Support steel frame vulnerability analysis method and system based on learning algorithm analysis

Abstract

The invention discloses a support steel frame vulnerability analysis method and system based on learning algorithm analysis, and relates to the technical field of support steel frame analys.The support steel frame design drawing with historical detection records is obtained and analyzed, a structure expression model is constructed, vulnerable blocks are divided, and structure parameters are configured; the incidence relation between the damage characteristics of the supporting steel frame and time and earthquake degrees is analyzed, and a damage dynamic expression model is established; performing similarity analysis on all the support steel frame structure expression models, performing classification to obtain a model set, and extracting a mark feature factor group; and in the actual vulnerability analysis, determining a corresponding support steel structure expression model set according to the matching condition of the support steel frame and the mark characteristic factor group, and performing vulnerability evaluation by utilizing a damage dynamic expression model in the model set. According to the technical scheme, the accuracy and efficiency of vulnerability evaluation are improved, and a scientific and effective technical means is provided for design, evaluation and maintenance of the supporting steel frame.

Description

Technical Field

[0001] The present invention relates to the technical field of supporting steel frame analysis, and in particular to a supporting steel frame vulnerability analysis method and system based on learning algorithm analysis. Background Art

[0002] In the field of building structural engineering, the support steel frame is a key component that bears and transmits loads. Its stability and safety are crucial to the durability of the entire structure. However, in practical applications, the support steel frame is often faced with a variety of complex external environments and load conditions, such as earthquakes and fatigue effects in long-term use. These factors may cause the performance of the steel frame structure to deteriorate or even be damaged. Traditional support steel frame design and evaluation methods mostly rely on empirical formulas, static analysis, and simple material mechanics tests. Although these methods can ensure the basic safety of the structure to a certain extent, it is difficult to comprehensively and accurately predict and evaluate the vulnerability of the steel frame under various complex working conditions.

[0003] In particular, for large and complex supporting steel frame structures, their stress states and damage modes are often highly nonlinear and uncertain, and traditional analysis methods often have difficulty capturing these subtle and critical changes. In addition, as the structure ages, factors such as material aging and environmental corrosion will gradually affect the performance of the steel frame, and these long-term effects are often simplified or ignored in traditional evaluation methods. Summary of the invention

[0004] The purpose of the present invention is to provide an analysis method and system capable of accurately analyzing the vulnerability of a supporting steel frame.

[0005] The present invention discloses a method for analyzing the vulnerability of a supporting steel frame based on a learning algorithm analysis, comprising: Obtain several supporting steel frame design drawings with historical inspection records, analyze each supporting steel frame design drawing, construct a supporting steel frame structure expression model, divide the supporting steel frame structure expression model into vulnerable blocks, and configure structural parameters for each vulnerable block; Analyze the historical detection records corresponding to each supporting steel frame structure expression model, determine the damage characteristics of the supporting steel frame in different time segments, and establish the corresponding relationship between the damage characteristics and the time segments; analyze the historical earthquake data, determine the earthquake severity parameters in different time segments, and establish the temporal correlation between the damage characteristics and the earthquake severity parameters based on the equivalent correspondence of the time segments; Analyze the damage characteristics, and based on the analysis results, adjust the damage expression of the corresponding supporting steel frame structure expression model, and based on the change of the damage characteristics over time, obtain the first dynamic expression of the supporting steel frame structure expression model, and based on the temporal correlation between the damage characteristics and the earthquake severity parameter, establish the temporal correlation between the earthquake severity parameter change and the first dynamic expression, and obtain the supporting steel structure damage dynamic expression model; Perform structural similarity analysis on all supporting steel frame structure expression models, and classify the supporting steel frame structure expression models based on the analysis results to obtain several supporting steel frame structure expression model sets, and perform signature feature factor analysis on each supporting steel frame structure expression model set to obtain several signature feature factor groups, and establish a correlation between the signature feature factor group and the supporting steel structure expression model set; When conducting vulnerability analysis on the supporting steel frame, the supporting steel structure expression model set corresponding to the supporting steel frame is determined based on the matching of different sign characteristic factor groups with the supporting steel frame, and the vulnerability of the supporting steel structure is expressed based on the supporting steel structure damage dynamic expression model corresponding to the supporting steel structure expression model set.

[0006] In some embodiments disclosed in the present invention, a method for dividing a supporting steel frame structure expression model into vulnerable blocks includes: Finite element analysis is performed on the supporting steel frame structure expression model, including meshing, dividing it into several model units, establishing the association relationship between adjacent model units, and configuring material properties for the model units, including elastic modulus, Poisson's ratio and density; Apply boundary conditions to the support structure expression model after meshing, and apply conventional loads to the support structure expression model after meshing; Based on historical earthquake data, the input parameters of the finite element analysis are determined, and the finite element results of the supporting steel frame structure expression model are obtained. Based on the finite element analysis results, the blocks on the supporting steel frame structure expression model where the stress is greater than or equal to the preset range are determined and recorded as damage concern blocks. Based on the type of structure of the damage concern blocks, the scope of the damage concern blocks is determined to obtain the vulnerable blocks.

[0007] In some embodiments disclosed in the present invention, the method for determining the damage characteristics of the supporting steel frame in different time periods includes: Analyze historical inspection records to determine the damage characteristics of different damage characteristic types, and establish damage characteristic comparison rules based on the structural impact of each damage characteristic type on the supporting steel frame. Based on the damage characteristic comparison rules, determine the negative structural impact parameters of the damage characteristics of different damage characteristic types in different vulnerable blocks; The negative structural impact parameters of different damage feature types corresponding to each vulnerable block are counted and configured in the vulnerable expression component of each vulnerable block.

