Furniture board structural performance detection system and detection method

Abstract

The present invention discloses a furniture panel structural performance detection system and detection method, comprising: a multi-source acquisition module: collecting panel structural data and environment-related data, extracting a first correlation parameter from the collected panel structural parameters, and extracting a second correlation parameter from the related environment parameters; an association acquisition module: performing cross-modal fusion analysis on the first correlation parameter and the second correlation parameter to obtain an optimized correlation index, and obtaining trend data from the optimized correlation index; a data correction module: collecting panel characteristic data, and optimizing the mutual correlation coefficient between the panel characteristic data and the trend data, performing correlation analysis based on the optimized mutual correlation coefficient to determine an attenuation mode, performing attenuation compensation on the collected real-time panel characteristic data based on the attenuation mode, obtaining corrected panel characteristic data, and ensuring that the corrected strength data is closer to the actual performance under the actual environment.

Description

Technical Field

[0001] The present invention relates to the technical field of data detection, and in particular to a furniture board structural performance detection system and method. Background Art

[0002] The development of furniture panel structural performance testing systems and methods is the result of growing market demand and technological advancement. As people's quality of life improves, the demand for furniture quality becomes increasingly stringent. As a key component of furniture, the structural performance of furniture panels directly determines the quality, safety, and service life of the furniture, prompting the continuous development of related testing technologies. Simultaneously, innovations in materials science have provided new test targets and challenges for the development of testing technology. The increasing impact of environmental factors on panel performance has also exposed the limitations of traditional testing methods. The rapid development of technologies such as artificial intelligence, sensors, and data processing has provided strong support for innovative breakthroughs in testing technology.

[0003] At present, existing detection systems usually use fixed algorithm models to process detection data. However, the performance of furniture boards will change with environmental changes, and fixed algorithms lack consideration of environmental factors. In an environment with high humidity, solid wood boards are easily affected by moisture, and the moisture content increases, resulting in dimensional expansion and decreased strength. For example, during the rainy season in the south, the moisture content of red oak boards may increase from the normal 12% to 20%. During detection, the fixed algorithm does not adjust the evaluation criteria for board strength and dimensional stability according to the change in moisture content, and still performs detection according to the parameters in a dry environment. This will overestimate the actual performance of the board, causing the manufactured furniture to deform, crack, and other problems during subsequent use. Therefore, a furniture board structural performance detection system and detection method are proposed to solve the above problems. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned objectives, the present invention proposes the following technical solutions:

[0005] A furniture board structural performance detection system, comprising:

[0006] Multi-source acquisition module: collects plate structure data and environment-related data, extracts first correlation parameters from the collected plate structure parameters, and extracts second correlation parameters from the related environment parameters;

[0007] Correlation acquisition module: performs cross-modal fusion analysis on the first correlation parameter and the second correlation parameter to obtain optimized correlation indicators, and obtains trend data from the optimized correlation indicators;

[0008] Data correction module: collects plate characteristic data and optimizes the correlation coefficient between the plate characteristic data and trend data. Based on the optimized correlation coefficient, correlation analysis is performed to determine the attenuation mode. Based on the attenuation mode, attenuation compensation is performed on the collected real-time plate characteristic data to obtain the corrected plate characteristic data.

[0009] Detection output module: obtains the detection results of furniture panels based on the corrected panel characteristic data.

[0010] The plate structure data acquisition process is as follows:

[0011] Use an industrial CT scanner to scan the plate in all directions to obtain the three-dimensional structural image data of the plate At the same time, use density measuring equipment to obtain the density values ​​of different positions of the board , using a 3D laser scanner to obtain the geometric shape data of the plate surface ;

[0012] The environment-related data collection method is:

[0013] Deploy a distributed environmental sensor network to collect ambient temperature data and relative humidity data .

