Quality Control Method and System for Blast-Relieved Perlite Boards Based on Load Tests

Abstract

The present application discloses a quality control method and system for explosion venting perlite boards based on load tests. The method includes: generating a quality control plan for optimizing the design of explosion venting perlite boards by using a quality control model according to the scanned images of the explosion venting perlite boards; the training method of the quality control model includes: generating a first quality control plan through a to-be-trained model based on the sample scanned images; adjusting the sample scanned images into optimized scanned images based on the first quality control plan; generating explosion relief channel morphology data based on the optimized scanned images; determining target simulation data through load tests based on the explosion relief channel morphology data, the target simulation data indicating explosion venting performance parameters related to the gel bead ratio, and the gel bead ratio being the ratio of the gelling material to the perlite mass of the optimized explosion venting perlite board; training the to-be-trained model based on the target simulation data and the performance requirements corresponding to the sample explosion venting perlite boards to obtain the quality control model, thereby being able to improve the design optimization efficiency of the explosion venting perlite boards.

Description

Technical Field

[0001] This application relates to the technical field of simulation design, and particularly to a quality control method and system for explosion venting perlite boards based on load tests. Background Art

[0002] As an important component for explosion protection and pressure relief in key facilities such as substations, the high-strength explosion venting perlite board plays an irreplaceable role in ensuring the safe and stable operation of the facilities. Its core function is to achieve rapid pressure relief through the structural design of the board to cope with explosion shocks. On the other hand, in order to adapt to the internal spaces of different key facilities (such as substations), in some cases, it is necessary to specially design corresponding explosion venting perlite boards for certain key facilities.

[0003] In the related art, in order to optimize the structural design of the explosion venting perlite board, usually, corresponding explosion venting perlite boards are prepared according to different designs first, and then the prepared explosion venting perlite boards are tested one by one to evaluate the advantages and disadvantages of the corresponding designs, so as to control the quality of the designed explosion venting perlite board. However, each test requires a large amount of manpower and material resources, and the test takes a long time, and there are also certain safety risks. Therefore, it seriously restricts the design optimization efficiency of the explosion venting perlite board. Summary of the Invention

[0004] To solve the above technical problems, the embodiments of this application propose a quality control method and system for explosion venting perlite boards based on load tests, which can improve the design optimization efficiency of the explosion venting perlite board.

[0005] In a first aspect, the embodiments of this application provide a quality control method for explosion venting perlite boards based on load tests, including:

[0006] According to the scanned image of the explosion venting perlite board to be optimized, using a quality control model, generate a quality control plan for optimizing the design of the explosion venting perlite board;

[0007] Among them, the training method of the quality control model includes:

[0008] Based on the sample scanned image of the sample explosion venting perlite board, generate a first quality control plan through a model to be trained;

[0009] Based on the first quality control plan, adjust the sample scanned image to an optimized scanned image, where the optimized scanned image corresponds to an optimized explosion venting perlite board, and the optimized explosion venting perlite board is obtained by optimizing the sample explosion venting perlite board according to the first quality control plan;

[0010] Based on the optimized scanned image, generate explosion relief channel morphology data;

[0011] Based on the pressure relief channel morphology data, determine the target simulation data through a load test, where the target simulation data indicates the explosion venting performance parameters related to the bead ratio, and the bead ratio is the ratio of the gelling material to the cenospheres of the optimized explosion venting cenosphere board;

[0012] Train the to-be-trained model based on the target simulation data and the performance requirements corresponding to the sample explosion venting cenosphere board to obtain the quality control model.

[0013] Optionally, the pressure relief channel morphology data includes multiple morphology data. The determining the target simulation data through a load test based on the pressure relief channel morphology data includes:

[0014] For each morphology data, if a simulation result corresponding to the morphology data is queried in the preset simulation dataset, determine the simulation data of the morphology data according to the queried simulation result; otherwise, conduct a load simulation test based on the morphology data to obtain the simulation data of the morphology data, where the simulation dataset includes at least one simulation result;

[0015] Determine the target simulation data based on the simulation data of each of the multiple morphology data.

[0016] Optionally, the conducting a load simulation test based on the morphology data to obtain the simulation data of the morphology data includes:

[0017] Generate a finite element model according to the morphology data;

[0018] Based on the finite element model, conduct a load simulation test using finite element analysis to obtain the explosion pressure release path distribution and explosion pressure release efficiency under the set explosion conditions;

[0019] Determine the simulation data of the morphology data based on the explosion pressure release path distribution and the explosion pressure release efficiency.

[0020] Optionally, the optimized scanned image includes the cross-sectional pore structure image of the optimized explosion venting cenosphere board. The generating the pressure relief channel morphology data based on the optimized scanned image includes:

[0021] Determine the pore structure parameters based on the cross-sectional pore structure image;

[0022] Generate a three-dimensional model of the pressure relief channel of the optimized explosion venting cenosphere board based on the pore structure parameters;

[0023] Determine the pressure relief channel morphology data of the optimized explosion venting cenosphere board based on the three-dimensional model of the pressure relief channel.

[0024] Optionally, the determining the pore structure parameters based on the cross-sectional pore structure image includes:

[0025] Remove the background area in the cross-sectional pore structure image to obtain a pore-related image;

[0026] Use a connected component analysis algorithm to determine the connected pores in the pore-related image;

[0027] Determine the pore size parameters of the connected pores, and determine the distribution of the connected pores of the connected pores in the cross-sectional pore structure image;

[0028] Based on the distribution of the connected pores and the pore size parameters, determine the permeability of the porous medium corresponding to the optimized explosion venting cenospheres board;

[0029] Based on the permeability of the porous medium, the distribution of the connected pores and the pore size parameters, generate the pore structure parameters corresponding to the optimized explosion venting cenospheres board.

[0030] Optionally, the determining the explosion venting channel morphology data of the optimized explosion venting cenospheres board based on the three-dimensional model of the explosion venting channel includes:

[0031] Based on the three-dimensional model of the explosion venting channel, determine the channel spatial distribution and channel geometric parameters;

[0032] Based on the three-dimensional model of the explosion venting channel and the channel connectivity characteristics extracted from the channel spatial distribution, use a graph theory algorithm to determine the channel network topology structure;

[0033] Based on the channel network topology structure and the channel geometric parameters, determine the explosion venting channel morphology data of the optimized explosion venting cenospheres board.

