Crack angle prediction method and system for three-cantilever beam structure based on multi-classification problem

Abstract

The present invention belongs to the field of building monitoring technology and discloses a method and system for predicting crack angles in a three-cantilever beam structure based on a multi-classification problem. The method and system comprise the following steps: obtaining a simulation data set of a three-cantilever beam identification structure under different crack conditions; based on the simulation data set, identifying multicollinearity between crack angle data and cantilever beam strain data through multivariate visualization and correlation analysis, and removing redundant variables to obtain an optimized data set; based on the optimized data set, constructing a crack angle prediction model using a variety of algorithms for multi-classification problems; obtaining the cantilever beam strain data and inputting it into the crack angle prediction model, and using the crack angle prediction model to predict the corresponding crack angle data. The present invention can not only monitor the dynamic changes of existing cracks under load, but also identify the direction of new cracks.

Description

Technical Field

[0001] The present invention relates to the technical field of building monitoring, and in particular to a method and system for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem. Background Art

[0002] Buildings are essential places or facilities where humans reside and engage in various life activities. Their safety directly impacts human survival and development. Therefore, accurate health monitoring of structures such as buildings and bridges is a key research area in modern engineering. The primary goal of building structural health monitoring is to collect information on a variety of physical quantities by installing multiple sensors on structures such as buildings, roads, and bridges. This information captures the structural responses of buildings, roads, and bridges to factors such as human and environmental stimuli, thereby assessing the extent of structural damage. This provides a reference for safety assessments and repairs for the human living environment, and a basis for predicting its future health. The core of structural health monitoring is information acquisition, analysis, and assessment. As the demand for a safer living environment continues to increase, sensing structural health through sensors is becoming an effective and widely used method. At the same time, traditional structural health monitoring methods and their heavy reliance on manual measurement are gradually being phased out. Intelligent sensors and AI-based massive data processing technologies are rapidly demonstrating their superiority and will play an increasingly important role in the field of structural health monitoring.

[0003] Currently, building structural health monitoring encompasses a wide range of topics, with cracks being a key target. Crack monitoring methods include sensors and fiber Bragg grating (FBG) technology. Sensor monitoring employs a variety of methods, with strain gauges being widely favored due to their simplicity. However, the existing method of attaching strain gauges to a dual-cantilever beam structure can identify the width of existing cracks but cannot predict the width and direction (angle) of new cracks.

[0004] Therefore, how to provide a crack angle prediction method and system for a three-cantilever beam structure based on multi-classification problems is an urgent problem to be solved. Summary of the Invention

[0005] The embodiments of the present invention provide a method and system for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem, so as to solve the problems in the prior art.

[0006] To provide a basic understanding of some aspects of the disclosed embodiments, the following is a brief summary. This summary is not intended to be an extensive review, identify key or critical elements, or delineate the scope of these embodiments. Its sole purpose is to present some concepts in a simplified form as a prelude to the detailed description that follows.

[0007] According to a first aspect of an embodiment of the present invention, a method for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem is provided.

[0008] In one embodiment, a method for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem includes the following steps:

[0009] Using the finite element method to perform static simulation on the three-cantilever beam identification body structure, and obtain simulation data sets of the three-cantilever beam identification body structure under different crack conditions; the simulation data sets include crack angle data and cantilever beam strain data;

[0010] Based on the simulation data set, multivariate visualization and correlation analysis were used to identify multicollinearity between the crack angle data and the cantilever beam strain data, and redundant variables were deleted to obtain an optimized data set.

[0011] Based on the optimized data set, a crack angle prediction model is constructed using a variety of algorithms for multi-classification problems.

[0012] The obtained cantilever beam strain data is input into the crack angle prediction model, and the corresponding crack angle data is predicted by the crack angle prediction model.

