Intelligent evaluation method for seismic damage steel reinforced concrete structure

Abstract

The application discloses a seismic damage steel reinforced concrete structure intelligent evaluation method and relates to the field of intelligent construction, and comprises the following steps: obtaining pre-earthquake point cloud data containing all monitoring points through three-dimensional laser scanning; obtaining post-earthquake point cloud data of a steel reinforced concrete component; calculating the inter-story drift index of the steel reinforced concrete component based on the post-earthquake point cloud data, and determining that the steel reinforced concrete component is macroscopically damaged when the inter-story drift index exceeds a limited threshold; obtaining the buckling parameter of the steel reinforced concrete component by comparing the horizontal displacement of the post-earthquake point cloud data relative to the pre-earthquake point cloud data for the rest of the steel reinforced concrete components; establishing a damage coefficient evaluation model based on a neural network, inputting the buckling parameter and the design parameter of the component into the damage coefficient evaluation model, and calculating the damage coefficient of the steel reinforced concrete component; and determining the damage state corresponding to the steel reinforced concrete component and the component repair scheme corresponding to the steel reinforced concrete component, so that the post-earthquake performance of the component can be accurately predicted and classified.

Description

Technical Field

[0001] This invention relates to the field of intelligent building construction. More specifically, this invention relates to an intelligent assessment method for seismically damaged steel-concrete composite structures. Background Technology

[0002] Steel-concrete composite members are composite structures formed by using steel as the skeleton and outer layer and concrete as the inner filling material. The steel-concrete composite members in the form of double steel plate concrete composite shear walls are representative. The steel plate-encased concrete structure combines the bending resistance of steel and the compressive strength of concrete, and is widely used in nuclear power plant containment structures and core tube and shear wall systems of super high-rise buildings.

[0003] When steel-concrete composite members experience small to medium magnitude earthquakes, they generally remain within the elastic working range with relatively small overall deformation. However, under the complex loads of an earthquake, the initial out-of-plane deformation of the outer steel plates in steel-concrete composite members, especially the outer steel plates of long steel-concrete composite walls, may be amplified by the seismic action due to processing and construction issues. This can lead to further out-of-plane buckling in localized areas. In the buckling zone, embedded fasteners such as tie bolts are prone to stress concentration when subjected to the interfacial shear and pull-out forces caused by the differential deformation between the steel plate and the concrete. This can lead to the steel plate entering the plastic stage or failing to hold itself in place with the internal concrete. As the outer steel plate separates from the core concrete, the synergistic effect between the outer steel plate and the concrete weakens. Although the steel-concrete composite member may not show obvious damage overall, its shear bearing capacity, stiffness, and subsequent energy dissipation capacity have been reduced to some extent, affecting the structure's performance reserves and safety margin in aftershocks or future earthquakes.

[0004] Currently, in the field of post-earthquake structural safety assessment and repair decision-making, there is a lack of a systematic, accurate, and efficient quantitative assessment method and technical standard for the hidden damage of steel-concrete composite members, especially the outer steel plates in double-plate shear walls. Traditional assessment methods rely heavily on engineers' experience in on-site macroscopic surveys and make rough judgments about the buckling area of ​​steel plates using simple tools such as hollow hammers. These methods are highly subjective and have low accuracy, making it difficult to accurately measure the amplitude of steel plate buckling and unable to establish a scientific quantitative correlation between such local damage and the overall performance degradation of the component. Assessors may tend to classify a large number of components with only local damage but still repairable as "requiring component replacement," thereby causing unnecessary demolition and reconstruction, resulting in increased repair costs, time delays, and waste of resources.

[0005] Therefore, there is a need to provide a new intelligent assessment method for earthquake-damaged steel-concrete composite structures. By establishing a scientific damage assessment model, the method can accurately predict and classify the overall post-earthquake performance status of steel-concrete composite components, providing objective and reliable data support and decision-making basis for repair strategies. Under the premise of ensuring structural safety, this method can maximize the economy and scientific nature of post-disaster repair work. Summary of the Invention

[0006] One objective of this invention is to provide an intelligent assessment method for earthquake-damaged steel-concrete composite structures, which, by establishing a scientific damage assessment model, enables accurate prediction and classification of the overall post-earthquake performance status of steel-concrete composite components.

