Foaming ceramic plate performance analysis method and device

Abstract

The invention relates to the technical field of laser welding, in particular to a foamed ceramic plate performance analysis method and device.The foamed ceramic plate performance analysis method comprises the steps that surface scanning is conducted on a foamed ceramic plate sample to obtain collected data and a stereoscopic image sequence, and a three-dimensional pore model is constructed according to a surface feature matrix and the stereoscopic image sequence; carrying out topology and geometric feature extraction processing on the three-dimensional pore model to obtain a pore feature vector, and carrying out thermal performance parameter derivation processing on the pore feature vector to generate a thermal conductivity prediction formula; according to the pore feature vector and a thermal conductivity prediction formula, iteratively calculating influence data of pores on stress distribution, and generating a mechanical property prediction formula; and constructing a performance index matrix according to the thermal conductivity prediction formula and the mechanical property prediction formula, and performing optimization processing on the performance index matrix to obtain a comprehensive performance score table. According to the method, through combination of microstructure analysis and macroscopic performance prediction, the accuracy and comprehensiveness of comprehensive performance evaluation of the foamed ceramic plate are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of ceramic plate performance analysis, and in particular to a method and device for analyzing the performance of a foamed ceramic plate. Background Art

[0002] Foamed ceramic panels are lightweight, high-strength, and high-temperature-resistant building materials that have been widely used in the construction, industrial, and decorative fields in recent years. Existing methods for analyzing the performance of foamed ceramic panels primarily rely on physical property testing and basic parameter measurements, such as experimental determination of density, compressive strength, and thermal conductivity, to obtain macroscopic performance data through destructive or non-destructive testing of samples. However, these methods face a significant technical challenge when analyzing the comprehensive performance of foamed ceramic panels: they are unable to fully characterize and predict the material's performance in actual application scenarios.

[0003] The performance analysis methods of existing technologies mainly focus on the measurement of a single performance parameter, such as a separate test of thermal conductivity or mechanical strength. Although this analysis method can provide quantitative data for specific performance, it lacks an in-depth analysis of the correlation between the microstructure of the material and the macroscopic performance. The performance of foamed ceramic panels depends not only on their macroscopic physical properties, but is also closely related to their microstructures such as the distribution, morphology and connectivity of their internal pores. For example, differences in pore structure significantly affect the thermal conductivity and mechanical response of the material, but existing technologies generally ignore the comprehensive impact of these microscopic features on performance. As a result, in actual applications, performance predictions based on a single parameter often deviate from actual performance, especially under complex working conditions, such as performance stability analysis under high temperature, high pressure or long-term loads. Existing methods lack systematic integrated analysis methods when processing multidimensional performance data and cannot provide comprehensive guidance for material optimization design.

[0004] Therefore, a new foam ceramic plate performance analysis method is urgently needed to solve the above problems. Summary of the Invention

[0005] The main purpose of the present invention is to provide a method and device for analyzing the performance of foamed ceramic panels, aiming to overcome the technical problem that the existing technology cannot fully characterize and predict the performance of foamed ceramic panels in actual application scenarios.

[0006] In order to achieve the above-mentioned invention problem, the present invention proposes a method for analyzing the performance of a foamed ceramic plate, the method comprising: Scanning the surface of the foamed ceramic plate sample to obtain collected data, and integrating the collected data to form a surface feature matrix; Scanning the surface pores of the foamed ceramic plate sample to obtain a stereoscopic image sequence, and constructing a three-dimensional pore model based on the surface feature matrix and the stereoscopic image sequence; Performing topological and geometric feature extraction processing on the three-dimensional pore model to obtain a pore feature vector, performing thermal performance parameter derivation processing on the pore feature vector to generate a thermal conductivity prediction formula; Iteratively calculating the influence data of pores on stress distribution based on the pore characteristic vector and thermal conductivity prediction formula to generate a mechanical property prediction formula; A performance index matrix is constructed according to the thermal conductivity prediction formula and the mechanical property prediction formula, and the performance index matrix is optimized to obtain a comprehensive performance score sheet.

[0007] Furthermore, the step of scanning the surface of the foamed ceramic plate sample to obtain collected data and integrating the collected data to form a surface feature matrix includes: The surface of the foamed ceramic plate sample is scanned using a scanning device to obtain a microstructure image set; extracting pore boundaries from the microstructure image set to generate a pore boundary image; Performing pore region segmentation processing on the pore boundary image, calculating the area ratio and shape parameters of each pore region, and generating a pore geometric feature set; Based on the energy dispersive spectroscopy technology, the foamed ceramic plate sample is synchronously scanned for element distribution, generating an element distribution data set containing the spatial coordinates and proportion data of the element distribution; The pore geometric feature set and the element distribution data set are subjected to multi-source data fusion processing to obtain a surface feature matrix.

[0008] Furthermore, the step of scanning the surface pores of the foamed ceramic plate sample to obtain a stereoscopic image sequence and constructing a three-dimensional pore model according to the surface feature matrix and the stereoscopic image sequence includes: Scan the foamed ceramic plate sample using an X-ray scanning device to collect multi-view projection data including pore and matrix information to obtain an initial stereoscopic image sequence; performing noise suppression processing on the initial stereoscopic image sequence to obtain a smoothed stereoscopic image sequence; Setting the surface feature matrix as a reference, performing density correction processing on the smoothed stereoscopic image sequence to obtain a corrected stereoscopic image sequence; Setting the surface feature matrix as a boundary constraint condition, performing pore segmentation processing on the corrected stereoscopic image sequence to obtain a pore image sequence including pores and matrix areas; The pore regions of the pore image sequence are converted into a three-dimensional grid structure to generate an initial three-dimensional pore model, and topology optimization is performed on the initial three-dimensional pore model to obtain a three-dimensional pore model.

[0009] Furthermore, the step of performing topological and geometric feature extraction processing on the three-dimensional pore model to obtain a pore feature vector includes: Performing spatial discretization processing on the three-dimensional pore model to generate a discrete grid data set including voxel units; Calculating the distance between each voxel unit in the discrete grid data set and the pore boundary, and performing geometric analysis processing to obtain pore diameter distribution data, wherein the value of each voxel represents the shortest distance to the pore wall; Performing connectivity analysis on the pore diameter distribution data to generate a pore network topology diagram comprising nodes and connectivity paths, wherein the connectivity paths are pore channels between nodes; The pore network topology graph is processed by dimensionality reduction and statistical analysis to obtain a pore characteristic vector.