[0014] The first associated parameter acquisition process is:

[0015] Through the image segmentation algorithm, the three-dimensional structure image data Separate the defective area and obtain the defect parameters ;

[0016] From geometric data Extract the flatness of the board surface , edge contour information data , obtain the optimized morphological index parameters through the mapping relationship between the flatness of the plate surface and the contour information data of the edge ;

[0017] Density values ​​from different locations Get the average density ;

[0018] The first associated parameter is expressed as ;

[0019] The second associated parameter acquisition process is:

[0020] Extract the temperature variation range within a specific time period , humidity change rate ;

[0021] Record a period of time Maximum internal ambient temperature and minimum value The difference between the two is the temperature range ;

[0022] By relative humidity for a period of time Continuous monitoring and calculation of humidity change rate ;

[0023] The second set of associated parameters is represented as .

[0024] The optimized morphological indicator parameters The acquisition process is:

[0025] The flatness of the board surface , edge contour information data Convert to point cloud data ;

[0026] Through a Model, integrate the attention mechanism module into the middle layer of the model to obtain Converged network model;

[0027] Point cloud data Input into the PointNet++ fusion network model to obtain the local feature vector ;

[0028] The network model fuses the processed local features to obtain global morphological features ;

[0029] Then the extracted global feature vector is analyzed to obtain the topological features ;

[0030] According to the extracted global morphological features and topological features , establish the mapping relationship between global morphological features and topological features;

[0031] Mapping the eigenvalue representing the continuity of edge contour in the topological feature to the smoothness index of the edge;

[0032] Mapping the components of the global eigenvector related to surface relief into surface smoothness indices;

[0033] The edge smoothness index and the surface flatness index together form a sub-index ;

[0034] According to the sub-indicators obtained by mapping , determine the weights of different sub-indicators according to the hierarchical analysis method, and then obtain the final optimized morphological index parameters through weighted summation .

[0035] The process of obtaining the optimization correlation index is as follows:

[0036] For the first associated parameter and the second associated parameter Perform feature encoding representation;

[0037] For the first associated parameter Defect parameters in , using convolutional neural network to extract spatial feature data and obtain the first eigenvector ;

[0038] Optimizing morphological indicator parameters for numerical parameters and average density , perform normalization to obtain the second eigenvector , directly take the first eigenvector Splice with the second eigenvector to get the first comprehensive feature representation ;

[0039] For the second associated parameter Temperature range and humidity change rate Directly perform normalized linear transformation to obtain the second comprehensive feature representation ;

[0040] Will and As the input of the multi-head attention mechanism, the optimized correlation index is obtained through the output of the multi-head attention mechanism .

[0041] The process of extracting the trend data is as follows:

[0042] Optimize related indicators Input into a capsule network model;

[0043] The optimized correlation index data obtained after preliminary mapping is allocated to different capsules according to the capsule rules in the capsule network model;

[0044] After the allocation is completed, different capsules output specific multidimensional vectors ;

[0045] For the The capsule in The multidimensional vector output at each time point is expressed as ;

[0046] Count the modulus and trend of the optimized correlation indicators in all capsules;

[0047] Arrange the modulus of each capsule at different time points into a matrix ,in , ;

[0048] At adjacent time points and , set up the The rate of change of the modulus of the capsule output vector :

[0049] when When The features captured by the capsules tend to increase. When , it means that the feature is attenuating, and the change rates of all capsules at different time points are sorted into a matrix ,in , ;

[0050] Finally, a data sequence is formed ,in, represents the importance matrix, Represents the trend matrix;

[0051] This data series is trend data.

[0052] The plate characteristic data includes tensile strength , compressive strength , flexural strength data .

[0053] The acquisition process of the attenuation mode is as follows:

[0054] Obtain relevant target characteristic index c of plate characteristic data;

[0055] Calculate the dynamic weight factor for each characteristic data according to the target characteristic index c , the formula is:

[0056]

[0057] in, is the mean square error of the neural network model for the target characteristic index c, is the weight coefficient;

[0058] Obtain the Pearson correlation coefficient between plate characteristic data and trend data :

[0059] Dynamic weight factor Assign Pearson correlation coefficient Get the optimized cross-correlation coefficient ;

[0060] Set a relevance threshold ;

[0061] when When , it is considered that there is a strong correlation between the plate characteristic data and the trend data;

[0062] when When , it is considered that there is a weak correlation between the plate characteristic data and the trend data;

[0063] Calculate the difference between trend data at adjacent time points , when at multiple consecutive time points , the trend data is considered to be in a decaying state;

[0064] For trend data that is in a decaying state and has a strong correlation Plate characteristic data;

[0065] like And the trend data decays, then the plate characteristic data appears to be enhanced;

[0066] like And the trend data decays, then the plate characteristic data decays;

[0067] According to the attenuation of the plate characteristic data, the attenuation mode of the plate is determined.

