grain composition optimization method and system for minimizing porosity

Abstract

The invention provides a cement particle grading optimization method and system for minimizing porosity, and relates to the technical field of cement material science and engineering, and the method comprises the following steps: obtaining cement particles with different particle size distributions; taking the cement particles as spheres, and generating a sphere set according to the cement particles with different particle size distribution; in the sphere set of each particle size distribution, sampling according to a proportion, and constructing a three-dimensional simulation microstructure model by using the sampled spheres; simulating a sphere accumulation process in a three-dimensional box by using a three-dimensional simulation microstructure model, calculating the porosity in the three-dimensional box in the simulation process, establishing a loss function according to the overlapping volume between spheres and the collision volume between the spheres and the boundary, and obtaining the porosity of cement proportions with different particle size distributions by using a gradient descent method; and adjusting the sampling proportion of each particle size distribution by adopting an optimization algorithm so as to minimize a loss function and realize the grain composition of the cement powder with the minimum porosity and different distribution so as to optimize the strength of the cement.

Description

Technical Field

[0001] The present disclosure relates to the field of cement material science and engineering technology, and in particular to a cement particle grading optimization method and system for minimizing porosity. Background Art

[0002] The statements in this section merely provide background information related to the present disclosure and do not necessarily constitute prior art.

[0003] The particle size, shape and distribution of cement particles have a crucial impact on the physical and mechanical properties of cement. Appropriate particle size can form a good skeleton structure in the cement paste, which helps to reduce porosity, increase the density of cement and concrete, and thus improve the strength, durability and impermeability of the material. On the contrary, unreasonable particle size will lead to more pores in the cement paste, thereby reducing the overall performance of the material, especially strength and durability.

[0004] Traditionally, the gradation of cement particles is obtained through physical methods such as grinding and screening. However, it is difficult to accurately control the particle size distribution of particle grading based on physical methods, and in actual production, the particle sizes of different particles will overlap or cross to a certain extent, making it difficult to form an ideal continuous gradation. In addition, the grinding and screening methods have certain limitations when adjusting the distribution of particles, and it is often difficult to accurately achieve the predetermined target particle size distribution. Therefore, the particle gradation obtained by traditional methods cannot effectively reduce the porosity, resulting in poor density and impermeability of cement products, which in turn affects the strength and durability of building materials.

[0005] In order to overcome the limitations of physical methods, empirical methods or simple theoretical models are usually used in the prior art to design the particle size distribution of cement. For example, the particle size distribution is designed by empirical formulas (such as Fuller's formula or Poisson's formula). However, most of these methods are based on macroscopic statistical models, ignoring the interaction between particles and the specific arrangement of particles in space, resulting in the difficulty of fully considering the pore distribution in the microstructure in the designed particle size distribution. In addition, most of these methods assume that the particles are idealized spheres or uniformly distributed. This assumption is far from the actual situation in practical applications, so it is impossible to minimize the porosity and it is difficult for the material performance to reach the optimal state.

[0006] In recent years, with the rapid development of computer technology and numerical simulation methods, it has become possible to design the microstructure of materials using computer simulation and optimization algorithms. Computer simulation technology can generate a three-dimensional structural model of the material in a virtual environment, simulate the particle stacking process and pore formation, and thus analyze and optimize the internal structure of the material.

[0007] However, although computer simulation technology has made significant progress, most of the current research focuses on the simulation of material microstructures, and there are still deficiencies in how to systematically combine simulation results with optimization algorithms to design cement particle grading. In particular, under the conditions of multi-dimensional and complex material parameters, it is difficult to achieve effective coupling between the simulation and optimization processes, and the optimization algorithms disclosed in the prior art, such as genetic algorithms, particle swarm optimization algorithms, simulated annealing and other intelligent optimization algorithms. These algorithms have been widely used in industries such as industry and manufacturing, and parameter adjustment of complex systems is achieved through iterative search and multi-objective optimization. However, there are still some challenges in the application of these optimization algorithms in the particle grading of cement materials. First, the design of cement particle grading belongs to a multi-variable and multi-constrained optimization problem. Factors such as particle shape, size, quantity, and distribution will affect porosity and material properties, making the optimization process more complicated. In addition, the parameter space involved in the optimization of cement particle grading is large, and traditional optimization algorithms are limited in computational efficiency and accuracy, making it difficult to quickly find the optimal particle grading solution. Summary of the invention

[0008] In order to solve the above problems, the present invention proposes a cement particle grading optimization method and system for minimizing porosity. By constructing a three-dimensional microstructure model under different particle size distributions, defining the loss function of porosity and particle overlapping volume, and using intelligent optimization methods such as genetic algorithms and gradient descent methods to continuously adjust the distribution ratio of particles to minimize porosity.