[0008] In some embodiments disclosed in the present invention, a method for adjusting the damage expression of a supporting steel frame structure expression model includes: The supporting steel frame structure expression model is analyzed, and a parallel central virtual line is set for the connecting rod of the supporting steel frame structure expression model, and a number of perpendicular parametric expression circles are made relative to the central virtual line, and the center of the parametric expression circle coincides with the central virtual line, and a parametric expression circle is set for the intersection of the central virtual line; Based on the negative structural impact parameters of the corresponding vulnerable block, the diameter of the parametric expression circle or parametric expression body is determined: ; in, is the diameter of a circle or a body expressed parametrically, is the diameter conversion adjustment factor, is the negative structural influence parameter, is the adjustment coefficient of the negative structural impact parameter, Adjustment constant for negative structural influence parameters.

[0009] In some embodiments disclosed in the present invention, a method for performing structural similarity analysis on all supporting steel frame structure expression models includes: A structural analysis space is established for the supporting steel frame structure expression model, and a number of spatial nodes are evenly set in the structural analysis space, and the spatial nodes mapped on the supporting steel frame structure expression model are triggered and marked as trigger mark points; Randomly select a number of trigger mark points and randomly combine them to obtain a number of trigger mark point groups, analyze the distances between the trigger mark points in each trigger mark point group, calculate the average value of the distances between the points, and record it as the distance between the mark points; A trigger mark point is randomly selected and recorded as an analysis trigger mark point. The distance between the trigger mark points next to the analysis trigger mark point is determined, and the average value of the distance between the points is calculated, which is recorded as the average distance between the points. The ratio of the average distance between the points to the distance between the marking points is calculated, which is recorded as the reference ratio of the distance between the points. If the reference ratio of the distance between the points is greater than or equal to the preset value, the corresponding analysis trigger mark point is marked and recorded as a high-density trigger mark point. The interconnected high-density trigger mark points are combined to obtain a high-density trigger mark point group, and the space corresponding to the high-density trigger mark point group is recorded as a structural high-density space; Performing spatial ratio analysis on the high-density structural spaces, and constructing spatial vector lines between the high-density structural spaces based on the order of the spatial ratios. The method for constructing the spatial vector lines includes determining the center points of the high-density structural spaces, and connecting the spatial vector lines between the center points. The different space vector lines are scaled and rotated for alignment adjustment, and the proximity between the space vector lines is compared. Based on the proximity, the supporting steel frame structure expression model is screened and classified.

[0010] In some embodiments disclosed in the present invention, the method for comparing the proximity between space vector lines includes: Determine the turning nodes of different space vector lines, quantitatively analyze the turning node distances between the turning nodes corresponding to the space vector lines, and determine the proximity between the space vector lines based on the turning node distances; The method for determining the degree of proximity includes determining the preset interval of distances between turning nodes to which the distance between each turning node belongs, and determining the sub-degree of proximity between the corresponding turning nodes based on the interval of distances between the turning nodes to which they belong, and if the sub-degree of proximity is greater than or equal to the preset value, the turning node is identified as a coincident turning node, and the degree of proximity between the space vector lines is determined based on the node ratio of the coincident turning nodes to all turning nodes; Among them, the expression for calculating the degree of proximity is: ; in, For the degree of proximity, For the The coincidence judgment function between the turning nodes, if the sub-closeness between the turning nodes is greater than or equal to the preset value, then Output 1, otherwise output 0. is the node ratio impact adjustment constant, It is the node proportion impact adjustment coefficient.

[0011] In some embodiments disclosed in the present invention, a method for performing a signature feature factor analysis on each supporting steel frame structure expression model set includes: Determine the number of high-density structural spaces in the supporting steel frame structure expression model, and identify it as the first characteristic factor; The turning angles of several spatial vector lines in the high-density structured space are determined, and based on the order of turning points, the turning angles are sorted to obtain a turning angle sequence, which is identified as the second signature characteristic factor.

[0012] In some embodiments disclosed in the present invention, the method of performing a signature feature factor analysis on each supporting steel frame structure expression model set further includes: The model space occupied volume of the supporting steel frame expression model and the concentrated supporting steel frame expression model is determined and identified as the third characteristic factor.

[0013] In some embodiments disclosed in the present invention, a support steel frame vulnerability analysis system based on learning algorithm analysis is also disclosed, including: The first module is used to obtain a number of supporting steel frame design drawings with historical inspection records, analyze each supporting steel frame design drawing, construct a supporting steel frame structure expression model, divide the supporting steel frame structure expression model into vulnerable blocks, and configure structural parameters for each vulnerable block; The second module is used to analyze the historical detection records corresponding to each supporting steel frame structure expression model, determine the damage characteristics of the supporting steel frame in different time segments, and establish the corresponding relationship between the damage characteristics and the time segments, analyze the historical earthquake data, determine the earthquake severity parameters in different time segments, and establish the temporal correlation between the damage characteristics and the earthquake severity parameters based on the equivalent correspondence of the time segments; The third module is used to analyze the damage characteristics, and based on the analysis results, adjust the damage expression of the corresponding supporting steel frame structure expression model, and based on the change of the damage characteristics over time, obtain the first dynamic expression of the supporting steel frame structure expression model, and based on the temporal correlation between the damage characteristics and the earthquake severity parameter, establish the temporal correlation between the earthquake severity parameter change and the first dynamic expression, and obtain the supporting steel structure damage dynamic expression model; The fourth module is used to perform structural similarity analysis on all supporting steel frame structure expression models, and based on the analysis results, classify the supporting steel frame structure expression models to obtain several supporting steel frame structure expression model sets, and perform a signature feature factor analysis on each supporting steel frame structure expression model set to obtain several signature feature factor groups, and establish an association relationship between the signature feature factor group and the supporting steel structure expression model set; The fifth module is used to perform vulnerability analysis on the supporting steel frame. Based on the matching of different sign characteristic factor groups with the supporting steel frame, the supporting steel structure expression model set corresponding to the supporting steel frame is determined, and based on the supporting steel structure damage dynamic expression model corresponding to the supporting steel structure expression model set, the vulnerability of the supporting steel structure is expressed.