[0010] Furthermore, the step of deriving thermal performance parameters of the pore characteristic vector to generate a thermal conductivity prediction formula includes: performing nonlinear normalization processing on the pore characteristic vector to generate a normalized characteristic matrix; Performing weight analysis on each feature of the standardized feature matrix, setting regularization parameters to rank the importance of features, and generating a key pore feature set; Constructing a nonlinear mapping relationship between the pore feature set and thermal conductivity to form a preliminary thermal conductivity model, and iteratively optimizing the preliminary thermal conductivity model to obtain an optimized thermal conductivity model; The weight coefficient of each feature is calculated according to the optimized thermal conductivity model, and the weight coefficient is subjected to feature mapping processing to form a thermal conductivity prediction formula.

[0011] Furthermore, the step of iteratively calculating the data on the influence of pores on stress distribution based on the pore characteristic vector and the thermal conductivity prediction formula to generate a mechanical property prediction formula includes: Performing finite element meshing processing on the three-dimensional pore model to obtain a finite element mesh model; Setting mechanical boundary conditions on the finite element mesh model according to the pore characteristic vector to obtain a mechanical simulation input model; Performing strain field iterative calculation processing on the mechanical simulation input model to obtain a stress distribution data set; The mechanical property parameters in the stress distribution data set are extracted, the mapping relationship between the pore characteristics and the mechanical property parameters is evaluated, and a mechanical property prediction formula is generated.

[0012] Furthermore, the step of constructing a performance index matrix based on the thermal conductivity prediction formula and the mechanical property prediction formula, and optimizing the performance index matrix to obtain a comprehensive performance score sheet includes: Extracting multiple performance parameters of the thermal conductivity prediction formula and the mechanical property prediction formula, unifying the multiple parameters, and generating an initial performance index matrix including multiple performance indexes; Normalizing the initial performance indicator matrix to obtain a standardized performance indicator matrix; quantifying the degree of correlation between each performance indicator in the standardized performance indicator matrix and the ideal performance based on a grey correlation analysis algorithm to obtain an indicator correlation vector; Analyzing the information entropy value of the indicator correlation vector, calculating the weight value of each indicator according to the information entropy value, and generating a performance indicator weight set; Performing weighted fusion processing on the standardized performance indicator matrix according to the performance indicator weight set to obtain a comprehensive performance score vector; Data formatting is performed on the comprehensive performance score vector to obtain the comprehensive performance score table.

[0013] The present invention also provides a foamed ceramic plate performance analysis device, comprising: A first acquisition module is used to scan the surface of the foamed ceramic plate sample to obtain acquisition data, and integrate the acquisition data to form a surface feature matrix; The second acquisition module is used to scan the surface pores of the foamed ceramic plate sample to obtain a stereoscopic image sequence, and to construct a three-dimensional pore model based on the surface feature matrix and the stereoscopic image sequence; a first derivation module, configured to extract topological and geometric features of the three-dimensional pore model to obtain a pore characteristic vector, derive thermal performance parameters of the pore characteristic vector, and generate a thermal conductivity prediction formula; A second derivation module iteratively calculates the influence data of pores on stress distribution based on the pore characteristic vector and the thermal conductivity prediction formula to generate a mechanical property prediction formula; The optimization module is used to construct a performance index matrix based on the thermal conductivity prediction formula and the mechanical property prediction formula, and optimize the performance index matrix to obtain a comprehensive performance score table.

[0014] The present invention further provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method when executing the computer program.

[0015] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the above method when executed by a processor.

[0016] Compared with the prior art, this application has the following beneficial effects: This application proposes a method and device for analyzing the performance of foamed ceramic panels. This method constructs a three-dimensional pore model using surface scanning and volumetric image sequences to capture the microscopic pore structure characteristics of the foamed ceramic panels. Compared to existing methods that focus solely on measuring single macroscopic parameters, this method can deeply explore the correlation between microscopic characteristics such as pore distribution, morphology, and connectivity and material properties. By extracting topological and geometric features and generating pore feature vectors, combined with thermal performance parameter derivation and iterative mechanical property calculations, thermal conductivity prediction formulas and mechanical property prediction formulas are generated, enabling accurate prediction of the thermal and mechanical properties of the material. By constructing a performance index matrix and performing optimization processing, a comprehensive performance score table is generated, providing a scientific quantitative basis for material design and optimization. This method not only addresses the shortcomings of existing technologies in analyzing the correlation between microstructure and macroscopic properties, but also effectively addresses the performance stability requirements under complex operating conditions (such as high temperature, high pressure, or long-term loads). This method significantly improves the reliability and efficiency of high-performance foamed ceramic panels in the architectural, industrial, and decorative fields. It provides more comprehensive and precise technical support for material research and development and engineering applications, promoting the optimized design and widespread application of foamed ceramic panels in multiple fields.

[0017] In summary, this application significantly improves the accuracy and comprehensiveness of the comprehensive performance evaluation of foamed ceramic panels by combining microstructure analysis with macroscopic performance prediction, thereby effectively solving the technical problem in the existing technology that is unable to fully characterize and predict the performance of materials in actual application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0019] The structures, proportions, sizes, etc. depicted in the drawings of this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with this technology. They are not intended to limit the conditions under which this application can be implemented, and therefore have no substantive technical significance. Any structural modifications, changes in proportional relationships, or adjustments in size, without affecting the efficacy and objectives that can be achieved by this application, should still fall within the scope of the technical contents disclosed in this application.

[0020] Figure 1 This is a schematic diagram of the steps of a foamed ceramic plate performance analysis method according to one embodiment of the present invention; Figure 2 This is a schematic block diagram of the structure of a foamed ceramic plate performance analysis device according to one embodiment of the present invention; Figure 3 is a schematic block diagram of the structure of a computer device according to an embodiment of the present invention; The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0022] Those skilled in the art will appreciate that, unless expressly stated otherwise, the singular forms "a", "an", "above", and "the" used herein may also include plural forms. It should be further understood that the term "comprising" used in the specification of the present invention refers to the presence of features, integers, steps, operations, elements, modules, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, modules, components, and / or groups thereof. It should be understood that when an element is said to be "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, "connected" or "coupled" as used herein may include wireless connections or wireless couplings. The term "and / or" used herein includes all or any module and all combinations of one or more associated listed items.

[0023] Those skilled in the art will understand that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art in the art to which this invention belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art and, unless specifically defined as such, will not be interpreted in an idealized or overly formal sense.

