Aerogel microstructure reconstruction method and system based on potential energy driving

Abstract

The invention relates to an aerogel microstructure reconstruction method and system based on potential energy driving, belongs to the technical field of aerogel materials, and solves the problems of low aerogel microstructure reconstruction precision and difficult porosity control in the prior art. The method comprises the steps of generating target particles based on aerogel particle characteristics, generating a current Brownian motion parameter based on a random diffusion model, calculating aggregation potential energy when the target particles displaced based on the current Brownian motion parameter are in contact with an existing aggregation particle cluster, and when a potential energy threshold value is reached, calculating the current Brownian motion parameter. Determining the target particles and the existing aggregated particle cluster as a current aggregated particle cluster; and calculating the particle number of the current aggregated particle cluster, calculating the current porosity of the current aggregated particle cluster when the contact particle number requirement is met, and outputting the microstructure of the current aggregated particle cluster as the microstructure of the reconstructed target aerogel if the target porosity is reached. And effective control and accurate reconstruction of the porosity of the aerogel microstructure are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of aerogel materials, and particularly relates to an aerogel microstructure reconstruction method and system based on potential energy driving. BACKGROUND

[0002] With the development of high-tech, aerogel materials have irreplaceable application value in key fields such as aerospace thermal protection, new energy battery separators, and high-end equipment insulation due to their unique nano-porous structure, extremely low density, high specific surface area, and excellent thermal / mechanical properties. Among them, the nano-porous-overlapping aerogel constructs a “cross-linking-pore” synergistic structure through controllable overlapping between particles, which not only significantly improves the mechanical stability of the material, but also optimizes the heat blocking efficiency, becoming a research hotspot of high-end functional materials.

[0003] However, the complexity of the nano-porous-overlapping aerogel microstructure (nanoscale particle size, dynamically changing particle overlapping interface, fractal characteristic pore distribution) is an industry technical difficulty for aerosol microstructure reconstruction and analysis. In the prior art, only “geometric contact (center distance ≤ particle radius and )” is used to determine particle aggregation, without considering that the particle aggregation of the nano-porous-overlapping aerogel essentially depends on the “weak attraction-stable cross-linking” mechanism, which deviates from the real physical properties of the aerosol and cannot provide a reliable structural basis for the heat transfer and mechanical property simulation of the aerogel. At the same time, the existing technology assumes that “particles have no overlap” in the aerogel, which cannot achieve accurate control of the porosity.

[0004] Therefore, there is an urgent need for an aerogel microstructure reconstruction method that is more consistent with the physical properties of the aerogel, which can provide a reliable structural basis for the heat transfer and mechanical property simulation of the aerogel material and can be widely applied to the analysis and research and development of the aerogel material. SUMMARY

[0005] In view of the above analysis, the embodiments of the present application aim to provide an aerogel microstructure reconstruction method and system based on potential energy driving to solve the problems of low reconstruction accuracy and difficult porosity control of the existing aerogel microstructure reconstruction.

[0006] In one aspect, the embodiments of the present application provide an aerogel microstructure reconstruction method based on potential energy driving, comprising the following steps: Step 1: generating a target particle based on the characteristics of the aerogel particles and generating current Brownian motion parameters based on a random diffusion model; Step 2: controlling the target particle to displace based on the current Brownian motion parameters and judging whether the target particle after displacement is in contact with an existing aggregated particle cluster; Step 3: when the target particle is in contact with the existing cluster of aggregated particles, calculating an aggregation potential energy of the target particle and the existing cluster of aggregated particles, and determining whether the aggregation potential energy reaches a potential energy threshold value; Step 4: when the aggregation potential energy reaches the potential energy threshold value, determining the target particle and the existing cluster of aggregated particles as a current cluster of aggregated particles; Step 5: calculating a particle number of the current cluster of aggregated particles, and determining whether the particle number meets a contact particle number requirement; Step 6: when the particle number meets the contact particle number requirement, calculating a current porosity of the current cluster of aggregated particles, and determining whether the current porosity reaches a target porosity; Step 7: if the target porosity is reached, outputting a microstructure of the current cluster of aggregated particles as a microstructure of a reconstructed target aerogel.

[0007] Further, the particle features include a particle size feature and a boundary of the simulation region; Generating a target particle based on aerogel particle features includes: obtaining the boundary of the simulation region and the particle size feature; generating the particle radius by using a normal distribution according to the particle size feature; dividing the simulation region into an internal sub-region and an edge sub-region based on the boundary, and generating an initial position in the internal sub-region or the edge sub-region; wherein the initial position is generated in the internal sub-region and the edge sub-region with the same probability; generating the target particle based on the initial position and the particle radius.

[0008] Further, determining whether the target particle after displacement is in contact with an existing cluster of aggregated particles includes: calculating a distance between the target particle after displacement and each aerogel particle in the existing cluster of aggregated particles; determining whether the distance reaches a contact distance threshold value; if the contact distance threshold value is reached, it is determined that the target particle after displacement is in contact with the existing cluster of aggregated particles.

[0009] Further, when the target particle is in contact with the existing cluster of aggregated particles, calculating an aggregation potential energy of the target particle and the existing cluster of aggregated particles, and determining whether the aggregation potential energy reaches a potential energy threshold value includes: extracting an aerogel particle in the existing cluster of aggregated particles that is in contact with the target particle as a contact particle; calculating the aggregation potential energy of the target particle and each of the contact particles using a L-J potential function, wherein the L-J potential function is expressed as: , wherein is the aggregation potential energy, is a L-J potential well depth, is a L-J zero potential point distance, and d is a distance between the target particle and the contact particle; determining whether the aggregation potential energy reaches the potential energy threshold.