[0007] To achieve these objectives and other advantages according to the present invention, the present invention provides an intelligent assessment method for seismically damaged steel-concrete composite structures, comprising the following steps:

[0008] S1. Multiple monitoring points are set on the outer steel plate surface of the steel-concrete composite structure in the building, and pre-earthquake point cloud data containing all monitoring points is obtained by three-dimensional laser scanning.

[0009] S2. Perform three-dimensional laser scanning on the same steel-concrete composite member after the earthquake to obtain post-earthquake point cloud data of the steel-concrete composite member;

[0010] S3. Based on the post-earthquake point cloud data, calculate the inter-story displacement index of the steel-concrete composite member. When the inter-story displacement index exceeds a predetermined threshold, determine that the steel-concrete composite member is a macroscopically damaged member.

[0011] S4. For steel-concrete composite members that are not determined to be macroscopically damaged, the pre-earthquake point cloud data and the post-earthquake point cloud data are aligned with the coordinates based on the monitoring points. By comparing the horizontal displacement of the post-earthquake point cloud data with that of the pre-earthquake point cloud data, the buckling parameters of the steel-concrete composite member are obtained. The buckling parameters include the buckling area ratio and the failure ratio of the embedded parts.

[0012] S5. Establish a damage coefficient evaluation model based on neural networks, input the buckling parameters and the design parameters of the component into the damage coefficient evaluation model, and calculate the damage coefficient of the steel-concrete composite member.

[0013] S6. Based on the value of the damage coefficient, determine the damage state of the steel-concrete composite member and the corresponding repair plan.

[0014] Preferably, in step S3, the bottom and top of the steel-concrete composite member are selected at heights of [missing information]. H / 100 to H / 150 The representative section was used to calculate the average value of all post-earthquake point cloud coordinates within the bottom section as the bottom reference coordinates. P b(x b、 y b、 z b ) The average value of all post-earthquake point cloud coordinates within the top section is calculated as the top reference coordinate. P t (x t、 y t、 z t ) Among them, the first inter-story drift angle of the steel-concrete composite member is calculated with the height direction as the Z-axis. Inter-layer displacement angle If the first inter-story drift angle Or the second-layer inter-layer displacement angle At that time, the steel-concrete composite member was determined to be a macroscopically damaged member.

[0015] Preferably, step S4 includes the following steps:

[0016] S41. For any point in the post-earthquake point cloud data P i (x i、 y i、 z i ) Perform elastic deformation correction. , This generates post-earthquake corrected coordinates, and all corrected points form corrected post-earthquake point cloud data.

[0017] S42. Establish a pre-earthquake coordinate system based on pre-earthquake point cloud data, define the coordinates of all monitoring points, and form a pre-earthquake coordinate set. And calculate the pre-earthquake distance between any two monitoring points. This forms a set of pre-earthquake distances. A post-earthquake coordinate system was established based on the post-earthquake corrected point cloud data, defining the coordinates of all monitoring points and forming a set of post-earthquake coordinates. Calculate the post-earthquake distance between any two monitoring points. This forms a set of post-earthquake distances. ,in, N To monitor the number of monitoring points, For the first l Pre-earthquake coordinates of the monitoring points For the first m Pre-earthquake coordinates of the monitoring points For the first nPre-earthquake coordinates of the monitoring points For the first l Post-earthquake corrected coordinates of monitoring points. For the first p Post-earthquake corrected coordinates of monitoring points. For the first q Post-earthquake corrected coordinates of monitoring points;

[0018] S43, Multi-point combination that iterates through all monitoring points From the pre-earthquake distance set Extract the corresponding pre-earthquake distance From the post-earthquake distance set Extract the corresponding post-earthquake distance Calculate the global distance error of this multi-point combination. Select the ones that make The smallest combination of multiple points is taken as the optimal reference point group, and the center point of the optimal reference point group is set as the alignment point. M The number of monitoring points within a multi-point combination. a 1 、a 2 、a M For monitoring point index, a i and a j This is an index of monitoring points within a multi-point combination;

[0019] S44. Align the center points of the pre-earthquake point cloud data and the corrected post-earthquake point cloud data. Identify the buckling region on the outer steel plate and calculate the buckling area ratio by the coordinate deviation of the points at the same height in the pre-earthquake point cloud data and the corrected post-earthquake point cloud data in the direction of the normal of the outer steel plate. Define the embedded parts within the buckling region as failures and calculate the failure ratio of the embedded parts.

