RC beam column joint failure form multi-parameter discrimination method, system, equipment and medium

Abstract

The invention discloses a multi-parameter distinguishing method, system and equipment for the failure form of an RC beam column joint and a medium, and belongs to the technical field of civil engineering. The method comprises the steps that design parameters and failure forms of RC beam-column joints are obtained, and an original sample set is established; fisher transformation is carried out on the original sample set to obtain a new sample set with low-dimensional features; performing classification function solution on the new sample set through a Bayes classification principle; establishing a multi-parameter classification discrimination equation corresponding to the classification result and the design parameters; and inputting the design parameters of the RC beam-column joint to be discriminated into the multi-parameter classification discrimination equation, and taking the failure form with the maximum value as the failure form of the RC beam-column joint to be discriminated. An existing analysis method only has a single parameter or two parameters, is used for dividing the failure forms of shear walls, beams and column components and is not suitable for RC beam column joints, the invention provides the RC beam column joint failure form multi-parameter judgment method based on Fisher transformation and the Bayes classification principle for the first time, the difficulty of multi-parameter analysis and division is effectively reduced, and the accuracy is improved.

Description

Technical Field

[0001] This invention relates to the field of civil engineering technology, and more specifically, to a multi-parameter method, system, device, and medium for determining the failure mode of RC beam-column joints. Background Technology

[0002] The various failure modes of reinforced concrete (RC) beam-column joints under lateral loads have different effects on structural performance. Therefore, accurately classifying the failure modes of members is the key to determining the deformation performance limits of members in structural performance design.

[0003] Currently, the classification of component failure modes is mainly done in two ways: One method is to classify the failure modes of components based on test results. This method is not predictive and cannot be directly applied to performance design.

[0004] Another approach combines experimental and finite element analysis methods to study the failure patterns of structural members, aiming to accurately and reasonably predict their failure modes based on design parameters. For example, a study on the failure mode classification of RC beams proposed a two-parameter criterion using the beam's shear span ratio and bending-shear ratio as parameters. While these methods demonstrate high accuracy in distinguishing between bending and shear failure in RC beams, their accuracy is lower when classifying beams exhibiting bending-shear failure. This is primarily because relying on single-parameter or two-parameter analysis for member identification and classification can lead to misclassification and omissions.

[0005] Using multi-parameter analysis to identify and classify the failure modes of structural members can improve the accuracy of the classification. However, considering the increase in parameters significantly increases the difficulty of the classification method. Therefore, existing analysis methods only offer single-parameter or two-parameter methods, and all are designed for the classification of failure modes in shear walls, beams, and columns. Currently, there is no effective method for classifying RC beam-column joints. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to address the above-mentioned shortcomings of the prior art. The purpose of the present invention is to provide a multi-parameter discrimination method for failure modes of RC beam-column joints, which can reduce the difficulty of multi-parameter analysis and improve the accuracy.

[0007] The second objective of this invention is to provide a multi-parameter discrimination system for the failure mode of RC beam-column joints.

[0008] The third objective of this invention is to provide a computer device.

[0009] The fourth objective of this invention is to provide a computer storage medium.

[0010] To achieve the first objective mentioned above, this invention provides a multi-parameter method for determining the failure mode of an RC beam-column joint, comprising the following steps: Step 1. Obtain the design parameters and failure modes of multiple RC beam-column joints to establish an original sample set; Step 2. Perform Fisher transform on the original sample set to obtain a new sample set with low-dimensional features; Step 3. Solve the classification function for the new sample set using the Bayes classification principle; Step 4. Establish a multi-parameter classification discriminant equation corresponding to the classification results and design parameters; Step 5. Input the design parameters of the RC beam-column joint to be identified into the multi-parameter classification and discrimination equation, and take the failure mode corresponding to the multi-parameter classification and discrimination equation with the largest value as the failure mode of the RC beam-column joint to be identified.

[0011] As a further improvement, the Fisher transform is: Calculate the within-group deviation matrix :

[0012] In the formula, c The total number of categories, i ∈1~ c ; S i Let be the intra-class deviation matrix of the i-th class; m i For the first i The mean vector of the class samples, x This is the original sample; D i To belong to category Sample subset; Calculate the inter-group deviation matrix :

[0013] In the formula, n i For the first i The number of samples in each class; m This is the vector of the overall mean of all samples; Calculate the transformation matrix , so that the criterion function Take the maximum value:

[0014] Direct the original sample set to Projection minimizes the differences within the same class and maximizes the differences between different classes to obtain a new sample set. : .

[0015] Furthermore, the Bayes discriminant function is:

[0016] If > For everything If established, then Return to kind.