[0010] Further, when the aggregation potential energy does not reach the potential energy threshold, generating a new Brownian motion parameter as a current Brownian motion parameter based on the random diffusion model; repeating steps 2-3.

[0011] Further, calculating a current porosity of the current aggregated particle cluster comprises: calculating a total volume of the current aggregated particle cluster; calculating a total overlapping volume of the current aggregated particle cluster; calculating the current porosity based on the total volume and the total overlapping volume.

[0012] Further, calculating a total overlapping volume of the current aggregated particle cluster comprises: selecting two aerogel particles from the current aggregated particle cluster as a current particle pair; calculating a distance between the current particle pair; determining an overlapping condition of the current particle pair based on the distance; selecting an overlapping volume calculation method based on the overlapping condition to calculate an overlapping volume of the current particle pair; determining the total overlapping volume of the current aggregated particle cluster according to the overlapping volumes of all the current particle pairs.

[0013] Further, selecting two aerogel particles from the current aggregated particle cluster as a current particle pair comprises: determining a particle spatial distribution density of the current aggregated particle cluster by a real-time density statistical method of a spatial grid; dividing the current aggregated particle cluster into a plurality of aggregated particle sub-clusters based on the particle spatial distribution density, wherein the aggregated particle sub-clusters comprise a plurality of aerogel particles; selecting one aerogel particle from the aggregated particle sub-clusters as a current particle, and selecting one aerogel particle from the current aggregated particle cluster other than the current particle as a pairing particle, and selecting the current particle and the pairing particle as a current particle pair.

[0014] Further, if the target porosity is not reached, the current aggregated particle cluster is taken as the existing aggregated particle cluster; generate a new aerogel particle as the target particle based on the particle characteristics, and generate a new Brownian motion parameter as a current Brownian motion parameter based on the random diffusion model; Steps 2-6 are repeated.

[0015] In another aspect, an embodiment of the present application provides an aerogel microstructure reconstruction system based on potential energy driving, comprising: A particle generation module for generating a target particle based on aerogel particle characteristics; A Brownian motion simulation module for generating a current Brownian motion parameter based on a random diffusion model, controlling the target particle to displace based on the current Brownian motion parameter, and determining whether the target particle after displacement is in contact with an existing aggregated particle cluster; A potential energy calculation module for calculating the aggregation potential energy of the target particle and the existing aggregated particle cluster when the target particle is in contact with the existing aggregated particle cluster, and determining whether the aggregation potential energy reaches a potential energy threshold; A particle cluster generation module for determining the target particle and the existing aggregated particle cluster as a current aggregated particle cluster when the aggregation potential energy reaches the potential energy threshold; The particle cluster generation module is further configured to calculate the number of particles in the current aggregated particle cluster, and determine whether the number of particles meets a contact particle number requirement; A porosity calculation module for calculating the current porosity of the current aggregated particle cluster when the number of particles meets the contact particle number requirement, and determining whether the current porosity reaches a target porosity; A microstructure determination module for outputting the microstructure of the current aggregated particle cluster as the microstructure of the reconstructed target aerogel when the target porosity is reached.

[0016] Compared with the prior art, the present application can achieve at least one of the following beneficial effects: 1. The introduction of aggregation potential, through the dual control mechanism of contact condition and potential threshold, restores the real aggregation mechanism of "weak attraction-stable crosslinking" of aerogel particles, improves the physical authenticity, realizes the accurate description of the physical mechanism of aerogel, makes the output aerogel microstructure more accurate, and provides a reliable structure basis for the simulation of heat transfer and mechanical properties of aerogel; through the contact particle number requirement, the porosity calculation frequency is reasonably controlled, and the consumption of computing resources is reduced.

[0017] 2. The physical properties of the aerogel particles are described by using the L-J potential function, the 12th order term represents the short-range repulsive force between particles to avoid structural distortion caused by excessive extrusion of particles, and the 6th order term represents the long-range attractive force between particles to accurately describe the weak interaction dominated by van der Waals force; by adjusting the depth of the potential well and the distance from the zero potential point, the interaction strength and range in the aggregation of different types of aerogel particles can be adapted.

[0018] 3. The particle overlap is considered, the particle overlap volume is introduced, and the repeated counting problem of the overlap volume is corrected to improve the accuracy of the porosity calculation, and the porosity reaches the target porosity as the reconstruction stop condition to realize accurate control of the porosity.

[0019] 4. The current aggregated particle cluster is divided into multiple aggregated particle sub-clusters based on the spatial distribution density of the particles, and the particle overlap volume of multiple sub-clusters can be calculated in parallel to effectively improve the porosity calculation efficiency of large-scale aerogel particle aggregation.

[0020] In the present application, the above technical solutions can be combined with each other to realize more preferred combination solutions. Other features and advantages of the present application will be described in the subsequent specification, and some advantages will become apparent from the specification, or will be understood by implementing the present application. The purpose and other advantages of the present application can be achieved and obtained from the contents specifically pointed out in the specification and the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0021] The accompanying drawings are included to provide a further understanding of the application and are incorporated herein and constitute a part of the application. The application will become more apparent from the accompanying drawings as follows. Figure 1 It is a flowchart of an aerogel microstructure reconstruction method based on potential energy driving in an embodiment of the present application. Figure 2 It is a visual example diagram of the aerogel microstructure output by the embodiment of the present application. Figure 3 It is a current porosity change curve example diagram of the embodiment of the present application. Figure 4 It is a schematic diagram of the main modules of an aerogel microstructure reconstruction system based on potential energy driving in an embodiment of the present application. Figure 5 It is a visual parameter input interface example diagram of the embodiment of the present application. DETAILED DESCRIPTION

[0022] The preferred embodiments of the present application will be specifically described below in conjunction with the drawings, wherein the drawings constitute a part of the present application and are used to illustrate the principles of the embodiments of the present application, and are not used to limit the scope of the present application.