[0009] According to some embodiments, the present disclosure adopts the following technical solutions: Methods for optimizing cement particle size distribution to minimize porosity include: Obtaining cement particle powders with different particle size distributions; Considering cement particles as spheres, a collection of spheres is generated based on cement particles with different particle size distributions; In each set of spheres with a particle size distribution, multiple spheres are sampled in proportion, and a three-dimensional simulated microstructure model is constructed using the multiple spheres; The sphere stacking process is simulated in a 3D box using a 3D simulated microstructure model. During the simulation, the porosity in the 3D box is calculated. A loss function is established based on the overlapping volume between spheres and the collision volume between spheres and the boundary. The porosity of cement mixes with different particle size distributions is obtained using the gradient descent method. The optimization algorithm is used to optimize the sampling ratio of each particle size distribution to minimize the loss function and achieve the particle gradation of cement powders with different distributions with minimum porosity to optimize the strength of cement.

[0010] According to some embodiments, the present disclosure adopts the following technical solutions: Cement particle size optimization system to minimize porosity, including: A data acquisition module, used for acquiring cement particle powders with different particle size distributions; A model building module is used to regard cement particles as spheres and generate a collection of spheres according to cement particles with different particle size distributions; In each set of spheres with a particle size distribution, multiple spheres are sampled in proportion, and a three-dimensional simulated microstructure model is constructed using the multiple spheres; A simulation calculation module is used to simulate the sphere stacking process in a three-dimensional box using a three-dimensional simulation microstructure model. During the simulation process, the porosity in the three-dimensional box is calculated, and a loss function is established based on the overlapping volume between the spheres and the collision volume between the spheres and the boundary. The porosity of cement mixes with different particle size distributions is obtained by a gradient descent method. The optimization module is used to optimize and adjust the sampling ratio of each particle size distribution using an optimization algorithm to minimize the loss function and achieve the particle grading of cement powders with different distributions of minimum porosity to optimize the strength of cement.

[0011] According to some embodiments, the present disclosure adopts the following technical solutions: A computer program product comprises a computer program, wherein when the computer program is executed by a processor, the method for optimizing cement particle gradation for minimizing porosity is implemented.

[0012] According to some embodiments, the present disclosure adopts the following technical solutions: A non-transitory computer-readable storage medium is used to store computer instructions. When the computer instructions are executed by a processor, the cement particle grading optimization method for minimizing porosity is implemented.

[0013] According to some embodiments, the present disclosure adopts the following technical solutions: An electronic device comprises: a processor, a memory and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory so that the electronic device implements the cement particle grading optimization method for minimizing porosity.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The cement particle grading optimization method for minimizing porosity disclosed in the present invention obtains cement clinker powder with different particle size distributions and models them as spheres to generate a three-dimensional microstructure model; then the porosity is calculated by the Monte Carlo sampling method, and a loss function with porosity, particle overlap volume and boundary collision volume as indicators is defined; the proportion of particles of each particle size is adjusted by an optimization algorithm such as a gradient descent method or a genetic algorithm to minimize the loss function. By optimizing the loss function, the porosity minimization is directly taken as the goal, which can effectively reduce the porosity of the cement slurry and improve the density of the material.

[0015] The cement particle grading optimization method for minimizing porosity disclosed in the present invention systematically adjusts the particle size distribution and proportion of cement particles by combining computer simulation with an intelligent optimization algorithm to reduce the porosity of the cement slurry and improve the strength and durability of the material.

[0016] The disclosed cement particle gradation optimization method for minimizing porosity can accurately control the gradation of cement particles to minimize porosity. The reduction of porosity helps to improve the strength and durability of cement and improve the physical and mechanical properties of the material, thereby preparing cement products with high density and excellent mechanical properties. The method is applicable to different types of cement and granular materials and has strong operability and broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings constituting a part of the present disclosure are used to provide a further understanding of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure.

