A method for identifying the spatiotemporal evolution of concrete damage and related equipment

Abstract

The present application relates to the technical field of concrete damage identification and provides a method for identifying the spatiotemporal evolution of concrete damage and related equipment. The method comprises: performing damage mapping based on all damage information to obtain an initial damage label for each sub-region in each historical time period; utilizing a support vector machine to quantitatively separate the damage information based on the initial damage label to obtain a final damage label; performing spatial damage analysis based on all damage information in each historical time period to obtain a damage grayscale matrix for each sub-region in each historical time period; constructing a three-dimensional damage model of the target concrete region in each historical time period based on the damage grayscale matrix and final damage label of all sub-regions in each historical time period, and serializing all three-dimensional damage models to obtain a spatiotemporal evolution damage model for the target concrete region. The method of the present application can improve the accuracy of identifying the spatiotemporal evolution of concrete damage.

Description

Technical Field

[0001] The present application relates to the technical field of concrete damage identification, and in particular to a method for identifying the spatiotemporal evolution of concrete damage and related equipment. Background Art

[0002] The core difficulty in intelligent monitoring of concrete structures lies in accurately capturing the multi-scale spatiotemporal evolution of damage. Initial damage inherent in concrete, such as aggregate interface defects and shrinkage cracks, forms a microcrack network under the coupling of service loads and the environment. Its three-dimensional dynamic expansion directly leads to material strength degradation and reduced durability. Existing monitoring methods are limited by their inadequate ability to resolve spatiotemporal dimensions, particularly the lack of online classification capabilities for crack propagation modes (tensile / shear). This makes it difficult to dynamically track the entire process from microcrack initiation and stable expansion to macrocracking, resulting in significant lags in early warning of structural performance degradation.

[0003] Current mainstream damage identification technologies, such as acoustic emission (AE) testing, suffer from a technical bottleneck in their ability to analyze multidimensional signals. Traditional AE data analysis relies on manual feature extraction and static parameter statistics (such as ring counts and energy accumulation), lacking the ability to intelligently integrate and analyze multidimensional signal features across time and space. Furthermore, existing detection methods exhibit significant deficiencies in spatial positioning accuracy and dynamic evolution characterization. This results in poor accuracy in identifying the spatiotemporal evolution of concrete damage. Summary of the Invention

[0004] The present application provides a method for identifying the spatiotemporal evolution of concrete damage and related equipment, which can solve the problem of poor accuracy in identifying the spatiotemporal evolution of concrete damage.

[0005] In a first aspect, the present application provides a method for identifying the spatiotemporal evolution of concrete damage, the method comprising:

[0006] Obtain damage information of each sub-area of ​​the target concrete area in multiple historical time periods;

[0007] Damage mapping is performed based on all damage information to obtain the initial damage label of each sub-region in each historical time period; the initial damage label is used to classify all damage information;

[0008] Using support vector machines, all damage information is quantitatively separated based on all initial damage labels to obtain the final damage label for each sub-region in each historical time period. The final damage label is used to describe the damage category of the sub-region in the historical time period.

[0009] For each historical time period, spatial damage analysis is performed based on all damage information in the historical time period to obtain the damage grayscale matrix of each sub-region in the historical time period. The damage grayscale matrix is ​​used to describe the damage degree of the sub-region.

[0010] Based on the damage grayscale matrix and final damage labels of all sub-regions in each historical time period, a three-dimensional damage model of the target concrete region in each historical time period is constructed. All three-dimensional damage models are serialized to obtain a spatiotemporal evolution damage model of the target concrete region. The three-dimensional damage model is used to describe the damaged sub-regions and damage categories of the target concrete region in the historical time period, and the spatiotemporal evolution damage model is used to describe the damage evolution process of the target concrete region in all historical time periods.

[0011] Optionally, damage mapping is performed based on all damage information to obtain the initial damage label of each sub-region in each historical time period, including:

[0012] Analyze each damage information separately to obtain the damage value corresponding to the damage information;

[0013] Initialize the weight matrix and standardize each damage value to obtain a standard damage value; multiple elements in the weight matrix correspond to multiple damage values ​​one by one;

[0014] The weight matrix is ​​iteratively updated using all standard damage values ​​to obtain the final weight matrix;

[0015] Cluster all damage values ​​according to all elements in the final weight matrix to obtain multiple clusters;

[0016] Perform secondary clustering on all clusters to obtain multiple target clusters;

[0017] Labels are assigned according to the target clusters corresponding to each damage information, and the initial damage labels of the sub-regions corresponding to each damage information in the corresponding historical time period are obtained.

[0018] Optionally, the weight matrix is ​​iteratively updated using all standard damage values ​​to obtain a final weight matrix, including:

[0019] The number of iterations increases by 1;

[0020] Determine whether the number of iterations reaches the preset number;

[0021] If so, the weight matrix is ​​used as the final weight matrix;

[0022] Otherwise, a standard damage value is randomly selected from all standard damage values ​​as the sample damage value;

[0023] Calculate the similarity between the sample damage value and each element in the weight matrix, and use the element with the largest similarity as the target element;

[0024] For each element in the weight matrix, determine whether the element is equal to the target element. If so, update the element according to the target element to obtain the updated element. Otherwise, use the element as the updated element.

