Purple soil red wall construction method for crack wall
The identification of cracked walls through the Grassmann manifold distance method and the LDA method solves the problem that the existing technology cannot identify and reinforce cracked walls, and achieves the improvement of the stability and aesthetics of the walls.
Patent Information
- Application Number
- CN202411817764.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-11
- Publication Date
- 2025-03-18
AI Technical Summary
The prior art cannot effectively identify and reinforce cracked walls, affecting the stability and aesthetics of the building.
The Grassmann manifold distance method is used in combination with the LDA method to identify wall cracks through data collection and processing, and construction and repairs are carried out based on the identification results.
It realizes accurate identification and repair of wall cracks, improves the stability and aesthetics of the wall, and extends the service life of the building.
Smart Images
Figure CN119641120A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of wall repair, and in particular to a purple clay red wall construction method for cracked walls. Background Art
[0002] Purple clay red wall is a common purple clay red wall finish for the exterior walls and courtyard walls of Confucian temples and temple buildings in southern my country. Its stability and durability issues have always been the focus of attention in the field of building repair and protection. Common problems with purple clay red walls include different material ratios and construction methods. The purple clay red walls are white or purple, with different colors, fading, rough surface, and different textures. These defects not only affect the beauty of the building, but also have a certain impact on the service life and safety of the building. At the same time, cracks often appear on the wall. The cracks in the concrete structure are of different sizes, and the cracks have a very large impact on the structural strength of the wall. If the wall cracks are not accurately identified, the reinforcement construction of the wall cannot be achieved. Therefore, there is an urgent need for a wall construction method that can effectively target cracked walls. Summary of the invention
[0003] The purpose of this application is to provide a method for constructing a purple clay red wall for cracked walls, which is used to at least solve the technical problem that the existing methods cannot identify and reinforce cracked walls. This application uses the Grassmann manifold distance method to accurately identify cracks in the wall and perform wall construction, thereby improving the stability of the wall and the economy of construction.
[0004] The present application provides a purple clay red wall construction method for cracked walls, comprising the following steps:
[0005] Collect concrete wall data and process the data to obtain processed wall crack data;
[0006] The impulse response pole feature space is constructed according to the processed wall crack data, and the feature space value is obtained;
[0007] The spectral clustering method based on Grassmann manifold distance metric is used to cluster the feature space values and identify the cracks in the concrete wall.
[0008] According to the identified cracks in the concrete wall, the concrete wall is repaired.
[0009] Furthermore, the process of collecting concrete wall data and processing the data to obtain processed wall crack data is as follows:
[0010] Transmitting high-frequency electromagnetic waves into the concrete wall space to obtain a concrete wall echo signal and obtain an echo signal data sample;
[0011] The LDA method is used to perform data dimension reduction processing on the echo signal data samples to obtain the echo signal after dimension reduction.
[0012] Furthermore, the process of using the LDA method to perform data dimensionality reduction processing on the echo signal data sample to obtain the echo signal after dimensionality reduction is:
[0013] Calculating the center of the echo signal data sample, and calculating the mean of the echo signal data sample according to the center;
[0014] A scattering matrix is calculated based on the center, and an intra-class scattering matrix is calculated based on the scattering matrix;
[0015] An inter-class scattering matrix is calculated based on the center and the mean, and an augmented matrix is calculated based on the intra-class scattering matrix and the inter-class scattering matrix;
[0016] Performing eigendecomposition on the augmented matrix to obtain eigenvalues and their corresponding eigenvectors, and sorting the eigenvectors to obtain a transformation matrix;
[0017] A dimension reduction data matrix is calculated based on the echo signal data samples and the transformation matrix to obtain low-dimensional data.