[0034] Optionally, the training of the to-be-trained model based on the target simulation data and the performance requirements corresponding to the sample explosion venting cenospheres board includes:

[0035] In the case that the explosion venting performance parameters indicated by the target simulation data do not meet the performance requirements, adjust the cenosphere ratio of the optimized explosion venting cenospheres board to adjust the optimized scan image, and thus re-execute the steps of generating the explosion venting channel morphology data and subsequent steps based on the optimized scan image until the performance requirements are met, and obtain the optimized scan image obtained last time;

[0036] Based on the comparison result between the optimized scan image obtained last time and the sample scan image, determine the second quality control scheme;

[0037] Based on the difference between the first quality control scheme and the second quality control scheme, train the to-be-trained model.

[0038] Optionally, adjusting the bead ratio of the optimized explosion - venting cenospheres board includes:

[0039] Based on the target simulation data, adjusting the bead ratio of the optimized explosion - venting cenospheres board.

[0040] Optionally, the target simulation data includes pressure fluctuation simulation data. Based on the target simulation data, adjusting the bead ratio of the optimized explosion - venting cenospheres board includes:

[0041] Performing cluster analysis on the pressure fluctuation simulation data to obtain at least one abnormal fluctuation feature;

[0042] Based on the at least one abnormal fluctuation feature, adjusting the bead ratio of the optimized explosion - venting cenospheres board.

[0043] In a second aspect, an explosion - venting cenospheres board quality control system based on a load test provided by an embodiment of the present application includes:

[0044] A quality control plan generation module, configured to generate a quality control plan for optimizing the design of the explosion - venting cenospheres board by using a quality control model according to a scanned image of the explosion - venting cenospheres board to be optimized;

[0045] Wherein, the training method of the quality control model includes:

[0046] Based on a sample scanned image of a sample explosion - venting cenospheres board, generating a first quality control plan through a model to be trained;

[0047] Based on the first quality control plan, adjusting the sample scanned image to an optimized scanned image, wherein the optimized scanned image corresponds to an optimized explosion - venting cenospheres board, and the optimized explosion - venting cenospheres board is obtained by optimizing the sample explosion - venting cenospheres board according to the first quality control plan;

[0048] Generating pressure - relief channel morphology data based on the optimized scanned image;

[0049] Based on the pressure - relief channel morphology data, determining target simulation data through a load test, wherein the target simulation data indicates explosion - venting performance parameters related to the bead ratio, and the bead ratio is the ratio of the gelling material of the optimized explosion - venting cenospheres board to the mass of cenospheres;

[0050] Training the model to be trained based on the target simulation data and the performance requirements corresponding to the sample explosion - venting cenospheres board to obtain the quality control model.

[0051] In summary, the embodiments of the present application have at least the following beneficial effects:

[0052] Using the embodiments of the present application, according to the scanned image of the blast venting cenosphere board to be optimized, a quality control scheme for optimizing the design of the blast venting cenosphere board is generated by using a quality control model; wherein, the training method of the quality control model includes: based on the sample scanned image of the sample blast venting cenosphere board, generating a first quality control scheme through a to-be-trained model; based on the first quality control scheme, adjusting the sample scanned image to an optimized scanned image, wherein the optimized scanned image corresponds to an optimized blast venting cenosphere board, and the optimized blast venting cenosphere board is obtained by optimizing the sample blast venting cenosphere board through the first quality control scheme; based on the optimized scanned image, generating blast venting channel morphology data; based on the blast venting channel morphology data, determining target simulation data through a load test, wherein the target simulation data indicates blast venting performance parameters related to the cenosphere ratio, and the cenosphere ratio is the ratio of the cementitious material to the cenosphere mass of the optimized blast venting cenosphere board; training the to-be-trained model based on the target simulation data and the performance requirements corresponding to the sample blast venting cenosphere board to obtain the quality control model, thereby improving the design optimization efficiency of the blast venting cenosphere board. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 FIG. is a schematic flowchart of a quality control method for a blast venting cenosphere board based on a load test provided by an embodiment of the present application;

[0054] Figure 2 FIG. is a schematic flowchart of a training method of a quality control model provided by an embodiment of the present application;

[0055] Figure 3 FIG. is a schematic structural diagram of a quality control system for a blast venting cenosphere board based on a load test provided by an embodiment of the present application;

[0056] Figure 4 FIG. is a schematic diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts shall fall within the protection scope of the present application.

[0058] In the description of the present application, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first", "second", "third", etc. may explicitly or implicitly include one or more of such features. In the description of the present application, unless otherwise specified, the meaning of "plural" is two or more. In the description of the present application, the term "comprising" and its variations are open-ended, that is, "including but not limited to". The term "based on" means "at least partially based on". The term "according to" means "at least partially according to". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments".

[0059] In the description of the present application, it should be noted that unless otherwise clearly defined and limited, the terms "installed", "connected", "coupled" should be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral connection; it may be a mechanical connection or an electrical connection; it may be directly connected or indirectly connected through an intermediate medium, and it may be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations.

[0060] In the description of the present application, it should be noted that unless otherwise defined, all technical and scientific terms used in the present application have the same meaning as commonly understood by those skilled in the technical field to which this application belongs. The terms used in the specification of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations.

[0061] In a first aspect, referring to Figure 1 , a schematic flowchart of a quality control method for an explosion venting floating bead board based on a load test provided by an embodiment of the present application is shown. The method includes step S101, which is as follows.

[0062] S101, according to the scanned image of the explosion venting floating bead board to be optimized, use a quality control model to generate a quality control plan for optimizing the design of the explosion venting floating bead board.

[0063] It should be noted that the quality control plan in this embodiment can usually optimize the internal structure design of the explosion venting floating bead board to be optimized, thereby changing the bead ratio of the explosion venting floating bead board.

[0064] In one example, the scanned image of the above-mentioned explosion venting cenosphere board can be obtained by a scanning device. Among them, the scanning device can include at least one of a scanning electron microscope, an atomic force microscope, X-ray computed tomography, and an optical microscope. The scanning electron microscope is usually used to obtain images of the microstructure of materials.

[0065] In one example, the scanned image of the above-mentioned explosion venting cenosphere board can be input into a quality control model based on a load test to obtain a quality control scheme output by the quality control model for optimizing the design of the explosion venting cenosphere board.

[0066] Among them, referring to Figure 2 , a schematic flowchart of the training method of the quality control model provided by an embodiment of the present application is shown. The training method of the quality control model includes steps S201 - S205, which are specifically as follows:

[0067] S201, based on the sample scanned image of the sample explosion venting cenosphere board, generate a first quality control scheme through a model to be trained.