[0025] Integrate all updated elements to obtain the updated weight matrix, and use the updated weight matrix as the weight matrix, returning to the step of the number of iterations plus 1.

[0026] Optionally, perform secondary clustering on all clusters to obtain multiple target clusters, including:

[0027] For each cluster, combine the elements of the final weight matrix corresponding to all damage values ​​in the cluster into a vector to obtain the weight vector of the cluster;

[0028] Calculate the cosine similarity between each two clusters based on all weight vectors;

[0029] All clusters are clustered according to all cosine similarities to obtain multiple target clusters.

[0030] Optionally, the final damage label is shear damage or tensile damage;

[0031] Using support vector machines, all damage information is quantitatively separated based on all initial damage labels to obtain the final damage labels for each sub-region in each historical time period, including:

[0032] Construct a Lagrangian generalized function based on all initial damage labels;

[0033] Solve the parameter solution that should be satisfied by minimizing the Lagrangian functional, and substitute the parameter solution into the Lagrangian functional to obtain the objective function;

[0034] Minimize the objective function to obtain the normal vector and the support vector machine threshold;

[0035] Construct a hyperplane based on the normal vector and the support vector machine threshold;

[0036] All damage information is mapped into the feature space, and all initial damage labels are quantitatively separated using a hyperplane in the feature space to obtain a directional separation result;

[0037] For each damage information, determine whether the damage information is on the first side of the hyperplane in the directional separation result. If so, the tensile damage is used as the final damage label of the sub-region corresponding to the damage information in the corresponding historical time period. Otherwise, the shear damage is used as the final damage label of the sub-region corresponding to the damage information in the corresponding historical time period.

[0038] Optionally, the Lagrangian functional is:

[0039] ;

[0040] in, represents the value of the Lagrange function, represents the normal vector of the hyperplane, represents the threshold of the support vector machine, represents the non-negative relaxation coefficient The non-negative penalty coefficient of and represents the Lagrange multiplier, Indicates the A non-negative penalty coefficient, Indicates the amount of damage information, and All are Lagrange multipliers, Indicates the The initial damage label corresponding to the damage information, Indicates the The damage value of each damage information.

[0041] Optional, parameter solution is:

[0042] ;

[0043] ;

[0044] ;

[0045] The objective function is:

[0046] ;

[0047] in, represents the damage weight matrix, Indicates the Lagrange multipliers, Indicates the The initial damage label corresponding to the damage information, Indicates the The damage value of the damage information, Represents a transpose operation.

[0048] Optionally, the hyperplane is:

[0049] ;

[0050] in, represents a vector on the hyperplane.

[0051] In a second aspect, the present application provides a device for identifying the spatiotemporal evolution of concrete damage, comprising:

[0052] An acquisition module is used to obtain damage information of each sub-area of ​​the target concrete area in multiple historical time periods;

[0053] The damage mapping module is used to perform damage mapping based on all damage information and obtain the initial damage label of each sub-region in each historical time period; the initial damage label is used to classify all damage information;

[0054] The quantitative separation module is used to quantitatively separate all damage information based on all initial damage labels using a support vector machine to obtain the final damage label for each sub-region in each historical time period. The final damage label is used to describe the damage category of the sub-region in the historical time period.

[0055] The damage analysis module is used to perform spatial damage analysis based on all damage information in each historical time period, and obtain the damage grayscale matrix of each sub-region in the historical time period; the damage grayscale matrix is ​​used to describe the damage degree of the sub-region;

[0056] A construction module is used to construct a three-dimensional damage model of the target concrete area in each historical time period based on the damage grayscale matrix and final damage label of all sub-areas in each historical time period, and serialize all three-dimensional damage models to obtain a spatiotemporal evolution damage model of the target concrete area; the three-dimensional damage model is used to describe the damaged sub-areas and damage categories of the target concrete area in the historical time period, and the spatiotemporal evolution damage model is used to describe the damage evolution process of the target concrete area in all historical time periods.

[0057] In a third aspect, an embodiment of the present application provides a terminal device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for identifying the spatiotemporal evolution of concrete damage when executing the above-mentioned computer program.

[0058] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium storing a computer program, which implements the above-mentioned method for identifying the spatiotemporal evolution of concrete damage when executed by a processor.