[0013] In one embodiment, the static simulation of the three-cantilever beam identification structure using the finite element method and obtaining the simulation data set of the three-cantilever beam identification structure under different crack conditions includes the following steps:

[0014] Construct a simulation measurement environment for the three-cantilever beam identification structure and the matching calibration structure, and determine the constraints of the three-cantilever beam identification structure;

[0015] Based on the pre-set simulation objectives, the boundary conditions of the three-cantilever beam identification structure and the matching calibration structure are determined;

[0016] Static simulation was performed in finite element analysis software, and strain data of the three-cantilever beam identification structure under different crack conditions were collected to obtain a simulation data set.

[0017] In one embodiment, performing static simulation in finite element analysis software and collecting strain data of the three-cantilever beam identification structure under different crack conditions to obtain a simulation data set includes the following steps:

[0018] Create a three-dimensional geometric model of the three-cantilever beam identification structure and the matching calibration structure in the finite element analysis software, and perform meshing on the geometric model;

[0019] Based on the structural constraints of the three-cantilever beam identifier, loads are applied to the three-cantilever beam identifier structure, and strain data at the cantilevers of the three-cantilever beam identifier structure are obtained to obtain a simulation data set.

[0020] In one embodiment, the load applied to the three-cantilever beam identification structure is a forced displacement load.

[0021] In one embodiment, the structural constraint condition of the three-cantilever beam identification body is to fix the width of the crack and change the angle of the crack.

[0022] In one embodiment, the method of identifying multicollinearity between the crack angle data and the cantilever beam strain data based on the simulation data set through multivariate visualization and correlation analysis, and deleting redundant variables to obtain an optimized data set includes the following steps:

[0023] For the simulation data set, visualization software is used to draw a corresponding relationship diagram between variables, and the determination coefficient is calculated by observing the distribution relationship and trend between each variable and combining it with the multiple regression model; the variables include the response variable vector and the regression variable;

[0024] The variance inflation factor of the regression variable is calculated based on the coefficient of determination, and whether there is multicollinearity between the regression variables is determined based on the preset threshold. If so, the regression variable is deleted; otherwise, the regression variable is retained.

[0025] The deleted or retained regression variables are integrated with the response variable vector to obtain an optimized data set.

[0026] In one embodiment, the method of constructing a crack angle prediction model based on an optimized data set and using multiple algorithms for multi-classification problems includes the following steps:

[0027] The crack angles in the optimized dataset are treated as discrete variables, and the crack angle range is divided into multiple subclasses according to the preset step size.

[0028] A classification model is constructed using a variety of algorithms for multi-classification problems. An optimized data set is divided into multiple subclasses according to the crack angle. The performance of the classification model is trained and evaluated to obtain performance evaluation indicators.

[0029] Based on the performance evaluation indicators, the performance of each classification model in the crack angle prediction task is analyzed, and according to the analysis results, the model with the best performance is selected as the final crack angle prediction model.

[0030] In one embodiment, the method of constructing a classification model using multiple algorithms for multi-classification problems, dividing an optimized data set into multiple subclasses according to crack angles, training and evaluating the performance of the classification model, and obtaining performance evaluation indicators includes the following steps:

[0031] The optimized data set is divided into a training set and a test set by dividing the crack angles into multiple subclasses according to a preset division ratio;

[0032] Using the training set, the extreme gradient boosting classification model, K-nearest neighbor classification model, decision tree classification model, random forest classification model and support vector machine classification model were used respectively;

[0033] The trained extreme gradient boosting classifier, K-nearest neighbor classifier, decision tree classifier, random forest classifier and support vector machine classifier are tested using the test set, and the performance of each classifier is evaluated based on the test results to obtain the performance evaluation indicators.

[0034] In one embodiment, the performance evaluation indicators include: precision, recall, and F1 score.

[0035] According to a second aspect of an embodiment of the present invention, a crack angle prediction system for a three-cantilever beam structure based on a multi-classification problem is provided.