[0023] One specific embodiment of the present application discloses an aerogel microstructure reconstruction method based on potential energy driving, as shown in Figure 1 includes Step 1: generating target particles based on aerogel particle characteristics, and generating current Brownian motion parameters based on a random diffusion model.

[0024] Specifically includes: Step S11: generating aerogel particles (i.e. target particles) based on aerogel particle characteristics. Based on the particle characteristic constraints of aerogel, the corresponding target particles are generated based on different types of organic aerogel (such as phenolic, melamine-formaldehyde, and silica aerogel), which ensures the high consistency of the generated target particles with the constituent particles of the aerogel to be reconstructed (i.e. target aerogel), and improves the physical authenticity of subsequent particle aggregation reconstruction.

[0025] Wherein, the particle characteristics include particle size characteristics and the boundary of the simulation region, and the target particles are generated based on the aerogel particle characteristics, specifically including: Firstly, the boundary and the particle size characteristics of the simulation region are obtained. Specifically, the boundary and the particle size characteristics can be extracted from a database or input through a visual interface, which is not limited here. Further, the obtained boundary and particle size characteristics can be verified for parameters to avoid errors in basic data. For example, the particle radius is set to 1m in the particle size characteristics, which does not conform to the usual size range of aerogel particles, or the target porosity is set to 1, which does not conform to the aggregation characteristics of aerogel, etc. will be prompted.

[0026] Secondly, according to the particle size characteristics, the particle radius is generated by using normal distribution. Specifically, in this embodiment, the particle size characteristics include the mean and standard deviation of the particle diameter, and the particle radius is generated by using normal distribution. For example, , wherein r is the particle radius, is an absolute value function, which ensures that the radius is positive, is a function for generating random samples conforming to normal distribution, mean is the mean of particle diameter, and std is the standard deviation.

[0027] Thirdly, based on the boundary, the simulation region is divided into an internal sub-region and an edge sub-region, and an initial position is generated in the internal sub-region or the edge sub-region; wherein the probability of generating the initial position in the internal sub-region and the edge sub-region is the same. Specifically, based on the boundary, the simulation region is divided into an internal sub-region and an edge sub-region, and the initial position is randomly generated in the internal sub-region or the edge sub-region, and the probability of generating the initial position in the internal sub-region or the edge sub-region is controlled to be the same. For example, the boundary is a cube with edge length d, then the internal sub-region is a cube with edge length The center of the internal sub-region is the center of the simulated region, and other regions are the edge sub-regions. The internal sub-regions are connected by... Uniformly distributed within the simulated sub-region, the edge sub-regions can be defined by randomly selecting x / y / z axis boundaries, with the corresponding axis coordinates fixed as "". " or" The remaining axis coordinates are evenly distributed to simulate the actual path of particle diffusion from the edge of the system to the interior during aerogel aggregation.

[0028] Finally, the target particle is generated based on the initial position and the particle radius. Based on the initial position and particle radius, an aerogel particle is generated with its center at the initial position and its radius equal to the particle radius; this is the target particle.

[0029] Understandably, each time a target particle is generated, it can be numbered, a particle index can be established, and the particle radius, initial position, and position after each displacement can be recorded, for example... Let be the particle radius of the target particle generated in the kth generation.

[0030] In this embodiment, target particles are generated one by one each time to achieve precise control of the aggregation process. This avoids structural distortion caused by the large-scale, unconstrained accumulation of aerogel particles without aggregation determination, and ensures that each aerogel particle undergoes subsequent physical mechanism verification.

[0031] Step S12: Generate the current Brownian motion parameters based on the stochastic diffusion model.

[0032] Specifically, a random diffusion model with a fixed displacement size is used to generate the current Brownian motion parameters, balancing physical realism and ease of use. Generate the current Brownian motion parameters, where, Given the current Brownian motion parameters, The standard deviation can be set based on the statistical characteristics of aerogel nanoparticles. This ensures that the single displacement of particles is much smaller than the particle diameter, avoiding non-physical "jump diffusion" and conforming to the slow diffusion characteristics of nanoparticles in sol-gel systems. This represents the number of random numbers. For example, That is, three independent random numbers are generated with a standard deviation of 1.0, which are then used to form a one-dimensional array of length 3. These three random numbers correspond to the displacement components of the particle in three-dimensional space (x-axis, y-axis, z-axis), and together they form a "three-dimensional displacement vector".

[0033] Step 2: Control the target particle to move based on the current Brownian motion parameters, and determine whether the target particle after displacement comes into contact with the existing aggregated particle cluster.

[0034] Specifically, based on the current Brownian motion parameters determined in step 1, the coordinate position information of the target particle is updated to realize the displacement of the target particle. For example, if the initial position coordinates of the target particle are (0, 0, 0) and the Brownian motion parameters are (1, 2, 3), the position coordinates of the target particle are updated to (1, 2, 3) to complete the displacement of the target particle.

[0035] Further, to avoid simulation abnormalities caused by particle overflow and ensure the integrity of the simulation system, the position of the particle exceeding the simulation area is clipped, and when the particle displacement exceeds the boundary of the simulation area, automatic clipping correction is performed. For example, under the periodic boundary condition, when the particle exceeds the physical boundary of the simulation area after Brownian motion, it reenters the simulation area from the opposite boundary. This scheme can eliminate the boundary effect and better represent an infinite material with a limited simulation area.