[0059] The above solution of the present application has the following beneficial effects:

[0060] In some embodiments of the present application, damage information for each sub-region of a target concrete region is obtained over multiple historical time periods. Damage mapping is then performed based on all the damage information to obtain an initial damage label for each sub-region in each historical time period. A support vector machine is then used to quantitatively separate all the damage information based on all the initial damage labels to obtain a final damage label for each sub-region in each historical time period. Spatial damage analysis is then performed for each historical time period based on all the damage information in that historical time period to obtain a damage grayscale matrix for each sub-region in that historical time period. Finally, a three-dimensional damage model of the target concrete region in each historical time period is constructed based on the damage grayscale matrices and final damage labels for all sub-regions in each historical time period. All three-dimensional damage models are then serialized to obtain a spatiotemporal evolution damage model for the target concrete region. Damage mapping can achieve preliminary differentiation of the damage information. Using a support vector machine to quantitatively separate the damage information and obtain the final damage label allows for further detailed classification of the damage information, improving the accuracy of the final damage label. Constructing a spatiotemporal evolution damage model based on the accurate final damage label can effectively improve the accuracy of identifying the spatiotemporal evolution of concrete damage.

[0061] Other beneficial effects of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0063] Figure 1 A flowchart of a method for identifying the spatiotemporal evolution of concrete damage provided in one embodiment of the present application;

[0064] Figure 2 A schematic diagram of the directional separation results provided in one embodiment of the present application;

[0065] Figure 3 A schematic diagram of a three-dimensional numerical model provided in one embodiment of the present application;

[0066] Figure 4 A grayscale image provided in an embodiment of the present application;

[0067] Figure 5 A grayscale image after binarization and visual differentiation provided in an embodiment of the present application;

[0068] Figure 6 A front view of a spatiotemporal evolution damage model provided in one embodiment of the present application;

[0069] Figure 7 A top view of a spatiotemporal evolution damage model provided in one embodiment of the present application;

[0070] Figure 8 A schematic diagram of the structure of a device for identifying the spatiotemporal evolution of concrete damage provided in one embodiment of the present application;

[0071] Figure 9 A schematic diagram of the structure of a terminal device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0072] In the following description, specific details such as specific system structures and techniques are provided for purposes of illustration rather than limitation to facilitate a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obscuring the description of the present application with unnecessary detail.

[0073] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0074] It will also be understood that the term "and / or" used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0075] As used in this specification and the appended claims, the term "if" can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]," depending on the context.

[0076] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0077] References to "one embodiment" or "some embodiments" in this specification mean that a particular feature, structure, or characteristic described in conjunction with that embodiment is included in one or more embodiments of the present application. Thus, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," and "in other embodiments" appearing in various places in this specification do not necessarily refer to the same embodiment, but rather mean "one or more but not all embodiments," unless otherwise specifically emphasized. The terms "including," "comprising," "having," and variations thereof all mean "including but not limited to," unless otherwise specifically emphasized.

[0078] In response to the problem of poor accuracy of the existing spatiotemporal evolution of concrete damage, an embodiment of the present application provides a method for identifying the spatiotemporal evolution of concrete damage. The spatiotemporal evolution identification method obtains damage information of each sub-region of the target concrete region in multiple historical time periods, then performs damage mapping based on all damage information to obtain the initial damage label of each sub-region in each historical time period, and then uses a support vector machine to quantitatively separate all damage information based on all initial damage labels to obtain the final damage label of each sub-region in each historical time period. Then, for each historical time period, spatial damage analysis is performed based on all damage information in the historical time period to obtain a damage grayscale matrix for each sub-region in the historical time period. Finally, based on the damage grayscale matrices and final damage labels of all sub-regions in each historical time period, a three-dimensional damage model of the target concrete region in each historical time period is constructed, and all three-dimensional damage models are serialized to obtain a spatiotemporal evolution damage model of the target concrete region. Among them, damage mapping of damage information can achieve preliminary differentiation of damage information. Using support vector machine to quantitatively separate damage information and obtain final damage labels can further classify damage information in detail and improve the accuracy of final damage labels. Based on accurate final damage labels, a spatiotemporal evolution damage model is constructed, which can effectively improve the recognition accuracy of spatiotemporal evolution of concrete damage.

[0079] Next, the method for identifying the spatiotemporal evolution of concrete damage provided in this application is exemplified.

[0080] like Figure 1 As shown, the method for identifying the spatiotemporal evolution of concrete damage provided by this application includes the following steps:

[0081] Step 11: Obtain damage information of each sub-area of ​​the target concrete area in multiple historical time periods.

[0082] The target concrete region is the concrete block (e.g., the concrete-cast portion of a bridge or building) where the spatiotemporal evolution of damage needs to be identified. The damage information is used to describe the state of subregions within the target concrete region (e.g., acoustic emission signals, internal 3D modeling, and piezoelectric vibration wave signals). Multiple historical time periods are set based on the actual time periods for damage analysis. The subregions can be obtained by spatially partitioning the target concrete region. For example, the target concrete region can be uniformly meshed, with the resulting meshed regions serving as subregions.

[0083] In some embodiments of the present application, damage information may be acquired using devices such as acoustic emission probe arrays, radars, and piezoelectric sensors.

[0084] Step 12: Perform damage mapping based on all damage information to obtain the initial damage label of each sub-region in each historical time period.

[0085] The above initial damage labels are used to classify all damage information, mainly to perform a preliminary classification of the status of all sub-areas in each historical time period based on the damage information.