[0018] Furthermore, the process of constructing the impulse response pole feature space based on the processed wall crack data and obtaining the feature space value is as follows:
[0019] Calculating the impulse response of the concrete wall according to the processed wall crack data;
[0020] Constructing a corresponding relationship between the impulse response, the attenuation factor and the resonant frequency according to the impulse response, and calculating the attenuation factor and the resonant frequency values;
[0021] The pole feature space of the concrete echo signal is constructed according to the attenuation factor and the resonant frequency, and the feature space value is obtained.
[0022] Furthermore, the spectral clustering method based on Grassmann manifold distance metric performs cluster analysis on feature space values, and the process of identifying concrete wall cracks is as follows:
[0023] Obtain the extreme feature data points of the concrete wall echo signal and obtain the number of clusters to be clustered;
[0024] Calculating the distance between the data points based on the Grassmann manifold method and constructing a similarity matrix;
[0025] A Laplace matrix is constructed according to the similarity matrix, and the maximum eigenvalue and the corresponding eigenvector of the Laplace matrix are solved, and a first matrix is constructed according to the eigenvector;
[0026] The row vectors of the construction matrix are normalized to obtain a second matrix, and each row of the second matrix is classified using a K-means method, and a concrete wall crack recognition result is obtained according to the classification result output.
[0027] Furthermore, the process of repairing the concrete wall according to the identified concrete wall cracks is as follows:
[0028] Mix red iron oxide, water and white latex in a ratio of 0.5:3:1 to obtain a red glue solution;
[0029] The red glue and the white lime of the finishing layer are mixed in a ratio of 1:0.5 to obtain a prepared red-lime glue mortar;
[0030] Use red lime to scrape the concrete wall twice, then use red lime mortar to apply two coats on the wall.
[0031] Furthermore, the two-pass scraping process is: scraping the plaster in a horizontal and vertical manner, and scraping off the excess red plaster in the joint and end process;
[0032] The pitting spots, pits, scratches and identified wall cracks on the wall should be re-scraped and leveled. After they are completely dry, they should be polished with fine sandpaper and the dust should be swept away until the surface is flat and smooth.
[0033] Furthermore, the finishing layer of white lime includes tempered putty or lime.
[0034] Furthermore, the preparation process of the red glue solution is: measure the water in the container according to the proportion, first slowly add the proportional amount of red iron oxide into the water with a spoon, stir while adding, until the addition is completed and the red iron oxide is completely dissolved in the water, and then add the white latex solution and stir it evenly to obtain the red glue solution.
[0035] Compared with the prior art, this application has achieved the following beneficial effects:
[0036] The present application discloses a purple clay red wall construction method for cracked walls. The cracks in the wall can be accurately identified by the Grassmann manifold distance method in combination with the LDA method, and the specific position and shape of the cracks in the wall can be obtained. Meanwhile, white latex is added to the purple clay red wall finishing layer (red glue) to enhance the adhesion, firmness and non-fading property of the red glue, and the red glue has certain waterproof properties. The red glue is added to the finishing gray tempered putty or lime to make the finishing gray have purple clay red color. The effect is better after applying the red glue, and the aesthetics of the purple clay red wall is significantly improved, the color difference error is small, the color does not fade, does not turn white or purple, and the wall is safe and durable. BRIEF DESCRIPTION OF THE DRAWINGS
[0037] Figure 1 A schematic diagram of a purple clay red wall construction method for cracked walls according to the present invention;
[0038] Figure 2 This is a schematic diagram of the purple clay red wall repair process of the present invention;
[0039] Figure 3 This is a schematic diagram of a purple soil red wall test sample of the present invention;
[0040] Figure 4 It is a schematic diagram of the finished product of purple clay red wall of the present invention. DETAILED DESCRIPTION
[0041] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0042] Embodiment 1
[0043] Please refer to Figure 1 As shown, this embodiment provides a purple soil red wall construction method for a cracked wall, comprising the following steps:
[0044] S101, collecting concrete wall data, and performing data processing to obtain processed wall crack data.