[0024] Reference Figure 1 The embodiment of the present invention provides a method for analyzing the performance of a foamed ceramic plate, comprising the following steps: S1: Scan the surface of the foamed ceramic plate sample to obtain collected data, and integrate the collected data to form a surface feature matrix; In step S1, high-precision instruments such as scanning electron microscopy (SEM) and energy dispersive spectroscopy (EDS) are used to scan the surface of the foamed ceramic board sample to obtain detailed information on its microstructure and chemical composition. The electron microscope scans the sample surface at a voltage of 10 kV and a high resolution of 0.5 nm, covering 20 different areas of the sample. These images generate two-dimensional images that document the pore morphology, size distribution, and surface texture of the foamed ceramic board. For example, the electron microscope can reveal pore shape (e.g., circular, elliptical, or irregular) and the connectivity between pores. EDS simultaneously collects chemical element distribution data on the sample surface, quantifying the chemical composition of the foamed ceramic board by analyzing the ratios of key elements such as Si, Al, and O. After acquiring the raw data, the SEM image is preprocessed. Edges are extracted using the Canny edge detection algorithm to identify pore boundary contours. Pore geometric characteristics, including pore area fraction and shape parameters (e.g., roundness or aspect ratio), are then calculated. The chemical composition data acquired by EDS are integrated with the pore geometric characteristics to form a multi-source dataset. In order to improve computational efficiency, by converting multi-source data sets into sparse matrix form, the storage space and computational time can be significantly reduced, thereby reducing the computational complexity from O(n 2 ) is reduced to O(nlogn) and a surface feature matrix is output, which mathematically integrates the geometric characteristics of the pores (such as area ratio and roundness) and chemical composition information (such as element ratio).

[0025] S2: Scanning the surface pores of the foamed ceramic plate sample to obtain a stereoscopic image sequence, and constructing a three-dimensional pore model based on the surface feature matrix and the stereoscopic image sequence; In step S2, the foamed ceramic board sample can be three-dimensionally scanned using X-ray micro-tomography (or other X-ray scanning equipment). The penetrating power of X-rays captures the internal pore structure of the sample, generating a series of continuous stereoscopic images that record the spatial distribution, morphology, and connectivity of the pores within the foamed ceramic board. After acquiring the stereoscopic image sequence, it is smoothed using a 3D Gaussian filter to remove noise caused by the scanning equipment or environmental interference. Density correction is performed on the stereoscopic image sequence using the surface feature matrix generated in step S1. Because foamed ceramic boards may have local density variations due to material composition or manufacturing process, the chemical composition data in the surface feature matrix (such as the ratios of elements such as Si, Al, and O) can be used as a reference to correct grayscale deviations caused by density variations in the stereoscopic images. The pores and matrix are segmented, and the complex geometric boundaries of the pores are captured through mathematical surface evolution, resulting in a clear pore-matrix interface. After image segmentation, the results are converted into a 3D pore model, which efficiently represents the pore geometry and topology, including key information such as pore volume and connectivity parameters. Parallel GPU acceleration can be used to distribute image processing and segmentation tasks across the GPU's multi-core processors, significantly reducing computation time.

[0026] S3: performing topological and geometric feature extraction processing on the three-dimensional pore model to obtain a pore feature vector, performing thermal performance parameter derivation processing on the pore feature vector to generate a thermal conductivity prediction formula; In step S3, the model is volumetrically meshed, dividing the 3D model into small volumetric cells. This can be implemented using Python combined with the MeshLab library, an open-source 3D mesh processing tool. By calling MeshLab's API, the model is divided into volumetric meshes. The size of each mesh cell can be adjusted based on the pore size range. The pore diameter distribution is calculated, and the geometric center of each pore is identified and the maximum distance from its boundary to the center is calculated, thereby obtaining a statistical distribution of pore diameters. For topological analysis, the connectivity paths and node degrees of the pore network are extracted. The pore network can be viewed as a graph structure consisting of pores (nodes) and connecting channels (edges). The connectivity paths between pores are identified, and the degree of each node (i.e., the number of channels connected to the node) and the connectivity index of the entire network are calculated. In the thermal performance parameter derivation stage, the pore eigenvectors are used to construct a thermal conductivity prediction formula. Thermal performance parameters of the foamed ceramic board, such as thermal diffusivity and specific heat capacity, are experimentally measured. Assume that the thermal diffusivity of a sample is measured using a laser thermal conductivity meter over a temperature range of 100-1000°C. Lasso regression is used to identify the pore characteristics that most significantly influence thermal conductivity, such as porosity and the number of interconnected paths. Lasso regression, by introducing an L1 regularization term, effectively eliminates irrelevant features and improves the model's interpretability. The support vector regression (SVR) algorithm is used to fit the nonlinear relationship between pore characteristics and thermal conductivity, training a formula to predict thermal conductivity. This formula might take the form: λ = f(porosity, interconnected paths, ...), where λ is the thermal conductivity.

[0027] S4: Iteratively calculating the influence data of pores on stress distribution according to the pore characteristic vector and the thermal conductivity prediction formula to generate a mechanical property prediction formula; In step S4, a small load (e.g., 0.1-150 mN) is applied to the material surface, and parameters such as indentation depth, hardness, and Young's modulus are recorded to quantify the mechanical properties of the material. The selection of these test points is not random, but is optimized based on the pore distribution characteristics extracted in step S3. For example, assuming that the pore characteristic vector in step S3 shows that the pores of a foamed ceramic board are mainly concentrated in the area close to the surface, then the test points should be preferentially selected at the junction of the pore-dense area and the non-dense area to capture the microscopic influence of the pores on the stress distribution, and a set of mechanical property data related to the pore distribution can be obtained. The influence of the pores on the stress distribution is quantified by computational methods, and the three-dimensional pore model generated in step S2 is converted into a finite element mesh. Specifically, the model contains the pore geometry information of the foamed ceramic board, and the STL model can be discretized into a finite element mesh. After the meshing is completed, the boundary conditions are set according to the pore characteristic vector. These boundary conditions include applied loads, constraints, and basic mechanical parameters of the material (such as Young's modulus and Poisson's ratio), which generate mechanical property prediction formulas that include functional relationships between stress concentration factors, fracture toughness, and pore eigenvectors.

[0028] S5: constructing a performance index matrix according to the thermal conductivity prediction formula and the mechanical property prediction formula, optimizing the performance index matrix, and obtaining a comprehensive performance score sheet.