[0014] The present invention discloses a method and system for analyzing the vulnerability of a supporting steel frame based on learning algorithm analysis, and relates to the technical field of supporting steel frame analysis. By acquiring and analyzing the design drawings of supporting steel frames with historical detection records, a structural expression model is constructed, and vulnerable blocks and structural parameters are divided; the correlation between the damage characteristics of the supporting steel frame and time and earthquake severity is analyzed, and a damage dynamic performance model is established; similarity analysis is performed on all supporting steel frame structural expression models, and a model set is obtained by classification, and a marker characteristic factor group is extracted; in actual vulnerability analysis, the corresponding supporting steel structure expression model set is determined according to the matching of the supporting steel frame with the marker characteristic factor group, and the damage dynamic performance model in the model set is used to perform vulnerability assessment. The above technical scheme of the present invention improves the accuracy and efficiency of vulnerability assessment, and provides a scientific and effective technical means for the design, assessment and maintenance of supporting steel frames.

[0015] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a method step diagram of a supporting steel frame vulnerability analysis method based on learning algorithm analysis disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The technical solution of the present invention is further described below through the accompanying drawings and embodiments.

[0018] The following will be combined with the accompanying drawings and specific embodiments to clearly and completely describe the technical solution of the present invention. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention and cannot be understood as limiting the scope of protection of the present invention. Those skilled in the art in this field can make some non-essential improvements and adjustments based on the content of the present invention described below. In the present invention, unless otherwise clearly specified and limited, the technical terms used in the present invention should be the common meanings understood by the technical personnel described in the present invention.

[0019] Example: The present invention discloses a method for analyzing the vulnerability of a supporting steel frame based on a learning algorithm analysis. Figure 1 ,include: Step S100, obtaining a number of supporting steel frame design drawings with historical inspection records, analyzing each supporting steel frame design drawing, constructing a supporting steel frame structure expression model, dividing the supporting steel frame structure expression model into vulnerable blocks, and configuring structural parameters for each vulnerable block.

[0020] This step is the initial stage of the method, and its main purpose is to establish a structural expression model of the supporting steel frame. First, by collecting the design drawings of the supporting steel frame with historical inspection records, these drawings contain the structural information, material information, size information, etc. of the steel frame. Then, these drawings are deeply analyzed using professional analysis software or algorithms to construct a structural expression model of the supporting steel frame. Next, according to the structural characteristics and stress conditions of the steel frame, the model is divided into vulnerable blocks so that the vulnerable parts of the steel frame can be analyzed more accurately in the future. Finally, structural parameters such as strength, stiffness, toughness, etc. are configured for each vulnerable block, which will serve as the basis for subsequent analysis.

[0021] Vulnerable blocks refer to areas in the structural expression model of the supporting steel frame that are easily damaged by external forces such as earthquakes, determined based on finite element analysis results and historical earthquake data. These areas are more likely to be damaged, such as cracks, deformation or destruction, when subjected to external loads due to stress concentration, weak material properties or structural design reasons. By dividing vulnerable blocks, the vulnerable parts of the steel frame can be analyzed more specifically, providing important guidance for subsequent structural reinforcement and seismic design.

[0022] In some embodiments disclosed in the present invention, a method for dividing a supporting steel frame structure expression model into vulnerable blocks includes: Step S101, finite element analysis is performed on the supporting steel frame structure expression model, including meshing, dividing into a plurality of model units, establishing associations between adjacent model units, and configuring material properties for the model units, including elastic modulus, Poisson's ratio and density.

[0023] In the process of dividing the supporting steel frame structure into vulnerable blocks, step S101 is a crucial step, which involves finite element analysis of the structural expression model. Finite element analysis is a numerical analysis method that simplifies complex problems into a series of simple problems to solve. In this step, the supporting steel frame structure first needs to be meshed and subdivided into several small model units. These model units form a whole by interrelationship, so that the mechanical behavior of the entire structure can be simulated. Subsequently, material properties such as elastic modulus, Poisson's ratio and density are configured for each model unit. These properties are key parameters for describing the mechanical properties of the material. Through finite element analysis, physical quantities such as stress and strain of each model unit under stress can be calculated, thereby providing basic data for the subsequent determination of vulnerable blocks. The advantage of finite element analysis is that it can adapt to various complex shapes and structures and has high calculation accuracy, so it has been widely used in engineering analysis.

[0024] In finite element analysis, the association between adjacent model units is achieved through meshing. First, the supporting steel frame structure is divided into several small model units, which are physically connected to each other. Then, the association between them is established by setting the nodes or interfaces between the units. This association ensures that when a unit is subjected to external loads, its physical quantities such as stress and strain can be transferred to adjacent units, thereby simulating the mechanical behavior of the entire structure. The establishment of this association is the basis of finite element analysis and the key to the subsequent determination of vulnerable blocks and vulnerability analysis. Step S102, applying boundary conditions to the meshed support structure expression model, and applying conventional loads to the meshed support structure expression model.

[0025] Step S102 is the process of applying boundary conditions and conventional loads to the supporting steel frame structure expression model after meshing based on finite element analysis. Boundary conditions refer to the setting of external constraints on the structure during the simulation analysis, which determines the displacement and deformation mode of the structure. In the analysis of the supporting steel frame structure, the setting of boundary conditions needs to accurately reflect the constraints of the structure in actual working conditions, such as fixed supports, sliding supports, etc. At the same time, it is also necessary to apply conventional loads to the structure to simulate the stress state of the structure in daily use. The application of conventional loads helps to understand the stress distribution and deformation of the structure under normal conditions, and provide a reference for subsequent seismic load analysis. By applying boundary conditions and conventional loads, the necessary input conditions can be provided for finite element analysis, thereby more accurately simulating the mechanical behavior of the structure.

[0026] Boundary conditions refer to the setting of external constraints on the structure during the simulation analysis process, which determines the displacement and deformation mode of the structure. In the analysis of supported steel frame structures, the application of boundary conditions needs to accurately reflect the constraints of the structure in actual working conditions, such as fixed supports, sliding supports, etc. This is usually achieved by applying constraints at specific locations in the model, such as limiting the displacement or rotation of certain nodes.