[0045] In this embodiment, in order to identify wall cracks, high-frequency electromagnetic waves are emitted into the concrete wall space of the building to obtain a concrete wall echo signal.
[0046] After obtaining the echo signal, the signal is subjected to dimensionality reduction processing. The specific process is as follows:
[0047] Input the initialization echo signal data sample of the concrete wall, the specific sample is X = {x1, x2, x3, Λ, x n};
[0048] This embodiment uses the LDA (Linear Discriminant Analysis) method to preprocess the data to achieve data compression processing. LDA, i.e., linear discriminant analysis, is a machine learning algorithm, specifically:
[0049] Assume that the number of inherent clusters of data X is K, and the center mk (k = 1, 2, ..., K) of a cluster Ck (k = 1, 2, ..., K) is specifically:
[0050]
[0051] Among them, n k is the number of samples in cluster Ck, x i Echo signal data.
[0052] Then the total mean m of all data samples in each cluster is calculated as:
[0053]
[0054] Among them, n is the total number of samples in the sample data.
[0055] For cluster Ck, the scattering matrix is calculated as:
[0056]
[0057] Where T is the transposed matrix.
[0058] The scattering matrix is processed to obtain the intra-class scattering matrix:
[0059]
[0060] Among them, the scattering matrix S k and S w They are all multidimensional matrices.
[0061] For the actual observed data, the inter-class scattering matrix is:
[0062]
[0063] The augmented matrix is combined with the intra-class scattering matrix and the inter-class scattering matrix. Specifically:
[0064]
[0065] Among them, G is the augmented matrix.
[0066] By decomposing the matrix, we get the decomposed eigenvalues and corresponding eigenvectors, and then sort them according to the size of the eigenvalues, and then construct the transformation matrix W, which is:
[0067] W=[e1,e2,Λ,e l ] (7)
[0068] Among them, e1 is the eigenvector corresponding to the largest eigenvalue, e2 is the eigenvector corresponding to the second largest eigenvalue, and e l is the eigenvector corresponding to the smallest eigenvalue of , and l is the number of eigenvectors.
[0069] Finally, the product of the observation data matrix X and the transformation matrix W is calculated to obtain the data dimension reduction result, which is:
[0070] X′=XW (8)
[0071] Among them, X′ is the result of data dimensionality reduction.
[0072] Through the above method, the original data is transformed into low-dimensional data, the data dimension reduction result is achieved, and the data set after dimension reduction is obtained.
[0073] S102, constructing an impulse response pole feature space according to the processed wall crack data, and obtaining a feature space value.
[0074] The impulse response p(t) of the concrete wall is obtained by calculation, which is:
[0075]
[0076] Where p′(t) is the early response of the impulse response, is the late response, t0 is the start time, R α is the complex amplitude of the αth resonance state, v is the velocity, t is the time, q α is the extreme point, specifically:
[0077] q α =h α +θ2πf α (10)
[0078] Among them, h α is the attenuation factor, θ is a constant, f α is the resonant frequency.
[0079] Finally, the pole feature space of the concrete echo signal is established, which is specifically expressed as: [H, F] = {(h1, f1), (h2, f2), Λ(h N ,f N )}.
[0080] By arranging N poles of the resonance state of the concrete wall echo signal in the two-dimensional pole feature space [H, F], the characteristic distribution of the concrete wall impulse response can be obtained, thereby completing the wall crack identification.
[0081] S103. A spectral clustering method based on Grassmann manifold distance metric is used to cluster feature space values and identify concrete wall cracks.
[0082] First, by inputting N concrete echo signal pole data points [H, F] = {(h1, f1), (h2, f2), Λ(h N ,f N )}, the number of clusters is k;
[0083] According to the Grassmann flow shape distance between two extreme feature data points, it is specifically:
[0084]
[0085] Among them, r, c are the extreme point data of Grassmann manifold, θ i is the principal angle between r and c.
