Building material quality detection system

Abstract

The invention relates to the technical field of quality detection, and particularly discloses a building material quality detection system, which comprises a material classification module for classifying according to the purposes of building materials and constructing a material defect set; the category determination module is used for determining key detection categories by using a principal component analysis method; the key detection module collects defect values of building materials of key monitoring categories in real time, and judges whether the change trend of the material defect ratio is reduced or not through a fitting curve; and the synchronous analysis module is used for calculating the average change rate of the limited material sequence and the adjacent batch of the material defect sequence if the material defect ratio is reduced, and judging whether the limited material ratio and the material defect ratio are synchronously reduced or not.

Description

Technical Field

[0001] The present invention relates to the technical field of quality inspection, and particularly relates to a building material quality inspection system. Background Art

[0002] At present, with the booming development of the construction industry, the quality of building materials plays an important role in the safety, durability and overall quality of construction projects. However, the quality of building materials varies greatly, bringing many potential risks to construction projects.

[0003] A Chinese patent application with the publication number CN113740173B discloses a building material detection system and method, including: a workbench is provided, a support plate is arranged at the upper end of the placement table, a detection air pump is arranged on the support plate, a detection probe is arranged at the lower end of the detection air pump, the lower end of the detection probe is conical, a positioning groove is opened at the upper end of the workbench, a baffle capable of moving relative to the positioning groove is arranged at the upper end of the workbench, a manipulator is arranged at the upper end of the workbench, the manipulator is provided with an elastic clamping assembly, the manipulator is provided with a horizontal moving assembly, a through hole is penetrated through the bottom wall of the positioning groove, an elastic pressing assembly is arranged in the through hole, and a lifting and moving assembly is arranged at the lower end of the elastic pressing assembly. It can realize the expansion test of the pipe orifice for various pipes with different specifications.

[0004] In the prior art, only a single performance of building materials is tested, and the overall quality of the materials is not comprehensively evaluated. If the material defect ratio and the defined material ratio are calculated, and then the material defect value is obtained to determine the key detection category, the quality of building materials can be comprehensively evaluated from multiple dimensions. For example, the prior art cannot judge the proportion of unqualified products in a batch of pipes and the situation of products that can be used under defined conditions. If the quantitative analysis of building materials is carried out to master the quality status of various materials, it can provide a basis for the selection of building materials.

[0005] In the prior art, the quality changes of different batches of materials are not tracked and analyzed, and the development trend of material quality cannot be analyzed. If the material defect values of different batches of building materials are collected in real time, a change curve is drawn, and the change trend of the material defect ratio is judged through linear regression analysis. When the material defect ratio increases abnormally, an early warning can be sent in time. It can also further analyze whether the material defect ratio and the defined material ratio decrease synchronously to discover potential quality risks in advance. For example, during the construction process, if the quality of pipes gradually deteriorates, it is difficult for the prior art to detect the overall trend change of the quality in time. If an early warning is sent to remind the construction party to take measures to avoid problems in the project due to the use of materials with declining quality.

[0006] Therefore, the present invention provides a building material quality inspection system. Summary of the Invention

[0007] The object of the present invention is to provide a building material quality detection system to solve the problems in the above background.

[0008] The object of the present invention can be achieved by the following technical solutions:

[0009] A building material quality detection system includes the following modules:

[0010] Material classification module: used to divide building materials into multiple categories, perform numerical analysis on the unqualified building defective materials of each category, obtain the material defect ratio and the limited material ratio, and construct a material defect set;

[0011] Category determination module: used to analyze and discriminate different categories of building materials in the material defect set by using the principal component analysis method to determine the key detection categories in different building material categories;

[0012] Key detection module: used to collect the material defect ratios of different batches of building materials in the key detection categories in real time, and judge whether the material defect ratio gradually decreases with the iteration of batches;

[0013] Synchronous analysis module: used to obtain the limited material ratio of the key monitoring category if the material defect ratio gradually decreases with the iteration of batches, and determine whether the material defect ratio and the limited material ratio decrease synchronously.