[0036] For the target particle after completing the displacement, the center coordinates of the target particle are updated, and it is judged whether the target particle is in contact with the existing aggregated particle cluster. Specifically, it includes: First, the distance between the target particle after displacement and each aerogel particle in the existing aggregated particle cluster is calculated. Specifically, the distance between the target particle and each aerogel particle in the existing aggregated particle cluster is calculated, and the minimum distance among them is selected as the distance between the target particle and the existing aggregated particle cluster. The distance is calculated by the three-dimensional Euclidean distance formula, , wherein is the distance between the target particle and the mth aerogel particle, (x k , y k , z k ) is the center coordinates of the target particle, and (x m , y m , z m ) is the center coordinates of the mth aerogel particle in the existing aggregated particle cluster.

[0037] Second, it is judged whether the distance reaches the contact distance threshold. Specifically, the contact distance threshold corresponding to each aerogel particle in the existing aggregated particle cluster is calculated, and the relationship between the distance and the contact distance threshold is judged one by one. The contact distance threshold is calculated by the formula , wherein is the contact distance threshold of the target particle and the mth aerogel particle, is the set distance correction coefficient, for example, 0.9, which is used to avoid structural distortion caused by excessive extrusion of particles and ensure the stability of the crosslinking state, is the particle radius of the target particle, is the particle radius of the mth aerogel particle.

[0038] ​It can be understood that the distance between the target particle and the aerogel particles in the existing aggregated particle cluster and the contact distance threshold can be calculated at the same time, it is judged whether the distance reaches the contact distance threshold, and the aerogel particles in the existing aggregated particle cluster are traversed one by one; or the distance between the target particle and all the aerogel particles in the existing aggregated particle cluster and the contact distance threshold can be calculated respectively first, and the corresponding relationship is recorded, and then the relationship between the distance and the contact distance threshold is judged one by one according to the corresponding relationship.

[0039] Finally, if the contact distance threshold is reached, it is determined that the target particle after displacement is in contact with the existing aggregated particle cluster. Specifically, if there is an aerogel particle in the existing aggregated particle cluster, the corresponding distance is less than the contact distance threshold, for example , it is determined that the target particle after displacement is in contact with the existing aggregated particle cluster.

[0040] It should be noted that if the aggregated particle cluster is not found, that is, there is only the target particle in the simulation region, for example, the number of target particle generations can be used for judgment, for example, when the number of generations is 1, there is only the target particle, the target particle is taken as the existing aggregated particle cluster, and at the same time, a new target particle is generated according to step 1.

[0041] In other embodiments, after the target particle is generated, it can be judged whether the target particle is in contact with the existing aggregated particle cluster, and for the non-contact, the position of the target particle is controlled based on the current Brownian motion parameters, and then it is judged whether the target particle after displacement is in contact with the existing aggregated particle cluster.

[0042] Step 3: When the target particle is in contact with the existing aggregated particle cluster, the aggregation potential energy of the target particle and the existing aggregated particle cluster is calculated, and it is judged whether the aggregation potential energy reaches the potential energy threshold.

[0043] Through the double judgment of contact judgment and potential energy threshold, the real aggregation behavior of the aerogel particles is restored, and the accurate determination of "stable crosslinking" is realized. When the target particle is in contact with the existing aggregated particle cluster, the aggregation potential energy of the target particle and the existing aggregated particle cluster is calculated, and it is judged whether the aggregation potential energy reaches the potential energy threshold. Specifically, it includes: First, the aerogel particles in the existing aggregated particle cluster that are in contact with the target particle are extracted as contact particles. Specifically, the aerogel particles in the existing aggregated particle cluster that have a distance from the target particle less than the contact distance threshold can be taken as the contact particles by using the judgment result of the distance between the target particle and the aerogel particles in the existing aggregated particle cluster and the contact distance threshold in step 2. In other embodiments, the contact judgment can be performed by setting different values for the collection distance correction coefficient according to the principle of step 2, which is not limited here.

[0044] Secondly, the aggregation potential energy corresponding to the target particle and each of the contacting particles is calculated using the LJ potential energy function; wherein, the LJ potential energy function is expressed as: , In the formula, To concentrate potential energy, the unit is joule, representing the strength of the attraction between particles. The LJ potential well depth is expressed in joules. LJ zero potential energy point distance, in nanometers, represents the distance between two particles when the aggregation potential energy is 0, and d is the distance between the target particle and the contacting particle, in nanometers.

[0045] in, The larger the value, the stronger the binding energy and the stronger the interaction between the two particles. This represents the distance between two particles when the aggregation potential energy is 0. It can be set according to the type of the target aerogel; for example, for carbon aerogel... Set it to 1.2. Set to 1.0. The spacing d is the same as the spacing in step 2. The calculation principle is the same, so it will not be repeated here.

[0046] The 12th term in the LJ potential energy function (i.e.) This describes the short-range repulsive force between the target particle and the contacting particles: when the particle spacing... At this time, the repulsive force increases rapidly, avoiding structural distortion caused by excessive particle compression, which matches the aggregation characteristic of aerogel particles of "contact but not dense"; the 6th term (i.e. Describes the long-range attractive force between particles: when the distance between particles... At this time, attraction plays a dominant role, accurately simulating the "weak attraction" effect dominated by van der Waals forces at the nanoscale, and accurately simulating the core physical mechanism between nanoscale aerogel particles; when At this time, the particles are in a net attraction state. Combined with the geometric spacing condition limited by the contact distance threshold, the accurate determination of "stable cross-linking" is achieved, which solves the problem of loose structure caused by "only geometric contact" in traditional methods.