[0027] Conventional loads refer to the conventional external forces that structures are subjected to in daily use, such as gravity, wind, etc. The purpose of applying conventional loads to the meshed support structure expression model is to simulate the stress state of the structure under normal conditions and understand its stress distribution and deformation in daily use. This is usually achieved by applying loads to the corresponding positions of the model, such as applying uniform loads or concentrated loads. By applying boundary conditions and conventional loads, the necessary input conditions can be provided for finite element analysis, thereby more accurately simulating the mechanical behavior of the structure.

[0028] Step S103, based on historical earthquake data, determine the input parameters of the finite element analysis, obtain the finite element results of the supporting steel frame structure expression model, based on the finite element analysis results, determine the blocks on the supporting steel frame structure expression model where the stress is greater than or equal to a preset range, record them as damage concern blocks, and based on the type of structure of the damage concern blocks, determine the scope of the damage concern blocks to obtain vulnerable blocks.

[0029] Step S103 is a process of determining vulnerable blocks on the supporting steel frame structure expression model based on the finite element analysis results and historical earthquake data. In this step, it is first necessary to determine the input parameters of the finite element analysis according to the historical earthquake data, such as the acceleration and frequency of the seismic wave, to simulate the stress of the structure under the earthquake. Then, the stress distribution and deformation of the structure under the earthquake load are calculated by finite element analysis. Next, according to the preset stress range, the blocks with stress values ​​greater than or equal to the range are screened out, and these blocks are regarded as damage concern blocks. The determination of damage concern blocks helps to narrow the scope of subsequent analysis and improve analysis efficiency. Finally, based on the type of structure of the damage concern block (such as beam-column joints, supporting components, etc.), combined with the mechanical properties of the structure and the stress characteristics under the earthquake, the scope of the vulnerable block is reasonably delineated. The determination of vulnerable blocks has important guiding significance for subsequent structural reinforcement and seismic design.

[0030] Step S200, analyze the historical detection records corresponding to each supporting steel frame structure expression model, determine the damage characteristics of the supporting steel frame in different time segments, and establish a corresponding relationship between the damage characteristics and the time segments, analyze the historical earthquake data, determine the earthquake severity parameters in different time segments, and establish a temporal correlation between the damage characteristics and the earthquake severity parameters based on the equivalent correspondence of the time segments.

[0031] This step aims to establish the temporal correlation between damage characteristics and earthquake severity parameters. First, the historical detection records corresponding to each supporting steel frame structure expression model are carefully analyzed to determine the damage characteristics of the supporting steel frame in different time periods, such as cracks, deformation, corrosion, etc. Then, the correspondence between these damage characteristics and time periods is established to understand the changes in steel frame damage over time. At the same time, historical earthquake data are analyzed to determine earthquake severity parameters in different time periods, such as magnitude, focal depth, and epicenter distance. Finally, by means of equivalent correspondence of time periods, the damage characteristics are temporally correlated with the earthquake severity parameters, providing a basis for the subsequent establishment of a dynamic damage performance model.

[0032] In some embodiments disclosed in the present invention, the method for determining the damage characteristics of the supporting steel frame in different time periods includes: Step S201, analyze the historical detection records to determine the damage characteristics of different damage characteristic types, and establish damage characteristic comparison rules based on the structural impact of each damage characteristic type on the supporting steel frame, and based on the damage characteristic comparison rules, determine the negative structural impact parameters of the damage characteristics of different damage characteristic types in different vulnerable blocks.

[0033] When evaluating the damage characteristics of the supporting steel frame structure over time, step S201 first focuses on the analysis of historical inspection records. The core of this step is to identify and summarize different types of damage characteristics, which may include cracks, corrosion, deformation, etc., which pose a potential threat to the integrity and stability of the structure. In order to deeply understand the specific impact of these damage characteristics on the supporting steel frame structure, it is necessary to evaluate the degree of weakening of the overall performance of the structure based on the type of each damage characteristic, and establish a set of damage characteristic comparison rules accordingly. This set of rules not only takes into account the severity of the damage itself, but also integrates factors such as its location and distribution in the structure and possible chain reactions. Through this rule, we can quantify the negative structural impact caused by different types of damage characteristics in different vulnerable blocks, and form a series of specific negative structural impact parameters. These parameters provide a key basis for the subsequent assessment of the vulnerability of the structure and the formulation of maintenance strategies.

[0034] Step S202 , counting negative structural impact parameters of different damage feature types corresponding to each vulnerable block, and configuring them in a vulnerability expression component of each vulnerable block.

[0035] Step S202 is to further characterize the vulnerability of the supporting steel frame in detail on the basis of step S201. In this step, we first statistically summarize the negative structural impact parameters of different damage feature types in each vulnerable block. This statistical process not only focuses on the impact of a single damage feature, but also considers the cumulative effect of multiple damage features in the same block, thereby more comprehensively reflecting the vulnerability of the block. Subsequently, these statistical results are configured into the vulnerability expression component of each vulnerable block to form a systematic vulnerability assessment system. As an intuitive display of structural vulnerability, the vulnerability expression component can help engineers and managers quickly identify the weak links of the structure and provide targeted guidance for subsequent maintenance, repair or reinforcement work. Through the implementation of this step, we can not only more accurately grasp the vulnerability distribution of the supporting steel frame structure, but also provide strong support for the long-term safety management of the structure.

[0036] Step S300, analyze the damage characteristics, and based on the analysis results, adjust the damage expression of the corresponding supporting steel frame structure expression model, and based on the temporal change of the damage characteristics, obtain the first dynamic expression of the supporting steel frame structure expression model, and based on the temporal correlation between the damage characteristics and the earthquake severity parameters, establish the temporal correlation between the earthquake severity parameter change and the first dynamic expression, and obtain the supporting steel structure damage dynamic expression model.