[0014] As a further technical solution of the present invention: the construction method of the material defect set is as follows:

[0015] Obtain the usage of the building, and divide the building materials into multiple categories according to different usage of the building materials;

[0016] From the building material database, obtain the number of unqualified building materials of each category of building materials in the same batch to obtain the number of defective materials;

[0017] Obtain the number of building materials with limited usage conditions in the same batch to obtain the number of limited materials, and obtain the number of each category of building materials in the same batch;

[0018] Take the ratio of the number of defective materials to the number of each category of building materials to obtain the material defect ratio;

[0019] Take the ratio of the number of limited materials to the number of each category of building materials to obtain the limited material ratio;

[0020] Based on the material defect ratio and the limited material ratio, obtain the material defect value through a weighted formula;

[0021] Calculate the material defect values of each category of building materials, sort the material defect values of each category of building materials in descending order, and construct a material defect set.

[0022] As a further technical solution of the present invention: the determination method of the key detection categories in the different building material categories is as follows:

[0023] Based on the material defect set, obtain the quantity of the specified materials actually used for the building materials of the same batch from the building material database to obtain the specified usage value;

[0024] Based on the specified usage value, construct a specified usage matrix, and use the principal component analysis method to obtain the comprehensive score of each building material;

[0025] Based on the comprehensive score of each building material, sort the building material categories corresponding to the material defect set, and determine the key detection categories among all the building material categories in the material defect set.

[0026] As a further technical solution of the present invention: the construction method of the specified usage matrix is as follows:

[0027] Obtain the specified usage value of each category of building materials, and process the ratio of the specified usage value to the number of specified materials to obtain the specified usage ratio;

[0028] Mark the specified usage ratio of each category of building materials as Xd i , where i is the number of the building material category of each type in the material defect set, and the value range of i is [1, m];

[0029] Obtain the specified usage ratios of all batches from the building material database, and mark the specified usage ratio of each type in each batch as Xd ij , where j is the batch number, and the value range of j is [1, n].

[0030] As a further technical solution of the present invention: the principal component analysis method is as follows:

[0031] Standardize the specified usage matrix to construct a standard specified usage matrix;

[0032] Based on the standard specified usage matrix, construct a covariance matrix through the covariance formula;

[0033] Perform eigenvalue decomposition on the covariance matrix to obtain the eigenvectors and corresponding eigenvalues of the covariance matrix;

[0034] Sum up all the eigenvalues to obtain the total eigenvalue sum, and process the ratio of each eigenvalue to the total eigenvalue sum to obtain the variance contribution rate;

[0035] Obtain the cumulative variance contribution rate of the preset k principal component numbers through the summation formula, and construct a principal component loading matrix P based on the eigenvectors corresponding to the k principal components;

[0036] Multiply the principal component load matrix by the standard limited dosage matrix to obtain the score matrix of k principal components;

[0037] Based on the score matrix of k principal components, obtain the comprehensive score of each building material through numerical analysis.

[0038] As a further technical solution of the present invention: the numerical analysis method for the comprehensive score of each building material is as follows:

[0039] Based on the ranking of each building material within the material defect set, obtain the weight coefficient;

[0040] Based on the weight coefficient, cumulative variance contribution rate, and principal component score matrix of each building material, obtain the comprehensive score of each building material through the product formula.

[0041] As a further technical solution of the present invention: the judgment method for whether the material defect ratio gradually decreases with the iteration of batches is as follows:

[0042] Based on the building materials in the key monitoring category, collect the material defect ratios of building materials in different batches in real time;

[0043] Using the batch as the X-axis and the material defect ratio as the Y-axis, plot the change curve of batch - material defect in the two-dimensional rectangular coordinate system;

[0044] Perform fitting analysis on the material defect ratios of building materials in different batches to obtain the fitting error value of the linear regression equation;

[0045] If the fitting error value of the linear regression equation is higher than the preset fitting error threshold, it indicates that the fitting effect of the linear regression equation on the change curve of batch - material defect meets the expectation;

[0046] If the fitting effect of the linear regression equation on the change curve of batch - material defect meets the expectation, obtain the slope of the linear regression equation;

[0047] If the slope of the linear regression equation is negative, it means that with the iteration of batches, the material defect ratio of building materials gradually decreases.

[0048] As a further technical solution of the present invention: the method for performing fitting analysis on the material defect ratios of building materials in different batches is as follows:

[0049] Use the linear regression equation to fit the change curve of batch - material defect;

[0050] Through the mean absolute error and mean square error formulas, obtain the mean absolute error and mean square error of the linear regression equation;

[0051] The mean absolute error and the mean square error are weighted and summed to obtain the fitting error value of the linear regression equation.

