Structural entity detection sampling method

Abstract

The invention discloses a structure entity detection sampling method. The method comprises the following steps: identifying a key force transmission component from a to-be-detected structure based on topological optimization; performing field entity detection on the key force transmission component to obtain detection data; establishing a structural numerical model based on the detection data, and performing structural analysis on the structural numerical model to obtain performance indexes; and carrying out structure reliability analysis based on the performance indexes. According to the method, the accuracy and efficiency of structure detection and identification can be improved, misjudgment caused by unreasonable sampling is reduced, and the detection cost is reduced.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of structural entity detection sampling, and particularly relates to a structural entity detection sampling method. BACKGROUND

[0002] In the field of construction engineering, structural detection and identification is a key link for evaluating the safety, durability and reliability of building structures. At present, the detection sampling of structural members mainly relies on the quantity requirements specified in the specifications, and adopts the method of random sampling. Specifically, the detection personnel randomly selects a certain number of members according to the specifications, and then detects the performance of these sampled members such as material strength and geometric size, and infers the performance state of the overall structure according to the sampling detection results. However, this traditional random sampling method has significant defects. Since the strength of building materials such as concrete itself has randomness, there are inherent differences in the actual performance of each member, and random sampling cannot ensure that the members that are crucial to the overall mechanical performance of the structure are extracted. More importantly, this method completely ignores the mechanical properties of the overall structure and the force transmission path. In practical application, each member in the structure is not equally important, and some members play a key role in load transmission and overall stability, while other members are relatively secondary. The traditional random sampling method lacks scientific theoretical guidance and is difficult to systematically identify and detect these key force transmission members, resulting in that the detection results are heavily dependent on random selection at the time of sampling, and the analysis conclusion is prone to large fluctuations, which cannot truly and stably reflect the actual reliability level of the structure. This uncertainty in the results caused by unreasonable sampling brings the risk of misjudgment to the structural safety evaluation, which may result in the failure to detect unsafe members or unnecessary reinforcement of safe structures, causing resource waste. In engineering practice, attempts have been made to increase the number of samples to alleviate this problem, but this will significantly increase the cost of manpower, material resources and time for detection, and reduce the detection efficiency. Therefore, there has been an urgent need in the field for a long time to develop a detection method that can scientifically identify key members and efficiently sample based thereon, so as to fundamentally improve the accuracy and efficiency of structural detection and identification. SUMMARY

[0003] To solve the above technical problems, the present application provides a structural entity detection sampling method, which can improve the accuracy and efficiency of structural detection and identification, reduce misjudgment caused by unreasonable sampling, and reduce detection cost.

[0004] To achieve the above purpose, the present application provides a structural entity detection sampling method, comprising: identifying key force transmission members from the structure to be detected based on topology optimization; performing on-site entity detection on the key force transmission members to obtain detection data; establishing a structure numerical model based on the detection data, and performing structural analysis on the structure numerical model to obtain performance indexes; performing structural reliability analysis based on the performance indexes.

[0005] Optionally, the identifying the key load transmission members from the structure to be detected based on the topology optimization comprises: establishing a digital model of the structure to be detected, the digital model comprising the number of spans, the number of layers, the node and element numbers, the load, the member cross-section properties, the material initial properties and the boundary constraints; setting topology optimization parameters, setting the volume fraction constraint of the structure according to the required number of members to be detected, and adopting a material interpolation scheme; performing topology optimization solving, using the relative density of the material as the design variable by the variable density method, integrating the global stiffness matrix, solving the displacement response, calculating the objective function and the constraint function, calculating the gradient, updating the design variable and judging the convergence, until the threshold value of the design variable is calculated according to the material volume fraction constraint after the convergence criterion is met, and the members with the material density greater than the threshold value are screened as the key load transmission members.

[0006] Optionally, the setting topology optimization parameters comprises: setting the volume fraction constraint according to the required number of members to be detected, the volume fraction being the ratio of the material volume of the detected members to the total material volume of the members; adopting the SIMP model as the material interpolation scheme, in which the material elastic modulus varies with the relative density and drives the intermediate density value to tend to 0 or 1 through a penalty factor.