[0020] Preferably, the damage coefficient assessment model established in step S5 includes the following steps:

[0021] A1. In the finite element calculation software, multiple standard finite element models of steel-concrete composite members with different design parameters are made. Based on the standard finite element models, random initial out-of-plane geometric defects are introduced on the surface of the outer steel plate to make a pre-damaged finite element model. The pre-damaged finite element model is subjected to horizontal reciprocating load simulation to obtain complete hysteretic response data and related mechanical characteristic parameters.

[0022] A2. Based on the aforementioned mechanical characteristic parameters, calculate the damage coefficient corresponding to each pre-damage finite element model using a preset damage model. DI The damage coefficient of the standard finite element model in the elastic stageDI The calculated value is zero;

[0023] A3. For each pre-damage finite element model, extract the buckling parameters and design parameters corresponding to the simulated loading, and form a training sample set for the neural network model together with the corresponding damage coefficient.

[0024] A4. Construct a neural network model with buckling parameters and design parameters as input layers and damage coefficient as output layer. Input the training sample set into the neural network model for supervised learning. Adjust the network weights and biases through the error backpropagation algorithm until the prediction error of the neural network model converges, and obtain the trained damage coefficient evaluation model.

[0025] Preferably, the damage model is:

[0026] ;

[0027] in, DI The damage coefficient is... β These are combination coefficients, ranging from 0.1 to 0.15.

[0028] To determine the yield bearing capacity of the pre-damaged finite element model under simulated loading, The yield displacement of the pre-damaged finite element model under simulated loading; The effective maximum displacement of the pre-damaged finite element model under simulated loading is given. The cumulative hysteresis energy dissipation of the pre-damaged finite element model under simulated loading, This refers to the limit displacement when a standard finite element model is subjected to monotonically pushed-over loading.

[0029] Preferably, when the damage coefficient DI ≤ 0.1 When the steel-concrete composite structure is determined to be basically intact, the corresponding repair plan is continued use; when 0.1 < DI ≤ 0.5 When the steel-concrete composite member is determined to have minor damage, the corresponding repair plan is to locally repair the buckling area; when 0.5 < DI ≤ 1.0 When the damage coefficient is determined to be moderate, the corresponding repair solution is overall reinforcement of the component; when the damage coefficient is... DI > 1.0 When the steel-concrete composite member is determined to be severely damaged, the corresponding repair solution is to replace the member.

[0030] Preferably, based on the aligned pre-earthquake point cloud data and the corrected post-earthquake point cloud data, the displacement deviation Δ in the direction normal to the outer steel plate is selected to be greater than the critical buckling displacement. U a The region is designated as the buckling region, where the critical buckling displacement is... , s For the design spacing of the embedded parts, tThe thickness of the outer steel plate. f The range of values ​​for the empirical coefficient 0.15～0.25；

[0031] buckling area ratio , The total area of ​​the buckling region. This represents the total surface area of ​​the outer steel plate.

[0032] Embedded component failure rate , This refers to the number of inserts located within the buckling region. This represents the total number of embedded parts on the outer steel plate.

[0033] Preferably, the buckling parameter also includes the area ratio of the most unfavorable weak constraint. Its calculation methods include:

[0034] The horizontal section of the component with the highest number of embedded failures is identified as the most unfavorable section.

[0035] At the most unfavorable section, the intersection of the effective embedded part and the outer steel plate, and the intersection of adjacent plates of the outer steel plate are considered as constraint points. Two adjacent constraint points are considered as one calculation unit. The formula for calculating the area of ​​the weakly constrained concrete region corresponding to each calculation unit is as follows: ,in θ 0 For the boundary curve tangent, b k For the first k The distance between constraint points on both sides of each computational unit, and the proportion of the area with the most unfavorable weak constraint. ,in This represents the cross-sectional area of ​​the most unfavorable section.