[0047] Finally, it is determined whether the aggregation potential energy reaches the potential energy threshold. Specifically, it is determined whether the aggregation potential energy between the target particle and each contacting particle reaches the potential energy threshold. In this embodiment, the potential energy threshold is set to 0.

[0048] Step 4: When the aggregation potential energy reaches the potential energy threshold, the target particle and the existing aggregated particle cluster are identified as the current aggregated particle cluster.

[0049] Specifically, when the aggregation potential energy of the target particle and each contact particle reaches the potential energy threshold, it is determined that the aggregation potential energy of the target particle and the existing aggregated particle cluster reaches the potential energy threshold, at this time, the target particle and the existing aggregated particle cluster form a "weakly attractive-stable crosslinking" aggregation state, and the target particle and the existing aggregated particle cluster are determined as the current aggregated particle cluster.

[0050] Further, when the aggregation potential energy does not reach the potential energy threshold, that is, the aggregation potential energy of the target particle and a certain contact particle does not reach the potential energy threshold, a new Brownian motion parameter is generated as the current Brownian motion parameter based on the random diffusion model, and steps 2-3 are repeatedly executed, that is, the displacement of the target particle is controlled based on the new Brownian motion parameter, and it is judged whether the target particle after displacement again contacts the existing aggregated particle cluster; when contacting, the aggregation potential energy of the target particle and the existing aggregated particle cluster is calculated, and it is judged whether the aggregation potential energy reaches the potential energy threshold, if reaching, step 5 is executed, and if not reaching, the cycle is executed again.

[0051] Step 5: Calculate the number of particles of the current aggregated particle cluster, and judge whether the number of particles meets the contact particle number requirement.

[0052] Specifically, the number of particles of the current aggregated particle cluster is calculated, and in the embodiment, the number of particles of the current aggregated particle cluster can be recorded as the number of times of generating the target particle. Whether the number of particles meets the contact particle number requirement is judged, and the contact particle number requirement is used to control the calculation frequency of the current porosity, so as to reduce the consumption of calculation resources. Through the contact particle number requirement, a dynamic value can be set, for example, when the number of particles does not exceed 1000, the contact particle number requirement is an integer multiple of the number of particles of 50, that is, the current porosity is calculated once every 50 new particles, and when the number of particles is more than 1000, the contact particle number requirement is an integer multiple of the number of particles of 20, so as to realize the calculation of the current porosity with wide interval in the early stage and relatively dense calculation of the current porosity in the later stage, and balance the calculation resources and the demand for accurate control of porosity. In other embodiments, time can also be controlled, which is not limited here.

[0053] Step 6: When the number of particles meets the contact particle number requirement, the current porosity of the current aggregated particle cluster is calculated, and it is judged whether the current porosity reaches the target porosity.

[0054] Specifically, it includes: Step S61: Calculate the total volume of the current aggregated particle cluster.

[0055] The total volume of the current aggregated particle cluster represents the sum of the volumes of each particle in the current aggregated particle cluster, which is calculated by the following formula: , In the formula, V is the total volume of the current aggregated particle cluster, and V i is the volume of the i th particle in the current aggregated particle cluster. Vtotal is the total volume, N represents the total number of aerogel particles in the current aggregated particle cluster, Rn is the particle radius of the nth aerogel particle.

[0056] Step S62: Calculate the total overlap volume of the current aggregated particle cluster.

[0057] Specifically, the total overlap volume of the current aggregated particle cluster is calculated by calculating the overlap volume of each two aerogel particles in the current aggregated particle cluster one by one.

[0058] First, any two aerogel particles in the current aggregated particle cluster are selected as the current particle pair. The overlap volume calculation can be directly performed on any two aerogel particles in the current aggregated particle cluster as the current particle pair.

[0059] Further, for the case of more particles and larger calculation task, grouping and parallel calculation can be performed according to the particle spatial distribution density, which specifically includes: 1) The particle spatial distribution density of the current aggregated particle cluster is determined by the real-time density statistical method of spatial grid. Specifically, the real-time density statistical method based on spatial grid is an effective spatial data analysis method, which can monitor and analyze the spatial distribution in real time through grid division and density estimation. In this embodiment, the particle spatial distribution density of the current aggregated particle cluster is determined by the real-time density statistical method of spatial grid, which can be realized by using the existing technology, and thus will not be described here.

[0060] 2) The current aggregated particle cluster is divided into multiple aggregated particle sub-clusters based on the particle spatial distribution density; wherein the aggregated particle sub-cluster includes multiple aerogel particles. Specifically, based on the particle spatial distribution density, the calculation task is divided and weighted according to the task mapping of load prediction. For example, the mapping relationship between the particle spatial distribution density and the number of task processes or the number of CPU cores of the computer, according to the particle spatial distribution density of the current aggregated particle cluster, the number of task processes or different computers are selected for grouping and parallel calculation. For the particle dense region, the calculation resource is increased, and for the sparse region, the calculation resource is reduced, which further improves the parallel calculation efficiency and memory usage efficiency, and solves the problem of uneven calculation load caused by particle aggregation in the aerogel reconstruction process.

[0061] 3) selecting one aerogel particle from the cluster of aggregated particles as a current particle, selecting one aerogel particle other than the current particle from the cluster of aggregated particles as a counterpart particle, and taking the current particle and the counterpart particle as a current particle pair. Specifically, for a cluster of aggregated particles, selecting one aerogel particle as a current particle, selecting one aerogel particle other than the current particle from the cluster of aggregated particles as a counterpart particle (i.e. the counterpart particle can or can not be in the same cluster of aggregated particles as the current particle), and taking the current particle and the counterpart particle as a current particle pair.