[0068] The process of obtaining the corrected plate characteristic data is as follows:

[0069] For the plate characteristic data with attenuation mode, a linear attenuation compensation model is established. ;

[0070] Assume plate characteristic data and trend data The attenuation relationship between them is: ,in, is the attenuation coefficient, indicating that there is attenuation in any type of plate characteristic data;

[0071] Collect plate characteristic data and corresponding trend data to form a series of sample points , the collected data are divided into a training set and a validation set as the training data of the linear compensation model;

[0072] Substitute the change in trend data into the established linear compensation relationship Calculate the predicted change in plate characteristic data ;

[0073] The predicted change The change in the actual plate characteristic data in the validation set Compare and minimize the loss function using the mean square error method;

[0074] After minimizing the loss function, the trained linear compensation model is obtained, and the predicted attenuation value is calculated based on the output results of the linear compensation model.

[0075] A method for testing the structural performance of furniture panels, comprising the following steps:

[0076] S1: Collect plate structure data and environment-related data, extract a first correlation parameter from the collected plate structure parameters, and extract a second correlation parameter from the related environment parameters;

[0077] S2: performing cross-modal fusion analysis on the first correlation parameter and the second correlation parameter to obtain an optimized correlation index, and obtaining trend data from the optimized correlation index;

[0078] S3: Collecting plate characteristic data, optimizing the correlation coefficient between the plate characteristic data and trend data, performing correlation analysis based on the optimized correlation coefficient to determine an attenuation mode, performing attenuation compensation on the collected real-time plate characteristic data based on the attenuation mode, and obtaining corrected plate characteristic data;

[0079] S4: Based on the corrected board characteristic data, obtain the test results of the furniture board.

[0080] The present invention has the following beneficial effects:

[0081] In this invention, the system first collects ambient temperature and humidity data in real time and performs cross-modal fusion analysis with the plate structure data. Trend data is extracted through a capsule network and a multi-head attention mechanism. Combined with dynamic weight factors to optimize correlation analysis, the system dynamically adjusts the detection parameters as the environment changes, avoiding overestimation of performance and solving the problem that the strength of the plate decreases after being damp, but the fixed algorithm cannot adapt.

[0082] Secondly, they used multiple sensors, such as industrial CT and 3D laser scanners, to collect data on the three-dimensional structure, density distribution, and surface geometry of the panels. This was combined with a distributed environmental sensor network to acquire time-series data on temperature and humidity. Furthermore, they used the PointNet++ fusion network model to extract global morphological features. They also used topological data analysis to quantify edge smoothness and surface flatness. This, combined with the dynamic weighted sub-indicators of the analytic hierarchy process, significantly improved the accuracy of the correlation analysis between structural defects and environmental impacts.

[0083] Finally, based on the Pearson correlation coefficient and the attenuation state of the trend data, the system identifies the attenuation mode of the plate characteristic data (tensile, compressive, and flexural strength) and corrects the real-time data through the linear compensation model. For example, during the rainy season, if the humidity change rate increases and the flexural strength Strong negative correlation with trend data , the system will automatically enhance the compensation to ensure that the corrected intensity data is closer to the actual performance in the actual environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 This is a system block diagram of a furniture board structural performance detection system and detection method proposed by the present invention;

[0085] Figure 2 This is a method step diagram of a furniture board structural performance detection system and detection method proposed by the present invention. DETAILED DESCRIPTION

[0086] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0087] Example 1

[0088] like Figure 1 As shown, the present invention proposes a furniture board structural performance detection system, comprising:

[0089] Multi-source acquisition module: collects plate structure data and environment-related data, extracts first correlation parameters from the collected plate structure parameters, and extracts second correlation parameters from the related environment parameters;

[0090] Plate structure data acquisition process:

[0091] Use an industrial CT scanner to scan the plate in all directions to obtain the three-dimensional structural image data of the plate At the same time, use density measuring equipment to obtain the density values ​​of different positions of the board , using a 3D laser scanner to obtain the geometric shape data of the plate surface ;