[0052] As a further technical solution of the present invention: The judgment method of whether the material defect ratio and the defined material ratio are synchronously reduced is as follows:

[0053] Obtain the defined material ratio of each batch in the key monitored category of building materials, and construct a defined material sequence;

[0054] Obtain the material defect ratio of each batch, and construct a material defect sequence;

[0055] Analyze the defined material sequence and the material defect sequence through the average rate of change to obtain the sequence change ratio;

[0056] If the sequence change ratio is within the preset change ratio range, it is considered that while the material defect ratio in the material defect sequence is gradually decreasing, the defined material ratio is synchronously decreasing.

[0057] As a further technical solution of the present invention: The method for obtaining the sequence change ratio is as follows:

[0058] Through the average rate of change formula, obtain the average rate of change Rx between adjacent batches of the defined material sequence and the average rate of change Ry between adjacent batches of the material defect sequence;

[0059] Through the ratio formula, obtain the ratio of the average rate of change Rx between adjacent batches of the defined material sequence to the average rate of change Ry of the adjacent batches of the material defect sequence, that is, the sequence change ratio.

[0060] The beneficial effects of the present invention:

[0061] (1) By classifying building materials in detail according to their uses and combining historical data to analyze the defect conditions of various building materials, constructing a material defect set, and using the principal component analysis method to determine the key detection categories in various building materials, it is beneficial to screen out the most concerned parts from numerous building material categories, reasonably allocate detection resources, improve detection efficiency, reduce unnecessary detection costs and time waste, and make the detection work of building material quality more targeted and efficient.

[0062] (2) Real-time monitor the building materials in the key monitored category. By plotting the batch - material defect change curve and fitting the linear regression equation, judge the change trend of the material defect ratio. If the material defect ratio decreases, judge whether the defined material ratio decreases synchronously, which can intuitively reflect the improvement of the overall quality of building materials and the reduction of potential quality risks. By analyzing the change trends of the material defect ratio and the defined material ratio, it is beneficial to analyze the development trend of building material quality problems from multiple dimensions and improve the efficiency of building material quality detection. Description of the Drawings

[0063] The present invention will be further described below in conjunction with the accompanying drawings.

[0064] Figure 1 It is a block diagram of a building material quality detection system of the present invention;

[0065] Figure 2 It is a flowchart of a material analysis model provided in the first embodiment of the present invention. Specific embodiments

[0066] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0067] The first embodiment

[0068] Please refer to Figure 1 As shown, the present invention is a building material quality detection system, including the following modules:

[0069] Material classification module: According to the different usage purposes of building materials, the building materials are divided into multiple categories, historical building material data is obtained, numerical analysis is performed on the unqualified building defect materials of each category, a material defect value is obtained, and a material defect set is constructed based on the material defect value;

[0070] Obtain the usage purpose of building materials, and divide the building materials into multiple categories according to the different usage purposes of building materials;

[0071] In some embodiments, according to the different purposes of building materials, the building materials are divided into multiple categories, including: load-bearing materials, space maintenance materials, decorative materials, functional guarantee materials, facility supporting materials;

[0072] Exemplarily, load-bearing materials are used to bear the load of buildings, including: concrete, steel, masonry;

[0073] Space maintenance materials are used to separate and enclose building spaces, including: bricks, blocks, wall panels in, and tiles, coils in roofing materials;

[0074] Decorative materials are used to enhance the internal and external beauty of buildings, including paints, wallpapers, stones, glass;

[0075] Functional guarantee materials are used for waterproofing, moisture-proofing, heat preservation, moisture preservation, sound insulation and sound absorption of buildings, including: waterproof coils, waterproof coatings, polystyrene boards, sound absorption boards;

[0076] Equipment supporting materials, including: steel pipes, plastic pipes, wires, cables, switches;

[0077] From the building materials database, obtain the number of unqualified building materials of each category of building materials in the same batch to get the number of defective materials, obtain the number of building materials with limited usage conditions in the same batch to get the number of limited materials, and obtain the number of each category of building materials in the same batch;