[0018] Figure 1 is a schematic diagram of cement particles with different particle size distributions according to an embodiment of the present disclosure; Figure 2 An initial spherical image of a sphere in three-dimensional space in a computer simulation of an embodiment of the present disclosure; Figure 3 The three-dimensional spherical image after being optimized by the method according to the embodiment of the present disclosure; Figure 4 Graph showing the relationship between porosity and optimization iteration number for an embodiment of the present disclosure. DETAILED DESCRIPTION

[0019] The present disclosure is further described below in conjunction with the accompanying drawings and embodiments.

[0020] It should be noted that the following detailed descriptions are all illustrative and are intended to provide further explanation of the present disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present disclosure belongs.

[0021] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present disclosure. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0022] Example 1 In one embodiment of the present disclosure, a method for optimizing cement particle gradation for minimizing porosity is provided. The method for optimizing cement particle gradation by computer simulation and intelligent optimization algorithm is used to minimize the porosity of cement paste, thereby improving the strength and durability of cement. The steps include: Step 1: Obtain cement particle powder with different particle size distributions; Step 2: Consider cement particles as spheres and generate a set of spheres according to cement particles with different particle size distributions; Step 3: In each set of spheres with a particle size distribution, multiple spheres are sampled in proportion, and a three-dimensional simulated microstructure model is constructed using the multiple spheres; Step 4: Use the three-dimensional simulation microstructure model to simulate the sphere stacking process in the three-dimensional box. During the simulation process, calculate the porosity in the three-dimensional box, establish a loss function based on the overlapping volume between the spheres and the collision volume between the spheres and the boundary, and obtain the porosity of cement mixes with different particle size distributions through the gradient descent method; Step 5: Use the optimization algorithm to optimize the sampling ratio of each particle size distribution to minimize the loss function and achieve the particle grading of cement powder with different distributions of minimum porosity to optimize the strength of cement.

[0023] As an embodiment, the present disclosure provides a cement particle grading optimization method for minimizing porosity, which constructs a three-dimensional microstructure model, defines a loss function of porosity and particle overlap volume, and uses intelligent optimization methods such as genetic algorithms and gradient descent methods to continuously adjust the distribution ratio of particles to minimize porosity. The specific implementation process is as follows: Step 1: Get cement particle powder with different particle size distribution. Select cement clinker block as raw material, use ball mill to grind the clinker block to achieve the set fineness, and sieve the ground clinker powder through sieves with different mesh sizes, and so on, to obtain cement particle powder with multiple different particle size distributions.

[0024] Specifically, first, high-quality cement clinker blocks are selected as raw materials. The clinker blocks are initially ground using a ball mill to achieve a certain fineness. The ground clinker powder is sieved through a series of sieves with different mesh sizes. For example, using sieves of 270 mesh and 340 mesh, the powder that can pass through 270 mesh but not 340 mesh is used as the first particle size distribution. By analogy, m clinker powders with different particle size distributions can be obtained. In order to ensure the differences between the distributions, the overlap between different distributions is minimized. Assuming that in an extreme case, if the two distributions are exactly the same, their ratio is meaningless, so the independence and purity of each particle size distribution must be strictly controlled during the screening process.

[0025] Step 2: Consider cement particles as spheres and generate a set of spheres based on cement particles with different particle size distributions.

[0026] Specifically, cement particles are regarded as spheres in the simulation, and a set of spheres of corresponding sizes is generated according to the different particle size distribution data obtained by screening. In order to simplify the calculation model, cement particles are regarded as spheres in the simulation. A set of spheres of corresponding sizes is generated according to the different particle size distribution data obtained by screening. It is assumed that the particles in each particle size distribution are uniformly distributed within their corresponding particle size range to ensure the accuracy and repeatability of the simulation.

[0027] Step 3: In each set of spheres of particle size distribution, multiple spheres are sampled in proportion, and a three-dimensional simulated microstructure model is constructed using the multiple spheres.