[0029] In step S5, when constructing the performance index matrix, the key performance parameters in the thermal conductivity prediction formula and the mechanical property prediction formula are extracted to form a multidimensional matrix. The performance index matrix contains multiple performance indicators, such as thermal conductivity, hardness, fracture toughness and compressive strength, etc. These values are integrated into a matrix form, each row represents a sample, and each column represents a performance index. The performance index matrix is optimized, and a comprehensive performance score sheet that can intuitively reflect the performance of each sample is generated by comprehensively analyzing the multidimensional performance indicators in the matrix. Specifically, the performance index matrix is normalized, and all indicators are converted into dimensionless relative values through normalization. The gray correlation between each indicator and the ideal performance is calculated. The ideal performance is set as the optimal value of each indicator (such as the lowest thermal conductivity, the highest hardness and fracture toughness). The gray correlation formula is used to calculate the degree of closeness of each sample to the ideal performance. The gray correlation formula is ),in is the sample normalization value, is the ideal value. The weight of each indicator is determined based on the entropy method, and the information content of the indicator is evaluated by calculating the entropy value of the indicator. The indicator with greater information content has a higher weight. The grey correlation degree is combined with the weight to calculate the comprehensive performance score of each sample, and a comprehensive performance score table is generated. Each row in the score table includes the sample number, the normalized value of each performance indicator, the grey correlation degree and the comprehensive score. Through the above optimization processing, the comprehensive performance score table can intuitively display the performance ranking of each sample. For example, sample 3 received the highest score due to its lower thermal conductivity (0.12 W / (m·K)) and higher hardness (510 MPa), indicating that it has the best balance between thermal insulation and load-bearing capacity.

[0030] In one embodiment, the step of scanning the surface of the foamed ceramic plate sample to obtain collected data and integrating the collected data to form a surface feature matrix includes: The surface of the foamed ceramic plate sample is scanned using a scanning device to obtain a microstructure image set; extracting pore boundaries from the microstructure image set to generate a pore boundary image; Performing pore region segmentation processing on the pore boundary image, calculating the area ratio and shape parameters of each pore region, and generating a pore geometric feature set; Based on the energy dispersive spectroscopy technology, the foamed ceramic plate sample is synchronously scanned for element distribution, generating an element distribution data set containing the spatial coordinates and proportion data of the element distribution; The pore geometric feature set and the element distribution data set are subjected to multi-source data fusion processing to obtain a surface feature matrix.

[0031] In the above-described embodiment, a scanning electron microscope (SEM) was used as the scanning device to perform multi-region scanning of a foamed ceramic plate sample at an operating voltage of 10 kV and a resolution of 0.5 nm. The scan covered at least 20 randomly selected areas on the sample surface, and a two-dimensional image of each area captured features such as pore morphology, surface texture, and microscopic defects. These images were spliced using a parallel processing algorithm to form a microstructure image set with uniform resolution. Noise suppression and edge enhancement were performed on the microstructure image set. An adaptive median filter algorithm was used for image preprocessing. This algorithm calculated the statistical characteristics of local pixels using a sliding window and dynamically adjusted the filter radius, effectively suppressing noise while preserving pore edge details. An improved Canny edge detection algorithm was applied, combined with the Sobel operator to calculate image gradients, extract pore boundaries, and generate a binary pore boundary image. Based on the pore boundary image, a region growing algorithm was used to segment the pore boundary image, identifying independent pore regions. The area, perimeter, and shape parameters of each pore, such as roundness, ellipticity, and convexity, were calculated to generate a pore geometric feature set. Simultaneously, energy dispersive spectroscopy (EDS) was used to perform simultaneous elemental distribution scans on the samples, focusing on quantifying the mass fractions and spatial distributions of key elements such as Si, Al, O, and Ca. Overlapping peaks were separated using a spectral peak decomposition algorithm, and background noise was corrected using Gaussian fitting. This generated an elemental distribution dataset containing the spatial coordinates and proportions of the elemental distributions. After obtaining the pore geometry feature set and elemental distribution dataset, the elemental distribution data were aligned with the pore geometry features at the pixel level to extract the correlation features between pore morphology and chemical composition. A fused feature matrix containing the geometric parameters and elemental distribution features was generated using sparse matrix operations. This fused feature matrix was then subjected to dimensionality reduction and normalization. A dimensionality reduction algorithm was used to map high-dimensional features to a low-dimensional space, preserving the topological structure of the pore geometry and chemical composition and maintaining the inherent correlations between features. The feature data were normalized using the Z-score normalization method to eliminate differences between different dimensions, generating a surface feature matrix containing comprehensive information on pore morphology and chemical composition.

[0032] In one embodiment, the step of scanning the surface pores of the foamed ceramic plate sample to obtain a stereoscopic image sequence and constructing a three-dimensional pore model based on the surface feature matrix and the stereoscopic image sequence includes: Scan the foamed ceramic plate sample using an X-ray scanning device to collect multi-view projection data including pore and matrix information to obtain an initial stereoscopic image sequence; performing noise suppression processing on the initial stereoscopic image sequence to obtain a smoothed stereoscopic image sequence; Setting the surface feature matrix as a reference, performing density correction processing on the smoothed stereoscopic image sequence to obtain a corrected stereoscopic image sequence; Setting the surface feature matrix as a boundary constraint condition, performing pore segmentation processing on the corrected stereoscopic image sequence to obtain a pore image sequence including pores and matrix areas; The pore regions of the pore image sequence are converted into a three-dimensional grid structure to generate an initial three-dimensional pore model, and topology optimization is performed on the initial three-dimensional pore model to obtain a three-dimensional pore model.

[0033] In the above-described embodiment, a sample is continuously scanned at a full 720-degree angle using an X-ray micro-tomography device, acquiring multi-view projection data containing pore and matrix information. During the scanning process, after X-rays penetrate the sample, the projection data records the density differences between different materials within the sample, reflecting the contrast between the pores and the matrix. The projection data is processed using a back-projection reconstruction algorithm to generate an initial stereoscopic image sequence. This sequence consists of a series of 2D slice images, each containing information about pore distribution and matrix density. Noise suppression is performed on the initial stereoscopic image sequence using a 3D Gaussian filter based on anisotropic diffusion. By adaptively adjusting the filter kernel size (ranging from 0.1 to 0.5 microns), noise is smoothed while preserving detailed information about pore edges. During the density correction stage, a surface feature matrix contains information about the density distribution of the sample surface. By combining this matrix with the smoothed stereoscopic image sequence, the grayscale values of the voxels in the image sequence can be adjusted to eliminate density differences caused by material inhomogeneity or scanning device deviation. The grayscale value of each voxel can be corrected through an iterative grayscale mapping algorithm. During pore segmentation, by introducing boundary constraints based on the surface feature matrix, the evolution rate of the level set function can be dynamically adjusted, thereby accurately distinguishing pores from the matrix region. During the segmentation process, the pore region is marked as 1 and the matrix region is marked as 0, so that the geometric shape and connectivity characteristics of the pores are fully preserved in the image sequence. The voxel data of the pore region is mapped to a triangular mesh structure to generate an initial three-dimensional pore model, which includes the pore volume, surface area, and preliminary connectivity parameters. The initial three-dimensional pore model is topologically optimized to calculate the pore connectivity index and pore wall thickness distribution, eliminate non-physical connections and tiny isolated pores, and combine the pore distribution information in the surface feature matrix to reconstruct the pore connectivity path. This optimization process retains the true geometric characteristics of the pores and improves the calculation accuracy of the connectivity parameters, enabling the three-dimensional pore model to accurately reflect the complex topological characteristics of the internal pore structure of the foamed ceramic board.