[0078] It should be noted that professional technicians in this field mark the materials with limited usage for building materials that do not meet the performance standards but can be used under limited usage conditions. For example, if the strength of structural steel is lower than the standard requirements, it can be used in non-critical stress-bearing parts;

[0079] Divide the number of defective materials by the number of each category of building materials to get the material defect ratio;

[0080] Divide the number of limited materials by the number of each category of building materials to get the limited material ratio;

[0081] Based on the material defect ratio and the limited material ratio, obtain the material defect value through a weighted formula;

[0082] Calculate the material defect value of each category of building materials, sort the material defect values of each category of building materials in descending order, and construct a material defect set;

[0083] Category determination module: Based on the material defect set, use the principal component analysis method to analyze and discriminate different categories of building materials in the material defect set to determine the key detection categories in different building material categories;

[0084] Based on the material defect set, obtain the number of limited materials actually used for the building materials in the same batch from the building materials database to get the limited usage value;

[0085] Obtain the limited usage value of each category of building materials, divide the limited usage value by the number of limited materials to get the limited usage ratio, and mark the limited usage ratio of each category of building materials as Xd i , where i is the number of each category of building material category in the material defect set, and the value range of i is [1, m];

[0086] Obtain the limited usage ratio of all batches from the building materials database, and mark the limited usage ratio of each type in each batch as Xd ij , where j is the batch number, and the value range of j is [1, n];

[0087] Using the principal component analysis method based on the limited dosage matrix, determine the key detection categories among all building material categories in the set of material defects.

[0088] Specifically, as Figure 2 shown, the construction method of the material analysis model is as follows:

[0089] Through the standardization formula: Standardize the limited dosage matrix, where Xd ij ′ is the standardized limited dosage ratio of each type in each batch, are the mean and variance of the limited dosage ratio of each type in the same batch respectively;

[0090] Based on the standardized limited dosage matrix, construct the standard limited dosage matrix Xb;

[0091] Based on the standard limited dosage matrix Xb, through the formula: Obtain the covariance value of the limited dosage ratio of each type in each batch of the standard limited dosage matrix Xb, n represents the total number of batches participating in the calculation, Xd′ ki 、Xd′ kj represent the limited dosage ratio of the i-th type in the k-th batch and the limited dosage ratio of the j-th type in the k-th batch of the standard limited dosage matrix Xb respectively;

[0092] Based on the covariance value Cov(Xd i ′ j ) of the limited dosage ratio of each type in each batch of the standard limited dosage matrix Xb, construct the covariance matrix Cov(Xb);

[0093] Perform eigenvalue decomposition on the covariance matrix Cov(Xb) to obtain the eigenvectors and corresponding eigenvalues of the covariance matrix;

[0094] Sum up all the eigenvalues to obtain the total eigenvalue, and perform a ratio process on each eigenvalue and the total eigenvalue to obtain the variance contribution rate, and mark the variance contribution rate as υ q , q is the number of the eigenvalue;

[0095] Through the formula: Obtain the cumulative variance contribution rate, where k is the number of principal components selected, the value range of k is [1,m], V k represents the sum of the variance contribution rates of the first k principal components;

[0096] Based on the eigenvectors corresponding to the k principal components, construct the principal component loading matrix P;

[0097] Multiply the principal component loading matrix by the standard limited dosage matrix to obtain the score matrix F of the k principal components iq, the matrix size is n*k;

[0098] It should be noted that the size of the standard limited dosage matrix is m*n, and the size of the score matrix F is a matrix of m*k;

[0099] Through the formula: Obtain the comprehensive score S of each building material i , where i is the number of the building material category, and α is the weight coefficient, which is determined by the ranking of each type of building material in the material defect set;

[0100] Based on the comprehensive score S of each building material in the material analysis model i , select the building material category with the lowest comprehensive score as the key detection category;

[0101] It should be noted that the lower the comprehensive score of the building material category, the fewer the actual number of limited materials used, indicating that the number of materials that can be used by the building material under limited conditions is small, which means that most of this kind of material does not meet the standard requirements under normal circumstances, and the potential unqualified rate is relatively high, and key monitoring is required to ensure the project quality.