[0007] Optionally, the performing topology optimization solving comprises: S1, integrating the global stiffness matrix of the structure based on the current design variable value and the penalty factor; S2, solving the equation set to obtain the global displacement response of the structure under the action of the load; S3, calculating the values of the objective function and the constraint function based on the displacement response and the design variable; S4, calculating the gradient of the objective function and the constraint function with respect to each design variable by a numerical method; S5, updating the design variable using an optimizer based on convex approximation; S6, repeating S1-S5 until the convergence criterion is met.

[0008] Optionally, the obtaining the detection data comprises: obtaining the material strength, the member geometric size, the initial deformation and defects, and the steel bar information of the key load transmission members according to the relevant specification requirements.

[0009] Optionally, the establishing a structure numerical model based on the detection data, and performing structural analysis on the structure numerical model to obtain performance indexes comprises: A model consistent with the structure to be detected is established by using finite element software; The detection data of the key force transmission components are assigned to the corresponding components in the model; For the components not detected, random simulation is adopted to perform probabilistic assignment based on prior information; The structural numerical model is subjected to structural analysis to simulate the performance response of the structure under the action of loads and extract performance indexes.

[0010] Optionally, the structural analysis of the structural numerical model comprises: Performing static or dynamic seismic analysis to simulate the performance response of the structure under the influence of uncertain factors; The performance indexes include the reaction force, displacement, inter-story drift angle and base shear of each node of the structure.

[0011] Optionally, the structural reliability analysis based on the performance indexes comprises: Extracting key performance indexes from the structural analysis results; According to a preset structure failure criterion, the number of structure failures in all simulation samples is counted; Based on the number of failures and the total number of samples, the structure failure probability and the structure reliability are calculated.

[0012] The technical effect of the present application is that a structure entity detection sampling method is disclosed, which identifies key force transmission components based on topology optimization and performs detection, significantly improves the accuracy of structure detection, makes the detection data more representative of the overall mechanical properties of the structure, reduces the misjudgment caused by unreasonable sampling, and thus makes the structure safety evaluation more reliable. At the same time, under the premise of ensuring accuracy, the method can minimize the detection workload, avoid blind detection of a large number of non-key components, effectively shorten the detection period, reduce the cost of manpower, material resources and time, and improve the detection efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0013] The accompanying drawings, which form a part of the present application, are intended to provide further understanding of the present application, and the illustrative embodiments of the present application and their description serve the purpose of explaining the present application. The accompanying drawings do not constitute an inappropriate limitation on the present application. In the drawings: Figure 1 A structure response uncertainty comparison chart of the conventional sampling and optimized sampling method of the embodiments of the present application; Figure 2 A sampling method convergence to true value comparison chart of the embodiments of the present application under different volume fractions; Figure 3 A comprehensive performance comparison chart of the sampling method of the embodiments of the present application in terms of uncertainty convergence and true convergence; Figure 4A flowchart of a structural entity detection sampling method is provided in an embodiment of the present application. DETAILED DESCRIPTION

[0014] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0015] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0016] As shown in the accompanying drawings, Figure 4 a structural entity detection sampling method is provided in the present embodiment, which comprises: identifying key load transmission members from a structure to be detected based on topology optimization; performing field entity detection on the key load transmission members to obtain detection data; establishing a numerical model of the structure based on the detection data, and performing structural analysis on the numerical model of the structure to obtain performance indicators; performing structural reliability analysis based on the performance indicators.

[0017] Further, identifying key load transmission members from a structure to be detected based on topology optimization comprises: establishing a digital model of the structure to be detected, the digital model comprising the number of spans, the number of layers, the node and element numbers, the load, the cross-sectional properties of the members, the initial properties of the materials, and the boundary constraints of the structure; setting topology optimization parameters, setting the volume fraction constraint of the structure according to the required number of member detection, and adopting a material interpolation scheme; performing topology optimization solving, using the variable density method to take the relative density of the material as the design variable, through cyclic iteration to integrate the global stiffness matrix, solve the displacement response, calculate the objective function and the constraint function, calculate the gradient, update the design variable and judge the convergence, until the threshold value of the design variable is calculated according to the material volume fraction constraint after the convergence criterion is met, and the members with a material density greater than the threshold value are selected as key load transmission members.