[0068] It should be noted that the first quality control scheme in this embodiment may refer to a scheme generated by the model to be trained according to the sample scanned image for optimizing the design of the sample explosion venting cenosphere board. Usually, it is to optimize the internal structure design of the sample explosion venting cenosphere board.

[0069] In one example, the sample scanned image of the above-mentioned sample explosion venting cenosphere board can be obtained by a scanning device. Among them, the scanning device can include at least one of a scanning electron microscope, an atomic force microscope, X-ray computed tomography, and an optical microscope.

[0070] In one example, the model to be trained may include a machine learning model to be trained, such as a convolutional neural network, a support vector machine, a random forest, etc. Among them, the convolutional neural network usually includes an input layer, a convolutional layer, an activation function layer, a pooling layer, a normalization layer, a fully connected layer, an output layer, etc.

[0071] In one example, the first quality control scheme can be generated by inputting the sample scanned image of the above-mentioned sample explosion venting cenosphere board into the above-mentioned model to be trained.

[0072] In one example, various scanned images described in any embodiment of the present application can be used to reflect at least one piece of information such as the cenosphere distribution, pore structure, glue - cenosphere ratio, and interface bonding state of the corresponding cenosphere board.

[0073] S202. Based on the first quality control scheme, adjust the sample scan image to an optimized scan image, where the optimized scan image corresponds to an optimized explosion - venting cenospheres board, and the optimized explosion - venting cenospheres board is obtained by optimizing the sample explosion - venting cenospheres board through the first quality control scheme.

[0074] In one example, the above - mentioned first quality control scheme can be used to optimize the internal structure design of the sample explosion - venting cenospheres board. Thus, the local image at the corresponding position in the sample scan image can be adjusted according to the optimized internal structure design, so that the optimized scan image can match the optimized explosion - venting cenospheres board, that is, the optimized scan image can be substantially the same as the image obtained by scanning the optimized explosion - venting cenospheres board. Here, the optimized scan image can be used to reflect the microstructure of the optimized explosion - venting cenospheres board.

[0075] In one example, the above - mentioned first quality control scheme can be used to indicate optimization parameters. For example, the optimization parameters can include at least one of the following: bead ratio, density distribution, porosity adjustment direction, etc.

[0076] S203. Generate explosion - relief channel morphology data based on the optimized scan image.

[0077] In one example, generating explosion - relief channel morphology data based on the optimized scan image can include: separating the solid part and the pore part in the optimized scan image through an image segmentation algorithm, and the image segmentation algorithm includes threshold segmentation, region growing, and / or deep - learning methods, etc. Extract pore network information from the pore part of the optimized scan image using a preset tool, and the preset tool can include ImageJ / FIJI (with various plugins for pore analysis), PoreSpy (which can be used for digital image analysis of porous media), etc. Calculate various morphological parameters for describing the characteristics of the explosion - relief channel according to the pore network information to form explosion - relief channel morphology data, and the morphological parameters can include at least one of the following: porosity (the ratio of pore volume to total volume), average pore diameter, pore distribution, connectivity, and the shortest path between connected pores.

[0078] S204. Determine target simulation data through a load test based on the explosion - relief channel morphology data, where the target simulation data indicates explosion - venting performance parameters related to the bead ratio, and the bead ratio is the ratio of the gelling material to the cenosphere mass of the optimized explosion - venting cenospheres board.

[0079] Among them, the explosion - venting cenospheres board is a special type of board, and its material usually includes a gelling material and lightweight aggregate. In this embodiment, the lightweight aggregate uses cenospheres (a kind of hollow micro - beads extracted from fly ash).

[0080] In one example, first, the CAD (Computer Aided Design) software or a dedicated porous medium modeling tool (such as PoreSpy, MATLAB, etc.) can be directly used to construct an accurate three-dimensional model based on the morphological data of the pressure relief channel. This three-dimensional model can be used to reflect the geometric characteristics of the internal pores of the material of the sample explosion venting perlite board, and the geometric characteristics include pore size, connectivity, distribution, etc. Then, a suitable finite element analysis software (such as general software like ANSYS, ABAQUS, COMSOL Multiphysics, etc.) is selected to simulate the three-dimensional model. During the simulation process, a series of designed loading scenarios are used to simulate different load conditions, so as to implement a load test and obtain target simulation data. Among them, the above different load conditions can include at least one of the following: static pressure loading, dynamic impact loading, etc. Among them, the static pressure loading can include pressure loading at different groups of constant pressure loading rates, and the dynamic impact loading can include impact loading corresponding to different groups of dynamic pressure loading rates. The dynamic pressure loading rate can dynamically output pressure according to a set pressure change function, and the pressure change function can be used to simulate the pressure situation during an explosion.

[0081] In one example, the above load test can be designed according to the relevant explosion protection / venting requirements of a specific substation.

[0082] It should be noted that since the sample explosion venting perlite board in this embodiment usually changes the bead ratio of the sample explosion venting perlite board when it is designed and optimized according to the first quality control scheme, therefore, in the case where the bead ratio of the sample explosion venting perlite board is changed, the bead ratio of the optimized explosion venting perlite board in this embodiment is the changed bead ratio of the sample explosion venting perlite board. Thus, the target simulation data can be used to indicate the explosion venting performance parameters of the optimized explosion venting perlite board at the current bead ratio after the above optimized explosion venting perlite board is obtained.

[0083] S205, based on the target simulation data and the performance requirements corresponding to the sample explosion venting perlite board, train the to-be-trained model to obtain the quality control model.

[0084] In this embodiment, it is possible to enable the model to be trained to strengthen its own recognition ability of the correlation between the bead ratio and the explosion venting performance parameters during the training process, so that the quality control model obtained through training can accurately and reliably complete the optimal design of the explosion venting float bead board to be optimized. In each design, only need to initially design an explosion venting float bead board, and then call the trained quality control model to optimize the relevant design of the explosion venting float bead board, avoiding the aimless preparation of multiple explosion venting float bead boards with different optimization directions and evaluating the advantages and disadvantages of different optimization directions through actual tests, thereby reducing the number of actual tests and improving the efficiency of the optimal design.