[0092] The process of obtaining the first associated parameter is:

[0093] Through the image segmentation algorithm, the three-dimensional structure image data Separate defective areas (such as identifying defects, voids, cracks, etc. inside the plate) and obtain defect parameters ;

[0094] From geometric data Extract the flatness of the board surface , edge contour information data , obtain the optimized morphological index parameters through the mapping relationship between the flatness of the plate surface and the contour information data of the edge ;

[0095] Density values ​​from different locations Get the average density ;

[0096] The first associated parameter is expressed as ;

[0097] Specifically, through the flatness of the plate surface , edge contour information data Get the optimized pattern indicator parameters The specific process is:

[0098] The flatness of the board surface , edge contour information data Convert to point cloud data ;

[0099] Through a PointNet++ model, an attention mechanism module is integrated into the middle layer of the model to obtain a PointNet++ fusion network model;

[0100] Specifically, the PointNet++ fusion network model is used. The traditional PointNet++ model excels at processing disordered point cloud data and extracting local and global features. On this basis, the attention mechanism module is integrated into the middle layer, enabling the network to automatically focus on key areas closely related to the flatness and edge contour of the plate when processing point cloud data.

[0101] Point cloud data Input into the PointNet++ fusion network model, the fusion network model first performs a preliminary extraction of the features of each point, uses grouping and feature extraction operations to extract local features of the point cloud at different scales, and aggregates features of each point in its neighborhood to obtain a local feature vector ;

[0102] The network model fuses the processed local features to obtain global morphological features that can represent the overall morphology of the plate. ;

[0103] Then the extracted global feature vector is analyzed to obtain the topological features ;

[0104] According to the extracted global morphological features and topological features , establish the mapping relationship between global morphological features and topological features;

[0105] Mapping the eigenvalue representing the continuity of edge contour in the topological feature to the smoothness index of the edge;

[0106] Mapping the components of the global eigenvector related to surface relief into surface smoothness indices;

[0107] The edge smoothness index and the surface flatness index together form a sub-index ;

[0108] According to the sub-indicators obtained by mapping , determine the weights of different sub-indicators according to the hierarchical analysis method, and then obtain the final optimized morphological index parameters through weighted summation , the formula is:

[0109]

[0110] in, The first Sub-indicators, is the number of sub-indicators, For the The weight of each sub-indicator;

[0111] The methods for collecting environment-related data are:

[0112] Deploy a distributed environmental sensor network to collect ambient temperature data and relative humidity data ;

[0113] The second associated parameter is obtained as follows:

[0114] Extract the temperature variation range within a specific time period , humidity change rate ;

[0115] Record a period of time Maximum internal ambient temperature and minimum value The difference between the two is the temperature range ;

[0116] Specifically, the temperature variation range has an important impact on the performance of the board. For example, large temperature variations may cause the board to expand and contract, thereby generating internal stress and affecting its structural stability.

[0117] By relative humidity for a period of time Continuous monitoring and calculation of humidity change rate ;

[0118] Specifically, the humidity change rate reflects the speed of change of the ambient humidity. Rapid humidity changes may cause the moisture content of the board to change rapidly, leading to problems such as deformation and cracking of the board.

[0119] These parameters constitute the second set of associated parameters .

[0120] Correlation acquisition module: performs cross-modal fusion analysis on the first correlation parameter and the second correlation parameter to obtain optimized correlation indicators, and obtains trend data from the optimized correlation indicators;

[0121] The process of obtaining the optimization-related indicators is as follows:

[0122] For the first associated parameter and the second associated parameter Perform feature encoding representation;

[0123] For the first associated parameter Defect parameters in , using convolutional neural network to extract spatial feature data and obtain the first eigenvector ;

[0124] Optimizing morphological indicator parameters for numerical parameters and average density , perform normalization to obtain the second eigenvector , directly take the first eigenvector Splice with the second eigenvector to get the first comprehensive feature representation ;

[0125] For the second associated parameter Temperature range and humidity change rate Directly perform normalized linear transformation to obtain the second comprehensive feature representation ;