[0086] Then, the Laplace matrix is established, specifically:
[0087] L=D -1 / 2 SD -1 / 2 (12)
[0088] Where D is a diagonal matrix, S is the similarity matrix, S∈R n×n .
[0089] Then, the k largest concrete wall echo signal pole eigenvalues of the L matrix are solved to obtain the corresponding echo signal pole eigenvectors (h1, f1), (h2, f2), K, (h k ,f k ), thus establishing the matrix [H, F] = {(h1, f1), (h2, f2), Λ (h k ,f k )}, where (h k ,f k ) is a column vector.
[0090] Then, the row vectors of [H,F] are normalized to obtain the matrix Y, where
[0091]
[0092] Each row of Y is regarded as a point in space, and the K-means method is used to classify it. If the i-th row of Y belongs to the j-th class, the concrete wall echo signal pole feature data point (h0, f0) is classified into the j-th class output data point, and the concrete wall crack recognition results (h1, f1), (h2, f2), K, (h k ,f k ).
[0093] S104: Repair the concrete wall according to the identified cracks in the concrete wall.
[0094] Specific as Figure 2 As shown, in this embodiment, red iron oxide, water and white latex are used to repair the concrete wall to obtain a purple soil red wall. The specific process is as follows:
[0095] Materials: Red iron oxide, water-based white latex, water, container, stirring tool
[0096] Red glue ratio test: The red glue is made of red iron oxide: water: white latex and other raw materials in 3 ratios of 0.4:2.5:0.8, 0.5:3:1, 0.6:3.5:1.2 for ratio test. After trial mixing, red iron oxide: water: white latex in 0.5:3:1 ratio is adopted, which has pure color, is not easy to fade after drying, has good adhesion and durability.
[0097] Preparation of red glue: Measure the water in the container according to the proportion, first slowly add the proportional amount of red iron oxide into the water with a spoon, stirring while adding, until the addition is complete and the red iron oxide is completely dissolved in the water, then add the white latex and stir thoroughly.
[0098] Preparation of red-lime mortar: The base mortar is tempered putty or lime (lime needs to be fully fermented, filtered to remove coarse particles, and paper towels are added): red glue = 1:0.5. Stir well and the prepared red-lime mortar must be used before the initial setting.
[0099] Scrape the first coat of red ash: Scrape horizontally and vertically during the red ash scraping process, and pay attention to scraping the red ash cleanly when joining and finishing.
[0100] Scrape the second coat of red lime: After scraping the first coat of red lime, use red lime to re-scrape the pockmarks, potholes, scratches, etc. on the wall. After it is completely dry, polish it with fine sandpaper and sweep away the dust until the surface is flat and smooth to meet the acceptance level.
[0101] Apply red-gray glue twice on the surface: Because the red-gray color will turn white after full scraping, it is necessary to apply red-gray glue. Stir it evenly before use. Do not add water at will. Its working consistency or viscosity should ensure that it does not drip or show brush marks during construction. And it should be used up within the specified time. The direction and length of the stroke should be consistent when applying. If the paint dries quickly, you should frequently apply a short brush and make the joints straight. The prepared purple clay red wall is like Figure 3 and 4 As shown, Figure 3 This is a schematic diagram of the purple clay red wall test sample. Figure 4 This is a schematic diagram of the finished product of purple clay and red wall.
[0102] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily mention changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims, and the above embodiments should be regarded as exemplary and non-restrictive.
Claims
1. A method for constructing a purple clay red wall for a cracked wall, characterized in that: The following steps are involved: High-frequency electromagnetic waves are emitted into the concrete wall space to collect concrete wall data, and data processing is performed to obtain processed wall crack data; Constructing an impulse response pole feature space according to the processed wall crack data, and obtaining a feature space value; A spectral clustering method based on Grassmann manifold distance metric is used to cluster the feature space values and identify the cracks in the concrete wall; According to the identified cracks in the concrete wall, the purple clay red wall construction method is used to repair the concrete wall.