[0086] In some embodiments of the present application, the step of performing damage mapping based on all damage information to obtain an initial damage label for each sub-region in each historical time period includes:

[0087] In the first step, each damage information is analyzed to obtain the damage value corresponding to the damage information.

[0088] The damage value is used to quantify the damage degree of the sub-region corresponding to the damage information.

[0089] For example, the damage value corresponding to the damage information can be obtained using methods such as acoustic emission calculation formulas, Fourier transforms, and Mel-value mapping. If the damage information is an acoustic emission signal, the AF value and RA value of the acoustic emission signal are calculated based on the rise time, acoustic emission amplitude, duration, and acoustic emission count of the acoustic emission signal. The AF value is the frequency characteristic of the acoustic emission signal, and the RA value reflects the relationship between the amplitude characteristic and the rise time. AF and RA are the damage values ​​of the damage information. The expression is:

[0090] ;

[0091] ;

[0092] In the second step, the weight matrix is ​​initialized and each damage value is normalized to obtain the standard damage value.

[0093] The multiple elements in the weight matrix correspond to the multiple damage values ​​one by one. The elements in the weight matrix are the weights of the damage values.

[0094] It should be noted that all weights in the weight matrix are divided into multiple weight components, each weight component corresponds to a preset label (such as the numbers 1, 0, -1), and each weight component includes at least one weight, that is, for a weight component, all weights divided into the weight component have the same preset label, and the number of weights and the value of the weights are set during initialization (usually randomly).

[0095] For example, the damage value may be normalized using an algorithm such as deviation normalization.

[0096] In the third step, all standard damage values ​​are used to iteratively update the weight matrix to obtain the final weight matrix.

[0097] First, the number of iterations is increased by 1.

[0098] Then, it is determined whether the number of iterations reaches the preset number.

[0099] If so, the weight matrix is ​​used as the final weight matrix.

[0100] Otherwise, a standard damage value is randomly selected from all standard damage values ​​as the sample damage value.

[0101] Then, the similarity between the sample damage value and each element in the weight matrix is ​​calculated, and the element with the largest similarity is used as the target element.

[0102] For each element in the weight matrix, determine whether the element is equal to the target element. If so, update the element according to the target element to obtain the updated element. Otherwise, use the element as the updated element.

[0103] Integrate all updated elements to obtain the updated weight matrix, and use the updated weight matrix as the weight matrix, returning to the step of the number of iterations plus 1.

[0104] It should be noted that the initial value of the number of iterations is 0. The similarity can be calculated based on the cosine similarity calculation formula, Euclidean distance, etc. The above-mentioned element is updated according to the target element to obtain the updated element. The specific expression is:

[0105] ;

[0106] in, represents the updated first elements, Indicates the weight matrix elements, represents the learning rate, represents the update coefficient adjusted according to the target element, Indicates the A damage value.

[0107] In the fourth step, all damage values ​​are clustered according to all elements in the final weight matrix to obtain multiple clusters.

[0108] Specifically, through the formula:

[0109] ;

[0110] Get the tag set .

[0111] in, Represents all damage values, Indicates the labels of the weights corresponding to all damage values ​​in the final weight matrix, Function is used to find the damage value Tags, represents the final weight matrix.

[0112] That is, through the above formula, the label of the element corresponding to the damage value in the final weight matrix is ​​used as the label of the damage value, and then all damage values ​​with the same label are regarded as a cluster.

[0113] The fifth step is to perform secondary clustering on all clusters to obtain multiple target clusters.

[0114] First, for each cluster, the elements of the final weight matrix corresponding to all damage values ​​in the cluster are combined into a vector to obtain the weight vector of the cluster.

[0115] Then, the cosine similarity between each two clusters is calculated based on all weight vectors.

[0116] Finally, all clusters are clustered according to all cosine similarities to obtain multiple target clusters.

[0117] For example, a density-based spatial clustering algorithm (DBSCAN) can be used for clustering operations, with clusters used as data points in DBSCAN and cosine similarity used as the distance between data points. All data points are clustered using the DBSCAN algorithm to obtain target clusters.

[0118] In the sixth step, labels are assigned according to the target clusters corresponding to each damage information to obtain the initial damage labels of the sub-regions corresponding to each damage information in the corresponding historical time period.

[0119] Exemplarily, a label for distinction is assigned to each target cluster (such as the numbers 1, 2, 3, which are used to distinguish different target clusters and are irrelevant to the labels in the final weight matrix), and the label of the target cluster is used as the label of the damage information corresponding to the damage value in the target cluster, and the label of the damage information is used as the initial damage label of the corresponding sub-region in the corresponding historical time period.

[0120] In step 13, a support vector machine is used to quantitatively separate all damage information based on all initial damage labels to obtain the final damage label of each sub-region in each historical time period.

[0121] The above-mentioned final damage label is used to describe the damage category of the sub-region in the historical time period. The damage category is shear damage or tensile damage, that is, the final damage label is shear damage or tensile damage.