[0036] Preferably, the damage coefficient assessment model employs a BP neural network model comprising an input layer, a hidden layer, and an output layer. The input layer of the BP neural network model includes multiple input neuron nodes, and the hidden layer has multiple hidden neuron nodes. Each input neuron node is connected to each of the hidden neuron nodes, and all hidden neuron nodes are connected to the output layer. The buckling parameters include the buckling area ratio, the embedding failure ratio, and the ratio of the most unfavorable weak constraint area. The design parameters include the outer steel plate thickness, the cross-sectional width, cross-sectional length, and height of the steel-concrete composite member, the embedding type code, the embedding design spacing, the concrete strength, and the steel plate yield strength. The buckling parameters and design parameters are used as input values ​​for the input neuron nodes, and the output results are compared with the corresponding damage coefficients. The algorithm optimizes the neural network model to obtain the damage coefficient assessment model.

[0037] The present invention has at least the following beneficial effects:

[0038] First, this invention forms a complete closed loop from data collection, preliminary screening, detailed analysis, intelligent evaluation to final decision-making, which elevates the original experience-based and fragmented evaluation process into an objective, systematic, and repeatable standardized method, effectively solving the problem of no clear rules to follow for the evaluation of steel-concrete composite components after an earthquake.

[0039] Secondly, this invention quantifies the internal damage of components that have not caused macroscopic damage, and can reveal hidden damage such as steel plate buckling and fastener failure that are difficult to detect by traditional methods, thereby achieving in-depth diagnosis of the true health status of steel-concrete composite components.

[0040] Third, this invention can provide engineers with clear and quantitative decision-making basis, which can prevent waste caused by over-repair of slightly damaged components and eliminate safety hazards caused by misjudgment of severely damaged components, thereby improving the economy and safety of post-earthquake repair work.

[0041] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Attached Figure Description

[0042] Figure 1 This is a technical flowchart of one technical solution of the present invention;

[0043] Figure 2 This is a schematic diagram of the initial buckling region of the pre-damage finite element model in one technical solution of the present invention;

[0044] Figure 3 This is a schematic diagram of the weakly constrained concrete region on the most unfavorable cross section in one technical solution of the present invention. Detailed Implementation

[0045] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can implement it based on the description.

[0046] It should be understood that terms such as “having,” “comprising,” and “including” as used herein do not exclude the presence or addition of one or more other elements or combinations thereof.

[0047] It should be noted that, unless otherwise specified, the experimental methods described in the following embodiments are conventional methods, and the reagents and materials mentioned are commercially available. In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "setting" should be interpreted broadly. For example, they can refer to fixed connection or setting, detachable connection or setting, or integral connection or setting. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances. The terms "lateral," "longitudinal," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this invention and simplifying the description. They do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention.

[0048] like Figure 1 As shown, this invention provides an intelligent assessment method for seismically damaged steel-concrete composite structures, comprising the following steps:

[0049] S1. Multiple monitoring points are set on the outer steel plate surface of the steel-concrete composite structure in the building. Pre-earthquake point cloud data containing all monitoring points is obtained by three-dimensional laser scanning. Specifically, before the earthquake, each component with steel-concrete composite structure is scanned before the earthquake. Several monitoring points are arranged on the outer steel plate surface of the steel-concrete composite structure to ensure that the scan data has sufficient feature points. The monitoring points are used for the alignment of the pre-earthquake point cloud data and the post-earthquake point cloud data. The number of monitoring points does not need to be too dense. The spacing between monitoring points can be set to more than 1m. The monitoring points are evenly distributed on the outer steel plate surface. This can be achieved by pasting target paper, engraving crosshairs, etc. Three-dimensional laser scanning can be achieved using equipment such as the FARO Focus series.

[0050] S2. After the earthquake, perform a three-dimensional laser scan on the same steel-concrete composite member to obtain post-earthquake point cloud data of the steel-concrete composite member. Specifically, in the building that has experienced an earthquake, repeat the scanning process of step S1, use the same equipment and collect point cloud data at similar stations. The obtained post-earthquake point cloud data will record the overall deformation of the corresponding steel-concrete composite member and the buckling deformation of the outer steel plate caused by the earthquake, providing current data support for subsequent comparative analysis.