[0036] In one embodiment, the three-cantilever beam structure crack angle prediction system based on multi-classification problem includes:

[0037] a data acquisition unit, configured to perform static simulation on the three-cantilever beam identification structure using a finite element method, and acquire a simulation data set of the three-cantilever beam identification structure under different crack conditions; the simulation data set includes crack angle data and cantilever beam strain data;

[0038] A data optimization unit is used to identify multicollinearity between crack angle data and cantilever beam strain data based on the simulation data set through multivariate visualization and correlation analysis, and to delete redundant variables to obtain an optimized data set;

[0039] A model building unit is used to build a crack angle prediction model based on an optimized data set and using a variety of algorithms for multi-classification problems;

[0040] The angle prediction unit is used to obtain the cantilever beam strain data and input it into the crack angle prediction model, and predict the corresponding crack angle data through the crack angle prediction model.

[0041] According to a third aspect of an embodiment of the present invention, a computer device is provided.

[0042] In some embodiments, the computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0043] According to a fourth aspect of embodiments of the present invention, a computer-readable storage medium is provided.

[0044] In one embodiment, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0045] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0046] This invention takes building structure health monitoring as its application background and carries out technical and theoretical research on multi-dimensional identification and event prediction of small gaps in structures based on sensor monitoring method. Unlike other methods of crack monitoring, the sensor monitoring method based on strain gauge strain principle is simple to deploy, stable and reliable in operation, low in cost, high in sensitivity and accuracy, and balanced in performance indicators. In view of the limitations of the double-cantilever beam single-dimensional identification method, a three-cantilever beam multi-dimensional identification method is proposed, which increases the measurement dimension. It can not only monitor the dynamic changes of existing cracks under load, but also identify the direction information of new cracks.

[0047] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0049] Figure 1 is a flow chart of a method for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem according to an exemplary embodiment;

[0050] Figure 2 is a principle block diagram of a crack angle prediction system for a three-cantilever beam structure based on a multi-classification problem according to an exemplary embodiment;

[0051] Figure 3 is a structural diagram of a computer device according to an exemplary embodiment;

[0052] Figure 4 is a line graph of evaluation indicators of a three-cantilever beam angle prediction model based on a multi-classification problem in a method for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem according to an exemplary embodiment;

[0053] Figure 5 It is a line graph of the number of correctly classified samples of the angle prediction model of three cantilever beams based on a multi-classification problem in a method for predicting crack angles of three cantilever beam structures based on a multi-classification problem according to an exemplary embodiment. DETAILED DESCRIPTION

[0054] The following description and accompanying drawings sufficiently illustrate the specific embodiments herein to enable those skilled in the art to practice them. Portions and features of some embodiments may be included in or substituted for portions and features of other embodiments. The scope of the embodiments herein includes the entire scope of the claims, including all available equivalents thereof. Herein, the terms "first," "second," and the like are used solely to distinguish one element from another and do not require or imply any actual relationship or order between these elements. In practice, the first element can also be referred to as the second element, and vice versa. Furthermore, the terms "comprise," "comprising," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a structure, device, or apparatus comprising a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such structure, device, or apparatus. Without further limitation, an element defined by the phrase "comprising a..." does not preclude the presence of other identical elements in the structure, device, or apparatus comprising the element. The various embodiments herein are described in a progressive manner, with each embodiment focusing on its differences from the other embodiments. Similar or identical parts between the various embodiments can be referenced to each other.

[0055] The terms "longitudinal", "transverse", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like used herein to indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are intended only to facilitate the description of this document and simplify the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on the present invention. In the description herein, unless otherwise specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, they can be mechanical or electrical connections, or they can be internal connections between two elements, they can be directly connected, or they can be indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to the specific circumstances.

[0056] As used herein, unless otherwise specified, the term "plurality" means two or more.

[0057] In this document, the character " / " indicates that the preceding and following objects are in an "or" relationship. For example, A / B means: A or B.

[0058] In this article, the term "and / or" is used to describe the association relationship between objects, indicating that three relationships can exist. For example, A and / or B means: A or B, or, A and B.

[0059] It should be understood that, although the various steps in the flowchart are shown in sequence as indicated by the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps may be performed in other orders. Moreover, at least a portion of the steps in the figure may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily performed at the same time, but may be performed at different times. The execution order of these sub-steps or stages is not necessarily to be performed in sequence, but may be performed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0060] Each module in the device or system of the present application can be implemented in whole or in part by software, hardware, or a combination thereof. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software so that the processor can call and execute the operations corresponding to the above modules.