[0034] In one embodiment, the step of performing topological and geometric feature extraction on the three-dimensional pore model to obtain a pore feature vector includes: Performing spatial discretization processing on the three-dimensional pore model to generate a discrete grid data set including voxel units; Calculating the distance between each voxel unit in the discrete grid data set and the pore boundary, and performing geometric analysis processing to obtain pore diameter distribution data, wherein the value of each voxel represents the shortest distance to the pore wall; Performing connectivity analysis on the pore diameter distribution data to generate a pore network topology diagram comprising nodes and connectivity paths, wherein the connectivity paths are pore channels between nodes; The pore network topology graph is processed by dimensionality reduction and statistical analysis to obtain a pore characteristic vector.

[0035] In the above-described embodiment, a three-dimensional pore model is imported into a computational framework based on volume geometry. Octree segmentation is used to recursively partition the model. By progressively subdividing the three-dimensional space into eight subregions, the side length of each voxel is kept within a range of 0.05 to 50 microns, capturing the microscopic details of the pore structure. During the partitioning process, the mesh density is adaptively adjusted based on the complexity of the pore boundaries. The mesh is denser in pore boundary regions to preserve geometric accuracy, while the mesh is sparser in uniform regions to optimize computational efficiency. Based on the generated discrete grid dataset, a distance transform is performed on the pore structure. For each voxel, the distance to the nearest pore boundary is calculated, generating a distance field matrix in which each voxel's value represents the shortest distance to the pore wall. In specific implementation, a distance field matrix with the same dimensions as the discrete grid dataset is initialized. Then, a parallel computation is performed to traverse each voxel and calculate its Euclidean distance to the pore boundary. After distance calculation, the distance values in the pore region are statistically analyzed to extract the distribution characteristics of pore diameters. Statistical parameters are generated by summarizing the values in the distance field matrix and stored as a pore diameter distribution dataset. Based on the pore diameter distribution data, the voxels in the pore region are divided into a set of nodes according to the diameter distribution data. Each node is defined as the local geometric center of a pore and can be determined using a clustering algorithm or voxel grouping method. Connected paths are defined as pore channels between nodes and can be identified using a depth-first search (DFS) algorithm. In the implementation, the voxels in the distance field matrix are traversed, and the voxels belonging to the same pore channel are determined based on the diameter distribution data. The degree (i.e., the number of connected paths) and the path length distribution of each node are calculated. A weighted undirected graph is constructed, where nodes represent pore centers and edges represent connected paths. Edge weights are assigned based on path length and pore diameter. After obtaining the pore network topology, the t-SNE (t-distributed stochastic neighbor embedding) algorithm is used to perform nonlinear dimensionality reduction on the high-dimensional features (such as node degree, path length, and pore diameter distribution) in the topology. This compresses the feature space from high dimensions to three or two dimensions, and statistical analysis of the pore characteristics is then performed. In implementation, the eigenvalues in the low-dimensional feature matrix are statistically calculated to extract statistical quantities such as mean, variance, interquartile range, skewness, and kurtosis. These statistics are then combined with the connectivity indices of the topology (such as average node degree and path density) to generate comprehensive parameters. Based on the pore feature parameter set, a feature integration algorithm is used to normalize and concatenate the feature data. The statistical parameters (mean, variance, interquartile range, etc.), topological parameters (node degree, connectivity index), and geometric parameters (pore diameter distribution) in the parameter set are normalized to ensure consistent dimensionality. Normalization can be performed using the z-score standardization method, which calculates the mean and standard deviation of each parameter for standardization, and uses the feature splicing algorithm to integrate the multidimensional parameters into a single feature vector to generate a pore feature vector.

[0036] In one embodiment, the step of deriving thermal performance parameters from the pore characteristic vector to generate a thermal conductivity prediction formula includes: performing nonlinear normalization processing on the pore characteristic vector to generate a normalized characteristic matrix; Performing weight analysis on each feature of the standardized feature matrix, setting regularization parameters to rank the importance of features, and generating a key pore feature set; Constructing a nonlinear mapping relationship between the pore feature set and thermal conductivity to form a preliminary thermal conductivity model, and iteratively optimizing the preliminary thermal conductivity model to obtain an optimized thermal conductivity model; The weight coefficient of each feature is calculated according to the optimized thermal conductivity model, and the weight coefficient is subjected to feature mapping processing to form a thermal conductivity prediction formula.

[0037] In the above embodiment, for features such as porosity, pore connectivity, and pore size distribution, appropriate transformation parameters are selected based on the statistical properties of the data to transform each feature into a normal distribution. This generates a standardized feature matrix, which is then subjected to feature screening to extract the key pore feature set most important for thermal conductivity prediction. By setting a regularization parameter (e.g., the λ value), the importance of features such as porosity, average pore diameter, and connectivity density is ranked. The ranking results reveal that certain features, such as porosity and connectivity density, may have a more significant impact on thermal conductivity, while others have a lesser influence. This process selects the features that contribute most to thermal conductivity prediction, generating a key pore feature set. Based on this key pore feature set, a nonlinear mapping relationship between pore features and thermal conductivity is constructed. This nonlinear model can be constructed using a support vector regression (SVR) algorithm, using a radial basis kernel (RBF) to perform a high-dimensional mapping of the feature space. By mapping low-dimensional features into a high-dimensional space, the RBF effectively captures the complex nonlinear relationship between pore features and thermal conductivity. During model construction, features from the key pore feature set (such as porosity and connectivity density) are used as input variables, and thermal conductivity is used as the output variable. A preliminary thermal conductivity model is fitted using the SVR algorithm. Parameters of this preliminary thermal conductivity model are optimized using the K-fold cross-validation method. Specifically, the dataset is divided into K subsets, with one subset used as the validation set in each iteration, and the remaining subsets used as the training set. Through repeated training and validation, the kernel parameters of the support vector regression (SVR) function (such as the width parameter σ of the RBF kernel) and the regularization parameters (such as the penalty coefficient C) are adjusted. K-fold cross-validation, through multiple iterations, comprehensively evaluates the model's performance on different data subsets, thereby preventing overfitting to specific training data and improving its generalization ability. After multiple rounds of iterative optimization, a thermal conductivity model with improved performance is generated. Based on the optimized thermal conductivity model, feature contribution analysis is further performed. Feature importance analysis algorithms are used to calculate the weight coefficients of each feature in the optimized model. These weight coefficients reflect the relative importance of features such as porosity, average pore diameter, and connected path density in thermal conductivity prediction. For example, porosity may have a higher weight, indicating that it contributes the most to thermal conductivity, followed by connected path density. By quantifying these weights, a set of thermal conductivity contribution weights is generated. The thermal conductivity contribution weight set is data formatted, and based on the weight coefficients and feature mapping relationship of the optimized thermal conductivity model, the model is converted into a structured mathematical expression. Specifically, the nonlinear mapping relationship between the weight coefficients and key pore features is organized into a formula form, such as expressing the relationship between thermal conductivity and features through polynomials or other mathematical forms.