[0062] Secondly, the distance between the current particle pair is calculated.

[0063] Specifically, the distance between the current particle pair is calculated using the Euclidean distance formula in three-dimensional space, where (x i , y i , z i ), (x j , y j , z j ) are the center coordinates of the two aerogel particles in the current particle pair.

[0064] Thirdly, the overlap condition of the current particle pair is determined based on the distance.

[0065] Specifically, when the distance is greater than or equal to the sum of the radii of the two aerogel particles in the current particle pair, the current particle pair is determined to be in a non-overlapping state; when the distance is less than or equal to the difference between the radii of the two aerogel particles, the current particle pair is determined to be in a completely overlapping state; when the distance is less than the sum of the radii of the two aerogel particles and greater than the difference between the radii of the two aerogel particles, the current particle pair is determined to be in a partially overlapping state.

[0066] Then, based on the overlap condition, the overlapping volume calculation method is selected to calculate the overlapping volume of the current particle pair.

[0067] Specifically, when the current particle pair is in a non-overlapping state, the overlapping volume is a default value, which is 0; when the current particle pair is in a completely overlapping state, the overlapping volume is calculated using the formula , where V is the overlapping volume of the current particle pair, , are the radii of the two aerogel particles in the current particle pair, i.e. the volume of the smaller aerogel particle is taken as the overlapping volume; when the current particle pair is in a partially overlapping state, the overlapping volume is calculated using the formula , where V , , , wherein a, b are auxiliary parameters in partially overlapped scenes, with units of nanometers, representing the "half long radius" and "half short radius" in the direction of the center line of the aerogel particle, respectively; is the height of the overlapping area, with units of nanometers, ensuring the authenticity of the three-dimensional volume calculation.

[0068] Finally, the total overlapping volume of the current aggregated particle cluster is determined according to the overlapping volumes of all the current particle pairs.

[0069] The total overlapping volume of the current aggregated particle cluster is the sum of the overlapping volumes of all the current particle pairs.

[0070] In other embodiments, the grouping parallel calculation can split the task by the "index of the current aggregated particle cluster". First, the coordinates and radius array of all the aggregated particles are obtained, and then the particle index is split into multiple continuous index groups according to the number of CPU cores. Each index group corresponds to a group of aerogel particles. When performing grouping parallel calculation, the compute_overlap_volumes function is independently called for each group to calculate the overlapping volume of each particle in the group with each aerogel particle in the current aggregated particle cluster (including all aerogel particles in the current group and other groups). The overlapping volume of each "in-group aerogel particle" with other "all aerogel particles" is calculated to ensure that no overlapping volume of any cross-group particle pair is missed, and all possible particle interactions are completely covered. The total overlapping volume is summarized as follows: after each index group completes the calculation, it returns an overlapping volume array of "in-group particle number x total particle number". First, the sum of each array is obtained to get the local overlapping volume of each group, and then the local sums of all groups are summarized. Since the overlapping volume of any two particles is calculated twice, the porosity needs to be divided by 2 to correct the repeated calculation in the subsequent calculation.

[0071] Step S63: calculating the current porosity based on the total volume and the total overlapping volume.

[0072] The current porosity is calculated based on the total volume and the total overlapping volume, and the following formula is used for calculation: , wherein, is the current porosity, is the total volume, is the total overlapping volume, which represents the sum of the overlapping volumes of all the aggregated particles, and since each aerogel particle is calculated twice, it is divided by 2 to correct, is the total volume of the simulation region, which is calculated based on the boundary.

[0073] For example, the boundary of the simulation region is a cube with a side length of l, and the total volume of the simulation region is ; the boundary is: a sphere with radius R, then the total volume of the simulation region .

[0074] Step 7: If the target porosity is reached, output the microstructure of the current aggregated particle cluster as the microstructure of the reconstructed target aerogel.

[0075] Specifically, when the current porosity reaches the target porosity, output the microstructure of the current aggregated particle cluster as the microstructure of the reconstructed target aerogel. The target porosity can be set according to the needs of the target aerogel, and when the difference between the current porosity and the target porosity is less than the error limit value, it is determined that the current porosity reaches the target porosity. For example, the error limit value is set to 0.01, and when , it is determined that the current porosity reaches the target porosity, where is the current porosity, is the target porosity.

[0076] Further, if the target porosity is not reached, the current aggregated particle cluster is taken as the existing aggregated particle cluster, a new aerogel particle is generated based on the particle characteristics as the target particle, and a new Brownian motion parameter is generated based on the random diffusion model as the current Brownian motion parameter; repeat steps 2-6.

[0077] Specifically, when the current porosity does not reach the target porosity, it means that the aggregation structure of the current aggregated particle cluster does not meet the requirements of the target aerogel. The current aggregated particle cluster is taken as the existing aggregated particle cluster, a new aerogel particle is generated based on the particle characteristics as the target particle, and a new Brownian motion parameter is generated based on the random diffusion model as the current Brownian motion parameter, starting a new cycle until the current porosity reaches the target porosity.

[0078] With the target porosity as the termination condition, the microstructure of the current aggregated particle cluster is output as the microstructure of the reconstructed target aerogel, ensuring that the reconstruction result accurately matches the requirements of the target aerogel.