[0122] In some embodiments of the present application, the step of using a support vector machine to quantitatively separate all damage information based on all initial damage labels to obtain a final damage label for each sub-region in each historical time period includes:

[0123] In the first step, a Lagrangian generalized function is constructed based on all initial damage labels.

[0124] Specifically, the Lagrange function is:

[0125] ;

[0126] in, represents the value of the Lagrange function, represents the normal vector of the hyperplane, represents the threshold of the support vector machine, represents the non-negative relaxation coefficient The non-negative penalty coefficient of and represents the Lagrange multiplier, Indicates the A non-negative penalty coefficient, Indicates the amount of damage information, and All are Lagrange multipliers, Indicates the The initial damage label corresponding to the damage information, Indicates the The damage value of each damage information.

[0127] The second step is to find the parameter solution that should be satisfied by minimizing the Lagrangian functional, and substitute the parameter solution into the Lagrangian functional to obtain the objective function.

[0128] Specifically, the parameter solution is:

[0129] ;

[0130] ;

[0131] ;

[0132] The objective function is:

[0133] ;

[0134] in, represents the damage weight matrix, Indicates the Lagrange multipliers, Indicates the The initial damage label corresponding to the damage information, Indicates the The damage value of the damage information, Represents a transpose operation.

[0135] The third step is to minimize the objective function and obtain the normal vector and the threshold of the support vector machine.

[0136] For example, a sequential minimum optimization algorithm may be used to minimize the objective function.

[0137] The fourth step is to construct a hyperplane based on the normal vector and the threshold of the support vector machine.

[0138] Specifically, the hyperplane is:

[0139] ;

[0140] in, represents a vector on the hyperplane.

[0141] In the fifth step, all damage information is mapped to the feature space, and all initial damage labels are quantitatively separated using a hyperplane in the feature space to obtain a directional separation result.

[0142] Specifically, the damage value of each damage information is calculated (calculated according to the process of calculating the damage value in step 12), and then the damage value is mapped to the feature space, and a hyperplane is set in the feature space. The hyperplane divides the feature space into two subspaces. All damage values ​​are separated in the two subspaces, and a directional separation result is obtained.

[0143] It should be noted that the above-mentioned directional separation results are used to describe the subspace in which each damage value is located, and the above-mentioned feature space is used to actually map the damage value and the hyperplane. It can be set according to the data format of the damage value, which can be a two-dimensional coordinate system. The hyperplane is a straight line in the two-dimensional coordinate system, and the damage value is a point in the two-dimensional coordinate system.

[0144] In the sixth step, for each damage information, determine whether the damage information is on the first side of the hyperplane in the directional separation result. If so, the tensile damage is used as the final damage label of the sub-region corresponding to the damage information in the corresponding historical time period. Otherwise, the shear damage is used as the final damage label of the sub-region corresponding to the damage information in the corresponding historical time period.

[0145] For example, if the damage values ​​are AF and RA, then the following can be constructed: Figure 2 The feature space of the two-dimensional coordinate system shown in the figure has a horizontal axis of RA, with a unit of ms / V, a vertical axis of AF, with a unit of kHZ, a diagonal line for a hyperplane, a side of the hyperplane close to the vertical axis for the first side, and a side close to the horizontal axis for the second side. The damage information corresponding to all damage values ​​on the first side is tensile damage, and the damage information corresponding to all damage values ​​on the second side is shear damage. The proportion of tensile damage is 38.3%, and the proportion of shear damage is 61.7%.

[0146] It can be understood that all the above operation processes are operation processes of a support vector machine.

[0147] Step 14: For each historical time period, spatial damage analysis is performed based on all damage information in the historical time period to obtain a damage grayscale matrix for each sub-region in the historical time period.

[0148] The damage grayscale matrix is ​​used to describe the damage degree of the sub-region, and the elements in the damage grayscale matrix are used to describe the damage degree at the sampling point in the sub-region.

[0149] For example, the damage degree of the sub-region can be obtained based on the damage value analysis of the damage information. If the damage information is an acoustic emission signal, the sub-region corresponding to the damage information is processed as follows:

[0150] Concrete three-point bending test model Figure 3 As shown in the figure, the dotted rectangle is the sub-area under study. The concrete mesh model of this sub-area is established based on the concrete characteristics, and a 100×120 mm calculation domain is generated, where Figure 3 a in the figure is the front view. Figure 3In it, b is the side view. Then, considering the scale of granite acoustic emission data and the computational resource constraints, the event number threshold N = 50 is determined. The contact radius R = 15 mm is defined, the characteristic length L = 1 mm is set, the height H = 100 mm is set, and the incremental gradient Δr = 0.01 mm is configured.