[0126] Specifically, this step is to convert parameters of different forms into feature vector forms suitable for subsequent attention mechanism processing;

[0127] Will and As the input of the multi-head attention mechanism, the optimized correlation index is obtained through the output of the multi-head attention mechanism ;

[0128] The process of extracting trend data is as follows:

[0129] The optimized correlation index data obtained through the multi-head attention transfer mechanism is input into a capsule network model;

[0130] The capsule network model maps the original optimized correlation index data to a new feature space, so that the potential features in the data can be better represented and processed;

[0131] The optimized correlation index data obtained after preliminary mapping is allocated to different capsules according to the capsule rules in the capsule network model;

[0132] Specifically, the capsule network model is based on the sensitivity and capture ability of each capsule to different types of indicator data. For example, features related to the internal texture structure of the board will be assigned to the capsule responsible for processing structural features; features related to environmental humidity and temperature changes will be assigned to the capsule responsible for processing corresponding environmental factors.

[0133] After the allocation is completed, different capsules will output specific multidimensional vectors ;

[0134] For the The capsule in The multidimensional vector output at each time point is expressed as ;

[0135] Observe the modulus and trend of the multidimensional vector. The modulus indicates the strength or importance of the feature captured by the capsule, while the trend of the vector reflects whether the feature pattern is increasing or decreasing.

[0136] Multidimensional vector The modulus length is expressed as: ,in, Represents a multidimensional vector Middle Components (i.e., the corresponding optimization correlation indicators);

[0137] The trend is reflected by the change of the modulus over time. and , compare In capsules The size of the corresponding optimization correlation index, if the time point The corresponding optimization correlation index is greater than The optimization correlation index of The optimized correlation index in each capsule is a decay trend;

[0138] Count the importance (modulus) and trend of the corresponding optimization correlation indicators in all capsules to form a data sequence ,in, represents the importance matrix, Represents the trend matrix, this data sequence is the trend data;

[0139] The specific process of obtaining the data sequence is as follows:

[0140] Arrange the modulus of each capsule at different time points into a matrix ,in , so that we can intuitively see the changes in the characteristic intensity of different capsules over time;

[0141] At adjacent time points and , set up the The rate of change of the modulus of the capsule output vector :

[0142]

[0143] when When The features captured by the capsules tend to increase; when When , it means that the feature is attenuating, and the change rates of all capsules at different time points are sorted into a matrix ,in .

[0144] Data correction module: collects plate characteristic data and optimizes the correlation coefficient between the plate characteristic data and trend data. Based on the optimized correlation coefficient, correlation analysis is performed to determine the attenuation mode. Based on the attenuation mode, attenuation compensation is performed on the collected real-time plate characteristic data to obtain the corrected plate characteristic data.

[0145] Plate properties including tensile strength , compressive strength , flexural strength data ;

[0146] Specifically, these three strength data intuitively demonstrate the load-bearing capacity of the board under different stress modes. In an environment with high humidity, the strength of solid wood boards will decrease after being damp. By collecting tensile strength, compressive strength, and bending strength data in real time, it can directly reflect the changes in the actual strength performance of the board;

[0147] Get the decay mode;

[0148] The process of obtaining the attenuation mode is:

[0149] First, the correlation coefficient between the plate characteristic data and the trend data is optimized;

[0150] By dynamically weighting the target characteristic indicators between the plate characteristic data and the environment, the traditional correlation coefficient calculation method is modified to obtain the optimized mutual correlation coefficient between the plate characteristic data and the trend data;

[0151] Obtain relevant target characteristic indicators of plate characteristic data:

[0152] Assume that the collected plate characteristic data is ;

[0153] Analyze the correlation between different characteristic data and environmental factors, and preliminarily determine the degree of correlation between them Then, a neural network model is used to take the environmental data and the characteristic data of the plate as the input layer, and the target characteristic index as the output layer, and the correlation degree is used to The network model is trained with historical data to learn the mapping relationship between them, and the model finally outputs the target characteristic index c;

[0154] Calculate the dynamic weight factor for each characteristic data according to the target characteristic index c , the formula is:

[0155]

[0156] in, is the mean square error of the neural network model for the target characteristic index c, is the weight coefficient;

[0157] Obtain the Pearson correlation coefficient between plate characteristic data and trend data :