[0051] S3. Based on the post-earthquake point cloud data, calculate the inter-story displacement index of the steel-concrete composite member. When the inter-story displacement index exceeds a predetermined threshold, the steel-concrete composite member is determined to be a macroscopically damaged member. Specifically, by calculating the inter-story displacement index and comparing it with a preset threshold, a rapid assessment and classification of the overall damage state of the member can be achieved. The inter-story displacement index can be a parameter in the relevant specifications for structural seismic performance assessment, such as the inter-story drift angle, which is the ratio of the relative displacement between the top and bottom of the steel-concrete composite member in the horizontal direction to the member's net height. The definition of macroscopic damage is that the overall deformation of the steel-concrete composite structure has exceeded the requirements of the building seismic code, requiring overall demolition and reconstruction. The member no longer needs local analysis. In addition to analyzing the inter-story displacement index, external visual observation can also be used. If the structure shows obvious damage such as large-area steel plate tearing or significant bulging, it can also be defined as macroscopic damage.

[0052] In this technical solution, since the overall deformation of steel-concrete composite members is not fixed by manual on-site measurement, and tools such as straightedges and plumb bobs are not accurate, the overall deformation degree of steel-concrete composite members under seismic action is obtained and quantified by post-earthquake point cloud data.

[0053] S4. For steel-concrete composite members not determined to have macroscopic damage, the pre-earthquake point cloud data and post-earthquake point cloud data are aligned based on the monitoring points. By comparing the horizontal displacement of the post-earthquake point cloud data with that of the pre-earthquake point cloud data, the buckling parameters of the steel-concrete composite member are obtained. The buckling parameters include the buckling area ratio and the failure ratio of the embedded parts. Specifically, based on preset monitoring points, an algorithm such as the iterative nearest point algorithm or its improved algorithm is used to find the optimal spatial transformation parameters to minimize the alignment error between the pre-earthquake and post-earthquake point cloud data. When a concrete member is subjected to seismic loading, and its outer steel plate undergoes out-of-plane buckling, the coordinates of the buckling area on the outer steel plate change along the normal direction. By calculating the coordinate deviation of the corresponding points in the post-earthquake point cloud data relative to the pre-earthquake point cloud data along the normal direction of the outer steel plate, and by setting a critical buckling displacement as a threshold based on plate and shell buckling theory, the buckling region can be identified. Then, its buckling area proportion can be calculated. If a buckling region contains the connection point between the embedded fastener and the steel plate, the constraint effect of the fastener on the steel plate at that point can be considered to have failed, thus obtaining the fastener failure proportion. By quantifying these two buckling parameters, the local damage, which is difficult to judge intuitively, is transformed into objective data, providing key input for subsequent damage coefficient calculations.

[0054] S5. Establish a damage coefficient assessment model based on a neural network. Input the buckling parameters and the component's design parameters into the damage coefficient assessment model to calculate the damage coefficient of the steel-concrete composite member. Specifically, the neural network consists of a large number of interconnected processing units. By adjusting the connection weights between neurons, it learns the complex nonlinear mapping relationship between the input parameters and the target output. During the training phase of the damage coefficient assessment model, a training dataset containing a large number of samples is needed. Each sample contains a set of input parameters and their corresponding damage coefficients. The training dataset can be determined through finite element numerical simulation or full-scale component experiments. The damage coefficient of the component model in the full-scale component or finite element numerical simulation can be obtained using methods such as... Park - Ang Damage models such as two-parameter models are used for calculation.

[0055] During training, optimization methods such as backpropagation are used to continuously adjust the weights within the network, gradually bringing the model's predicted values ​​closer to the true values ​​until the prediction error converges to an acceptable range. In the application phase, the buckling parameters of the component to be evaluated and its design parameters are used as input features, and a damage coefficient is ultimately output.

[0056] S6. Based on the damage coefficient value, determine the damage state of the steel-concrete composite member and the corresponding repair plan. Specifically, in accordance with the design guidelines of "no damage in minor earthquakes, repairable in moderate earthquakes, and no collapse in major earthquakes" in seismic requirements, the damage coefficient can be divided into four levels. The first level of damage coefficient corresponds to a structure that is basically intact and does not require repair. The second level of damage coefficient corresponds to a structure with minor damage, requiring reinforcement of the buckling area. The third level of damage coefficient corresponds to a structure with significant damage, requiring cutting and reinforcement of the buckling area or the area where the embedded parts have failed, and secondary grouting of newly formed pores between the outer steel plate and the concrete. The fourth level of damage coefficient indicates that the outer steel plate and embedded parts of the steel-concrete composite structure were severely damaged in the previous earthquake due to uneven stress development, and the synergistic stress-bearing effect between the outer steel plate structure and the internal concrete has basically disappeared, requiring the removal of the outer steel plate and special replacement and repair.