[0085] In one example, based on the target simulation data and the performance requirements corresponding to the sample explosion venting float bead board, training the model to be trained to obtain the quality control model may include: calculating a first loss through a loss function based on the difference between the explosion venting performance parameters indicated by the target simulation data and the performance requirements corresponding to the sample explosion venting float bead board, and training the model to be trained according to the first loss to obtain the quality control model. Among them, during the training process, the model to be trained can be trained using the backpropagation algorithm according to the first loss. Among them, the loss function may include mean square error, cross-entropy loss, etc.

[0086] In one example, the differences described in various embodiments of the present application can be calculated respectively through at least one of the following: the similarity between semantic / image features, the relative difference between index values. Among them, the similarity can be calculated through cosine similarity, Euclidean distance, etc.

[0087] In an optional implementation manner, the explosion venting channel morphology data includes multiple morphology data, and based on the explosion venting channel morphology data, determining the target simulation data through a load test includes:

[0088] For each morphology data, if a simulation result corresponding to the morphology data is queried in a preset simulation data set, then determine the simulation data of the morphology data according to the queried simulation result, otherwise, conduct a load simulation test according to the morphology data to obtain the simulation data of the morphology data, where the simulation data set includes at least one simulation result;

[0089] Based on the simulation data of each of the multiple morphology data, determine the target simulation data.

[0090] In one example, the load simulation tests in various embodiments of the present application can be designed according to the relevant explosion protection / venting requirements of a specific substation.

[0091] In one example, determining the simulation data of the form data according to the queried simulation results may include: directly using the queried simulation results as the simulation data of the form data.

[0092] In one example, determining the target simulation data based on the simulation data of each of the multiple form data may include: forming the target simulation data from the simulation data of each of the multiple form data.

[0093] In one example, the preset simulation data set may further include at least one preset form data corresponding to each simulation result. Thus, it is possible to match, in the preset simulation data set, the preset form data that has the highest similarity to the form data and is higher than the set similarity threshold, and use the simulation result corresponding to the matched preset form data as the queried simulation result.

[0094] In one example, the above-mentioned pressure relief channel form data includes multiple form data, and the multiple form data may be obtained by dividing the pressure relief channel form data according to different local region images in the optimized scan image. In other words, each form data at this time may correspond to a local region image in the optimized scan image. Correspondingly, since the training model to be trained usually needs to be iteratively trained, at least one simulation result in the simulation data set may include: the form data and its simulation data generated during the completed training rounds of the iterative training.

[0095] Here, it should be understood that a scan image usually contains a large amount of information, and the higher the scan accuracy, the greater the amount of information. On the other hand, during the training process, usually a very large number of iterative training rounds are required. If, in each round of training, the pressure relief channel form data of the entire optimized scan image is used to determine the target simulation data through a load simulation test, the computational amount is extremely large. And this embodiment actually equivalently reuses the simulation data generated in the previous completed training rounds, thereby reducing the number of load simulation tests and reducing the computational amount brought by excessive load simulation tests.

[0096] It should also be noted here that different rounds of training in the above iterative training may be completed according to different sample scan images, and the above different sample scan images usually have some different regions, that is, some other regions may be the same. At this time, after being optimized and designed by the first quality control scheme, different optimized scan images can be obtained, but there may still be some regions that are the same between different optimized scan images. The reused simulation data is this part of the same region.

[0097] In an optional implementation manner, performing a load simulation test according to the form data to obtain the simulation data of the form data includes:

[0098] Generate a finite element model based on the morphological data;

[0099] Based on the finite element model, perform a load simulation test using finite element analysis to obtain the explosion pressure release path distribution and explosion pressure release efficiency under the set explosion conditions;

[0100] Based on the explosion pressure release path distribution and the explosion pressure release efficiency, determine the simulation data of the morphological data.

[0101] In one example, the finite element analysis can select a suitable finite element analysis software (such as general software like ANSYS, ABAQUS, COMSOL Multiphysics, etc.) to perform a load simulation test on the finite element model. During the simulation, the load simulation test is realized according to the designed set explosion conditions, and the set explosion conditions can characterize the load conditions. The load conditions can include dynamic impact loading. Among them, the dynamic impact loading can include impact loading corresponding to multiple groups of dynamic pressure loading rates, and the dynamic pressure loading rate can dynamically output pressure according to a set pressure change function, and the pressure change function can be used to simulate the pressure situation when an explosion occurs.

[0102] In this way, during the simulation, the explosion pressure release path distribution and explosion pressure release efficiency under each explosion condition can be collected. Here, the explosion pressure release path distribution can be used to characterize the release path and the distribution of the release path when the optimized explosion venting perlite board bears and releases the explosion shock pressure, and the explosion pressure release efficiency can be used to characterize the explosion shock pressure rate corresponding to each position on each release path.

[0103] In one example, generating a finite element model based on the morphological data can include: generating a finite element model according to the morphological data by using the above finite element analysis software.

[0104] In one example, determining the simulation data of the morphological data based on the explosion pressure release path distribution and the explosion pressure release efficiency can include: constituting the simulation data of the morphological data from the explosion pressure release path distribution and the explosion pressure release efficiency.

[0105] In an optional implementation manner, the optimized scanned image includes a cross-sectional pore structure image of the optimized explosion venting perlite board, and generating the explosion venting channel morphological data based on the optimized scanned image includes:

[0106] Determine the pore structure parameters based on the cross-sectional pore structure image;

[0107] Generate a three-dimensional model of the explosion venting channel of the optimized explosion venting perlite board based on the pore structure parameters;

[0108] Based on the three-dimensional model of the pressure relief channel, determine the pressure relief channel morphological data of the optimized explosion-proof cenospheres board.

[0109] In one example, the above pore structure parameters can be used to indicate at least one of the following: connected pore information, pore distribution characteristics.

[0110] In one example, based on the cross-sectional pore structure image, determining pore structure parameters may include: detecting pores and their geometric structure parameters from the cross-sectional pore structure image through a target detection algorithm, and then generating pore structure parameters according to the detected pores and their geometric structure parameters.

[0111] In one example, based on the pore structure parameters, generating the three-dimensional model of the pressure relief channel of the optimized explosion-proof cenospheres board may include: directly using CAD software or specialized porous media modeling tools (such as PoreSpy, MATLAB, etc.) to construct the three-dimensional model of the pressure relief channel of the optimized explosion-proof cenospheres board according to the pore structure parameters.