[0061] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0062] Figure 1 An embodiment of the method for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem of the present invention is shown.

[0063] In this optional embodiment, the method for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem includes the following steps:

[0064] Step S101, using the finite element method to perform static simulation on the three-cantilever beam identification body structure, and obtaining a simulation data set of the three-cantilever beam identification body structure under different crack conditions; the simulation data set includes crack angle data and cantilever beam strain data;

[0065] Step S102 , based on the simulation data set, identifying multicollinearity between the crack angle data and the cantilever beam strain data through multivariate visualization and correlation analysis, and deleting redundant variables to obtain an optimized data set;

[0066] Step S103, based on the optimized data set, a crack angle prediction model is constructed using a variety of algorithms for multi-classification problems;

[0067] Step S104: The obtained cantilever beam strain data is input into a crack angle prediction model, and the corresponding crack angle data is predicted by the crack angle prediction model.

[0068] It should be noted that the finite element analysis software used in the present invention includes SolidWorks and Ansys. No matter which one is used for statics simulation, as long as the premise assumptions are reasonable and consistent with the setting parameters, the simulation results are basically the same. Traditional finite element analysis objects are mainly models based on structural mechanics, but with the increasing application of multi-physics field simulation, simulations based on fluid mechanics, thermodynamics, electromagnetic fields, etc. are also widely used. It should be pointed out that the finite element analysis process under different application fields is basically the same, that is, establishing the model geometry, clarifying the constraint relationship between the geometric structure components, defining materials and loads, meshing, and solving and outputting according to the simulation parameter type.

[0069] In this optional embodiment, the static simulation of the three-cantilever beam identification body structure using the finite element method and obtaining the simulation data set of the three-cantilever beam identification body structure under different crack conditions includes the following steps:

[0070] Construct a simulation measurement environment for the three-cantilever beam identification structure and the matching calibration structure, and determine the constraints of the three-cantilever beam identification structure;

[0071] Based on the pre-set simulation objectives, the boundary conditions of the three-cantilever beam identification structure and the matching calibration structure are determined;

[0072] Static simulation was performed in finite element analysis software, and strain data of the three-cantilever beam identification structure under different crack conditions were collected to obtain a simulation data set.

[0073] Among them, the three-cantilever beam identification body structure is the main body of the three-cantilever beam multi-dimensional identification body mechanical structure. It consists of a base plate, spring steel sheet, pressure block, PCB bearing studs, mounting feet and other parts. The simplified structure material of the three-cantilever beam identification body is ordinary carbon steel.

[0074] The matching calibration structure includes two symmetrically arranged calibration platforms and an array of calibration threaded holes uniformly distributed with the diameter of the simplified structure mounting threaded holes as the spacing. The direction of the crack is changed by changing the matching relationship between the simplified structure mounting holes and the calibration threaded hole array; the width of the crack is changed by changing the displacement between the calibration platforms. Changes in matching parameters and displacement parameters during the simulation are collectively referred to as changes in test conditions. The simplified structure is matched and constrained by the base mounting threaded holes and a symmetrical thread in the calibration threaded hole array; the bottom angle planes of the simplified structure are matched and constrained by the calibration platform through contact surfaces; and the calibration platform is matched and constrained by the contact surfaces. Once the constraint relationship is established, it means that the matching is rigidly connected during the simulation, that is, no relative displacement occurs.

[0075] In this optional embodiment, performing static simulation in finite element analysis software and collecting strain data of the three-cantilever beam identification structure under different crack conditions to obtain a simulation data set includes the following steps:

[0076] Create a three-dimensional geometric model of the three-cantilever beam identification structure and the matching calibration structure in the finite element analysis software, and perform meshing on the geometric model;

[0077] Based on the structural constraints of the three-cantilever beam identifier, loads are applied to the three-cantilever beam identifier structure, and strain data at the cantilevers of the three-cantilever beam identifier structure are obtained to obtain a simulation data set.

[0078] In one embodiment, the load applied to the three-cantilever beam identification structure is a forced displacement load.