[0151] Then, based on the spatial coordinate information of the acoustic emission events detected by the acoustic emission signal (i.e., the damaged parts in the sub-region) and the parameters set above, the spatial b value of the concrete (i.e., the damage degree, the smaller the spatial b value, the greater the crack development degree of the concrete) is calculated. The specific steps are as follows:

[0152] Based on the spatial coordinate data of the acoustic emission events, the sub-region of the mid-span section is selected as the calculation domain for the acoustic emission b value, and the calculation is performed according to the following process:

[0153] According to the L-index criterion, the reference point coordinate matrix A (i.e., the coordinate matrix required for the sampling points) is constructed, and k discrete reference points (i.e., sampling points) are generated within the target sub-region according to the reference point coordinate matrix A;

[0154] Starting from the first reference point, the following operations are performed on each reference point in turn:

[0155] Set the initial search radius r = 0.01 m;

[0156] Establish a cylindrical spatial domain with the reference point as the center, the search radius r as the radius, and the height H;

[0157] Extract the number of acoustic emission events n in the time window t within this spatial domain;

[0158] If n ≥ N threshold, the spatial b value is obtained by least squares fitting;

[0159] If n < N, the search radius is iteratively expanded in steps of Δr = 0.01 m, and the step of establishing a cylindrical spatial domain with the reference point as the center, the search radius r as the radius, and the height H is returned;

[0160] Traverse all reference points, integrate the spatial b values of all reference points into a matrix to obtain the damage grayscale matrix.

[0161] Among them, the formula for obtaining the spatial b value by least squares fitting is as follows:

[0162] ;

[0163] Among them, is the amplitude of the acoustic emission (AE, Acoustic Emission) event, is the number of AE events with an amplitude greater than , is an empirical constant, is the spatial b value (damage degree) of the AE event, Calculates the base 10 logarithm.

[0164] In step 15, based on the damage grayscale matrix and final damage labels of all sub-regions in each historical time period, a three-dimensional damage model of the target concrete region in each historical time period is constructed, and all three-dimensional damage models are serialized to obtain a spatiotemporal evolution damage model of the target concrete region.

[0165] The above-mentioned 3D damage model is used to describe the damaged sub-regions and damage categories of the target concrete area in the historical time period. The spatiotemporal evolution damage model is used to describe the damage evolution process of the target concrete area in all historical time periods. The spatiotemporal evolution damage model is a combination of all 3D damage models arranged in chronological order from early to late.

[0166] Specifically, first, each damage grayscale matrix is ​​displayed as a graph to obtain a grayscale image; then, for each historical time period, all the grayscale images in the historical time period are stacked according to the position of each sub-region to obtain the initial three-dimensional damage model of the target concrete area in the historical time period. The watershed segmentation algorithm is then used to separate the damage grayscale matrix to divide the severely damaged sub-region and other sub-regions in the initial three-dimensional damage model. The threshold binarization method is then used to set the grayscale value of the severely damaged sub-region to 1 and the grayscale value of other sub-regions to 0. According to the final damage label, the shear damage and tensile damage sub-regions in the initial three-dimensional damage model are visually distinguished (such as using different colors to represent them) to obtain a three-dimensional damage model; finally, all three-dimensional damage models are sorted in order from early to late in all historical time periods to obtain a spatiotemporal evolution damage model of the target concrete area.

[0167] The method of the present application is illustrated below with reference to a specific example.

[0168] A concrete block is processed by the method of the present application to obtain a damage grayscale matrix of a sub-region in the concrete block. The grayscale image obtained by displaying the damage grayscale matrix as a graph is as follows: Figure 4 As shown, the grayscale image after threshold binarization and visual distinction is as follows Figure 5 As shown, the front view of the spatiotemporal evolution damage model after stacking and sorting all grayscale images is shown in Figure 6 As shown, Figure 6 a represents the front view of the three-dimensional damage model corresponding to 0-300 seconds (S), Figure 6 b shows the front view of the three-dimensional damage model corresponding to 300-600 seconds (S), Figure 6c represents the front view of the three-dimensional damage model corresponding to 600-900 seconds (S), and the top view of the spatiotemporal evolution damage model is shown in Figure 7 As shown, Figure 7 a represents the top view of the three-dimensional damage model corresponding to 0-300 seconds (S), Figure 7 b shows the top view of the three-dimensional damage model corresponding to 300-600 seconds (S). Figure 7 c shows the top view of the three-dimensional damage model corresponding to 600-900 seconds (S).

[0169] It is worth mentioning that damage mapping can achieve preliminary differentiation of damage information. Using support vector machines to quantitatively separate damage information and obtain final damage labels allows for further detailed classification of damage information, improving the accuracy of the final damage labels. Constructing a spatiotemporal evolution damage model based on these accurate final damage labels effectively improves the accuracy of the spatiotemporal evolution of concrete damage. This represents a technological leap from "passive perception" to "active prediction" of concrete damage, providing a basis for intelligent decision-making in engineering stability control.