[0158] Dynamic weight factor Assign Pearson correlation coefficient Get the optimized cross-correlation coefficient ;

[0159] Set a relevance threshold , used to judge the strength of the correlation, when When , it is considered that there is a strong correlation between the plate characteristic data and the trend data;

[0160] when When , it is considered that there is a weak correlation between the plate characteristic data and the trend data;

[0161] Continuously monitor trend data To determine whether it shows a decay trend, calculate the difference between trend data at adjacent time points. , if at multiple consecutive time points , the trend data is considered to be in a decaying state;

[0162] For trend data in a decaying state and with strong correlation ( ) of the plate characteristic data, and further analyze its changes. And the trend data decays, then the plate characteristic data appears to be enhanced;

[0163] like And the trend data decays, then the plate characteristic data decays;

[0164] Determine the attenuation mode of the plate according to the attenuation of the plate characteristic data;

[0165] Further traverse the tensile strength of the plate characteristic data , compressive strength , flexural strength data Correlation with trend data, if And the trend data continues to decay, which means that there is a decay pattern in the tensile strength, recorded as , indicating that “compressive strength-trend data has decay”;

[0166] For the plate characteristic data that is determined to have attenuation mode, a linear attenuation compensation model is established ;

[0167] Assume plate characteristic data and trend data The attenuation relationship between them is: ,in, is the attenuation coefficient, indicating that there is attenuation in any type of plate characteristic data;

[0168] Collect plate characteristic data and corresponding trend data within a random time span to form a series of sample points , the collected data are divided into a training set and a validation set as the training data of the linear compensation model;

[0169] According to the ratio of 7:3, the first 70% of the data is used as the training set for fitting, and the last 30% of the data is used as the validation set;

[0170] Using the validation set data, substitute the change in trend data into the established linear compensation relationship Calculate the predicted change in plate characteristic data , and then compared with the actual plate characteristic data changes in the validation set Compare and minimize the loss function using the mean square error method;

[0171] After minimizing the loss function, the trained linear compensation model is obtained. The predicted attenuation value is calculated based on the output of the linear compensation model, and then the real-time plate data is corrected, which is expressed as;

[0172] Specifically, the predicted attenuation value is calculated based on the output of the linear compensation model, which is obtained by the trained linear compensation model and is used to reflect the expected reduction in the plate characteristic data due to the change in the trend data;

[0173] Specifically, in the established linear compensation relationship, the change in trend data collected in real time is substituted into the model, and the calculated change in the plate characteristic data is the predicted attenuation value, which reflects the model trained based on historical data and the estimation of the attenuation degree of plate characteristic data under the current trend data change.

[0174] Detection output module: obtains the detection results of furniture panels based on the corrected panel characteristic data;

[0175] In the revised board characteristic data, the board structure data can reflect the inherent properties of the board itself, while the environmental data reflects the external conditions of the board. Only by mastering these two aspects of data can we fully understand the performance of the board in actual use. For example, the internal structure of solid wood boards will be affected by different temperature and humidity environments, which will change the performance and affect the accuracy of detection.

[0176] Since the performance of the board will change due to various factors such as the environment, and the original board structure data cannot accurately reflect its current true performance, by performing attenuation compensation on the real-time collected board characteristic data based on the attenuation mode, these data can be corrected to make them more in line with the actual situation. After correcting the collected strength data, directly performing routine testing on it can obtain more accurate furniture board test results.

[0177] Example 2

[0178] like Figure 2 As shown, the present invention proposes a method for detecting the structural performance of furniture panels, comprising:

[0179] S1: Collect plate structure data and environment-related data, extract a first correlation parameter from the collected plate structure parameters, and extract a second correlation parameter from the related environment parameters;

[0180] S2: performing cross-modal fusion analysis on the first correlation parameter and the second correlation parameter to obtain an optimized correlation index, and obtaining trend data from the optimized correlation index;

[0181] S3: Collecting plate characteristic data, optimizing the correlation coefficient between the plate characteristic data and trend data, performing correlation analysis based on the optimized correlation coefficient to determine an attenuation mode, performing attenuation compensation on the collected real-time plate characteristic data based on the attenuation mode, and obtaining corrected plate characteristic data;

[0182] S4: Based on the corrected board characteristic data, obtain the test results of the furniture board.