[0037] The core of this step is to establish a dynamic performance model of damage to the supporting steel structure. First, the damage characteristics obtained in step S200 are deeply analyzed to understand the degree of its influence on the performance of the steel frame structure. Then, the damage expression of the supporting steel frame structure expression model is adjusted according to the analysis results to make it more realistically reflect the damage of the steel frame in actual use. Next, based on the change of the damage characteristics over time, the first dynamic performance of the supporting steel frame structure expression model is obtained, that is, the change of steel frame damage over time. Finally, combined with the temporal correlation between the damage characteristics and the earthquake severity parameters, the temporal correlation between the change of the earthquake severity parameters and the first dynamic performance is established, thereby obtaining the dynamic performance model of damage to the supporting steel structure.

[0038] In some embodiments disclosed in the present invention, a method for adjusting the damage expression of a supporting steel frame structure expression model includes: Step S301, analyze the supporting steel frame structure expression model, set a parallel center virtual line for the connecting rod of the supporting steel frame structure expression model, and make a number of perpendicular parametric expression circles relative to the center virtual line, and the center of the parametric expression circle coincides with the center virtual line, and set a parametric expression circle for the intersection of the center virtual line.

[0039] Step S302, based on the negative structural impact parameter of the corresponding vulnerable block, determine the diameter of the parameter expression circle or parameter expression body: .

[0040] in, is the diameter of a circle or a body expressed parametrically, is the diameter conversion adjustment factor, is the negative structural influence parameter, is the adjustment coefficient of the negative structural impact parameter, Adjustment constant for negative structural influence parameters.

[0041] Diameter conversion adjustment factor: This is a factor used to convert the negative structural influence parameter into the parametric expression circle or parametric expression round body diameter. Ensure that the converted diameter can accurately reflect the size of the negative structural influence parameter.

[0042] Negative structural impact parameter adjustment coefficient: This is a coefficient used to adjust the size of the negative structural impact parameter. It may take into account multiple factors, such as the severity, location, distribution, etc. of the damage characteristics, and the degree of impact of these factors on the overall performance of the structure. By adjusting this coefficient, the impact of different damage characteristics on the structural vulnerability can be more accurately assessed.

[0043] Negative structural influence parameter influence adjustment constant: This is a constant used when calculating negative structural influence parameters. It may be used to ensure the accuracy and consistency of the calculation results, or to adjust the calculation results to conform to the actual situation. The specific value of this constant may be determined based on experimental data, experience or theoretical analysis.

[0044] Step S400, performing structural similarity analysis on all supporting steel frame structure expression models, and based on the analysis results, classifying the supporting steel frame structure expression models to obtain a number of supporting steel frame structure expression model sets, and performing a signature feature factor analysis on each supporting steel frame structure expression model set to obtain a number of signature feature factor groups, and establishing an association relationship between the signature feature factor group and the supporting steel structure expression model set.

[0045] This step aims to classify the supporting steel frame structure expression models and extract the signature feature factor groups. First, a structural similarity analysis is performed on all supporting steel frame structure expression models, and they are classified into several supporting steel frame structure expression model sets according to the model's geometric shape, material type, connection method and other characteristics. Then, a signature feature factor analysis is performed on each model set to extract several signature feature factor groups that can represent the characteristics of the model set. These signature feature factor groups will serve as an important basis for subsequent vulnerability analysis. Finally, an association relationship between the signature feature factor group and the supporting steel structure expression model set is established, so that a model set matching a specific signature feature factor group can be quickly found in subsequent analysis.

[0046] In some embodiments disclosed in the present invention, a method for performing structural similarity analysis on all supporting steel frame structure expression models includes: Step S401, a structural analysis space is established for the supporting steel frame structure expression model, and a number of spatial nodes are evenly set in the structural analysis space, and the spatial nodes mapped on the supporting steel frame structure expression model are triggered and marked as trigger mark points.

[0047] Step S402, randomly select a number of trigger mark points, and randomly combine them to obtain a number of trigger mark point groups, analyze the distances between the trigger mark points in each trigger mark point group, calculate the average value of the distances between the points, and record it as the distance between the mark points.

[0048] Step S403, randomly select a trigger mark point and record it as a trigger mark point for analysis, determine the distance between the trigger mark points next to the trigger mark point for analysis, and calculate the average value of the distance between the points, which is recorded as the average distance between the points, and calculate the ratio of the average distance between the points to the distance between the marking points, which is recorded as the reference ratio of the distance between the points. If the reference ratio of the distance between the points is greater than or equal to the preset value, mark the corresponding trigger mark point for analysis and record it as a high-density trigger mark point.

[0049] Step S404: Combine the interconnected high-density trigger mark points to obtain a high-density trigger mark point group, and record the space corresponding to the high-density trigger mark point group as a structural high-density space.

[0050] Step S405, performing spatial ratio analysis between the high-density structural spaces, and constructing spatial vector lines between the high-density structural spaces based on the order of the spatial ratios. The method for constructing the spatial vector lines includes determining the center points of the high-density structural spaces and connecting the spatial vector lines between the center points.

[0051] Structural high-density spaces refer to areas formed by high-density trigger mark points in the supporting steel frame structure expression model. These areas have high density and complexity in structure. The spatial ratio analysis between structural high-density spaces is to compare the form, size and proportional relationship between different high-density spaces. This analysis helps to understand the overall layout and characteristics of the structure, as well as the interaction and influence between different parts. Through spatial ratio analysis, the characteristic factors of the structure can be further extracted, providing an important basis for subsequent vulnerability analysis.

[0052] Step S406, scaling and rotating the different space vector lines for alignment adjustment, and comparing the proximity between the space vector lines, and based on the proximity, screening and classifying the supporting steel frame structure expression models.

[0053] In some embodiments disclosed in the present invention, the method for comparing the proximity between space vector lines includes: Step S4061, determining the turning nodes of different space vector lines, and quantitatively analyzing the turning node distances between the turning nodes corresponding to the space vector lines, and determining the proximity between the space vector lines based on the turning node distances.