[0112] In one example, based on the three-dimensional model of the pressure relief channel, determining the pressure relief channel morphological data of the optimized explosion-proof cenospheres board may include: extracting pore network information from the three-dimensional model of the pressure relief channel, and calculating various morphological parameters for describing the characteristics of the pressure relief channel according to the pore network information to form the pressure relief channel morphological data. The morphological parameters may include at least one of the following: porosity (the ratio of pore volume to total volume), average pore diameter, pore distribution, connectivity, and the shortest path between connected pores.

[0113] In an alternative implementation, the determining of pore structure parameters based on the cross-sectional pore structure image includes:

[0114] Removing the background area in the cross-sectional pore structure image to obtain a pore-related image;

[0115] Using a connected component analysis algorithm to determine the connected pores in the pore-related image;

[0116] Determining the pore diameter parameters of the connected pores, and determining the connected pore distribution of the connected pores in the cross-sectional pore structure image;

[0117] Based on the connected pore distribution and the pore diameter parameters, determining the permeability of the porous medium corresponding to the optimized explosion-proof cenospheres board;

[0118] Based on the permeability of the porous medium, the connected pore distribution, and the pore diameter parameters, generating the pore structure parameters corresponding to the optimized explosion-proof cenospheres board.

[0119] In one example, based on the connected pore distribution and the pore size parameters, determining the permeability of the porous medium corresponding to the optimized explosion venting cenospheres board may include: calculating the porosity based on the connected pore distribution, and using the Kozeny-Carman model to determine the permeability of the porous medium corresponding to the optimized explosion venting cenospheres board based on the average pore size characterized by the pore size parameters and the porosity. Among them, the Kozeny-Carman model (constructed based on the Kozeny-Carman equation) is a classical model for estimating the permeability of porous media.

[0120] Among them, the mathematical expression of the Kozeny-Carman model includes:

[0121]

[0122] Among them, is the permeability of the porous medium, is the porosity (dimensionless), is the average pore size, is the tortuosity (dimensionless) preset in the Kozeny-Carman model, and usually takes values between 1.5 and 4.

[0123] In one example, based on the connected pore distribution and the pore size parameters, determining the permeability of the porous medium corresponding to the optimized explosion venting cenospheres board may also include: using a permeability prediction model to predict the permeability of the porous medium corresponding to the optimized explosion venting cenospheres board based on the connected pore distribution and the pore size parameters. Among them, the permeability prediction model can be a model that has been trained to have the prediction ability with the connected pore distribution and pore size parameters as the model input and the permeability of the porous medium as the model output. During specific training, sample connected pore distribution and sample pore size parameters can be used as sample data (the sample data also carries the corresponding permeability label expected, and the label represents the corresponding expected permeability of the porous medium), and a general training algorithm (such as the embodiments related to the training method of the above-mentioned model to be trained in this application) can be used to train the model so that the trained model can have the above-mentioned ability. It is easy to understand that the sample data in this embodiment can be experimental data measured in advance through multiple corresponding experiments. For example, on a certain explosion venting cenospheres board, measure its current connected pore distribution and pore size parameters, and then test / calculate its current permeability of the porous medium to form sample data.

[0124] In one example, the pore size parameters may include pore size shape parameters (such as diameter length, geometric shape, etc.).

[0125] In one example, based on the permeability of the porous medium, the distribution of the connected pores, and the pore size parameters, generating the pore structure parameters corresponding to the optimized explosion venting cenospheres board may include: constituting the pore structure parameters corresponding to the optimized explosion venting cenospheres board with the permeability of the porous medium, the distribution of the connected pores, and the pore size parameters.

[0126] In an alternative embodiment, the determining the explosion venting channel morphology data of the optimized explosion venting cenospheres board based on the three-dimensional model of the explosion venting channel includes:

[0127] Determining the channel spatial distribution and channel geometric parameters based on the three-dimensional model of the explosion venting channel;

[0128] Determining the channel network topology using a graph theory algorithm based on the three-dimensional model of the explosion venting channel and the channel connectivity characteristics extracted from the channel spatial distribution;

[0129] Determining the explosion venting channel morphology data of the optimized explosion venting cenospheres board based on the channel network topology and the channel geometric parameters.

[0130] In one example, determining the channel spatial distribution and channel geometric parameters based on the three-dimensional model of the explosion venting channel may include: identifying each channel in the three-dimensional model of the explosion venting channel, taking the spatial distribution of each channel in the three-dimensional model of the explosion venting channel as the channel spatial distribution, and taking the geometric parameters of each channel in the three-dimensional model of the explosion venting channel as the channel geometric parameters.

[0131] In one example, determining the channel network topology using a graph theory algorithm based on the three-dimensional model of the explosion venting channel and the channel connectivity characteristics extracted from the channel spatial distribution may include: identifying each pore point (including branch points and end points) and the edges between each pore point in the three-dimensional model of the explosion venting channel according to the channel connectivity characteristics, and then generating the channel network topology using a graph theory algorithm based on each pore point and the edges between each pore point.

[0132] In one example, determining the explosion venting channel morphology data of the optimized explosion venting cenospheres board based on the channel network topology and the channel geometric parameters may include: constituting the explosion venting channel morphology data of the optimized explosion venting cenospheres board with the channel network topology and the channel geometric parameters.

[0133] In an alternative embodiment, the training the to-be-trained model based on the target simulation data and the performance requirements corresponding to the sample explosion venting cenospheres board includes:

[0134] In the case where the explosion relief performance parameters indicated by the target simulation data do not meet the performance requirements, adjusting the rubber-bead ratio of the optimized explosion relief floating bead plate to adjust the optimized scan image, thereby re-executing the steps of generating pressure relief channel morphology data based on the optimized scan image and subsequent steps until the performance requirements are met, and obtaining the last optimized scan image;

[0135] Determining a second quality control solution based on a comparison result between the optimized scan image obtained last time and the sample scan image;

[0136] The to-be-trained model is trained based on the difference between the first quality control scheme and the second quality control scheme.

[0137] It should be noted that the comparison result between the last optimized scanning image obtained and the sample scanning image in this embodiment can be used to characterize the difference between the sample scanning image and the last optimized scanning image obtained and whose performance meets the requirements, so that the structural differences of the floating bead boards corresponding to the two can be determined based on the difference, and then a second quality control scheme can be generated based on the structural difference, so that the second quality control scheme can be used to optimize the design of the sample explosion-proof floating bead board to the floating bead board corresponding to the last optimized scanning image obtained.