[0018] Further, setting topology optimization parameters comprises: setting the volume fraction constraint according to the required number of detection, the volume fraction being the ratio of the material volume of the detected members to the total material volume of the members; adopting a SIMP model as the material interpolation scheme, in which the elastic modulus of the material changes with the relative density and drives the intermediate density value to tend to 0 or 1 through a penalty factor.

[0019] Specifically, the implementation process of this embodiment includes: According to the specifications, the volume fraction constraint of the structure (volume of inspected component material / total volume of component material) is set based on the number of component inspections. The SIMP model is used as the material interpolation scheme, and its mathematical expression is as follows: ; in, It is the interpolated elastic modulus; It is the minimum elastic modulus (to avoid numerical singularities); It is the relative density, which is used as a design variable and its value ranges from 0 to 1. p It is a penalty factor (usually greater than 1, which can make the intermediate density value tend to 0 or 1). It is the original elastic modulus of the material.

[0020] Furthermore, performing topology optimization solutions includes: S1. The global stiffness matrix of the integrated structure is based on the current design variable values ​​and the penalty factor; S2. Solve the system of equations to obtain the global displacement response of the structure under load; S3. Calculate the values ​​of the objective function and constraint function based on the displacement response and design variables; S4. Use numerical methods to calculate the gradients of the objective function and constraint functions with respect to each design variable; S5. Update design variables using an optimizer based on convex approximation; S6. Repeat S1-S5 until the convergence criterion is met.

[0021] Specifically, the implementation process of this embodiment includes: The topology optimization of the structure is performed using the variable density method, treating the relative density of the material as a continuous design variable between 0 and 1. When the relative density is 0, the component is discarded; when it is 1, the component is retained. An iterative loop is used to solve the optimization problem, with the penalty factor of the SIMP model gradually increasing in each iteration. p This makes the intermediate density between 0 and 1 more inclined towards 0 or 1. Repeat this process 10 times: Stiffness matrix integration: based on current design variable values ​​and penalty factors for the current stage. p The global stiffness matrix of the entire structure is obtained by integration. ; Solve the system of equations The structure under load is obtained. F Global displacement response under action U .

[0022] Based on displacement response U Given the current design variables, calculate the current values ​​of the objective function and constraint functions.

[0023] Numerical methods (such as the finite difference method) are used to calculate the gradients of the objective function and constraint functions with respect to each design variable.

[0024] Design variable update: Input the function values ​​and their gradients calculated in the previous steps into an optimizer based on convex approximation (such as the moving asymptote method) to solve an approximate subproblem, thereby obtaining a set of updated and better design variables.

[0025] Convergence check: until the convergence criterion (such as the KKT optimality condition) is met, otherwise return and re-enter the loop.

[0026] Finally, the method of calculating the threshold of design variables based on the material volume fraction constraint can be used to select components with material density greater than the threshold, and obtain the optimal structural form under the required number of components, i.e., the key components.

[0027] Furthermore, the acquisition of test data includes: obtaining the material strength, component geometry, initial deformation and defects, and reinforcement information of key force transmission components in accordance with relevant specifications.

[0028] Furthermore, based on the detection data, a structural numerical model is established, and structural analysis is performed on the structural numerical model to obtain performance indicators, including: A model consistent with the structure to be tested is established using finite element software; Assign the detection data of key force transmission components to the corresponding components in the model; For components that were not detected, random simulation was used to assign probabilistic values ​​based on prior information; Structural analysis is performed on the numerical model of the structure to simulate the performance response of the structure under load and to extract performance indicators.

[0029] Specifically, the implementation process of this embodiment includes: A model is built using finite element software, and a consistent model is established for the structure. The real data of the detected key components are assigned to the corresponding components in the model. For the remaining components that were not detected, their material properties and other parameters are assigned probabilistically using random simulation (such as Monte Carlo method) based on prior information (such as material factory data, specifications, etc.).

[0030] Furthermore, structural analysis of the numerical model of the structure includes: Perform static or dynamic seismic analysis to simulate the performance response of the structure under the influence of uncertain factors; The extracted performance indicators include the reaction force, displacement, inter-story drift angle, and base shear force at each node of the structure.

[0031] Specifically, the implementation process of this embodiment includes: Structural analysis, such as static or dynamic seismic analysis, is performed on the established probabilistic model to simulate the performance response of the structure under the influence of various uncertain factors.