[0057] This technical solution establishes a systematic intelligent assessment of earthquake-damaged steel-concrete composite members, encompassing data acquisition, damage identification, and repair decision-making. By combining 3D laser scanning technology with intelligent algorithms, a tiered assessment mechanism is formed. First, macroscopically damaged members are rapidly screened using inter-story displacement indicators. Then, local buckling damage is accurately identified through point cloud data comparison. Finally, a neural network model integrates multiple parameters to precisely quantify the degree of damage. This method effectively solves the problems of low efficiency and strong subjectivity in traditional assessment methods, improving the objectivity and accuracy of assessment results. It provides a reliable technical basis for post-earthquake repair decisions, achieving an optimal balance between safety and economy.

[0058] In another technical solution, in step S3, a height of [missing information] is selected at the bottom and top of the steel-concrete composite member, respectively. H / 100 to H / 150 The representative section was used to calculate the average value of all post-earthquake point cloud coordinates within the bottom section as the bottom reference coordinates. P b (x b、 y b、 z b ) The average value of all post-earthquake point cloud coordinates within the top section is calculated as the top reference coordinate. P t (x t、 y t、 z t ) Among them, the first inter-story drift angle of the steel-concrete composite member is calculated with the height direction as the Z-axis. Inter-layer displacement angle If the first inter-story drift angle Or the second-layer inter-layer displacement angle When the steel-concrete composite member is determined to be a macroscopically damaged member, by selecting representative sections and calculating the average coordinate value, the interference of local deformation and measurement errors at the bottom and top of the outer steel plate on the overall displacement calculation can be effectively eliminated, and the deformation characteristics of the member in different directions can be fully reflected.

[0059] In this technical solution, based on the seismic design codes of various countries, the limit value of the elasto-plastic inter-story drift angle of steel-concrete structures is specified as 1 / 120 to 1 / 50. Therefore, 1% is set as the threshold for judging macroscopic failure of steel-concrete composite members.

[0060] For example, the net height of a certain double-steel plate concrete wall H =3100mm, select representative sections of 30mm height at the bottom and top respectively, with the bottom reference coordinates being... P b (1015.2, 998.7, 25.3) The top reference coordinates are P t (1018.5, 1001.2, 3075.1) Calculate the first inter-story drift angle θ in = (1018.5-1015.2) / 3050 = 0.11% Inter-story drift angle θ out = (1001.2-998.7) / 3050=0.08%Since both displacement angles are less than the specified threshold, it is determined that the component has not suffered macroscopic damage and can proceed to the next step of detailed evaluation.

[0061] In another technical solution, step S4 includes the following steps:

[0062] S41. For any point in the post-earthquake point cloud data P i (x i、 y i、 z i ) Perform elastic deformation correction. , This process generates post-earthquake corrected coordinates, and all corrected points form corrected post-earthquake point cloud data. Specifically, for steel-concrete composite members that have not experienced macroscopic damage, the post-earthquake point cloud data is corrected to include only local deformation by deducting the displacement component caused by their overall deformation. This separates the overall deformation from local damage, preventing slight tilting of the member after the earthquake from interfering with subsequent local buckling identification and improving the accuracy of local damage identification.

[0063] S42. Establish a pre-earthquake coordinate system based on pre-earthquake point cloud data, define the coordinates of all monitoring points, and form a pre-earthquake coordinate set. And calculate the pre-earthquake distance between any two monitoring points. This forms a set of pre-earthquake distances. A post-earthquake coordinate system was established based on the post-earthquake corrected point cloud data, defining the coordinates of all monitoring points and forming a set of post-earthquake coordinates. Calculate the post-earthquake distance between any two monitoring points. This forms a set of post-earthquake distances. ,in, N To monitor the number of monitoring points, For the first l Pre-earthquake coordinates of the monitoring points For the first m Pre-earthquake coordinates of the monitoring points For the first n Pre-earthquake coordinates of the monitoring points For the first l Post-earthquake corrected coordinates of monitoring points. For the first p Post-earthquake corrected coordinates of monitoring points. For the first q The post-earthquake corrected coordinates of the monitoring points specifically involve exporting the pre-earthquake coordinate data of all monitoring points from the pre-earthquake point cloud data, and the pre-earthquake coordinates of each monitoring point. Next, calculate the theoretical Euclidean distance between any two monitoring points. This forms a set of pre-earthquake distances. The post-earthquake coordinates of each monitoring point are: The calculation between any two monitoring points also adopts the Euclidean distance calculation method. This step makes full use of the geometric characteristics of the distance metric having rotation and translation invariance, effectively overcoming the problem of inconsistent external coordinate systems caused by differences in scanning angle and device pose.