[0183] In the application, several formulas involved are calculated by taking their numerical values ​​after removing the dimensions, and the formulas are established by collecting a large amount of data and performing software simulation to obtain a formula for the most recent real situation. Some coefficients or weights in the formulas are set by technical personnel in this field according to actual conditions, so they will not be elaborated here.

[0184] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. Those skilled in the art will appreciate that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0185] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A furniture board structural performance detection system, characterized in that: include: Multi-source acquisition module: collects plate structure data and environment-related data, extracts first correlation parameters from the collected plate structure parameters, and extracts second correlation parameters from the related environment parameters; Correlation acquisition module: performs cross-modal fusion analysis on the first correlation parameter and the second correlation parameter to obtain optimized correlation indicators, and obtains trend data from the optimized correlation indicators; The first associated parameter acquisition process is: Through the image segmentation algorithm, the defective area of ​​the three-dimensional structure image data I is separated to obtain the defect parameters Extract the flatness of the plate surface from the geometric shape data S Edge contour information data Obtain optimized morphological index parameters through the mapping relationship between the flatness of the plate surface and the contour information data of the edge Get the average density from the density values ​​p at different locations The first associated parameter is expressed as The second associated parameter acquisition process is: Extract the temperature variation range within a specific time period ΔT Humidity change rate Record the maximum value of the ambient temperature within a period of time ΔT h max and the minimum value h min The difference between the two is the temperature range By continuously monitoring the relative humidity R for a period of time ΔT, the humidity change rate is calculated The second set of associated parameters is represented as The optimized morphological indicator parameters The acquisition process is: The flatness of the board surface Edge contour information data Convert to point cloud data Through a PointNet++ model, an attention mechanism module is integrated into the middle layer of the model to obtain a PointNet++ fusion network model; Input the point cloud data D into the PointNet++ fusion network model to obtain the local feature vector a; The network model fuses the processed local features to obtain global morphological features Then the extracted global feature vector is analyzed to obtain the topological features According to the extracted global morphological features and topological features Establish the mapping relationship between global morphological features and topological features; Mapping the eigenvalue representing the continuity of edge contour in the topological feature to the smoothness index of the edge; Mapping the components of the global eigenvector related to surface relief into surface smoothness indices; The edge smoothness index and the surface flatness index together form the sub-index x i ; According to the mapping of each sub-index x i , determine the weights of different sub-indicators according to the hierarchical analysis method, and then obtain the final optimized morphological index parameters through weighted summation Data correction module: collects plate characteristic data and optimizes the correlation coefficient between the plate characteristic data and trend data. Based on the optimized correlation coefficient, correlation analysis is performed to determine the attenuation mode. Based on the attenuation mode, attenuation compensation is performed on the collected real-time plate characteristic data to obtain the corrected plate characteristic data. Detection output module: obtains the detection results of furniture panels based on the corrected panel characteristic data.

2. A furniture board structural performance detection system according to claim 1, characterized in that: The plate structure data acquisition process is as follows: An industrial CT scanner is used to scan the plate in all directions to obtain the plate's three-dimensional structural image data I. A density measurement device is used to obtain the density values ​​p at different positions on the plate, and a three-dimensional laser scanner is used to obtain the plate's surface geometric shape data S. The environment-related data collection method is: A distributed environmental sensor network is deployed to collect ambient temperature data h and relative humidity data R.