[0017] Furthermore, the multi-parameter classification discriminant equation is:

[0018] In the formula, For the first A multi-parameter classification and discrimination equation, i ∈1~ c , For the first One design parameter, These are the coefficients corresponding to the design parameters. It is a constant.

[0019] Furthermore, the category conditional probability of each damaged morphology sample is calculated using the Bayes formula for the new sample set:

[0020] In the formula, For class conditional probability density; Representing the Prior probabilities of class states; For feature vectors x The marginal probability density; Combined with the formula Find the Bayes classification function; Finally, the coefficients corresponding to the design parameters are calculated based on the conditional probabilities of each failure mode sample and the Bayes classification function fitting.

[0021] Furthermore, the design parameters include the axial compression ratio. Shear-compression ratio Concrete strength and stirrup characteristic value The failure modes include beam end failure, node core area shear failure, beam end yield core area shear failure, and column end yield core area shear failure.

[0022] Furthermore, let the axial compression ratio correspond shear-compression ratio correspond Concrete strength correspond , characteristic value of stirrup correspond ; Cause beam end failure Shear failure in the core area of ​​the node corresponds to Shear failure in the yield core region at the beam end Shear failure in the core region at the column tip yielding zone corresponds to ; The multi-parameter classification discriminant equation is obtained:

[0023]

[0024]

[0025] .

[0026] To achieve the second objective mentioned above, this invention provides a multi-parameter discrimination system for the failure mode of RC beam-column joints, comprising: The acquisition module is used to acquire the design parameters and failure modes of multiple RC beam-column joints and establish an original sample set; The Fisher transform module is used to perform a Fisher transform on the original sample set to obtain a new sample set with low-dimensional features; The Bayes classification and discrimination module is used to solve the classification function for the new sample set based on the Bayes classification principle. The equation building module is used to build multi-parameter classification and discrimination equations corresponding to the classification results and design parameters. The prediction and discrimination module is used to input the design parameters of the RC beam-column joint to be judged into the multi-parameter classification and discrimination equation, and take the failure mode corresponding to the multi-parameter classification and discrimination equation with the largest value as the failure mode of the RC beam-column joint to be judged.

[0027] To achieve the above-mentioned objective three, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-mentioned method for multi-parameter discrimination of failure modes of RC beam-column joints.

[0028] To achieve the fourth objective mentioned above, the present invention provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-mentioned method for multi-parameter discrimination of failure modes of RC beam-column joints.

[0029] Beneficial effects Compared with the prior art, the advantages of this invention are as follows: 1. Existing analysis methods only offer single-parameter or two-parameter approaches (multi-parameter analysis is too difficult), and are all designed for the failure mode classification of shear walls, beams, and columns, not applicable to RC beam-column joints. This invention, for the first time, proposes a Bayes classification method based on Fisher transform, which considers the influence of the interrelationships between different parameters on failure mode determination. It fully utilizes the information from each parameter, providing a mathematically rigorous classification method. This offers a mathematical and quantitative approach to classifying the failure modes of beam-column joints, ensuring the reliability of the results. The classification and discrimination results obtained using the Fisher transform and Bayes classification method have significant statistical meaning, providing a reliable theoretical basis for the classification of beam-column joint failure modes. Furthermore, it significantly reduces the difficulty of multi-parameter analysis and, compared to traditional single- and two-parameter analyses, greatly improves the accuracy of multi-parameter discrimination.

[0030] 2. Based on the selection of four parameter combinations—axial compression ratio, shear compression ratio, concrete strength, and stirrup characteristic value—this invention establishes a corresponding multi-parameter classification and discrimination equation, which can quickly determine the failure mode of beam-column joint members, achieving the goal of better dividing the failure modes of beam-column joints and clarifying the intervals corresponding to different failure modes. Attached Figure Description

[0031] Figure 1 This is a flowchart of the present invention. Detailed Implementation

[0032] The present invention will be further described below with reference to specific embodiments shown in the accompanying drawings.

[0033] See Figure 1 A multi-parameter method for determining the failure mode of RC beam-column joints includes the following steps: Step 1. Obtain the design parameters and failure modes of multiple RC beam-column joints and establish an original sample set; the original sample set can be obtained through earthquake damage investigation, experimental research, or finite element analysis, etc. Step 2. Perform Fisher transform on the original sample set to obtain a new sample set with low-dimensional features; Step 3. Solve the classification function for the new sample set using the Bayes classification principle; Step 4. Establish a multi-parameter classification discriminant equation corresponding to the classification results and design parameters; Step 5. Input the design parameters of the RC beam-column joint to be identified into the multi-parameter classification and discrimination equation, and take the failure mode corresponding to the multi-parameter classification and discrimination equation with the largest value as the failure mode of the RC beam-column joint to be identified.