[0079] Further, the microstructure output supports multiple formats to meet subsequent analysis needs, such as a particle characteristics CSV file containing core parameters such as particle ID, three-dimensional coordinates (X / Y / Z), diameter, radius, etc., which can be directly imported into third-party software such as ANSYS and COMSOL for heat transfer / mechanical performance simulation; a porosity evolution CSV file containing simulation steps, aggregated particle number, real-time porosity, and deviation from the target porosity, etc. data, facilitating analysis of the porosity convergence process; a simulation log file recording parameter configuration, simulation progress, key calculation results, etc. information, facilitating problem tracing and simulation reproduction.

[0080] Further, the embodiment method supports particle motion visualization, for example, three-dimensional microstructure visualization, which can show the spatial distribution of aggregated particles through matplotlib 3D plotting technology, as shown in Figure 2 The porosity evolution curve visualization outputs the current porosity and particle number at each time, and forms a change curve of the current porosity and the calculation number, as shown in Figure 3 The horizontal coordinate step represents the calculation number of the current porosity, and the target porosity and the allowable deviation range are marked to intuitively verify the convergence of the simulation.

[0081] The embodiment of the present application provides an aerogel microstructure reconstruction system based on potential energy driving, as shown in Figure 4 The embodiment of the present application provides an aerogel microstructure reconstruction system based on potential energy driving, as shown in The particle generation module is configured to generate a target particle based on aerogel particle characteristics; The Brownian motion simulation module is configured to generate current Brownian motion parameters based on a random diffusion model, control the target particle to displace based on the current Brownian motion parameters, and judge whether the target particle after displacement contacts an existing aggregated particle cluster; The potential energy calculation module is configured to calculate the aggregation potential energy of the target particle and the existing aggregated particle cluster when the target particle contacts the existing aggregated particle cluster, and judge whether the aggregation potential energy reaches a potential energy threshold; The particle cluster generation module is configured to determine the target particle and the existing aggregated particle cluster as a current aggregated particle cluster when the aggregation potential energy reaches the potential energy threshold; The particle cluster generation module is further configured to calculate the particle number of the current aggregated particle cluster, and judge whether the particle number meets a contact particle number requirement; The porosity calculation module is configured to calculate the current porosity of the current aggregated particle cluster when the particle number meets the contact particle number requirement, and judge whether the current porosity reaches a target porosity; The microstructure determination module is configured to output the microstructure of the current aggregated particle cluster as the microstructure of the reconstructed target aerogel when the target porosity is reached.

[0082] Further, the particle generation module is further configured to generate a new target particle based on particle characteristics again when the particle number does not meet the contact particle number requirement, or the current porosity does not reach the target porosity.

[0083] The particle generation module is further configured to provide a visual parameter input interface, as shown in Figure 5As shown, an intuitive input space (numerical box) is provided to realize the setting management of key parameters, obtain the boundary and particle size characteristics of the simulation area, and further determine the initial position and particle radius of the target particle. The restrictions used by the module can also be input, such as the potential energy threshold, target porosity, number of aggregated particle clusters, etc., and the input parameters can be set as a template for subsequent use.

[0084] The Brownian motion simulation module is also used to generate new Brownian motion parameters based on the random diffusion model when the target particle does not contact the existing aggregated particle cluster or when the aggregation potential energy does not reach the potential energy threshold, and to control the displacement of the target particle based on the new Brownian motion parameters, to determine whether the target particle contacts the existing aggregated particle cluster after displacement again, and if not, to repeatedly generate Brownian motion parameters and control the displacement of the target particle until the target particle contacts the existing aggregated particle cluster.

[0085] In summary, the method and system for reconstructing the microstructure of aerogel based on potential energy driving according to the embodiments of the present application have at least one of the following beneficial effects: 1. The introduction of aggregation potential energy restores the real aggregation mechanism of "weak attraction-stable crosslinking" of aerogel particles through a double control mechanism of contact conditions and potential energy thresholds, improves physical authenticity, realizes accurate description of the physical mechanism of aerosol, and makes the output aerogel microstructure more accurate, providing a reliable structural basis for the simulation of the heat transfer and mechanical properties of aerogel; the calculation frequency of porosity is reasonably controlled through the requirement of the number of contact particles, reducing the consumption of computing resources.

[0086] 2. The L-J potential energy function is used to describe the physical properties of "weak attraction-stable crosslinking" of aerogel particles, the 12th order term represents the short-range repulsive force between particles to avoid structural distortion caused by excessive compression of particles, and the 6th order term represents the long-range attractive force between particles to accurately describe the weak interaction dominated by van der Waals force; by adjusting the potential well depth and the distance from the zero potential point, the interaction strength and range in the aggregation of different types of aerogel particles can be adapted.

[0087] 3. The particle overlap volume is introduced to consider the particle overlap and correct the repeated counting problem of the overlap volume, improving the accuracy of porosity calculation, and taking the porosity reaching the target porosity as the stopping condition for reconstruction to realize accurate control of porosity.

[0088] 4. The current aggregated particle cluster is divided into multiple aggregated particle clusters based on the spatial distribution density of particles, and multiple sub-clusters can calculate the particle overlap volume in parallel, effectively improving the porosity calculation efficiency when a large number of aerogel particles are aggregated.

[0089] Those skilled in the art can understand that all or part of the processes of the above-mentioned embodiment methods can be completed by instructing the relevant hardware by a computer program, and the program can be stored in a computer readable storage medium. The computer readable storage medium is a disk, an optical disk, a read-only memory, a random access memory, etc.

[0090] The above description is merely preferred specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application.