[0028] Specifically, assuming that the particles in each particle size distribution are uniformly distributed within their corresponding particle size range, a set ratio is sampled from each particle size distribution to obtain a total of N spheres, and a three-dimensional simulated microstructure model is constructed of the N spheres according to random or specific rules. The spheres are placed in a computer-simulated three-dimensional box to simulate the actual stacking process.

[0029] A total of N spheres are sampled from each particle size distribution at a certain ratio. These spheres are placed in a computer-simulated three-dimensional box according to random or specific rules (such as placement in order by particle size) to simulate the actual stacking process. The size and boundary conditions of the box are set according to actual production needs to ensure the representativeness and practicality of the simulation results.

[0030] Step 4: Use the three-dimensional simulation microstructure model to simulate the sphere stacking process in a three-dimensional box. During the simulation process, calculate the porosity in the three-dimensional box, establish a loss function based on the overlapping volume between the spheres and the collision volume between the spheres and the boundary, and obtain the porosity of cement mixes with different particle size distributions through the gradient descent method.

[0031] Calculate the porosity and loss function. During the simulation, calculate the porosity inside the box, which is the ratio of the volume not occupied by the spheres to the total volume of the box. Also, consider the overlap volume between the spheres and the collision volume between the spheres and the boundaries, which will affect the actual packing density.

[0032] Step 401: Porosity refers to the ratio of the volume not occupied by particles to the total volume of the cement paste. In order to accurately estimate the porosity, this embodiment adopts the Monte Carlo sampling method, including: Step 4011: Random point generation Let sample be the number of samples and dim be the dimension. First, generate a uniformly distributed random number moment with dimensions sample and dim. u , and map it to the upper and lower bounds Get random sampling points random_points within the defined range:

[0033]

[0034] Step 4012: Distance matrix calculation Use torch.cdist to calculate the distance matrix between random_points and the given sphere centers :

[0035] Among them, random_points represents a set of random points sampled from the unit cube by uniform distribution, and centers is the given center of the sphere.

[0036] Step 4013: Sphere internal judgment, distance_matrix Subtract the radius radii of each sphere to determine whether the sampling point is inside the sphere:

[0037] Step 4014: Calculate density loss density_loss , which is the sum of all overlapping distances, and divided by the number of samples:

[0038] This method provides an accurate estimate of porosity through the statistical proportion of a large number of randomly sampled points.

[0039] Step 402: The overlapping volume between spheres refers to the volume of the overlapping portion of two spheres in space. In order to accurately calculate this volume, the following steps are used: Step 4021: Neighbor sphere determination For each sphere, the neighbor list Determine the center position of the neighboring spheres around it .

[0040] Step 4022: Distance calculation Calculate the distance between each pair of neighbor spheres The formula is:

[0041] in, and Respectively i and j The center coordinate vector of the sphere.

[0042] Step 4023: Overlap volume calculation According to the radius of the sphere and Calculate overlap volume :

[0043] This formula represents the overlapping volume when two spheres intersect .

[0044] if , then the calculated is the overlapping volume.

[0045] if , then the spheres do not intersect and the overlapping volume .

[0046] Step 4024: Overlapping volume smoothing In order to better control the penalty effect of overlapping volume, the calculated overlapping volume is smoothed using the `softplus function to obtain the overlapping volume loss , the formula is:

[0047] Among them, get_slope(epoch) controls the degree of smoothing, Represents the volume of any sphere directly overlapping, volume_sum represents the total volume of the spheres, and epoch represents the number of iterations. It is used to normalize the overlapping volume. Through the above steps, the overlapping volume loss quantifies the degree of mutual overlap of each pair of spheres, helping the model reduce the overlap between spheres.

[0048] Step 403: The boundary collision volume refers to the volume of the part of the sphere that exceeds the boundary of the simulation box when the sphere collides with the boundary of the simulation box.

[0049] Step 4031: Boundary detection Check whether each sphere exceeds the boundary of the box. Define the upper and lower bounds of the box u_bound and l_bound, and calculate the upper_bound of the sphere beyond the upper boundary. The formula is as follows:

[0050] If the sphere position exceeds the upper bound, keep the value; otherwise, set the penalty value to 0.

[0051] Otherwise, the excess is set to 0.