2. The purple clay red wall construction method according to claim 1, characterized in that: The process of collecting concrete wall data and processing the data to obtain processed wall crack data is as follows: Transmitting high-frequency electromagnetic waves into the concrete wall space to obtain a concrete wall echo signal and obtain an echo signal data sample; The LDA method is used to perform data dimension reduction processing on the echo signal data samples to obtain the echo signal after dimension reduction.
3. The purple clay red wall construction method according to claim 2, characterized in that: The process of using the LDA method to perform data dimensionality reduction processing on the echo signal data sample to obtain the echo signal after dimensionality reduction is as follows: Calculating the center of the echo signal data sample, and calculating the mean of the echo signal data sample according to the center; A scattering matrix is calculated based on the center, and an intra-class scattering matrix is calculated based on the scattering matrix; An inter-class scattering matrix is calculated based on the center and the mean, and an augmented matrix is calculated based on the intra-class scattering matrix and the inter-class scattering matrix; Performing eigendecomposition on the augmented matrix to obtain eigenvalues and their corresponding eigenvectors, and sorting the eigenvectors to obtain a transformation matrix; A dimension reduction data matrix is calculated based on the echo signal data samples and the transformation matrix to obtain low-dimensional data.
4. The purple clay red wall construction method according to claim 1, characterized in that: The process of constructing the impulse response pole feature space based on the processed wall crack data and obtaining the feature space value is as follows: Calculating the impulse response of the concrete wall according to the processed wall crack data; Constructing a corresponding relationship between the impulse response, the attenuation factor and the resonant frequency according to the impulse response, and calculating the attenuation factor and the resonant frequency values; The concrete echo signal pole feature space is constructed according to the attenuation factor and the resonance frequency. Get the feature space value.
5. The purple clay red wall construction method according to claim 1, characterized in that: The spectral clustering method based on Grassmann manifold distance metric performs cluster analysis on feature space values and identifies concrete wall cracks in the following process: Obtain the extreme feature data points of the concrete wall echo signal and obtain the number of clusters to be clustered; Calculating the distance between the data points based on the Grassmann manifold method and constructing a similarity matrix; A Laplace matrix is constructed according to the similarity matrix, and the maximum eigenvalue and the corresponding eigenvector of the Laplace matrix are solved, and a first matrix is constructed according to the eigenvector; The row vectors of the construction matrix are normalized to obtain a second matrix, and each row of the second matrix is classified using a K-means method, and a concrete wall crack recognition result is obtained according to the classification result output.
6. The purple clay red wall construction method according to claim 1, characterized in that: The process of repairing the concrete wall according to the identified concrete wall cracks is as follows: Mix red iron oxide, water and white latex in a ratio of 0.5:3:1 to obtain a red glue solution; The red glue and the white lime of the finishing layer are mixed in a ratio of 1:0.5 to obtain a prepared red-lime glue mortar; Use red lime to scrape the concrete wall twice, then use red lime mortar to apply two coats on the wall.
7. The purple clay red wall construction method according to claim 6, characterized in that: The two-pass scraping process is: scraping the plaster horizontally and vertically, and scraping off the excess red plaster in the process of jointing and finishing; The pitting spots, pits, scratches and identified wall cracks on the wall should be re-scraped and leveled. After they are completely dry, they should be polished with fine sandpaper and the dust should be swept away until the surface is flat and smooth.
8. The purple clay red wall construction method according to claim 6, characterized in that: The white ash of the finishing layer includes tempered putty or lime.
9. The purple clay red wall construction method according to claim 6, characterized in that: The preparation process of the red glue solution is as follows: the water in the container is measured in proportion, and the proportional amount of red iron oxide is slowly added into the water with a spoon, stirring while adding until the addition is completed and the red iron oxide is completely dissolved in the water, and then the white latex is added and stirred evenly to obtain the red glue solution.