[0038] In another embodiment, the thermal conductivity prediction formula is: ; Where κ is the thermal conductivity, measured in W / (m·K), which represents the material's ability to conduct heat and is the output of the formula. ϕ is the porosity, which represents the ratio of the pore volume to the total volume of the material and ranges from [0 to 1]. The density of connected paths indicates the complexity of pore connected paths. It is defined as the average number of connected paths per unit volume and is usually normalized to [0, 1]. It is the uniformity of pore size distribution, which characterizes the statistical dispersion of pore size. 、 、 α is a nonlinear adjustment parameter that controls the exponential decay effect of porosity on thermal conductivity. Its value range is [0.5, 2] and is determined through model optimization. β is a logarithmic enhancement factor for the connected path density. Its value range is [1, 5] and is used to enhance the nonlinear effect of the connected path on thermal conductivity. γ is a logistic adjustment parameter for the pore size distribution. Its value range is [1, 10] and is used to control the nonlinear contribution of the size distribution to thermal conductivity. ϵ is a dimensionless model error correction term that represents random error or noise not captured by the model. It is typically close to 0 and is determined during the model optimization process. is an exponential decay function, which describes the nonlinear inhibitory effect of porosity on thermal conductivity. High porosity will lead to a significant decrease in thermal conductivity. It is a logarithmic function that captures the nonlinear enhancement effect of the connectivity path density on thermal conductivity and emphasizes the role of connectivity paths in heat conduction. is the square term of the porosity, which enhances the nonlinear effect at high porosity values. A sigmoid function is used to simulate the nonlinear effect of pore size distribution uniformity on thermal conductivity, smoothing the transition between different uniformity levels. By combining exponential, logarithmic, and sigmoid functions, the complex nonlinear relationship between pore characteristics and thermal conductivity is captured, better adapting to the randomness and complexity of pore structures.

[0039] In one embodiment, the step of iteratively calculating the data on the influence of pores on stress distribution based on the pore characteristic vector and the thermal conductivity prediction formula to generate a mechanical property prediction formula includes: Performing finite element meshing processing on the three-dimensional pore model to obtain a finite element mesh model; Setting mechanical boundary conditions on the finite element mesh model according to the pore characteristic vector to obtain a mechanical simulation input model; Performing strain field iterative calculation processing on the mechanical simulation input model to obtain a stress distribution data set; The mechanical property parameters in the stress distribution data set are extracted, the mapping relationship between the pore characteristics and the mechanical property parameters is evaluated, and a mechanical property prediction formula is generated.

[0040] In the above embodiment, the pore feature vector is nonlinearly normalized to generate a standardized feature matrix. Common normalization methods can be used to map each dimension of the pore feature vector to the interval [0, 1] to generate the standardized feature matrix. A weight analysis is performed on the standardized feature matrix, and a regularization parameter is set to rank the features according to their importance. Lasso regression can be used to introduce an L1 regularization parameter to select features that contribute most to thermal conductivity, such as the complexity of pore connectivity and the uniformity of pore size distribution. A key pore feature set is generated by calculating the regression coefficient for each feature and ranking them. A nonlinear mapping relationship between the pore feature set and thermal conductivity is established. Adaptive mesh refinement techniques can be used to perform high-resolution segmentation of densely populated areas and generate a finite element mesh model containing node coordinates and cell connectivity information. In specific implementations, a machine learning algorithm (such as a support vector machine or neural network) can be used to fit the relationship between the key pore feature set and thermal conductivity to generate a preliminary nonlinear mapping model. For example, a finite element mesh model can be used to simulate the heat flow distribution of the pore structure under different temperature gradients using the heat conduction equation, and relevant thermal conductivity data can be extracted as a training set. Based on this data, the preliminary thermal conductivity model can be optimized using a backpropagation algorithm to gradually approximate the true thermal conductivity distribution. The preliminary thermal conductivity model is then iteratively optimized. Through iterative calculations, the model parameters can be fine-tuned, such as adjusting the hidden layer weights of a neural network or the kernel parameters of a support vector machine, to minimize prediction error. Bayesian inference algorithms can be used to assess model uncertainty by constructing a posterior probability distribution. For example, by sampling the randomness of pore characteristics (such as porosity fluctuations or the random distribution of connectivity paths) multiple times, a set of thermal conductivity predictions is generated, thereby optimizing the model's robustness. The optimized thermal conductivity model can better adapt to the complexity and randomness of pore structures, ensuring the reliability of predictions under different operating conditions. Based on the optimized thermal conductivity model, weight coefficients for each feature are calculated and feature mapping is performed to form a thermal conductivity prediction formula. Based on the optimized thermal conductivity model, the weight coefficients for each key pore feature can be extracted. For example, by analyzing the weight matrix of a neural network or the coefficients of a regression model, the contribution of features such as porosity and connectivity paths to thermal conductivity can be determined. Subsequently, these weight coefficients are combined with the pore eigenvectors through eigenmapping processing to generate a mathematical expression for the structural policing.

[0041] In one embodiment, the step of constructing a performance index matrix based on the thermal conductivity prediction formula and the mechanical property prediction formula, and optimizing the performance index matrix to obtain a comprehensive performance score sheet includes: Extracting multiple performance parameters of the thermal conductivity prediction formula and the mechanical property prediction formula, unifying the multiple parameters, and generating an initial performance index matrix including multiple performance indexes; Normalizing the initial performance indicator matrix to obtain a standardized performance indicator matrix; quantifying the degree of correlation between each performance indicator in the standardized performance indicator matrix and the ideal performance based on a grey correlation analysis algorithm to obtain an indicator correlation vector; Analyzing the information entropy value of the indicator correlation vector, calculating the weight value of each indicator according to the information entropy value, and generating a performance indicator weight set; Performing weighted fusion processing on the standardized performance indicator matrix according to the performance indicator weight set to obtain a comprehensive performance score vector; Data formatting is performed on the comprehensive performance score vector to obtain the comprehensive performance score table.