[0052] Calculate the sphere's excess below the lower boundary, lower_bound:

[0053] If the sphere position exceeds the lower bound, keep the value; otherwise, set the penalty value to 0.

[0054] Among them, centers represents the center coordinates of each sphere, which is a three-dimensional coordinate array.

[0055] radii represents the radius of each sphere, which determines the extent of the sphere in three-dimensional space.

[0056] u _bound and l _bound represents the upper and lower bounds of the cube space, defining the maximum and minimum boundary coordinates of the space respectively.

[0057] Step 4032: Boundary loss calculation For all spheres beyond the boundary, the boundary loss bound_loss is calculated, which reflects the degree of overlap between the sphere and the boundary. The specific formula is:

[0058] The softplus function is used to smooth the boundary loss, and bound_slope controls the degree of smoothing. The above steps provide a way to estimate the boundary collision volume of the sphere in the simulation box through upper and lower boundary detection and loss calculation.

[0059] Step 404: In order to efficiently and accurately calculate the loss function L, the present disclosure uses automatic differentiation and tensor calculation methods based on PyTorch to implement the following loss function calculation process. The following are the key steps and formulas for calculating the loss function L:

[0060] in: , They respectively represent the weight parameters that control the influence of the three loss terms.

[0061] Through the above implementation, the loss function L can comprehensively evaluate the porosity, overlapping volume and boundary collision volume of the current particle grading scheme, thereby guiding the optimization algorithm to adjust the particle grading to achieve the goal of minimizing the porosity.

[0062] Step 5: Use the optimization algorithm to optimize the sampling ratio of each particle size distribution to minimize the loss function and achieve the particle grading of cement powder with different distributions of minimum porosity to optimize the strength of cement.

[0063] Using intelligent optimization algorithms (such as genetic algorithms, particle swarm optimization algorithms, etc.), the sampling ratio of each particle size distribution is adjusted to minimize the loss function L. In each iteration, the sampling ratio is updated, and sampling and simulation are performed again until the loss function converges or the set number of iterations is reached. Finally, the optimal particle grading scheme that minimizes the porosity is obtained, which corresponds to the optimal ratio of cement powder with different particle size distributions.

[0064] Example 2 In one embodiment of the present disclosure, a cement particle grading optimization system for minimizing porosity is provided, comprising: A data acquisition module, used for acquiring cement particle powders with different particle size distributions; A model building module is used to regard cement particles as spheres and generate a collection of spheres according to cement particles with different particle size distributions; In each set of spheres with a particle size distribution, multiple spheres are sampled in proportion, and a three-dimensional simulated microstructure model is constructed using the multiple spheres; A simulation calculation module is used to simulate the sphere stacking process in a three-dimensional box using a three-dimensional simulation microstructure model. During the simulation process, the porosity in the three-dimensional box is calculated, and a loss function is established based on the overlapping volume between the spheres and the collision volume between the spheres and the boundary. The porosity of cement mixes with different particle size distributions is obtained by a gradient descent method. The optimization module is used to optimize and adjust the sampling ratio of each particle size distribution using an optimization algorithm to minimize the loss function and achieve the particle grading of cement powders with different distributions of minimum porosity to optimize the strength of cement.

[0065] Example 3 In one embodiment of the present disclosure, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the method for optimizing cement particle grading for minimizing porosity is implemented.

[0066] Example 4 In one embodiment of the present disclosure, a non-transitory computer-readable storage medium is provided, wherein the non-transitory computer-readable storage medium is used to store computer instructions, and when the computer instructions are executed by a processor, the cement particle grading optimization method for minimizing porosity is implemented.

[0067] Example 5 In one embodiment of the present disclosure, an electronic device is provided, comprising: a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory, so that the electronic device executes the cement particle grading optimization method for minimizing porosity.

[0068] The present disclosure is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present disclosure. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0069] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0070] Although the above describes the specific implementation methods of the present disclosure in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present disclosure. Technical personnel in the relevant field should understand that on the basis of the technical solution of the present disclosure, various modifications or variations that can be made by those skilled in the art without creative work are still within the scope of protection of the present disclosure.