[0102] The technical solution of this embodiment is: according to different usage purposes of building materials, the building materials are divided into multiple categories, historical building material data is obtained, numerical analysis is performed on the unqualified building defect materials of each category to obtain material defect values, based on the material defect values, a material defect set is constructed, and based on the material defect set, the principal component analysis method is used to analyze and discriminate different categories of building materials in the material defect set to determine the key detection categories in different building material categories, which is beneficial to concentrating resources on key monitoring of materials with high potential quality risks and improving the detection efficiency.

[0103] Embodiment Two

[0104] As Figure 1 shown, a building material quality detection system further includes the following modules:

[0105] Key detection module: Based on the key monitoring category, it is used to collect the material defect ratio of different batches of building materials in real time and judge whether the material defect ratio gradually decreases with the iteration of batches;

[0106] Based on the building materials in the key monitoring category, collect the material defect ratio of different batches of building materials in real time;

[0107] In some embodiments, a material detection device is used to mark the defective building materials, and through numerical analysis of the defective building materials, the material defect ratio is obtained;

[0108] Taking the batch as the X-axis and the material defect ratio as the Y-axis, plot the change curve of batch - material defect in a two-dimensional rectangular coordinate system;

[0109] Use the linear regression equation: Fit the change curve of batch - material defect;

[0110] where a 1 and b 1 are the slope and intercept of the linear regression equation respectively, is the fitted value of the linear regression equation;

[0111] Through the formula: Obtain the mean absolute error MAE of the linear regression equation, where h is the number of each coordinate point of the linear regression equation and z is the total number of coordinate points,

[0112] Through the formula: Obtain the mean square error MSE of the linear regression equation;

[0113] Perform weighted summation processing on the mean absolute error and the mean square error to obtain the fitting error value of the linear regression equation;

[0114] Compare the fitting error value of the linear regression equation with the preset fitting error threshold to determine whether the fitting effect of the linear regression equation on the change curve of batch - material defect meets the expectation;

[0115] If the fitting error value of the linear regression equation is lower than the preset fitting error threshold, it indicates that the fitting error value of the linear regression equation is within the expected range, and the fitting effect of the linear regression equation on the change curve of batch - material defect meets the expectation;

[0116] If the fitting error value of the linear regression equation is higher than the preset fitting error threshold, it indicates that the fitting error value of the linear regression equation is not within the expected range, and the fitting effect of the linear regression equation on the change curve of batch - material defect does not meet the expectation;

[0117] If the fitting effect of the linear regression equation on the change curve of batch - material defect meets the expectation, determine whether the material defect ratio gradually increases as the batch iterates;

[0118] Specifically, the method for determining whether the material defect ratio gradually increases as the batch iterates is as follows:

[0119] Obtain the slope a 1 of the linear regression equation. If the slope a 1 is positive, that is, whether the material defect ratio gradually increases as the batch iterates, send a material defect warning to the system;

[0120] If the slope a 1 is negative, that is, whether the material defect ratio of building materials gradually decreases as the batches iterate;

[0121] It should be noted that when the fitting effect of the linear regression equation meets the expectation and the slope is positive, it indicates that the material defect ratio is gradually increasing. As an early warning signal for the quality of building materials, it reminds quality managers to pay timely attention to the building materials in this key monitoring category and take corresponding measures for improvement to prevent the problem from deteriorating further, thus avoiding potential safety hazards to the construction project caused by material quality problems;

[0122] Synchronization analysis module: used to obtain the defined material ratio of the key monitoring category if the material defect ratio gradually decreases as the batches iterate, and determine whether the material defect ratio and the defined material ratio decrease synchronously;

[0123] If the material defect ratio of the building materials in the key monitoring category gradually decreases as the batches iterate, obtain the defined material ratio of each batch in the building materials of the key monitoring category, and mark the defined material ratio of each batch as d j , where j is the batch number, and the value range of j is, construct the defined material sequence,

[0124] Obtain the material defect ratio of each batch, and mark the material defect ratio of each batch as c j , construct the material defect sequence;

[0125] Through the formula: Obtain the average change rate Rx between adjacent batches of the defined material sequence, where t j The manufacturing time of each batch of building materials;

[0126] Through the formula: Obtain the average change rate Ry of adjacent batches of the material defect sequence;

[0127] Through the formula: Obtain the ratio of the average change rate Rx between adjacent batches of the defined material sequence and the average change rate Ry between adjacent batches of the material defect sequence, that is, the sequence change ratio R;