[0032] Furthermore, structural reliability analysis based on the aforementioned performance indicators includes: Extract key performance indicators from structural analysis results; The number of structural failures in all simulation samples was counted according to the preset structural failure criteria. The structural failure probability and structural reliability are calculated based on the number of failures and the total number of samples.

[0033] Specifically, the implementation process of this embodiment includes: Key performance indicators are extracted from the simulation results, such as the reaction force, displacement, inter-story drift angle, and base shear force at each node of the structure.

[0034] The statistical distribution of performance indicators was analyzed, and the number of structural failures in all simulation samples was counted based on preset structural failure criteria (e.g., inter-story drift angle exceeding the limit). n .

[0035] Number of times n Perform statistical analysis to calculate structural reliability: ; ; in, This represents the probability of structural failure. n The number of failures. N For the total number, For structural reliability.

[0036] An application example of this invention: Step 1: Establish and analyze the original structural model. 1.1 A two-dimensional planar frame structure with 10 floors and 8 spans is set as the research object.

[0037] 1.2 Assign material properties to all components of the model. Assume that the concrete strength of each component is consistent. Use the Burr distribution to simulate the random distribution of concrete strength. Assign material properties to each component and apply loads according to the specifications.

[0038] 1.3 Use finite element software (such as OpenSees) to perform pushover analysis on the benchmark model, and record its displacement-base reaction curve and maximum base reaction, etc., as a performance standard for subsequent comparison.

[0039] Step 2: Structural Optimization 2.1 Based on the original structure, establish a digital model, including defining the number of spans, number of stories, node and unit numbers, loads, component section properties, initial material properties, boundary constraints, and degrees of freedom of the structure.

[0040] 2.2 Setting the material volume fraction: In this embodiment, we calculated nine cases with volume fractions ranging from 10% to 90%, and set the objective function as the total strain energy of the structure.

[0041] 2.2 Run the topology optimization program to calculate the optimal structural layout under different volume fraction constraints from 10% to 90%, and record and save the results for the next step of analysis.

[0042] Step 3: Construct and analyze the "optimized sampling group" and the "traditional sampling control group". 3.1 Optimize the sampling group: 3.1.1 Based on the optimization results of step 2, select the key components for the corresponding cases.

[0043] 3.1.2 Establish 1000 analysis models using OpenSees. In each model, assign the selected key components the original structural component attributes obtained in step 1; the attributes of the remaining unselected components are then randomly assigned using the Burr distribution.

[0044] 3.1.3 Pushover analysis was performed on each of the 1000 models.

[0045] 3.2 Traditional sampling control group: 3.2.1 Randomly select the same number of components as the optimized sampling group.

[0046] 3.2.2 Using the same method as the optimized sampling group, 1000 models were established. Randomly selected components were assigned the corresponding component attributes of the original structure, and the remaining components were randomly assigned values ​​and Pushover analysis was performed.

[0047] Step 4: Statistical analysis and comparative verification of the results. 4.1 Data extraction: The base node reaction forces of the optimized sampling group and the traditional control group were statistically analyzed, and the displacement-base reaction force curve and the maximum base reaction force of each model were calculated.

[0048] 4.2 Comparative Analysis: Calculate and compare the standard deviation and variance of the maximum basal reaction force results for the optimized sampling group and the traditional control group.

[0049] 4.3 Compare the statistical results of the two groups with the performance data of the “real” structure obtained in step one.

[0050] 4.4 Conclusion: By comparing the data dispersion (standard deviation / variance), the superiority of the topology optimization-based sampling method over the traditional random sampling method in reducing the uncertainty of structure detection is verified, as well as the reliability of its evaluation results with the real data.

[0051] like Figures 1-3 The results show that the topology-optimized sampling method outperforms traditional sampling methods in terms of both uncertainty (reflected by the standard deviation) and accuracy (reflected by comparison with real structural data) of the maximum base reaction force. As the volume fraction increases, the convergence speed of the data results in terms of uncertainty and convergence to the true value is significantly faster than that of traditional methods, and the difference widens with increasing volume fraction.