[0079] In this optional embodiment, the structural constraint condition of the three-cantilever beam identification body is to fix the width of the crack and change the angle of the crack.

[0080] Specifically, while maintaining the crack width, the crack orientation is adjusted by changing the matching relationship between the simplified structure's mounting holes and the calibration threaded hole array. The crack orientation requires the use of a reference cantilever beam and a reference zero angle. Cantilever beams A and B, as well as 0° and 180°, serve as reference points for describing the crack orientation. The crack orientation is parallel to the beam width and perpendicular to the beam length, with the intersection located at the center of the intersection of the beam width and length.

[0081] In this optional embodiment, the steps of identifying multicollinearity between the crack angle data and the cantilever beam strain data through multivariate visualization and correlation analysis based on the simulation data set, and deleting redundant variables to obtain the optimized data set include the following steps:

[0082] For the simulation data set, visualization software is used to draw a corresponding relationship diagram between variables, and the determination coefficient is calculated by observing the distribution relationship and trend between each variable and combining it with the multiple regression model; the variables include the response variable vector and the regression variable;

[0083] The variance inflation factor of the regression variable is calculated based on the coefficient of determination, and whether there is multicollinearity between the regression variables is determined based on the preset threshold. If so, the regression variable is deleted; otherwise, the regression variable is retained.

[0084] The deleted or retained regression variables are integrated with the response variable vector to obtain an optimized data set.

[0085] The calculation formula for the variance inflation factor of the regression variable based on the coefficient of determination is:

[0086] ;

[0087] Where, Regression variables j The variance inflation factor, Regression variables j The coefficient of determination of .

[0088] Furthermore, if the regressors involved in building a predictive model are orthogonal, a loose inference can be drawn, meaning that the regressors have little mutual influence and are therefore suitable for the model. In most cases, the regressors are not orthogonal, but the lack of orthogonality is not a serious problem that limits the quality of the predictive model. However, if the regressors have a near-linear relationship, multicollinearity may exist, which can lead to misleading predictions and incorrect conclusions.

[0089] In order to check the multicollinearity of variable data, a multiple regression model is first introduced:

[0090] ;

[0091] Where y is the response variable vector, X is the regression variable matrix, and β is the coefficient vector. is a random error, where The mean is 0 and the variance is Normal distribution. If there is a set that is not all zero t 1 , t 2 ，t 3 ,......, t p , satisfying the equation:

[0092] ;

[0093] Where, p Indicates the upper limit of accumulation operation ( j =1~ p ), represents the regression coefficient.

[0094] The regression variables X can be considered to be linearly correlated. If the subsets of X satisfy the above equation, then the correlation coefficient matrix The rank of is less than p, and does not exist. Assume that this is approximately true for some subset of X, then There will be a linear correlation and multicollinearity problem.

[0095] Variance inflation factor is usually used to test multicollinearity. For the correlation coefficient matrix, we have:

[0096] ;

[0097] The J-th diagonal element of the correlation coefficient matrix C It can be written as:

[0098] ;

[0099] Where, Yes The coefficient of determination obtained when regressing on the other (p-1) regressors. If the regressors are nearly orthogonal, then Smaller, Close to 1; if there is a nearly linear relationship between the regression variables, then Close to 1, Instead, it becomes larger. Since the variance of the j-th regression coefficient is ,in Can be seen as a slope Least squares estimation Variance Increased impact factor, thus As a variance inflation factor To test for multicollinearity. According to the multicollinearity diagnostic criteria, if the variance inflation factor is less than 10, it can be considered that there is no multicollinearity problem between the regression variables. The smaller the index, the better. Conversely, if it is greater than 10, it indicates that there is a multicollinearity problem between the regression variables, and the severity of the problem increases with the increase of this index.

[0100] In addition, there are several common methods to solve multicollinearity:

[0101] (1) Increase the amount of data for regression variables: The biggest problem with multicollinearity is that it increases the slope. The least squares estimate of variance Increasing the amount of data for regression variables can reduce the variance of parameter estimates overall, thereby reducing the impact of multicollinearity; but on the other hand, the increase in data volume is limited by more practical factors, such as cost or the scope of the design model.