[0054] Among them, the method for determining the degree of proximity includes judging the preset turning node distance interval to which the distance between each turning node belongs, and determining the sub-proximity between the corresponding turning nodes based on the corresponding turning node distance interval. If the sub-proximity is greater than or equal to the preset value, the turning node is determined to be a coincident turning node. Based on the proportion of coincident turning nodes to all turning nodes, the degree of proximity between the spatial vector lines is determined.

[0055] Turning node distance refers to the distance between adjacent turning nodes in a space vector line. In the structural similarity analysis of the supporting steel frame structure expression model, the similarity between different models can be quantified by determining the turning nodes of the space vector line and calculating the distance between them. Turning node distance is an important indicator for evaluating model similarity, which reflects the differences and changes in the details of the model in structure. By comparing the turning node distances of different models, the similarities and differences between them can be judged more accurately.

[0056] Sub-proximity is a quantitative evaluation of the distance between each turning node when comparing the proximity between space vector lines. It reflects the similarity between two space vector lines at a specific turning node. By calculating the sub-proximity, the similarities and differences between different space vector lines can be analyzed in more detail. In the structural similarity analysis of the supporting steel frame structure expression model, the sub-proximity is an important component of evaluating the similarity of the model, which helps to more accurately judge the consistency between different models.

[0057] Among them, the expression for calculating the degree of proximity is: .

[0058] in, For the degree of proximity, For the The coincidence judgment function between the turning nodes, if the sub-closeness between the turning nodes is greater than or equal to the preset value, then Output 1, otherwise output 0. is the node ratio impact adjustment constant, It is the node proportion impact adjustment coefficient.

[0059] In some embodiments disclosed in the present invention, a method for performing a signature feature factor analysis on each supporting steel frame structure expression model set includes: Step S406, determining the number of structural high-density spaces of the structural high-density spaces of the supporting steel frame expression model in the supporting steel frame structure expression model concentration, and identifying it as the first marker characteristic factor.

[0060] Step S407, determine several turning angles of the space vector line of the high-density structure space, and sort the turning angles based on the turning point order to obtain a turning angle sequence, which is identified as the second marker feature factor.

[0061] In some embodiments disclosed in the present invention, the method of performing a signature feature factor analysis on each supporting steel frame structure expression model set further includes: Step S408, determining the model space occupied volume of the supporting steel frame expression model in the supporting steel frame structure expression model concentration, and identifying it as the third marker feature factor.

[0062] The model space occupied volume refers to the volume occupied by the supporting steel frame structure expression model in three-dimensional space. This volume can be obtained by calculating the geometric size and shape of the model. The model space occupied volume is an important indicator for evaluating the size and complexity of the model, which reflects the occupancy of the model in the physical space. In the structural similarity analysis of the supporting steel frame structure expression model, the model space occupied volume can be used as one of the characteristic factors to compare the similarities and differences between different models. At the same time, it can also provide important information about the model size and shape for subsequent vulnerability analysis.

[0063] Step S500, when performing vulnerability analysis on the supporting steel frame, based on the matching of different marker characteristic factor groups with the supporting steel frame, a supporting steel structure expression model set corresponding to the supporting steel frame is determined, and based on the supporting steel structure damage dynamic expression model corresponding to the supporting steel structure expression model set, the vulnerability of the supporting steel structure is expressed.

[0064] This step is the final application stage of the method, that is, to conduct vulnerability analysis of the supporting steel frame. First, the matching situation of the characteristic factor group of the supporting steel frame to be analyzed and the established characteristic factor group is analyzed. Then, the supporting steel structure expression model set corresponding to the supporting steel frame to be analyzed is determined according to the degree of matching. Next, the supporting steel structure damage dynamic performance model corresponding to the model set is used to express the vulnerability of the supporting steel frame to be analyzed, that is, to predict its damage and performance changes under different earthquake severity parameters. Finally, according to the results of the vulnerability analysis, corresponding maintenance, repair or reinforcement suggestions are proposed to ensure the safety and stability of the supporting steel frame.

[0065] For the supporting steel frame that needs to be analyzed for vulnerability, we first need to re-express its structural characteristics according to the established analysis process (including establishing structural analysis space, marking trigger points, determining high-density trigger mark point groups, constructing structural high-density space, drawing spatial vector lines, and screening and classifying models). Then, we comprehensively compare this new expression with the existing supporting steel structure expression model set. The comparison process involves multiple dimensions, such as the distribution characteristics of the high-density trigger mark point group, the morphological proportion of the structural high-density space, and the specific parameters of the spatial vector line, and similarity calculation or pattern matching algorithms are used to quantify the similarity between the two. Finally, based on the preset matching threshold or standard, we can determine which model or models in the model set have a higher matching situation with the new supporting steel frame, thereby providing an accurate model basis for subsequent vulnerability analysis. This process not only ensures the pertinence and accuracy of vulnerability analysis, but also realizes the effective combination of theoretical models and actual engineering practice.

[0066] In some embodiments disclosed in the present invention, a support steel frame vulnerability analysis system based on learning algorithm analysis is also disclosed, including: The first module is used to obtain a number of supporting steel frame design drawings with historical inspection records, analyze each supporting steel frame design drawing, construct a supporting steel frame structure expression model, divide the supporting steel frame structure expression model into vulnerable blocks, and configure structural parameters for each vulnerable block; The second module is used to analyze the historical detection records corresponding to each supporting steel frame structure expression model, determine the damage characteristics of the supporting steel frame in different time segments, and establish the corresponding relationship between the damage characteristics and the time segments, analyze the historical earthquake data, determine the earthquake severity parameters in different time segments, and establish the temporal correlation between the damage characteristics and the earthquake severity parameters based on the equivalent correspondence of the time segments; The third module is used to analyze the damage characteristics, and based on the analysis results, adjust the damage expression of the corresponding supporting steel frame structure expression model, and based on the change of the damage characteristics over time, obtain the first dynamic expression of the supporting steel frame structure expression model, and based on the temporal correlation between the damage characteristics and the earthquake severity parameter, establish the temporal correlation between the earthquake severity parameter change and the first dynamic expression, and obtain the supporting steel structure damage dynamic expression model; The fourth module is used to perform structural similarity analysis on all supporting steel frame structure expression models, and based on the analysis results, classify the supporting steel frame structure expression models to obtain several supporting steel frame structure expression model sets, and perform a signature feature factor analysis on each supporting steel frame structure expression model set to obtain several signature feature factor groups, and establish an association relationship between the signature feature factor group and the supporting steel structure expression model set; The fifth module is used to perform vulnerability analysis on the supporting steel frame. Based on the matching of different sign characteristic factor groups with the supporting steel frame, the supporting steel structure expression model set corresponding to the supporting steel frame is determined, and based on the supporting steel structure damage dynamic expression model corresponding to the supporting steel structure expression model set, the vulnerability of the supporting steel structure is expressed.