[0042] In the above embodiment, multiple performance parameters are extracted from the thermal conductivity prediction formula and the mechanical property prediction formula. The thermal conductivity value is extracted from the thermal conductivity prediction formula, for example, by parsing the formula output or calling a related interface to obtain numerical data. Simultaneously, values such as hardness, fracture toughness, and compressive strength are extracted from the mechanical property prediction formula to ensure data integrity and accuracy. Data cleansing is performed to check data validity, remove outliers or missing values, and align the data formats of the outputs from different formulas to ensure that all parameters match a unified sample set. These parameters are then organized into a multidimensional matrix, with each row representing a sample and each column corresponding to a performance indicator (such as thermal conductivity, hardness, etc.), thereby generating an initial performance indicator matrix. After generating the initial performance indicator matrix, it is normalized to eliminate differences in dimensionality and numerical ranges between different performance indicators, resulting in a standardized performance indicator matrix. A linear normalization algorithm can be used to map each element in the matrix to the interval [0, 1]. This transformation normalizes the values of each performance indicator to the same range, eliminating dimensionality effects and ensuring numerical comparability in subsequent analysis. Using a gray relational analysis algorithm, the correlation between each performance indicator in the standardized performance indicator matrix and the ideal performance is quantified. An ideal performance reference sequence is constructed, consisting of the theoretically optimal values of each performance indicator. For example, thermal conductivity might take its minimum value to represent optimal thermal insulation, while hardness and compressive strength might take their maximum values to represent optimal mechanical properties. The ideal performance reference sequence can be constructed based on theoretical values, experimental data, or industry standards. The gray relational coefficient is calculated for each sample in the standardized performance indicator matrix and the ideal performance reference sequence. Specifically, for each element in the matrix, the absolute difference between its value and the corresponding ideal value is calculated, and the correlation coefficient is calculated using the resolution coefficient (typically 0.5). The correlation coefficient reflects the degree of proximity of each sample to a particular performance indicator. By taking a weighted average of the correlation coefficients for each performance indicator (the weights can be initially set equal), the overall correlation value for that indicator is obtained. Finally, an indicator correlation vector is generated, where each element of the vector corresponds to the correlation of a performance indicator. Information entropy analysis is performed on the indicator correlation vector. Information entropy is used to quantify data uncertainty. Lower entropy values indicate a higher information contribution of the indicator and, therefore, a higher weight should be assigned. Each element in the indicator correlation vector is normalized, and its proportion to the total is calculated. The information entropy of each performance indicator is calculated using the entropy formula: E = -k∑(p·ln(p)), where p is the normalized correlation value and k is a constant (1 / ln(n) is used to normalize the entropy). By comparing the entropy values of each indicator, its weight is calculated using the formula w = (1 - E) / (n - ∑E), where n is the number of indicators. After the weight calculation is complete, a set of performance indicator weights is generated, where each weight value reflects the relative importance of the corresponding performance indicator to the overall performance.For example, if the entropy value of thermal conductivity is low, it indicates a high information contribution, and its weight will be increased accordingly. The standardized performance indicator matrix is weighted and fused according to the performance indicator weight set. For each row (Samples) of the matrix, the column values (performance indicators) are multiplied by the corresponding weight values and then summed to obtain the comprehensive score for that sample. The comprehensive performance score vector is formatted, associating each score in the comprehensive performance score vector with a sample identifier (such as sample number) and other relevant information (such as performance indicator name), generating a multi-field table. For example, each row of the table might contain fields such as sample number, thermal conductivity value, hardness value, and comprehensive score. Data field mapping is used to define the table structure and store the data in a standard format, ensuring that the output comprehensive performance score table is clear and easy to use. This table provides an intuitive representation of the performance evaluation of foam ceramic panels and can be used for material selection or design optimization.

[0043] Reference Figure 2 , a foamed ceramic plate performance analysis device, comprising: The first acquisition module 100 is used to scan the surface of the foamed ceramic plate sample to obtain acquired data, and integrate the acquired data to form a surface feature matrix; The second acquisition module 200 is used to scan the surface pores of the foamed ceramic plate sample to obtain a stereoscopic image sequence, and to construct a three-dimensional pore model based on the surface feature matrix and the stereoscopic image sequence; A first derivation module 300 is configured to extract topological and geometric features from the three-dimensional pore model to obtain a pore characteristic vector, derive thermal performance parameters from the pore characteristic vector, and generate a thermal conductivity prediction formula. The second derivation module 400 iteratively calculates the influence data of pores on stress distribution based on the pore characteristic vector and the thermal conductivity prediction formula to generate a mechanical property prediction formula; The optimization module 500 is used to construct a performance index matrix according to the thermal conductivity prediction formula and the mechanical property prediction formula, and optimize the performance index matrix to obtain a comprehensive performance score table.

[0044] Reference Figure 3 In the embodiment of the present application, a computer device is also provided. The computer device may be a server, and its internal structure may be as follows: Figure 3As shown. The computer device includes a processor, a memory, a network interface and a database connected via a system bus. The processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as a database of a method for analyzing the performance of foamed ceramic plates. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, a method for analyzing the performance of foamed ceramic plates is implemented.

[0045] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, a method for analyzing the performance of a foamed ceramic plate is implemented, comprising the steps of: performing a surface scan on a foamed ceramic plate sample to obtain collected data, and integrating the collected data to form a surface feature matrix; performing a surface pore scan on the foamed ceramic plate sample to obtain a stereoscopic image sequence, and constructing a three-dimensional pore model based on the surface feature matrix and the stereoscopic image sequence; performing topological and geometric feature extraction processing on the three-dimensional pore model to obtain a pore feature vector, and performing thermal performance parameter derivation processing on the pore feature vector to generate a thermal conductivity prediction formula; iteratively calculating data on the influence of pores on stress distribution based on the pore feature vector and the thermal conductivity prediction formula to generate a mechanical property prediction formula; constructing a performance index matrix based on the thermal conductivity prediction formula and the mechanical property prediction formula, and optimizing the performance index matrix to obtain a comprehensive performance score sheet.

[0046] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed, the computer program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media provided herein and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0047] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for analyzing the performance of a foamed ceramic plate, characterized in that: include: Scanning the surface of the foamed ceramic plate sample to obtain collected data, and integrating the collected data to form a surface feature matrix; Scanning the surface pores of the foamed ceramic plate sample to obtain a stereoscopic image sequence, and constructing a three-dimensional pore model based on the surface feature matrix and the stereoscopic image sequence; Performing topological and geometric feature extraction processing on the three-dimensional pore model to obtain a pore feature vector, performing thermal performance parameter derivation processing on the pore feature vector to generate a thermal conductivity prediction formula; Iteratively calculating the influence data of pores on stress distribution based on the pore characteristic vector and thermal conductivity prediction formula to generate a mechanical property prediction formula; A performance index matrix is constructed according to the thermal conductivity prediction formula and the mechanical property prediction formula, and the performance index matrix is optimized to obtain a comprehensive performance score sheet.