Claims

1. A method for reconstructing microstructure of aerogels based on potential energy driving, characterized in that, The method comprises the following steps: Step 1: generating a target particle based on the characteristics of the aerogel particles, and generating current Brownian motion parameters based on a random diffusion model; Step 2: controlling the target particle to displace based on the current Brownian motion parameters, and judging whether the target particle after displacement is in contact with an existing aggregated particle cluster; Step 3: when the target particle is in contact with the existing aggregated particle cluster, calculating the aggregation potential energy of the target particle and the existing aggregated particle cluster, and judging whether the aggregation potential energy reaches a potential energy threshold; Step 4: when the aggregation potential energy reaches the potential energy threshold, determining the target particle and the existing aggregated particle cluster as a current aggregated particle cluster; Step 5: calculating the number of particles of the current aggregated particle cluster, and judging whether the number of particles meets a contact particle number requirement; Step 6: when the number of particles meets the contact particle number requirement, calculating the current porosity of the current aggregated particle cluster, and judging whether the current porosity reaches a target porosity; Step 7: if the target porosity is reached, outputting the microstructure of the current aggregated particle cluster as the microstructure of the reconstructed target aerogel.

2. The method of claim 1, wherein, The particle characteristics include particle size characteristics and a boundary of a simulation region; The generating of the target particle based on the characteristics of the aerogel particles comprises: obtaining the boundary of the simulation region and the particle size characteristics; generating the particle radius by using a normal distribution according to the particle size characteristics; dividing the simulation region into an internal sub-region and an edge sub-region based on the boundary, and generating an initial position in the internal sub-region or the edge sub-region; wherein the initial position is generated in the internal sub-region and the edge sub-region with the same probability; generating the target particle based on the initial position and the particle radius.

3. The method of claim 1, wherein, The judging of whether the target particle after displacement is in contact with the existing aggregated particle cluster comprises: calculating the distance between the target particle after displacement and each aerogel particle in the existing aggregated particle cluster; judging whether the distance reaches a contact distance threshold; if the contact distance threshold is reached, it is determined that the target particle after displacement is in contact with the existing aggregated particle cluster.

4. The method of claim 1, wherein, When the target particle is in contact with the existing aggregated particle cluster, the calculation of the aggregation potential energy of the target particle and the existing aggregated particle cluster, and the judgment of whether the aggregation potential energy reaches a potential energy threshold, comprise: extracting the aerogel particles in the existing aggregated particle cluster that are in contact with the target particle as contact particles; calculating the aggregation potential energy of the target particle and each contact particle by using an L-J potential energy function; wherein the L-J potential energy function is expressed as: wherein is the potential of aggregation, is the L-J potential well depth, is the L-J zero potential point distance, d is the distance between the target particle and the contact particle; judging whether the aggregation potential energy reaches the potential energy threshold.

5. The method according to claim 1, wherein: when the aggregation potential energy does not reach the potential energy threshold, generating new Brownian motion parameters as current Brownian motion parameters based on the random diffusion model; repeating steps 2-3.

6. The method of claim 1, wherein, The calculation of the current porosity of the current aggregated particle cluster comprises: calculating the total volume of the current aggregated particle cluster; calculating the total overlapping volume of the current aggregated particle cluster; calculating the current porosity based on the total volume and the total overlap volume.

7. The method of claim 6, wherein, calculating the total overlap volume of the current aggregated particle cluster, comprising: selecting two aerogel particles from the current aggregated particle cluster as a current particle pair; calculating the distance of the current particle pair; judging the overlap condition of the current particle pair based on the distance; selecting an overlap volume calculation method based on the overlap condition, and calculating the overlap volume of the current particle pair; determining the total overlap volume of the current aggregated particle cluster according to the overlap volumes of all the current particle pairs.

8. The method of claim 7, wherein, selecting two aerogel particles from the current aggregated particle cluster as a current particle pair, comprising: determining the particle spatial distribution density of the current aggregated particle cluster by a real-time density statistical method of a spatial grid; dividing the current aggregated particle cluster into a plurality of aggregated particle sub-clusters based on the particle spatial distribution density; wherein the aggregated particle sub-cluster comprises a plurality of aerogel particles; selecting one aerogel particle from the aggregated particle sub-cluster as a current particle, and selecting one aerogel particle from the current aggregated particle cluster other than the current particle as a counterpart particle, and taking the current particle and the counterpart particle as a current particle pair.

9. The method of claim 1, wherein, if the target porosity is not reached, taking the current aggregated particle cluster as the existing aggregated particle cluster; generating a new aerogel particle as the target particle based on the particle characteristics, and generating a new Brownian motion parameter as a current Brownian motion parameter based on the random diffusion model; repeating steps 2-6.

10. A potential energy driven aerogel microstructure reconfiguration system, comprising: comprising: a particle generation module configured to generate a target particle based on aerogel particle characteristics; a Brownian motion simulation module configured to generate a current Brownian motion parameter based on a random diffusion model, control the target particle to displace based on the current Brownian motion parameter, and judge whether the target particle after displacement contacts an existing aggregated particle cluster; a potential energy calculation module configured to calculate the aggregation potential energy of the target particle and the existing aggregated particle cluster when the target particle contacts the existing aggregated particle cluster, and judge whether the aggregation potential energy reaches a potential energy threshold; a particle cluster generation module configured to determine the target particle and the existing aggregated particle cluster as a current aggregated particle cluster when the aggregation potential energy reaches the potential energy threshold; the particle cluster generation module is further configured to calculate the number of particles of the current aggregated particle cluster, and judge whether the number of particles meets a contact particle number requirement; a porosity calculation module configured to calculate a current porosity of the current aggregated particle cluster when the number of particles meets the contact particle number requirement, and judge whether the current porosity reaches a target porosity; a microstructure determination module configured to output the microstructure of the current aggregated particle cluster as the microstructure of a reconstructed target aerogel when the target porosity is reached.

Images