[0067] The present invention discloses a method and system for analyzing the vulnerability of a supporting steel frame based on learning algorithm analysis, and relates to the technical field of supporting steel frame analysis. By acquiring and analyzing the design drawings of supporting steel frames with historical detection records, a structural expression model is constructed, and vulnerable blocks and structural parameters are divided; the correlation between the damage characteristics of the supporting steel frame and time and earthquake severity is analyzed, and a damage dynamic performance model is established; similarity analysis is performed on all supporting steel frame structural expression models, and a model set is obtained by classification, and a marker characteristic factor group is extracted; in actual vulnerability analysis, the corresponding supporting steel structure expression model set is determined according to the matching of the supporting steel frame with the marker characteristic factor group, and the damage dynamic performance model in the model set is used to perform vulnerability assessment. The above technical scheme of the present invention improves the accuracy and efficiency of vulnerability assessment, and provides a scientific and effective technical means for the design, assessment and maintenance of supporting steel frames.

[0068] Through the description of the above implementation methods, those skilled in the art can clearly understand that the present invention can be implemented by hardware, or by software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each implementation scenario of the present invention.

[0069] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for analyzing the vulnerability of supporting steel frames based on learning algorithm analysis, characterized in that: include: Obtain several supporting steel frame design drawings with historical inspection records, analyze each supporting steel frame design drawing, construct a supporting steel frame structure expression model, divide the supporting steel frame structure expression model into vulnerable blocks, and configure structural parameters for each vulnerable block; Analyze the historical detection records corresponding to each supporting steel frame structure expression model, determine the damage characteristics of the supporting steel frame in different time segments, and establish the corresponding relationship between the damage characteristics and the time segments; analyze the historical earthquake data, determine the earthquake severity parameters in different time segments, and establish the temporal correlation between the damage characteristics and the earthquake severity parameters based on the equivalent correspondence of the time segments; Analyze the damage characteristics, and based on the analysis results, adjust the damage expression of the corresponding supporting steel frame structure expression model, and based on the change of the damage characteristics over time, obtain the first dynamic expression of the supporting steel frame structure expression model, and based on the temporal correlation between the damage characteristics and the earthquake severity parameter, establish the temporal correlation between the earthquake severity parameter change and the first dynamic expression, and obtain the supporting steel structure damage dynamic expression model; Perform structural similarity analysis on all supporting steel frame structure expression models, and classify the supporting steel frame structure expression models based on the analysis results to obtain several supporting steel frame structure expression model sets, and perform signature feature factor analysis on each supporting steel frame structure expression model set to obtain several signature feature factor groups, and establish a correlation between the signature feature factor group and the supporting steel structure expression model set; When conducting vulnerability analysis on the supporting steel frame, the supporting steel structure expression model set corresponding to the supporting steel frame is determined based on the matching of different sign characteristic factor groups with the supporting steel frame, and the vulnerability of the supporting steel structure is expressed based on the supporting steel structure damage dynamic expression model corresponding to the supporting steel structure expression model set.

2. The method for analyzing the vulnerability of a supporting steel frame based on learning algorithm analysis according to claim 1 is characterized in that: The method of dividing the vulnerable blocks of the supporting steel frame structure expression model includes: Finite element analysis is performed on the supporting steel frame structure expression model, including meshing, dividing it into several model units, establishing the association relationship between adjacent model units, and configuring material properties for the model units, including elastic modulus, Poisson's ratio and density; Apply boundary conditions to the support structure expression model after meshing, and apply conventional loads to the support structure expression model after meshing; Based on historical earthquake data, the input parameters of the finite element analysis are determined, and the finite element results of the supporting steel frame structure expression model are obtained. Based on the finite element analysis results, the blocks on the supporting steel frame structure expression model where the stress is greater than or equal to the preset range are determined and recorded as damage concern blocks. Based on the type of structure of the damage concern blocks, the scope of the damage concern blocks is determined to obtain the vulnerable blocks.

3. The method for vulnerability analysis of supporting steel frame based on learning algorithm analysis according to claim 1 is characterized in that: Methods for determining damage characteristics of supporting steel frames at different time intervals include: Analyze historical inspection records to determine the damage characteristics of different damage characteristic types, and establish damage characteristic comparison rules based on the structural impact of each damage characteristic type on the supporting steel frame. Based on the damage characteristic comparison rules, determine the negative structural impact parameters of the damage characteristics of different damage characteristic types in different vulnerable blocks; The negative structural impact parameters of different damage feature types corresponding to each vulnerable block are counted and configured in the vulnerable expression component of each vulnerable block.

4. The method for analyzing the vulnerability of a supporting steel frame based on learning algorithm analysis according to claim 3 is characterized in that: The methods for adjusting the damage expression of the supporting steel frame structure expression model include: The supporting steel frame structure expression model is analyzed, and a parallel central virtual line is set for the connecting rod of the supporting steel frame structure expression model, and a number of perpendicular parametric expression circles are made relative to the central virtual line, and the center of the parametric expression circle coincides with the central virtual line, and a parametric expression circle is set for the intersection of the central virtual line; Based on the negative structural impact parameters of the corresponding vulnerable block, the diameter of the parametric expression circle or parametric expression body is determined: ; in, is the diameter of a circle or a body expressed parametrically, is the diameter conversion adjustment factor, is the negative structural influence parameter, is the adjustment coefficient of the negative structural impact parameter, Adjustment constant for negative structural influence parameters.