[0064] S43, Multi-point combination that iterates through all monitoring points From the pre-earthquake distance set Extract the corresponding pre-earthquake distance From the post-earthquake distance set Extract the corresponding post-earthquake distance Calculate the global distance error of this multi-point combination. Select the ones that make The smallest combination of multiple points is taken as the optimal reference point group, and the center point of the optimal reference point group is set as the alignment point. M The number of monitoring points within a multi-point combination. a 1 、a 2 、a M For monitoring point index, a i and a j For the indexing of monitoring points within a multi-point combination, since buckling may occur at various locations of the outer steel plate under earthquake action, the existing point cloud alignment method is not ideal, and may be inaccurate due to slight displacement of the original monitoring points. This step can exclude those monitoring points that may have local displacement after the earthquake, thereby ensuring the reliability and representativeness of the benchmark, and significantly improving the accuracy and robustness of point cloud registration. In this technical solution, the number of monitoring points on the surface of the steel-concrete composite member can be controlled between 20 and 50, and modern computers can quickly complete the selection of the optimal benchmark point group.

[0065] Traditional point cloud registration algorithms, such as the iterative nearest-point algorithm, assume that only rigid body transformation exists between two point clouds. However, after an earthquake, the widespread local out-of-plane buckling of the outer steel plates of structural members results in a complex superposition of "global rigid body transformation" and "local non-rigid deformation" between the pre- and post-earthquake point clouds. If traditional algorithms are forcibly applied, the local deformation in the buckling region will act as severe noise interference in the registration process, causing deviations in the solved spatial transformation matrix and systematically contaminating the entire coordinate system, making subsequent buckling displacement calculations lose their accurate reference. The "optimal reference point group" point cloud alignment method breaks away from the framework of traditional algorithms. It no longer attempts to bridge the positional differences of all points. The algorithm traverses all possible combinations of monitoring points and selects the set of points with the smallest global distance error as the alignment reference. It automatically identifies and excludes monitoring points located within the buckling region, ensuring that the spatial coordinate transformation is based on an undamaged and stable structural reference. This lays a reliable coordinate system foundation for subsequent high-precision identification of the buckling region and calculation of buckling parameters. Furthermore, given the limited number of pre-set monitoring points in engineering practice, this traversal algorithm fully meets the real-time requirements of field applications in terms of computational efficiency, combining theoretical rigor with engineering feasibility.

[0066] S44. Align the center points of the pre-earthquake point cloud data and the corrected post-earthquake point cloud data. Based on the coordinate deviation of the points at the same height in the pre-earthquake point cloud data and the corrected post-earthquake point cloud data in the direction of the wall normal, identify the buckling area on the outer steel plate and calculate the buckling area ratio. Define the embedded parts within the buckling area as failures and calculate the failure ratio of the embedded parts.

[0067] In another technical solution, step S5 of establishing the damage coefficient assessment model includes the following steps:

[0068] A1. In the finite element calculation software, multiple standard finite element models of steel-concrete composite members with different design parameters are made. Based on the standard finite element models, random initial out-of-plane geometric defects are introduced on the surface of the outer steel plate to make a pre-damaged finite element model. The pre-damaged finite element model is subjected to horizontal reciprocating load simulation to obtain complete hysteretic response data and related mechanical characteristic parameters.