[0138] In one example, the above performance requirements may include at least one of the following: whether the rising speed of the pressure in the internal channel of the plate exceeds the maximum rising speed threshold when venting (pressure relief), whether there is excessive local oscillation when under pressure (for example, exceeding the maximum value of the oscillation change per unit time), and whether there is a pressure relief delay during pressure relief (for example, whether the pressure relief is completed within the set time).

[0139] In one example, adjusting the bead ratio of the optimized explosion-venting floating bead board may include adjusting the bead ratio of the optimized explosion-venting floating bead board according to a preset adjustment strategy. The preset adjustment strategy may include increasing the bead ratio to enhance the overall strength and adhesion of the material, or decreasing the bead ratio to increase porosity to enhance pressure relief efficiency.

[0140] In one example, training the model to be trained based on the difference between the first quality control scheme and the second quality control scheme may include: calculating a second loss using a loss function based on the difference, and training the model to be trained based on the second loss to obtain the quality control model. During the training process, the model to be trained may be trained using a backpropagation algorithm based on the second loss. The loss function may include mean square error, cross entropy loss, etc.

[0141] In an alternative embodiment, adjusting the bead ratio of the optimized explosion - venting cenospheres board includes:

[0142] Based on the target simulation data, adjusting the bead ratio of the optimized explosion - venting cenospheres board.

[0143] In an example, based on the target simulation data, adjusting the bead ratio of the optimized explosion - venting cenospheres board may include: selecting a target adjustment strategy from the above - mentioned preset adjustment strategies based on the difference between the explosion - venting performance parameters indicated by the target simulation data and the performance requirements, and adjusting the bead ratio of the optimized explosion - venting cenospheres board according to the target adjustment strategy.

[0144] In an alternative embodiment, the target simulation data includes pressure - fluctuation simulation data. Based on the target simulation data, adjusting the bead ratio of the optimized explosion - venting cenospheres board includes:

[0145] Performing clustering analysis on the pressure - fluctuation simulation data to obtain at least one abnormal - fluctuation feature;

[0146] Based on the at least one abnormal - fluctuation feature, adjusting the bead ratio of the optimized explosion - venting cenospheres board.

[0147] In an example, the clustering analysis may include the K - Means clustering algorithm.

[0148] In an example, the abnormal - fluctuation feature can be used to indicate at least one of the following: whether there is a sudden pressure rise / drop, whether there is an intensified local oscillation, and whether there is a delay in pressure relief.

[0149] In an example, when the abnormal - fluctuation feature indicates an intensified local oscillation, if there is a severe initial oscillation, it may indicate that there may be a blockage at the pore entrance (due to too small pore entrances and / or uneven pore - entrance distribution, etc.). At this time, the design of the optimized explosion - venting cenospheres board can be adjusted according to the pore - entrance size and / or pore - entrance distribution, thereby adjusting the bead ratio. If there is a severe mid - stage oscillation, it may indicate that there may be a fracture in the internal channel and / or the internal channel is too narrow. At this time, the internal - channel design of the optimized explosion - venting cenospheres board can be adjusted.

[0150] In an example, based on the at least one abnormal - fluctuation feature, adjusting the bead ratio of the optimized explosion - venting cenospheres board may include: when the abnormal - fluctuation feature indicates a severe pressure - relief oscillation, the bead ratio can be appropriately increased to enhance the strength of the pore wall and reduce the fluctuations caused by local collapse. When the abnormal - fluctuation feature indicates a slow pressure relief, the bead ratio can be appropriately decreased to restore the porosity and connectivity and improve the pressure - relief efficiency.

[0151] Second aspect, correspondingly, the embodiments of the present application further provide a quality control system for explosion venting perlite boards based on load tests, which can implement all processes of the quality control method for explosion venting perlite boards based on load tests provided in the above embodiments.

[0152] Refer to Figure 3 , which shows a schematic structural diagram of the quality control system for explosion venting perlite boards based on load tests provided in the embodiments of the present application. The quality control system for explosion venting perlite boards based on load tests includes:

[0153] A quality control plan generation module 301, configured to generate a quality control plan for optimizing the design of the explosion venting perlite board by using a quality control model according to the scanned image of the explosion venting perlite board to be optimized;

[0154] Among them, the training method of the quality control model includes:

[0155] Based on the sample scanned image of the sample explosion venting perlite board, generate a first quality control plan through the model to be trained;

[0156] Based on the first quality control plan, adjust the sample scanned image to an optimized scanned image, where the optimized scanned image corresponds to an optimized explosion venting perlite board, and the optimized explosion venting perlite board is obtained by optimizing the sample explosion venting perlite board according to the first quality control plan;

[0157] Generate pressure relief channel morphology data based on the optimized scanned image;

[0158] Based on the pressure relief channel morphology data, determine target simulation data through load tests, where the target simulation data indicates explosion venting performance parameters related to the bead ratio, and the bead ratio is the ratio of the gelling material to the perlite mass of the optimized explosion venting perlite board;

[0159] Train the model to be trained based on the target simulation data and the performance requirements corresponding to the sample explosion venting perlite board to obtain the quality control model.

[0160] In an optional implementation manner, the pressure relief channel morphology data includes multiple morphology data. Based on the pressure relief channel morphology data, determining the target simulation data through load tests includes:

[0161] For each morphology data, if a simulation result corresponding to the morphology data is queried in a preset simulation dataset, determine the simulation data of the morphology data according to the queried simulation result; otherwise, perform a load simulation test according to the morphology data to obtain the simulation data of the morphology data, where the simulation dataset includes at least one simulation result;

[0162] Determine the target simulation data based on the simulation data of each of the multiple morphology data.

[0163] In an optional implementation manner, the performing a load simulation test according to the morphology data to obtain the simulation data of the morphology data includes:

[0164] Generate a finite element model according to the morphology data;

[0165] Based on the finite element model, perform a load simulation test using finite element analysis to obtain the explosion pressure release path distribution and explosion pressure release efficiency under set explosion conditions;

[0166] Determine the simulation data of the morphology data based on the explosion pressure release path distribution and the explosion pressure release efficiency.

[0167] In an optional implementation manner, the optimized scan image includes the cross-sectional pore structure image of the optimized explosion venting perlite board, and the generating the explosion venting channel morphology data based on the optimized scan image includes:

[0168] Determine pore structure parameters based on the cross-sectional pore structure image;

[0169] Generate a three-dimensional model of the explosion venting channel of the optimized explosion venting perlite board based on the pore structure parameters;

[0170] Determine the explosion venting channel morphology data of the optimized explosion venting perlite board based on the three-dimensional model of the explosion venting channel.