[0128] Compare the sequence change ratio with the preset change ratio range to determine whether the defined material ratio decreases synchronously while the material defect ratio in the material defect sequence gradually decreases;

[0129] If the sequence change ratio is within the preset change ratio range, it is considered that the defined material ratio decreases synchronously while the material defect ratio in the material defect sequence gradually decreases;

[0130] If the sequence change ratio is not within the preset change ratio range, it is considered that while the material defect ratio in the material defect sequence is gradually decreasing, the change trend of the defined material ratio is not synchronous;

[0131] It should be noted that the sequence change ratio can intuitively reflect the change trends of two variables within a certain range, making the changes in the material defect ratios and the defined material ratios of different batches comparable;

[0132] Exemplarily, if the material defect ratio decreases from 5% to 3% between adjacent batches, the average change rate is -2%; the defined material ratio decreases from 4% to 2%, and the average change rate is -2%. This indicates that between these two batches, both the material defect ratio and the defined material ratio are decreasing at the same rate, quantitatively reflecting the degree of synchronous decrease of the material defect ratio and the defined material ratio, which is convenient for comparison and analysis;

[0133] It should be noted that a decrease in the material defect ratio means that during production or use, the proportion of building materials with quality problems decreases, that is, more building materials meet the quality standards for normal use. A decrease in the defined material ratio means that the proportion of the quantity of building materials that originally did not meet the performance standards but could be used under defined conditions also decreases, indicating that the building materials are generally closer to meeting the standard requirements, and the potential quality risks are further reduced;

[0134] The technical solution of this embodiment is: based on the key monitoring category, used to collect the material defect ratios of different batches of building materials in real time, determine whether the material defect ratio gradually decreases as the batches iterate, and whether the material defect ratio of the key monitoring category of building materials gradually decreases. If the material defect ratio gradually decreases as the batches iterate for the key monitoring category of building materials, obtain the defined material ratio of the key monitoring category, determine whether the material defect ratio and the defined material ratio decrease synchronously, and quantify the data on the quality change of building materials from two dimensions of the material defect ratio and the defined material ratio, making the evaluation dimension of building material quality detection more comprehensive.

[0135] The above has described a detailed description of an embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the application of the present invention should still fall within the scope covered by the present invention.

Claims

1. A building material quality detection system, characterized in that: Includes the following modules: Material classification module: used to classify building materials into multiple categories, perform numerical analysis on unqualified building defective materials of each category, obtain material defect ratio and limited material ratio, and construct a material defect set; Category determination module: used to analyze and distinguish different categories of building materials in the material defect set using the principal component analysis method, and determine the key detection categories in different building material categories; Key detection module: used to collect the material defect ratio of different batches of building materials in the key detection category in real time, and determine whether the material defect ratio gradually decreases with the iteration of batches; Synchronous analysis module: used to obtain the limited material ratio of the key monitoring category if the material defect ratio gradually decreases with the iteration of batches, and determine whether the material defect ratio and the limited material ratio decrease synchronously.

2. A building material quality detection system according to claim 1, characterized in that: The material defect set is constructed as follows: Obtain the usage of building materials, and classify the building materials into multiple categories according to their different usages; From the building material database, obtain the number of unqualified building materials of each category of building materials in the same batch to obtain the number of defective materials; Obtain the quantity of building materials with limited use conditions in the same batch, obtain the number of limited materials, and obtain the quantity of each type of building materials in the same batch; The number of defective materials is compared with the number of each type of building materials to obtain the material defect ratio; The number of limited materials is processed by ratio with the number of each type of building materials to obtain the limited material ratio; Based on the material defect ratio and the limited material ratio, the material defect value is obtained through a weighted formula; The material defect value of each category of building materials is calculated, and the material defect values ​​of each category of building materials are sorted in order from large to small to construct a material defect set.

3. A building material quality detection system according to claim 2, characterized in that: The key testing categories in the different building material categories are determined as follows: Based on the material defect set, the quantity of limited materials actually used in the same batch of building materials is obtained from the building material database to obtain the limited usage value; Based on the limited usage values, a limited usage matrix was constructed, and the principal component analysis method was used to obtain the comprehensive score of each building material; Based on the comprehensive score of each building material, the building material categories corresponding to the material defect set are sorted, and the key inspection categories among all building material categories in the material defect set are determined.