[0170] like Figure 8 As shown, an embodiment of the present application provides a device for identifying the spatiotemporal evolution of concrete damage. The device 800 for identifying the spatiotemporal evolution of concrete damage includes:

[0171] An acquisition module 801 is used to acquire damage information of each sub-region of a target concrete region in multiple historical time periods;

[0172] The damage mapping module 802 is used to perform damage mapping based on all damage information to obtain an initial damage label for each sub-region in each historical time period; the initial damage label is used to classify all damage information;

[0173] The quantitative separation module 803 is used to use a support vector machine to quantitatively separate all damage information based on all initial damage labels to obtain the final damage label of each sub-region in each historical time period; the final damage label is used to describe the damage category of the sub-region in the historical time period;

[0174] The damage analysis module 804 is used to perform spatial damage analysis based on all damage information in each historical time period, and obtain a damage grayscale matrix for each sub-region in the historical time period; the damage grayscale matrix is ​​used to describe the damage degree of the sub-region;

[0175] Construction module 805 is used to construct a three-dimensional damage model of the target concrete area in each historical time period based on the damage grayscale matrix and final damage label of all sub-areas in each historical time period, and serialize all three-dimensional damage models to obtain a spatiotemporal evolution damage model of the target concrete area; the three-dimensional damage model is used to describe the damaged sub-areas and damage categories of the target concrete area in the historical time period, and the spatiotemporal evolution damage model is used to describe the damage evolution process of the target concrete area in all historical time periods.

[0176] It should be noted that the information interaction, execution process, etc. between the above-mentioned devices / units are based on the same concept as the method embodiment of this application. Their specific functions and technical effects can be found in the method embodiment section and will not be repeated here.

[0177] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0178] like Figure 9 As shown, an embodiment of the present application provides a terminal device. The terminal device D10 of this embodiment includes: at least one processor D100 ( Figure 9 Only one processor is shown in the figure), a memory D101, and a computer program D102 stored in the memory D101 and executable on the at least one processor D100, wherein the processor D100 implements the steps of any of the above-mentioned method embodiments when executing the computer program D102.

[0179] Specifically, when the processor D100 executes the computer program D102, it obtains damage information for each sub-region of the target concrete region over multiple historical time periods, then performs damage mapping based on all damage information to obtain an initial damage label for each sub-region in each historical time period. A support vector machine is then used to quantitatively separate all damage information based on all initial damage labels to obtain a final damage label for each sub-region in each historical time period. A spatial damage analysis is then performed for each historical time period based on all damage information in that historical time period to obtain a damage grayscale matrix for each sub-region in that historical time period. Finally, a three-dimensional damage model of the target concrete region in each historical time period is constructed based on the damage grayscale matrices and final damage labels for all sub-regions in each historical time period. All three-dimensional damage models are serialized to obtain a spatiotemporal evolution damage model for the target concrete region. Damage mapping of the damage information enables preliminary differentiation of the damage information, while quantitative separation of the damage information using the support vector machine to obtain the final damage label enables further detailed classification of the damage information, improving the accuracy of the final damage label. Constructing a spatiotemporal evolution damage model based on the accurate final damage label effectively improves the accuracy of the spatiotemporal evolution of concrete damage.

[0180] The processor D100 may be a central processing unit (CPU), or may be another general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. A general-purpose processor may be a microprocessor or any conventional processor.

[0181] In some embodiments, the memory D101 may be an internal storage unit of the terminal device D10, such as a hard disk or memory of the terminal device D10. In other embodiments, the memory D101 may also be an external storage device of the terminal device D10, such as a plug-in hard disk, a smart memory card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal device D10. Furthermore, the memory D101 may include both an internal storage unit of the terminal device D10 and an external storage device. The memory D101 is used to store an operating system, application programs, a boot loader, data, and other programs, such as the program code of the computer program. The memory D101 may also be used to temporarily store data that has been output or is about to be output.

[0182] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned various method embodiments can be implemented.

[0183] An embodiment of the present application provides a computer program product. When the computer program product is run on a terminal device, the terminal device can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0184] If the integrated unit is implemented as a software functional unit and sold or used as a standalone product, it can be stored in a computer-readable storage medium. Based on this understanding, the present application implements all or part of the process steps in the above-mentioned method embodiments by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include at least any entity or device capable of carrying the computer program code to the apparatus / terminal device for identifying the spatiotemporal evolution of concrete damage, a recording medium, computer memory, read-only memory (ROM), random access memory (RAM), an electrical carrier signal, a telecommunications signal, and a software distribution medium. Examples include a USB flash drive, a removable hard drive, a magnetic disk, or an optical disk.