Claims

1. A method for optimizing cement particle size distribution to minimize porosity, characterized in that: include: Obtaining cement particle powders with different particle size distributions; Considering cement particles as spheres, a collection of spheres is generated based on cement particles with different particle size distributions; In each set of spheres with a particle size distribution, multiple spheres are sampled in proportion, and a three-dimensional simulated microstructure model is constructed using the multiple spheres; The sphere stacking process is simulated in a 3D box using a 3D simulated microstructure model. During the simulation, the porosity in the 3D box is calculated. A loss function is established based on the overlapping volume between spheres and the collision volume between spheres and the boundary. The porosity of cement mixes with different particle size distributions is obtained using the gradient descent method. The optimization algorithm is used to optimize the sampling ratio of each particle size distribution to minimize the loss function and achieve the particle gradation of cement powders with different distributions with minimum porosity to optimize the strength of cement.

2. The method for optimizing cement particle size distribution to minimize porosity according to claim 1, characterized in that: Cement clinker blocks are selected as raw materials, and a ball mill is used to grind the clinker blocks for the first time to achieve a set fineness. The ground clinker powder is sieved through sieves of different mesh sizes, and so on, to obtain cement particle powders with multiple different particle size distributions.

3. The method for optimizing cement particle size distribution to minimize porosity according to claim 1, characterized in that: In the simulation, cement particles are regarded as spheres, and a set of spheres of corresponding sizes is generated according to the different particle size distribution data obtained by screening. It is assumed that the particles in each particle size distribution are evenly distributed in their corresponding particle size range. Samples are taken from each particle size distribution according to a set ratio to obtain a total of N spheres. The N spheres are used to construct a three-dimensional simulated microstructure model according to random or specific rules, and are placed in a three-dimensional box simulated by a computer to simulate the actual stacking process.

4. The method for optimizing cement particle size distribution to minimize porosity according to claim 1, wherein: During the simulation, the porosity of the 3D box is calculated, which is the ratio of the volume not occupied by the spheres to the total volume of the box, taking into account the overlap volume between the spheres and the collision volume between the spheres and the boundary. and Calculate overlap volume : in, is the distance between each pair of neighbor spheres; and is the radius of the sphere; this formula represents the overlapping volume when two spheres intersect ; if , then the calculated is the overlapping volume; if , then the spheres do not intersect and the overlapping volume .

5. The method for optimizing cement particle size distribution to minimize porosity according to claim 1, characterized in that: The loss function calculation process is implemented using automatic differentiation and tensor calculation methods based on PyTorch. The following are the key steps and formulas for calculating the loss function. in: , They represent the weight parameters that control the influence of the three loss terms.

6. The method for optimizing cement particle size distribution to minimize porosity according to claim 1, characterized in that: In each iteration, the sampling ratio is updated, and sampling and simulation are performed again until the loss function converges or the set number of iterations is reached; ultimately, the optimal particle grading scheme that minimizes the porosity is obtained, which corresponds to the optimal ratio of cement powder with different particle size distributions.

7. A cement particle size optimization system for minimizing porosity, characterized in that: include: A data acquisition module, used for acquiring cement particle powders with different particle size distributions; A model building module is used to regard cement particles as spheres and generate a collection of spheres according to cement particles with different particle size distributions; In each set of spheres with a particle size distribution, multiple spheres are sampled in proportion, and a three-dimensional simulated microstructure model is constructed using the multiple spheres; A simulation calculation module is used to simulate the sphere stacking process in a three-dimensional box using a three-dimensional simulation microstructure model. During the simulation process, the porosity in the three-dimensional box is calculated, and a loss function is established based on the overlapping volume between the spheres and the collision volume between the spheres and the boundary. The porosity of cement mixes with different particle size distributions is obtained by a gradient descent method. The optimization module is used to optimize and adjust the sampling ratio of each particle size distribution using an optimization algorithm to minimize the loss function and achieve the particle grading of cement powders with different distributions of minimum porosity to optimize the strength of cement.

8. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method for optimizing cement particle grading for minimizing porosity according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium is used to store computer instructions, and when the computer instructions are executed by a processor, the cement particle grading optimization method for minimizing porosity as described in any one of claims 1-6 is implemented.

10. An electronic device, characterized in that: include: A processor, a memory and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory so that the electronic device executes the cement particle grading optimization method for minimizing porosity as described in any one of claims 1-6.

Images