[0171] In an optional implementation manner, the determining the pore structure parameters based on the cross-sectional pore structure image includes:

[0172] Remove the background area in the cross-sectional pore structure image to obtain a pore-related image;

[0173] Use a connected component analysis algorithm to determine the connected pores in the pore-related image;

[0174] Determine the pore diameter parameters of the connected pores, and determine the connected pore distribution of the connected pores in the cross-sectional pore structure image;

[0175] Determine the permeability of the porous medium corresponding to the optimized explosion venting perlite board based on the connected pore distribution and the pore diameter parameters;

[0176] Generate the pore structure parameters corresponding to the optimized explosion venting perlite board based on the permeability of the porous medium, the connected pore distribution and the pore diameter parameters.

[0177] In an alternative embodiment, determining the pressure relief channel morphology data of the optimized explosion venting cenospheres board based on the three-dimensional model of the pressure relief channel includes:

[0178] Based on the three-dimensional model of the pressure relief channel, determining the channel spatial distribution and channel geometric parameters;

[0179] Based on the three-dimensional model of the pressure relief channel and the channel connectivity features extracted from the channel spatial distribution, using graph theory algorithms to determine the channel network topology structure;

[0180] Based on the channel network topology structure and the channel geometric parameters, determining the pressure relief channel morphology data of the optimized explosion venting cenospheres board.

[0181] In an alternative embodiment, training the to-be-trained model based on the target simulation data and the performance requirements corresponding to the sample explosion venting cenospheres board includes:

[0182] In the case where the explosion venting performance parameters indicated by the target simulation data do not meet the performance requirements, adjusting the bead ratio of the optimized explosion venting cenospheres board to adjust the optimized scan image, and thus re-executing the steps of generating the pressure relief channel morphology data and subsequent steps based on the optimized scan image until the performance requirements are met, and obtaining the optimized scan image obtained last;

[0183] Based on the comparison result between the optimized scan image obtained last and the sample scan image, determining the second quality control scheme;

[0184] Based on the difference between the first quality control scheme and the second quality control scheme, training the to-be-trained model.

[0185] In an alternative embodiment, adjusting the bead ratio of the optimized explosion venting cenospheres board includes:

[0186] Based on the target simulation data, adjusting the bead ratio of the optimized explosion venting cenospheres board.

[0187] In an alternative embodiment, the target simulation data includes pressure fluctuation simulation data, and based on the target simulation data, adjusting the bead ratio of the optimized explosion venting cenospheres board includes:

[0188] Performing cluster analysis on the pressure fluctuation simulation data to obtain at least one abnormal fluctuation feature;

[0189] Based on the at least one abnormal fluctuation feature, adjusting the bead ratio of the optimized explosion venting cenospheres board.

[0190] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method for quality control of explosion venting cenospheres board based on load test described in any one of the above are implemented.

[0191] In a fourth aspect, an embodiment of the present application provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, the steps of the method for quality control of explosion venting cenospheres board based on load test described in any one of the above are implemented.

[0192] In a fifth aspect, an embodiment of the present application provides a computer device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the steps of the method for quality control of explosion venting cenospheres board based on load test described in any one of the above are implemented.

[0193] See Figure 4 , the computer device of this embodiment includes: a processor 401, a memory 402, and a computer program stored in the memory 402 and executable on the processor 401, such as a quality control program for explosion venting cenospheres board based on load test. When the processor 401 executes the computer program, the steps in each of the above embodiments of the method for quality control of explosion venting cenospheres board based on load test are implemented, such as Figure 1 the step S101 shown.

[0194] Exemplarily, the computer program may be divided into one or more modules / units, and the one or more modules / units are stored in the memory 402 and executed by the processor 401 to complete the present application. The one or more modules / units may be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program in the computer device.

[0195] The computer device may be a computing device such as a desktop computer, a notebook, a palm computer, and a cloud server. The computer device may include, but is not limited to, a processor 401 and a memory 402. Those skilled in the art can understand that the schematic diagram is only an example of the computer device, and does not constitute a limitation on the computer device. It may include more or fewer components than shown, or combine certain components, or different components. For example, the computer device may further include input / output devices, network access devices, a bus, etc.

[0196] The processor 401 may be a Central Processing Unit (CPU), or may also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor, or the processor 401 may also be any conventional processor, etc. The processor 401 is the control center of the computer device, and connects various parts of the entire computer device through various interfaces and lines.

[0197] The memory 402 can be used to store the computer programs and / or modules. The processor 401 realizes various functions of the computer device by running or executing the computer programs and / or modules stored in the memory 402, and by invoking the data stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.). In addition, the memory 402 may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a Flash Card, at least one magnetic disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0198] Among them, if the modules / units integrated in the computer device are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, to implement all or part of the processes in the above-described embodiment methods of this application, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor 401, the steps of the above-described various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0199] In summary, the embodiments of this application have at least the following beneficial effects:

[0200] By using the embodiments of this application, according to the scanned image of the blast relief perlite board to be optimized, a quality control scheme for optimizing the design of the blast relief perlite board is generated by using a quality control model; wherein, the training method of the quality control model includes: based on the sample scanned image of the sample blast relief perlite board, generating a first quality control scheme through a to-be-trained model; based on the first quality control scheme, adjusting the sample scanned image to an optimized scanned image, wherein the optimized scanned image corresponds to an optimized blast relief perlite board, and the optimized blast relief perlite board is obtained by optimizing the sample blast relief perlite board through the first quality control scheme; based on the optimized scanned image, generating blast relief channel morphology data; based on the blast relief channel morphology data, determining target simulation data through a load test, wherein the target simulation data indicates blast relief performance parameters related to the bead ratio, and the bead ratio is the ratio of the gelling material to the perlite mass of the optimized blast relief perlite board; based on the target simulation data and the performance requirements corresponding to the sample blast relief perlite board, training the to-be-trained model to obtain the quality control model, thereby being able to improve the design optimization efficiency of the blast relief perlite board.

[0201] Through the description of the above embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus a necessary hardware platform. Of course, it can also be implemented entirely through hardware. Based on such an understanding, all or part of the technical solution of this application that contributes to the background art can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM (Read-Only Memory), RAM (Random Access Memory), magnetic disk, optical disc, etc., and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0202] The above is the preferred embodiment of this application. It should be noted that for those of ordinary skill in the art, without departing from the principle of this application, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of this application.