[0034] Specifically, the Fisher transformation is: (1) Calculate the within-group deviation matrix :

[0035] In the formula, c The total number of categories, i ∈1~ c ; S i Let be the intra-class deviation matrix of the i-th class; m i For the first i The mean vector of the class samples, x This is the original sample; D i To belong to category Sample subset. Training samples are ( for 3D row vectors > They belong to, respectively There are 10 different categories, that is, the size of which is 1000. sample subset Category .

[0036] (2) Calculate the inter-group deviation matrix :

[0037] In the formula, n i For the first i The number of samples in each class; m This is the vector of the overall mean of all samples.

[0038] (3) Calculate the transformation matrix , so that the criterion function Take the maximum value: Criterion function: In the formula, W : Projection direction vector, used to project the original high-dimensional sample into a low-dimensional space; W T BW : Projected between-group deviations (scalar); W T S W W : Within-group deviation after projection (scalar).

[0039]

[0040] (4) Move the original sample set to Projection minimizes the differences within the same class and maximizes the differences between different classes, resulting in a new sample set. : .

[0041] The original sample is transformed by steps (1), (2), and (3) so that it is projected into a new sample space. The new sample set after projection is obtained according to step (4), thereby converting the representation mode in the high-dimensional measurement space into the mode in the low-dimensional feature space, completing a key step.

[0042] The Bayes discriminant function is:

[0043] If > For everything If established, then Return to kind. For the first i There are 10 categories, totaling 100 categories. c indivual.

[0044] The coefficient vector of the first-order term, also known as the weight vector, represents the direction of the normal vector of the discriminant hyperplane in the feature space after the inverse matrix transformation of the covariance matrix.

[0045] : constant term, from the first i Prior probability of a class p(w i ) It consists of terms such as the mean vector and the logarithm of the covariance matrix, and is related to... x Irrelevant constants.

[0046] The multi-parameter classification discriminant equation is:

[0047] In the formula, For the first A multi-parameter classification and discrimination equation, i ∈1~ c , For the first One design parameter, These are the coefficients corresponding to the design parameters. It is a constant.

[0048] Calculate the coefficients corresponding to the design parameters ( )as follows: (1) First, calculate the conditional probability of each damaged morphology sample using Bayes' formula on the new sample set:

[0049] In the formula, For class conditional probability density; Representing the Prior probabilities of class states; For feature vectors x The marginal probability density; (2) Combine with the formula Find the Bayes classification function; (3) Finally, the coefficients corresponding to the design parameters are calculated based on the conditional probabilities of each failure mode sample and the Bayes classification function. The coefficients corresponding to the design parameters can be calculated using SPSS data analysis software or EXCEL software.

[0050] Preferably, the design parameters include the axial compression ratio. Shear-compression ratio Concrete strength and stirrup characteristic value The failure modes include beam end failure, joint core zone shear failure, beam end yield core zone shear failure, and column end yield core zone shear failure.

[0051] Let the axial compression ratio correspond shear-compression ratio correspond Concrete strength correspond , characteristic value of stirrup correspond ; Cause beam end failure Shear failure in the core area of ​​the node corresponds to Shear failure in the yield core region at the beam end Shear failure in the core region at the column tip yielding zone corresponds to .

[0052] The multi-parameter classification discriminant equation is obtained:

[0053]

[0054]

[0055] .

[0056] In one embodiment, four failure modes—beam-end failure, shear failure, beam-end yield-shear failure, and column-end yield-shear failure—are defined as categories 1-4, respectively. After classifying the failure modes of 206 beam-column nodes through steps 1-3, a sample dataset of beam-column nodes with known categories is obtained. Of these, 56 samples belong to category 1, 30 to category 2, 96 to category 3, and 24 to category 4. The mean vectors for each category are denoted as follows: m i ( i =1,2,3,4).

[0057] The coefficients corresponding to the calculated design parameters are shown in Table 1.

[0058]

[0059] Table 1 axial compression ratio will be used Shear-compression ratio Concrete strength and stirrup characteristic value The classification functions corresponding to the four failure modes characterized by the four parameters are expressed as follows:

[0060]

[0061]

[0062]

[0063] It can be used to quickly determine the failure mode of joint members in beams and columns. After substituting into the calculation, the following results are obtained: Four function values ​​are used to make classification decisions; that is, the sample is determined to belong to a category based on the magnitude of the obtained classification function value. If the obtained... The maximum value indicates beam end failure. The maximum value indicates shear failure. The maximum value indicates beam end yield shear failure. The maximum value indicates column end yield shear failure.