2. The foamed ceramic plate performance analysis method according to claim 1, characterized in that: The step of scanning the surface of the foamed ceramic plate sample to obtain collected data and integrating the collected data to form a surface feature matrix includes: The surface of the foamed ceramic plate sample is scanned using a scanning device to obtain a microstructure image set; extracting pore boundaries from the microstructure image set to generate a pore boundary image; Performing pore region segmentation processing on the pore boundary image, calculating the area ratio and shape parameters of each pore region, and generating a pore geometric feature set; Based on the energy dispersive spectroscopy technology, the foamed ceramic plate sample is synchronously scanned for element distribution, generating an element distribution data set containing the spatial coordinates and proportion data of the element distribution; The pore geometric feature set and the element distribution data set are subjected to multi-source data fusion processing to obtain a surface feature matrix.

3. The foamed ceramic board performance analysis method according to claim 1, characterized in that: The step of scanning the surface pores of the foamed ceramic plate sample to obtain a stereoscopic image sequence, and constructing a three-dimensional pore model according to the surface feature matrix and the stereoscopic image sequence comprises: Scan the foamed ceramic plate sample using an X-ray scanning device to collect multi-view projection data including pore and matrix information to obtain an initial stereoscopic image sequence; performing noise suppression processing on the initial stereoscopic image sequence to obtain a smoothed stereoscopic image sequence; Setting the surface feature matrix as a reference, performing density correction processing on the smoothed stereoscopic image sequence to obtain a corrected stereoscopic image sequence; Setting the surface feature matrix as a boundary constraint condition, performing pore segmentation processing on the corrected stereoscopic image sequence to obtain a pore image sequence including pores and matrix areas; The pore regions of the pore image sequence are converted into a three-dimensional grid structure to generate an initial three-dimensional pore model, and topology optimization is performed on the initial three-dimensional pore model to obtain a three-dimensional pore model.

4. The method for analyzing properties of a foamed ceramic plate according to claim 1, wherein: The step of performing topological and geometric feature extraction processing on the three-dimensional pore model to obtain a pore feature vector includes: Performing spatial discretization processing on the three-dimensional pore model to generate a discrete grid data set including voxel units; Calculating the distance between each voxel unit in the discrete grid data set and the pore boundary, and performing geometric analysis processing to obtain pore diameter distribution data, wherein the value of each voxel represents the shortest distance to the pore wall; Performing connectivity analysis on the pore diameter distribution data to generate a pore network topology diagram comprising nodes and connectivity paths, wherein the connectivity paths are pore channels between nodes; The pore network topology graph is processed by dimensionality reduction and statistical analysis to obtain a pore characteristic vector.

5. The foamed ceramic plate performance analysis method according to claim 1, characterized in that: The step of deriving thermal performance parameters from the pore characteristic vector to generate a thermal conductivity prediction formula includes: performing nonlinear normalization processing on the pore characteristic vector to generate a normalized characteristic matrix; Performing weight analysis on each feature of the standardized feature matrix, setting regularization parameters to rank the importance of features, and generating a key pore feature set; Constructing a nonlinear mapping relationship between the pore feature set and thermal conductivity to form a preliminary thermal conductivity model, and iteratively optimizing the preliminary thermal conductivity model to obtain an optimized thermal conductivity model; The weight coefficient of each feature is calculated according to the optimized thermal conductivity model, and the weight coefficient is subjected to feature mapping processing to form a thermal conductivity prediction formula.

6. The foamed ceramic board performance analysis method according to claim 1, characterized in that: The step of iteratively calculating the data on the influence of pores on stress distribution based on the pore characteristic vector and the thermal conductivity prediction formula to generate a mechanical property prediction formula includes: Performing finite element meshing processing on the three-dimensional pore model to obtain a finite element mesh model; Setting mechanical boundary conditions on the finite element mesh model according to the pore characteristic vector to obtain a mechanical simulation input model; Performing strain field iterative calculation processing on the mechanical simulation input model to obtain a stress distribution data set; The mechanical property parameters in the stress distribution data set are extracted, the mapping relationship between the pore characteristics and the mechanical property parameters is evaluated, and a mechanical property prediction formula is generated.

7. The foamed ceramic board performance analysis method according to claim 1, characterized in that: The step of constructing a performance index matrix according to the thermal conductivity prediction formula and the mechanical property prediction formula, and optimizing the performance index matrix to obtain a comprehensive performance score sheet includes: Extracting multiple performance parameters of the thermal conductivity prediction formula and the mechanical property prediction formula, unifying the multiple parameters, and generating an initial performance index matrix including multiple performance indexes; Normalizing the initial performance indicator matrix to obtain a standardized performance indicator matrix; quantifying the degree of correlation between each performance indicator in the standardized performance indicator matrix and the ideal performance based on a grey correlation analysis algorithm to obtain an indicator correlation vector; Analyzing the information entropy value of the indicator correlation vector, calculating the weight value of each indicator according to the information entropy value, and generating a performance indicator weight set; Performing weighted fusion processing on the standardized performance indicator matrix according to the performance indicator weight set to obtain a comprehensive performance score vector; Data formatting is performed on the comprehensive performance score vector to obtain the comprehensive performance score table.

8. A foamed ceramic plate performance analysis device, characterized in that: include: A first acquisition module is used to scan the surface of the foamed ceramic plate sample to obtain acquisition data, and integrate the acquisition data to form a surface feature matrix; The second acquisition module is used to scan the surface pores of the foamed ceramic plate sample to obtain a stereoscopic image sequence, and to construct a three-dimensional pore model based on the surface feature matrix and the stereoscopic image sequence; a first derivation module, configured to extract topological and geometric features of the three-dimensional pore model to obtain a pore characteristic vector, derive thermal performance parameters of the pore characteristic vector, and generate a thermal conductivity prediction formula; A second derivation module iteratively calculates the influence data of pores on stress distribution based on the pore characteristic vector and the thermal conductivity prediction formula to generate a mechanical property prediction formula; The optimization module is used to construct a performance index matrix based on the thermal conductivity prediction formula and the mechanical property prediction formula, and optimize the performance index matrix to obtain a comprehensive performance score table.

9. A computer device, characterized in that: The method comprises a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images