[0052] This invention presents a sampling method for frame structure inspection based on topology optimization. By leveraging a topology optimization algorithm, it can accurately identify key force-transmitting components. Using these components as sampling points makes the inspection data more representative of the overall mechanical performance of the structure, significantly improving the accuracy of the inspection results. Compared to traditional methods, it reduces misjudgments caused by unreasonable sampling, making structural safety assessments more reliable and providing solid data support for subsequent structural maintenance and reinforcement decisions. Secondly, inspection efficiency is greatly improved. Traditional sampling may blindly inspect a large number of components, consuming significant manpower, resources, and time. This invention, while ensuring accuracy, minimizes the workload, shortens the inspection cycle, and reduces inspection costs.

[0053] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A sampling method for structural entity detection, characterized in that, include: Key force-transmitting components are identified from the structure to be tested based on topology optimization; On-site physical testing was conducted on the key force transmission components to obtain test data; A structural numerical model is established based on the detection data, and structural analysis is performed on the structural numerical model to obtain performance indicators. Structural reliability analysis is performed based on the aforementioned performance indicators.

2. The structural entity inspection sampling method as described in claim 1, characterized in that, Key force-transmitting components identified from the structure under test based on topology optimization include: A digital model of the structure to be inspected is established. The digital model includes the number of spans, number of stories, node and unit numbers, loads, component cross-sectional properties, initial material properties, and boundary constraints of the structure. Set topology optimization parameters, set volume fraction constraints for the structure according to the number of component inspections required by the specification, and adopt a material interpolation scheme; The topology optimization solution is performed, and the relative density of the material is used as the design variable by using the variable density method. The global stiffness matrix is ​​integrated through iterative loops, the displacement response is solved, the objective function and constraint function are calculated, the gradient is calculated, the design variables are updated and the convergence is judged until the convergence criterion is met. The threshold of the design variables is calculated based on the material volume fraction constraint, and the components with material density greater than the threshold are selected as key force transmission components.

3. The structural entity inspection sampling method as described in claim 2, characterized in that, Setting topology optimization parameters includes: The volume fraction constraint is set according to the number of tests required by the specification, whereby the volume fraction is the ratio of the volume of the tested component material to the total volume of the component material; The SIMP model is used as the material interpolation scheme, in which the elastic modulus of the material varies with the relative density and drives the intermediate density value to tend to 0 or 1 through a penalty factor.

4. The structural entity inspection sampling method as described in claim 2, characterized in that, Performing topology optimization solutions includes: S1. The global stiffness matrix of the integrated structure is based on the current design variable values ​​and the penalty factor; S2. Solve the system of equations to obtain the global displacement response of the structure under load; S3. Calculate the values ​​of the objective function and constraint function based on the displacement response and design variables; S4. Use numerical methods to calculate the gradients of the objective function and constraint functions with respect to each design variable; S5. Update design variables using an optimizer based on convex approximation; S6. Repeat S1-S5 until the convergence criterion is met.

5. The structural entity inspection sampling method as described in claim 1, characterized in that, The acquisition of test data includes: obtaining the material strength, geometric dimensions, initial deformation and defects, and reinforcement information of key force transmission components in accordance with relevant specifications.

6. The structural entity inspection sampling method as described in claim 1, characterized in that, Based on the detection data, a structural numerical model is established, and structural analysis is performed on the structural numerical model to obtain performance indicators, including: A model consistent with the structure to be tested is established using finite element software; Assign the detection data of key force transmission components to the corresponding components in the model; For components that were not detected, random simulation was used to assign probabilistic values ​​based on prior information; Structural analysis is performed on the numerical model of the structure to simulate the performance response of the structure under load and to extract performance indicators.

7. The structural entity inspection sampling method as described in claim 6, characterized in that, Structural analysis of the numerical model of the structure includes: Perform static or dynamic seismic analysis to simulate the performance response of the structure under the influence of uncertain factors; The extracted performance indicators include the reaction force, displacement, inter-story drift angle, and base shear force at each node of the structure.

8. The structural entity inspection sampling method as described in claim 1, characterized in that, Structural reliability analysis based on the aforementioned performance indicators includes: Extract key performance indicators from structural analysis results; The number of structural failures in all simulation samples was counted according to the preset structural failure criteria. The structural failure probability and structural reliability are calculated based on the number of failures and the total number of samples.

Images