3. A furniture board structural performance detection system according to claim 1, characterized in that: The process of obtaining the optimization correlation index is as follows: Perform feature coding representation on the first associated parameter K1 and the second associated parameter K2; For the defect parameter in the first correlation parameter K1 Use convolutional neural network to extract spatial feature data and obtain the first feature vector F1; Optimizing morphological indicator parameters for numerical parameters and average density Normalization is performed to obtain the second eigenvector F2, and the first eigenvector F1 is directly concatenated with the second eigenvector to obtain the first comprehensive feature representation. The temperature variation range of the second correlation parameter K2 and humidity change rate Directly perform normalized linear transformation to obtain the second comprehensive feature representation Will and As the input of the multi-head attention mechanism, the optimized correlation index is obtained through the output of the multi-head attention mechanism 4. A furniture board structural performance detection system according to claim 1, characterized in that: The process of extracting the trend data is as follows: Optimize related indicators Input into a capsule network model; The optimized correlation index data obtained after preliminary mapping is allocated to different capsules according to the capsule rules in the capsule network model; After the allocation is completed, different capsules output specific multidimensional vectors v; The multidimensional vector output by the mth capsule at the jth time point is represented as v mj ; Count the modulus and trend of the optimized correlation indicators in all capsules; The modulus of each capsule at different time points is organized into a matrix A, where At adjacent time points j and j+1, let the change rate of the modulus of the output vector of the mth capsule be p mj : When p mj When >0, it means that the features captured by the mth capsule are increasing. mj When <0, it indicates that the feature is attenuating. The change rates of all capsules at different time points are organized into a moment B, where B mj =p mj , Finally, a data sequence T = [A, B] is formed, where A represents the importance matrix and B represents the trend matrix; This data series is trend data.

5. The furniture board structure performance detection system according to claim 1, characterized in that: The plate characteristic data includes tensile strength y1, compressive strength y2, and bending strength data y3.

6. A furniture board structural performance detection system according to claim 1, characterized in that: The acquisition process of the attenuation mode is as follows: Obtain relevant target characteristic index c of plate characteristic data; According to the target characteristic index c, the dynamic weight factor w is calculated for each characteristic data. The formula is expressed as: w=αc+(1-α)(1-MSE c ) Among them, MSE c is the mean square error of the neural network model for the target characteristic index c, and α is the weight coefficient; Obtain the Pearson correlation coefficient ρ between the plate characteristic data and trend data YT : Assign a dynamic weight factor w to the Pearson correlation coefficient ρ YT Get the optimized cross-correlation coefficient Set a relevance threshold when When , it is considered that there is a strong correlation between the plate characteristic data and the trend data; when When , it is considered that there is a weak correlation between the plate characteristic data and the trend data; Calculate the difference between trend data at adjacent time points ΔT = T t+1 -T t , when ΔT<0 at multiple consecutive time points, the trend data is considered to be in a decaying state; For trend data that is in a decaying state and has a strong correlation Plate characteristic data; like And the trend data decays, then the plate characteristic data appears to be enhanced; like And the trend data decays, then the plate characteristic data decays; According to the attenuation of the plate characteristic data, the attenuation mode of the plate is determined.

7. The furniture board structure performance detection system according to claim 1, characterized in that: The process of obtaining the corrected plate characteristic data is as follows: For the plate characteristic data with confirmed attenuation mode, a linear attenuation compensation model ΔY is established; Assume that the attenuation relationship between the plate characteristic data Y = [y1, y2, y3] and the trend data T is: in, is the attenuation coefficient, indicating that there is attenuation in any type of plate characteristic data; Collect plate characteristic data and corresponding trend data to form a series of sample points Divide the collected data into a training set and a validation set as training data for the linear compensation model; Substitute the change in trend data into the established linear compensation relationship Calculate the predicted change in plate characteristic data The predicted change The change in the actual plate characteristic data in the validation set Compare and minimize the loss function using the mean square error method; After minimizing the loss function, the trained linear compensation model is obtained, and the predicted attenuation value is calculated based on the output results of the linear compensation model.

8. A method for testing the structural performance of furniture panels, using the testing system according to any one of claims 1 to 7, characterized in that: The implementation steps include: S1: Collect plate structure data and environment-related data, extract a first correlation parameter from the collected plate structure parameters, and extract a second correlation parameter from the related environment parameters; S2: performing cross-modal fusion analysis on the first correlation parameter and the second correlation parameter to obtain an optimized correlation index, and obtaining trend data from the optimized correlation index; S3: Collecting plate characteristic data, optimizing the correlation coefficient between the plate characteristic data and trend data, performing correlation analysis based on the optimized correlation coefficient to determine an attenuation mode, performing attenuation compensation on the collected real-time plate characteristic data based on the attenuation mode, and obtaining corrected plate characteristic data; S4: Based on the corrected board characteristic data, obtain the test results of the furniture board.

Images