[0102] (2) Reducing the types of regression variables: When there are three or more types of regression variables, removing one of the two regression variables with multicollinearity problems may solve the problem of multicollinearity. However, when removing variables, variable information of the prediction model will be lost. Once important variables are removed, the quality of the prediction model will be immediately affected.

[0103] (3) Data dimensionality reduction of regression variables: Data dimensionality reduction refers to the process of using two or more original regression variables to generate new regression variables through some operation to replace the original regression variables. Data dimensionality reduction requires that the emergence of new variables and the elimination of original variables should minimize information loss. A common data dimensionality reduction method is principal component analysis.

[0104] (4) Adding penalty terms: For linear regression models, it is a common method to eliminate multicollinearity by adding penalty terms (regularization terms) to the loss function (cost function). The penalty terms include: the L0 norm penalty term restricts the parameters that are not equal to 0, that is, the number of parameters that are not equal to 0 is limited to a certain range to achieve the purpose of restricting the model; the L1 norm penalty term requires the sum of the parameter values to be limited to a certain range, that is, the sum of the parameters must be less than a certain value. The L1 norm penalty is also called parameter sparsity penalty; the L2 norm penalty term adds a square term to avoid the problem of positive and negative offset. The L2 norm penalty is also called weight decay penalty. Among the multiple regression algorithms selected by the invention, LASSO regression uses the L1 norm penalty term; Ridge regression uses the L2 norm penalty term.

[0105] (5) Ostrich policy: The essence of the ostrich policy is to ignore the problem of multicollinearity. If the prediction model is not affected by the multicollinearity of the known data, then the multicollinearity problem can be ignored. The rationality of the ostrich policy also shows from another perspective that the multicollinearity problem between regression variables cannot be used as a sufficient and necessary condition for the inapplicability of the prediction model.

[0106] In this optional embodiment, the construction of a crack angle prediction model based on the optimized data set and using multiple algorithms for multi-classification problems includes the following steps:

[0107] The crack angles in the optimized dataset are treated as discrete variables, and the crack angle range is divided into multiple subclasses according to the preset step size.

[0108] A classification model is constructed using a variety of algorithms for multi-classification problems. An optimized data set is divided into multiple subclasses according to the crack angle. The performance of the classification model is trained and evaluated to obtain performance evaluation indicators.

[0109] Based on the performance evaluation indicators, the performance of each classification model in the crack angle prediction task is analyzed, and according to the analysis results, the model with the best performance is selected as the final crack angle prediction model.

[0110] In this optional embodiment, the classification model is constructed using multiple algorithms for multi-classification problems, and an optimized data set is divided into multiple subclasses according to crack angles. The performance of the classification model is trained and evaluated to obtain performance evaluation indicators, which includes the following steps:

[0111] The optimized data set is divided into a training set and a test set by dividing the crack angles into multiple subclasses according to a preset division ratio;

[0112] Using the training set, the extreme gradient boosting classification model, K-nearest neighbor classification model, decision tree classification model, random forest classification model and support vector machine classification model were used respectively;

[0113] The trained extreme gradient boosting classifier, K-nearest neighbor classifier, decision tree classifier, random forest classifier and support vector machine classifier are tested using the test set, and the performance of each classifier is evaluated based on the test results to obtain the performance evaluation indicators.

[0114] In this optional embodiment, the performance evaluation indicators include: precision, recall and F1 score.

[0115] It should be noted that in the process of establishing the angle prediction model of the three-cantilever beam multi-dimensional recognition method, five algorithms for multi-classification problems, including the extreme gradient boosting classifier, were used.

[0116] Without considering cross-validation, directly establish prediction models for the above five classifiers and perform training and verification to obtain the corresponding accuracy Accuracy , precision Precision , recall rate Recall And the number of samples that each classifier correctly classifies for the test set data, Figure 4-Figure 5 The evaluation indicators and line graphs of the number of correctly classified samples are given respectively.