4. A building material quality detection system according to claim 3, characterized in that: The limited usage matrix is ​​constructed as follows: Obtain the limited usage value of each type of building material, and perform ratio processing on the limited usage value and the limited material number to obtain the limited usage ratio; The limited usage ratio of each type of building material is marked as Xd i , where i is the number of each type of building material category in the material defect set, and the value range of i is [1,m]; From the building materials database, obtain the limited usage ratios of all batches, and mark the limited usage ratios of each type of each batch as Xd ij , where j is the batch number and the value range of j is [1,n].

5. A building material quality detection system according to claim 3, characterized in that: The principal component analysis method is: Standardize the limited usage matrix and construct a standard limited usage matrix; Based on the standard limited dosage matrix, the covariance matrix is ​​constructed through the covariance formula; Perform eigendecomposition on the covariance matrix to obtain the eigenvectors and corresponding eigenvalues ​​of the covariance matrix; All eigenvalues ​​are summed to obtain the total eigenvalue, and each eigenvalue is ratioed to the total eigenvalue to obtain the variance contribution rate; The cumulative variance contribution rate of the preset k principal components is obtained through the summation formula, and the principal component loading matrix P is constructed based on the eigenvectors corresponding to the k principal components; The main component loading matrix is ​​multiplied by the standard limit usage matrix to obtain the score matrix of k main components; Based on the score matrix of k principal components, the comprehensive score of each building material was obtained through numerical analysis.

6. A building material quality detection system according to claim 5, characterized in that: The numerical analysis method for the comprehensive score of each building material is as follows: Based on the ranking of each building material in the material defect set, a weight coefficient is obtained; Based on the weight coefficient, cumulative variance contribution rate, and principal component score matrix of each building material, the comprehensive score of each building material is obtained through the product formula.

7. A building material quality detection system according to claim 1, characterized in that: The method for judging whether the material defect ratio is gradually reduced with the iteration of batches is as follows: Based on key monitoring categories of building materials, real-time collection of material defect ratios of different batches of building materials; With batch as X-axis and material defect ratio as Y-axis, draw the batch-material defect change curve in a two-dimensional rectangular coordinate system; Fitting analysis is performed on the material defect ratios of different batches of building materials to obtain the fitting error value of the linear regression equation; If the fitting error value of the linear regression equation is higher than the preset fitting error threshold, it indicates that the fitting effect of the linear regression equation on the batch-material defect variation curve is in line with expectations; If the fitting effect of the linear regression equation on the batch-material defect variation curve meets expectations, obtain the slope of the linear regression equation; If the slope of the linear regression equation is negative, it means that the material defect ratio of the building materials gradually decreases with the iteration of batches.

8. A building material quality detection system according to claim 7, characterized in that: The method for fitting analysis of the material defect ratios of different batches of building materials is as follows: Use linear regression equation to fit the batch-material defect variation curve; Obtain the mean absolute error and mean square error of the linear regression equation through the mean absolute error and mean square error formulas; The mean absolute error and the mean square error are weighted and summed to obtain the fitting error value of the linear regression equation.

9. A building material quality detection system according to claim 1, characterized in that: The method for judging whether the material defect ratio and the limited material ratio are reduced synchronously is as follows: Obtain the restricted material ratio of each batch of key monitoring category building materials and construct a restricted material sequence; Obtain the material defect ratio of each batch and construct a material defect sequence; The average change rate of the limited material sequence and material defect sequence is analyzed to obtain the sequence change ratio; If the sequence variation ratio is within the preset variation ratio range, it is considered that the material defect ratio in the material defect sequence is gradually decreasing while the limited material ratio is synchronously decreasing.

10. A building material quality detection system according to claim 9, characterized in that: The sequence change ratio is obtained as follows: The average change rate Rx between adjacent batches of the defined material sequence and the average change rate Ry between adjacent batches of the material defect sequence are obtained through the average change rate formula; Through the ratio formula, the ratio of the average change rate Rx between adjacent batches of the defined material sequence to the average change rate Ry between adjacent batches of the material defect sequence, that is, the sequence change ratio, is obtained.

Images