[0064] A multi-parameter discrimination system for failure modes of RC beam-column joints includes: The acquisition module is used to acquire the design parameters and failure modes of multiple RC beam-column joints and establish an original sample set; The Fisher transform module is used to perform a Fisher transform on the original sample set to obtain a new sample set with low-dimensional features; The Bayes classification and discrimination module is used to solve the classification function for the new sample set based on the Bayes classification principle. The equation building module is used to build multi-parameter classification and discrimination equations corresponding to the classification results and design parameters. The prediction and discrimination module is used to input the design parameters of the RC beam-column joint to be judged into the multi-parameter classification and discrimination equation, and take the failure mode corresponding to the multi-parameter classification and discrimination equation with the largest value as the failure mode of the RC beam-column joint to be judged.

[0065] A computer device includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the above-mentioned multi-parameter discrimination method for failure modes of RC beam-column joints.

[0066] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the above-mentioned multi-parameter discrimination method for failure modes of RC beam-column joints.

[0067] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention, and these will not affect the effectiveness of the implementation of the present invention or the practicality of the patent.

Claims

1. A method for judging the failure mode of an RC beam-column joint, characterized in that, Includes the following steps: Step 1. Obtain the design parameters and failure modes of multiple RC beam-column joints to establish an original sample set; Step 2. Perform Fisher transform on the original sample set to obtain a new sample set with low-dimensional features; Step 3. Solve the classification function for the new sample set using the Bayes classification principle; Step 4. Establish a multi-parameter classification discriminant equation corresponding to the classification results and design parameters; Step 5. Input the design parameters of the RC beam-column joint to be identified into the multi-parameter classification and discrimination equation, and take the failure mode corresponding to the multi-parameter classification and discrimination equation with the largest value as the failure mode of the RC beam-column joint to be identified.

2. The multi-parameter discrimination method for failure modes of RC beam-column joints according to claim 1, characterized in that, The Fisher transform is: Computing the within-group dispersion matrix : In the formula, c The total number of categories, i ∈1~ c ; S i Let be the intra-class deviation matrix of the i-th class; m i For the first i The mean vector of the class samples, x This is the original sample; D i To belong to category Sample subset; Computing an inter-group dispersion matrix : wherein n i is the number of samples of the first i class; m is the total mean vector of all samples; Computing the transformation matrix maximizing the criterion function taking the maximum Direct the original sample set to Projection minimizes the differences within the same class and maximizes the differences between different classes, resulting in a new sample set. : 。 3. The method of claim 2, wherein, The Bayes discriminant function is: If > For everything If established, then Return to kind.

4. The method of claim 1, wherein, The multi-parameter classification discriminant equation is: In the formula, For the first A multi-parameter classification and discrimination equation, i ∈1~ c , For the first One design parameter, These are the coefficients corresponding to the design parameters. It is a constant.

5. The method of claim 4, wherein, The conditional probability of each damaged morphology sample is calculated using Bayes' formula for the new sample set: where is the class conditional probability density; represents the first prior probability of the class state; is the marginal probability density of the feature vector x; Recombining the formula The Bayes classification function is found; Finally, the coefficients corresponding to the design parameters are calculated based on the conditional probabilities of each failure mode sample and the Bayes classification function.

6. The method of claim 4, wherein the method is characterized by: The design parameters include axial compression ratio. Shear-compression ratio Concrete strength and stirrup characteristic value The failure modes include beam end failure, node core area shear failure, beam end yield core area shear failure, and column end yield core area shear failure.

7. The method of claim 6, wherein, Let the axial compression ratio correspond shear-compression ratio correspond Concrete strength correspond , characteristic value of stirrup correspond ; Beam end failure corresponds to Node core zone shear failure corresponds to Beam end yield core zone shear failure corresponds to Column end yield core zone shear failure corresponds to ; The multi-parameter classification discriminant equation is obtained: 。 8. An RC beam-column joint failure mode multi-parameter discrimination system, characterized in that, include: The acquisition module is used to acquire the design parameters and failure modes of multiple RC beam-column joints and establish an original sample set; The Fisher transform module is used to perform a Fisher transform on the original sample set to obtain a new sample set with low-dimensional features; The Bayes classification and discrimination module is used to solve the classification function for the new sample set based on the Bayes classification principle. The equation building module is used to build multi-parameter classification and discrimination equations corresponding to the classification results and design parameters. The prediction and discrimination module is used to input the design parameters of the RC beam-column joint to be judged into the multi-parameter classification and discrimination equation, and take the failure mode corresponding to the multi-parameter classification and discrimination equation with the largest value as the failure mode of the RC beam-column joint to be judged. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. When the processor executes the computer program, it implements the multi-parameter discrimination method for failure modes of RC beam-column joints as described in any one of claims 1-7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by the processor, it implements the multi-parameter discrimination method for failure mode of RC beam-column joint as described in any one of claims 1-7.

Images