[0117] Among the angle prediction models established using five classifiers for a multi-classification problem, the K-nearest neighbor classifier achieved the highest precision, approximately 98.6%, while the support vector machine classifier achieved the lowest precision, approximately 1.4%. The recall results were consistent with the precision; the K-nearest neighbor classifier achieved the highest accuracy, approximately 98.8%, while the support vector machine classifier achieved the lowest accuracy, approximately 0.1%. The highest number of correctly classified samples was 479 (K-nearest neighbor classifier, decision tree classifier, and random forest classifier), while the lowest was only 7 (support vector machine classifier). In summary, it is feasible to establish a three-cantilever beam angle prediction model using the multi-dimensional recognition method for a multi-classification problem using the K-nearest neighbor classifier, and the evaluation metrics meet the error requirements for crack angle measurement in practical engineering applications. Both the decision tree classifier and the random forest classifier are alternative options for establishing this prediction model.

[0118] Figure 2 An embodiment of the three-cantilever beam structure crack angle prediction system based on multi-classification problem of the present invention is shown.

[0119] In this optional embodiment, the three-cantilever beam structure crack angle prediction system based on multi-classification problem includes:

[0120] The data acquisition unit 201 is used to perform static simulation on the three-cantilever beam identification body structure using the finite element method, and obtain simulation data sets of the three-cantilever beam identification body structure under different crack conditions; the simulation data sets include crack angle data and cantilever beam strain data;

[0121] The data optimization unit 202 is configured to identify multicollinearity between the crack angle data and the cantilever beam strain data based on the simulation data set through multivariate visualization and correlation analysis, and to delete redundant variables to obtain an optimized data set;

[0122] A model building unit 203 is used to build a crack angle prediction model based on the optimized data set using multiple algorithms for multi-classification problems;

[0123] The angle prediction unit 204 is used to obtain the cantilever beam strain data and input it into the crack angle prediction model, and predict the corresponding crack angle data through the crack angle prediction model.

[0124] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory, and a network interface connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The database of the computer device is used to store static information and dynamic information data. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the steps of the above-mentioned method embodiment are implemented.

[0125] Those skilled in the art will understand that Figure 3 The structure shown in the figure is merely a block diagram of a portion of the structure related to the solution of the present invention and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0126] In addition, the present invention also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiment when executing the computer program.

[0127] In addition, the present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiment are implemented.

[0128] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware using a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes in the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0129] The present invention is not limited to the structures described above and shown in the drawings, and various modifications and changes can be made without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for predicting crack angles of three-cantilever beam structures based on multi-classification problems, characterized by: The following steps are involved: A finite element method is used to perform static simulation of a three-cantilever beam identifier structure, and a simulation data set of the three-cantilever beam identifier structure under different crack conditions is obtained; the simulation data set includes crack angle data and cantilever beam strain data; based on the constraints of the three-cantilever beam identifier structure, a load is applied to the three-cantilever beam identifier structure, and strain data at the cantilevers of the three-cantilever beam identifier structure is obtained; the constraints of the three-cantilever beam identifier structure are fixed crack width and variable crack angle; Based on the simulation data set, multivariate visualization and correlation analysis were used to identify multicollinearity between the crack angle data and the cantilever beam strain data, and redundant variables were deleted to obtain an optimized data set. Based on the optimized data set, a crack angle prediction model is constructed using a variety of algorithms for multi-classification problems. The obtained cantilever beam strain data is input into the crack angle prediction model, and the corresponding crack angle data is predicted by the crack angle prediction model.

2. The method for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem according to claim 1 is characterized in that: The method of performing static simulation on the three-cantilever beam identification structure using the finite element method and obtaining simulation data sets of the three-cantilever beam identification structure under different crack conditions includes the following steps: Construct a simulation measurement environment for the three-cantilever beam identification structure and the matching calibration structure, and determine the constraints of the three-cantilever beam identification structure; Based on the pre-set simulation objectives, the boundary conditions of the three-cantilever beam identification structure and the matching calibration structure are determined; Static simulation was performed in finite element analysis software, and strain data of the three-cantilever beam identification structure under different crack conditions were collected to obtain a simulation data set.