Claims

1. A method for controlling the quality of explosion-proof floating bead boards based on load testing, characterized in that: include: Based on the scanned image of the explosion-venting floating bead board to be optimized, using the quality control model, a quality control plan for optimizing the design of the explosion-venting floating bead board is generated; The training method of the quality control model includes: Based on the sample scanned image of the sample explosion-proof floating bead plate, a first quality control solution is generated through the to-be-trained model; Based on the first quality control scheme, the sample scan image is adjusted to an optimized scan image, wherein the optimized scan image corresponds to an optimized explosion-venting floating bead board, and the optimized explosion-venting floating bead board is obtained by optimizing the sample explosion-venting floating bead board through the first quality control scheme; generating pressure relief channel morphological data based on the optimized scan image; Based on the pressure relief channel morphology data, target simulation data is determined through a load test, wherein the target simulation data indicates explosion relief performance parameters related to the glue-bead ratio, and the glue-bead ratio is the ratio of the mass of the cementitious material to the floating beads of the optimized explosion relief floating bead board; Based on the target simulation data and the performance requirements corresponding to the sample explosion-proof floating bead board, the model to be trained is trained to obtain the quality control model.

2. The method according to claim 1, characterized in that The pressure relief channel morphology data includes a plurality of morphology data, and the target simulation data is determined by a load test based on the pressure relief channel morphology data, including: For each morphological data, if a simulation result corresponding to the morphological data is obtained by querying a preset simulation data set, then the simulation data of the morphological data is determined according to the queried simulation result; otherwise, a load simulation test is performed according to the morphological data to obtain the simulation data of the morphological data, wherein the simulation data set includes at least one simulation result; Target simulation data is determined based on the simulation data of each of the plurality of morphological data.

3. The method according to claim 2, characterized in that The performing of a load simulation test according to the morphological data to obtain simulation data of the morphological data includes: generating a finite element model based on the morphological data; Based on the finite element model, a load simulation test is performed using finite element analysis to obtain the explosion pressure release path distribution and explosion pressure release efficiency under set explosion conditions; Based on the explosion pressure release path distribution and the explosion pressure release efficiency, simulation data of the morphological data is determined.

4. The method according to claim 1, wherein The optimized scan image includes a cross-sectional pore structure image of the optimized explosion-relief floating bead plate, and generating pressure relief channel morphology data based on the optimized scan image includes: determining pore structure parameters based on the cross-sectional pore structure image; Based on the pore structure parameters, a three-dimensional model of the pressure relief channel of the optimized explosion relief floating bead plate is generated; Based on the three-dimensional model of the pressure relief channel, the morphological data of the pressure relief channel of the optimized explosion-relief floating bead board is determined.

5. The method according to claim 4, characterized in that The determining of pore structure parameters based on the cross-sectional pore structure image includes: removing a background area from the cross-sectional pore structure image to obtain a pore-related image; Using a connected domain analysis algorithm to determine connected pores in the pore-related image; determining pore size parameters of the connected pores, and determining a connected pore distribution of the connected pores in the cross-sectional pore structure image; Determining the porous medium permeability corresponding to the optimized explosion-relief floating bead plate based on the connected pore distribution and the pore size parameters; Based on the permeability of the porous medium, the connected pore distribution and the pore size parameters, the pore structure parameters corresponding to the optimized explosion-venting floating bead plate are generated.

6. The method according to claim 4, characterized in that The step of determining the morphological data of the pressure relief channel of the optimized explosion relief floating bead plate based on the three-dimensional model of the pressure relief channel includes: Determining channel spatial distribution and channel geometric parameters based on the three-dimensional model of the pressure relief channel; Determining the channel network topology using a graph theory algorithm based on the three-dimensional model of the pressure relief channel and the channel connectivity features extracted from the channel spatial distribution; Based on the channel network topology and the channel geometric parameters, the pressure relief channel morphological data of the optimized explosion relief floating bead board is determined.

7. The method according to any one of claims 1 to 6, characterized in that The training of the to-be-trained model based on the target simulation data and the performance requirements corresponding to the sample explosion-relief floating bead board includes: In the case where the explosion relief performance parameters indicated by the target simulation data do not meet the performance requirements, adjusting the rubber-bead ratio of the optimized explosion relief floating bead plate to adjust the optimized scan image, thereby re-executing the steps of generating pressure relief channel morphology data based on the optimized scan image and subsequent steps until the performance requirements are met, and obtaining the last optimized scan image; Determining a second quality control solution based on a comparison result between the optimized scan image obtained last time and the sample scan image; The to-be-trained model is trained based on the difference between the first quality control scheme and the second quality control scheme.

8. The method according to claim 7, characterized in that The adjusting of the rubber-bead ratio of the optimized explosion-relief floating bead board includes: Based on the target simulation data, the rubber-bead ratio of the optimized explosion-relief floating bead board is adjusted.

9. The method according to claim 8, characterized in that The target simulation data includes pressure fluctuation simulation data, and adjusting the rubber-bead ratio of the optimized explosion-relief floating bead board based on the target simulation data includes: Performing cluster analysis on the pressure fluctuation simulation data to obtain at least one abnormal fluctuation feature; Based on the at least one abnormal fluctuation feature, the rubber-bead ratio of the optimized explosion-relief floating bead board is adjusted.

10. A quality control system for explosion-proof floating bead boards based on load testing, characterized in that: include: A quality control scheme generating module is used to generate a quality control scheme for optimizing the design of the explosion-venting floating bead board based on a scanned image of the explosion-venting floating bead board to be optimized and using a quality control model; The training method of the quality control model includes: Based on the sample scanned image of the sample explosion-proof floating bead plate, a first quality control solution is generated through the to-be-trained model; Based on the first quality control scheme, the sample scan image is adjusted to an optimized scan image, wherein the optimized scan image corresponds to an optimized explosion-venting floating bead board, and the optimized explosion-venting floating bead board is obtained by optimizing the sample explosion-venting floating bead board through the first quality control scheme; generating pressure relief channel morphological data based on the optimized scan image; Based on the pressure relief channel morphology data, target simulation data is determined through a load test, wherein the target simulation data indicates explosion relief performance parameters related to the glue-bead ratio, and the glue-bead ratio is the ratio of the mass of the cementitious material to the floating beads of the optimized explosion relief floating bead board; Based on the target simulation data and the performance requirements corresponding to the sample explosion-proof floating bead board, the model to be trained is trained to obtain the quality control model.

Images