5. The method for analyzing the vulnerability of a supporting steel frame based on learning algorithm analysis according to claim 1 is characterized in that: The methods for structural similarity analysis of all supporting steel frame structure expression models include: A structural analysis space is established for the supporting steel frame structure expression model, and a number of spatial nodes are evenly set in the structural analysis space, and the spatial nodes mapped on the supporting steel frame structure expression model are triggered and marked as trigger mark points; Randomly select a number of trigger mark points and randomly combine them to obtain a number of trigger mark point groups, analyze the distances between the trigger mark points in each trigger mark point group, calculate the average value of the distances between the points, and record it as the distance between the mark points; A trigger mark point is randomly selected and recorded as an analysis trigger mark point. The distance between the trigger mark points next to the analysis trigger mark point is determined, and the average value of the distance between the points is calculated, which is recorded as the average distance between the points. The ratio of the average distance between the points to the distance between the marking points is calculated, which is recorded as the reference ratio of the distance between the points. If the reference ratio of the distance between the points is greater than or equal to the preset value, the corresponding analysis trigger mark point is marked and recorded as a high-density trigger mark point. The interconnected high-density trigger mark points are combined to obtain a high-density trigger mark point group, and the space corresponding to the high-density trigger mark point group is recorded as a structural high-density space; Performing spatial ratio analysis on the high-density structural spaces, and constructing spatial vector lines between the high-density structural spaces based on the order of the spatial ratios. The method for constructing the spatial vector lines includes determining the center points of the high-density structural spaces, and connecting the spatial vector lines between the center points. The different space vector lines are scaled and rotated for alignment adjustment, and the proximity between the space vector lines is compared. Based on the proximity, the supporting steel frame structure expression model is screened and classified.

6. The method for analyzing the vulnerability of a supporting steel frame based on learning algorithm analysis according to claim 5 is characterized in that: Methods for comparing the proximity between space vector lines include: Determine the turning nodes of different space vector lines, quantitatively analyze the turning node distances between the turning nodes corresponding to the space vector lines, and determine the proximity between the space vector lines based on the turning node distances; The method for determining the degree of proximity includes determining the preset interval of distances between turning nodes to which the distance between each turning node belongs, and determining the sub-degree of proximity between the corresponding turning nodes based on the interval of distances between the turning nodes to which they belong, and if the sub-degree of proximity is greater than or equal to the preset value, the turning node is identified as a coincident turning node, and the degree of proximity between the space vector lines is determined based on the node ratio of the coincident turning nodes to all turning nodes; Among them, the expression for calculating the degree of proximity is: ; in, For the degree of proximity, For the The coincidence judgment function between the turning nodes, if the sub-closeness between the turning nodes is greater than or equal to the preset value, then Output 1, otherwise output 0. is the node ratio impact adjustment constant, It is the node proportion impact adjustment coefficient.

7. The method for analyzing the vulnerability of a supporting steel frame based on learning algorithm analysis according to claim 5 is characterized in that: The method of performing characteristic factor analysis on each supporting steel frame structure expression model set includes: Determine the number of high-density structural spaces in the supporting steel frame structure expression model, and identify it as the first characteristic factor; The turning angles of several spatial vector lines in the high-density structured space are determined, and based on the order of turning points, the turning angles are sorted to obtain a turning angle sequence, which is identified as the second signature characteristic factor.

8. The method for analyzing the vulnerability of a supporting steel frame based on learning algorithm analysis according to claim 7 is characterized in that: The method of performing a signature factor analysis on each supporting steel frame structure expression model set also includes: The model space occupied volume of the supporting steel frame expression model and the concentrated supporting steel frame expression model is determined and identified as the third characteristic factor.

9. The support steel frame vulnerability analysis system based on learning algorithm analysis is characterized by: A method for performing a vulnerability analysis of a supporting steel frame according to any one of claims 1 to 8, comprising: The first module is used to obtain a number of supporting steel frame design drawings with historical inspection records, analyze each supporting steel frame design drawing, construct a supporting steel frame structure expression model, divide the supporting steel frame structure expression model into vulnerable blocks, and configure structural parameters for each vulnerable block; The second module is used to analyze the historical detection records corresponding to each supporting steel frame structure expression model, determine the damage characteristics of the supporting steel frame in different time segments, and establish the corresponding relationship between the damage characteristics and the time segments, analyze the historical earthquake data, determine the earthquake severity parameters in different time segments, and establish the temporal correlation between the damage characteristics and the earthquake severity parameters based on the equivalent correspondence of the time segments; The third module is used to analyze the damage characteristics, and based on the analysis results, adjust the damage expression of the corresponding supporting steel frame structure expression model, and based on the change of the damage characteristics over time, obtain the first dynamic expression of the supporting steel frame structure expression model, and based on the temporal correlation between the damage characteristics and the earthquake severity parameter, establish the temporal correlation between the earthquake severity parameter change and the first dynamic expression, and obtain the supporting steel structure damage dynamic expression model; The fourth module is used to perform structural similarity analysis on all supporting steel frame structure expression models, and based on the analysis results, classify the supporting steel frame structure expression models to obtain several supporting steel frame structure expression model sets, and perform a signature feature factor analysis on each supporting steel frame structure expression model set to obtain several signature feature factor groups, and establish an association relationship between the signature feature factor group and the supporting steel structure expression model set; The fifth module is used to perform vulnerability analysis on the supporting steel frame. Based on the matching of different sign characteristic factor groups with the supporting steel frame, the supporting steel structure expression model set corresponding to the supporting steel frame is determined, and based on the supporting steel structure damage dynamic expression model corresponding to the supporting steel structure expression model set, the vulnerability of the supporting steel structure is expressed.

Images