3. The method for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem according to claim 2 is characterized in that: The statics simulation is performed in the finite element analysis software, and the strain data of the three-cantilever beam identification structure under different crack conditions are collected to obtain the simulation data set including: A three-dimensional geometric model of the three-cantilever beam identification structure and the matching calibration structure is created in the finite element analysis software, and the geometric model is meshed.

4. The method for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem according to claim 3 is characterized in that: The load applied to the three-cantilever beam identification structure is a forced displacement load.

5. The method for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem according to claim 1, characterized in that: The method of identifying multicollinearity between the crack angle data and the cantilever beam strain data through multivariate visualization and correlation analysis based on the simulation data set and deleting redundant variables to obtain the optimized data set includes the following steps: For the simulation data set, visualization software is used to draw a corresponding relationship diagram between variables, and the determination coefficient is calculated by observing the distribution relationship and trend between each variable and combining it with the multiple regression model; the variables include the response variable vector and the regression variable; The variance inflation factor of the regression variable is calculated based on the coefficient of determination, and whether there is multicollinearity between the regression variables is determined based on the preset threshold. If so, the regression variable is deleted; otherwise, the regression variable is retained. The deleted or retained regression variables are integrated with the response variable vector to obtain an optimized data set.

6. The method for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem according to claim 1, characterized in that: The method of constructing a crack angle prediction model based on an optimized data set and using a variety of algorithms for multi-classification problems includes the following steps: The crack angles in the optimized dataset are treated as discrete variables, and the crack angle range is divided into multiple subclasses according to the preset step size. A classification model is constructed using a variety of algorithms for multi-classification problems. An optimized data set is divided into multiple subclasses according to the crack angle. The performance of the classification model is trained and evaluated to obtain performance evaluation indicators. Based on the performance evaluation indicators, the performance of each classification model in the crack angle prediction task is analyzed, and according to the analysis results, the model with the best performance is selected as the final crack angle prediction model.

7. The method for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem according to claim 6, characterized in that: The method of constructing a classification model using a variety of algorithms for multi-classification problems, dividing an optimized data set into multiple subclasses according to crack angles, training and evaluating the performance of the classification model, and obtaining performance evaluation indicators includes the following steps: The optimized data set is divided into a training set and a test set by dividing the crack angles into multiple subclasses according to a preset division ratio; Using the training set, the extreme gradient boosting classification model, K-nearest neighbor classification model, decision tree classification model, random forest classification model and support vector machine classification model were used respectively; The trained extreme gradient boosting classifier, K-nearest neighbor classifier, decision tree classifier, random forest classifier and support vector machine classifier are tested using the test set, and the performance of each classifier is evaluated based on the test results to obtain the performance evaluation indicators.

8. The method for predicting crack angles of a three-cantilever beam structure based on a multi-classification problem according to claim 7, characterized in that: The performance evaluation indicators include: precision, recall and F1 score.

9. A three-cantilever beam structure crack angle prediction system based on multi-classification problem, characterized by: include: a data acquisition unit configured to perform static simulation of the three-cantilever beam identifier structure using a finite element method and obtain a simulation data set of the three-cantilever beam identifier structure under different crack conditions; the simulation data set includes crack angle data and cantilever beam strain data; apply a load to the three-cantilever beam identifier structure based on a constraint condition of the three-cantilever beam identifier structure, and obtain strain data at the cantilever of the three-cantilever beam identifier structure; the constraint condition of the three-cantilever beam identifier structure is to fix the crack width and change the crack angle; A data optimization unit is used to identify multicollinearity between crack angle data and cantilever beam strain data based on the simulation data set through multivariate visualization and correlation analysis, and to delete redundant variables to obtain an optimized data set; A model building unit is used to build a crack angle prediction model based on an optimized data set and using a variety of algorithms for multi-classification problems; The angle prediction unit is used to obtain the cantilever beam strain data and input it into the crack angle prediction model, and predict the corresponding crack angle data through the crack angle prediction model.

Images