[0185] In the above embodiments, the description of each embodiment has its own focus. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0186] Those skilled in the art will appreciate that the units and algorithm steps of each example 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. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0187] The above is a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles described in the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for identifying the spatiotemporal evolution of concrete damage, characterized in that: include: Obtain damage information of each sub-area of ​​the target concrete area in multiple historical time periods; Perform damage mapping based on all damage information to obtain an initial damage label for each sub-region in each historical time period; The initial damage label is used to classify all damage information; Using a support vector machine, all damage information is quantitatively separated according to all initial damage labels to obtain the final damage label of each sub-region in each historical time period; The final damage label is used to describe the damage category of the sub-region in the historical time period; For each historical time period, spatial damage analysis is performed based on all damage information in the historical time period to obtain a damage grayscale matrix for each sub-region in the historical time period; the damage grayscale matrix is ​​used to describe the damage degree of the sub-region; Based on the damage grayscale matrix and final damage labels of all sub-regions in each historical time period, a three-dimensional damage model of the target concrete region in each historical time period is constructed, and all three-dimensional damage models are serialized to obtain a spatiotemporal evolution damage model of the target concrete region; the three-dimensional damage model is used to describe the damaged sub-regions and damage categories of the target concrete region in the historical time period, and the spatiotemporal evolution damage model is used to describe the damage evolution process of the target concrete region in all historical time periods; The damage mapping is performed based on all damage information to obtain the initial damage label of each sub-region in each historical time period, including: Analyze each piece of damage information to obtain a damage value corresponding to the damage information; Initializing a weight matrix and normalizing each of the damage values ​​to obtain a standard damage value; wherein the multiple elements in the weight matrix correspond one-to-one to the multiple damage values; Iteratively updating the weight matrix using all standard damage values ​​to obtain a final weight matrix; Clustering all damage values ​​according to all elements in the final weight matrix to obtain multiple clusters; Perform secondary clustering on all clusters to obtain multiple target clusters; Labels are assigned according to the target clusters corresponding to each damage information to obtain the initial damage labels of the sub-regions corresponding to each damage information in the corresponding historical time period.

2. The method for identifying the spatiotemporal evolution of concrete damage according to claim 1, characterized in that: The weight matrix is ​​iteratively updated using all standard damage values ​​to obtain a final weight matrix, including: The number of iterations increases by 1; Determine whether the number of iterations reaches a preset number; If yes, the weight matrix is ​​used as the final weight matrix; Otherwise, a standard damage value is randomly selected from all standard damage values ​​as the sample damage value; Calculating the similarity between the sample damage value and each element in the weight matrix, and taking the element with the greatest similarity as the target element; For each element in the weight matrix, determine whether the element is equal to the target element; if so, update the element according to the target element to obtain an updated element; otherwise, use the element as the updated element; All updated elements are integrated to obtain an updated weight matrix, and the updated weight matrix is ​​used as the weight matrix, and the process returns to the step of increasing the number of iterations by 1.

3. The method for identifying the spatiotemporal evolution of concrete damage according to claim 1, characterized in that: The secondary clustering is performed on all clusters to obtain multiple target clusters, including: For each cluster, combining the elements of the final weight matrix corresponding to all damage values ​​in the cluster into a vector to obtain a weight vector of the cluster; Calculate the cosine similarity between each two clusters based on all weight vectors; All clusters are clustered according to all cosine similarities to obtain multiple target clusters.

4. The method for identifying the spatiotemporal evolution of concrete damage according to claim 1, characterized in that: The final damage label is shear damage or tensile damage; The support vector machine is used to quantitatively separate all damage information according to all initial damage labels to obtain the final damage label of each sub-region in each historical time period, including: Construct a Lagrangian generalized function based on all initial damage labels; Solving the parameter solution that the Lagrangian functional should satisfy for minimization, and substituting the parameter solution into the Lagrangian functional to obtain the objective function; Minimizing the objective function to obtain a normal vector and a support vector machine threshold; constructing a hyperplane according to the normal vector and a threshold of a support vector machine; Mapping all damage information to a feature space, and using the hyperplane in the feature space to quantitatively separate all initial damage labels to obtain a directional separation result; For each damage information, determine whether the damage information is on the first side of the hyperplane in the directional separation result. If so, the tensile damage is used as the final damage label of the sub-region corresponding to the damage information in the corresponding historical time period. Otherwise, the shear damage is used as the final damage label of the sub-region corresponding to the damage information in the corresponding historical time period.

5. The method for identifying the spatiotemporal evolution of concrete damage according to claim 4, characterized in that: The Lagrangian functional is: ； in, represents the value of the Lagrange function, represents the normal vector of the hyperplane, represents the threshold of the support vector machine, represents the non-negative relaxation coefficient The non-negative penalty coefficient of and represents the Lagrange multiplier, Indicates the A non-negative penalty coefficient, Indicates the amount of damage information, and All are Lagrange multipliers, Indicates the The initial damage label corresponding to the damage information, Indicates the The damage value of each damage information.

6. The method for identifying the spatiotemporal evolution of concrete damage according to claim 5, characterized in that: The parameter solution is: ； ； ； The objective function is: ； in, represents the damage weight matrix, Indicates the Lagrange multipliers, Indicates the The initial damage label corresponding to the damage information, Indicates the The damage value of the damage information, Represents a transpose operation.

7. The method for identifying the spatiotemporal evolution of concrete damage according to claim 6, characterized in that: The hyperplane is: ； in, represents a vector on the hyperplane.

8. A terminal device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for identifying the spatiotemporal evolution of concrete damage according to any one of claims 1 to 7 is implemented.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for identifying the spatiotemporal evolution of concrete damage according to any one of claims 1 to 7 is implemented.

Images