[0069] Specifically, multiple standard finite element models of steel-concrete composite members with different design parameters can be established in finite element calculation software such as ABAQUS or ANSYS. The design parameters include, but are not limited to, the thickness of the outer steel plate, the cross-sectional width, cross-sectional length, height of the steel-concrete composite member, the type code of the embedding member, the design spacing of the embedding member, the concrete strength, and the yield strength of the steel plate. In terms of defining the material properties and selecting the material constitutive model, a bilinear or trilinear kinematic hardening model that can reflect the Bauschinger effect can be selected for steel, and a plastic damage model that can simulate compressive crushing and tensile cracking can be selected for concrete. The contact between concrete and steel plate is defined as a normal "hard" contact and a tangential friction action is defined. The ends of the embedding members are bound to the steel plate to simulate welding or high-strength bolt connections. The part of the embedding member embedded in the concrete is bound or constrained by a connector between the embedding member and the concrete.

[0070] In this technical solution, such as Figure 2 As shown, the initial buckling regions of two pre-damaged finite element models corresponding to the same standard finite element model can be obtained by independently applying an initial axial compression or eccentric compression load to the steel plate element in the standard finite element model, so that the outer steel plate forms an initial buckling deformation that cannot be elastically recovered. Each standard finite element model can obtain several pre-damaged finite element models with different buckling region locations and buckling area ratios through different initial loading.

[0071] After the pre-damage finite element model is established, a horizontal reciprocating load that meets the requirements of the seismic code is applied to the top of the model. The load loading regime can be force-controlled or displacement-controlled. The loading amplitude increases in multiples of the component's yield displacement until the component's bearing capacity drops to 85% of the peak bearing capacity. Through this process, complete hysteretic response data and mechanical characteristic parameters, including load-displacement curves, stiffness degradation curves, and energy dissipation, are obtained from the pre-damage finite element model. The reason for using horizontal reciprocating load simulation is that it can accurately reproduce the actual stress state of the component under seismic action, including typical hysteretic characteristics such as strength degradation, stiffness degradation, and pinching effect.

[0072] A2. Based on the aforementioned mechanical characteristic parameters, calculate the damage coefficient corresponding to each pre-damage finite element model using a preset damage model. DI The damage coefficient of the standard finite element model in the elastic stage DI The calculated value is zero. Specifically, commonly used damage models include... Park - Ang Two-parameter model, modified Park - AngModels, etc., combine the monotonic damage caused by maximum deformation and the fatigue damage caused by energy dissipation through a linear weighted formula, thereby more comprehensively capturing the complex nonlinear behaviors of components under cyclic loading, such as stiffness degradation, strength decay and pinching effect, overcoming the limitations of evaluation based solely on deformation or energy as a single indicator.

[0073] In the specific calculation, the mechanical characteristic parameters extracted from step A1, such as yield bearing capacity, yield displacement, effective maximum displacement, cumulative hysteretic energy dissipation, critical failure displacement and energy dissipation, etc., are substituted into the mathematical expression of the selected damage model to calculate the unique value corresponding to each pre-damage finite element model. DI As a special case, when the standard finite element model, without introducing any damage, is analyzed within the elastic range, its deformation is far from yielding, and the hysteresis loop area is zero. Therefore, the calculated value... DI The value is defined as zero.

[0074] A3. For each pre-damaged finite element model, extract the buckling parameters and design parameters corresponding to the model before simulated loading. These, along with the corresponding damage coefficients, form the training sample set for the neural network model. Specifically, extract the initial state parameters of the pre-damaged finite element model before simulated loading. The buckling area ratio is used to quantify the range of initial defects, and the fastener failure ratio is used to quantify the initial damage of the buckling-restrained member. Design parameters include the outer steel plate thickness, the cross-sectional width, cross-sectional length, and height of the steel-concrete composite member, the fastener type code, the fastener design spacing, the concrete strength, and the steel plate yield strength. Use the extracted buckling parameters and design parameters as a set of input features, and compare them with the damage coefficients corresponding to the model in step A2. DI Pair the samples one by one to form training samples for the neural network model. The number of training samples should be sufficient to cover all possible parameter combinations, and it is generally recommended to have no less than 200 sets to ensure that the trained neural network model has good generalization ability.

[0075] A4. Construct a neural network model with buckling parameters and design parameters as input layers and damage coefficient as output layer. Input the training sample set into the neural network model for supervised learning. Adjust the network weights and biases through the error backpropagation algorithm until the prediction error of the neural network model converges, and obtain the trained damage coefficient evaluation model.

[0076] In another technical solution, the damage model is:

[0077]

[0078] in, DI The damage coefficient is... β These are combination coefficients, ranging from 0.1 to 0.15.

[0079] For the yield bearing capacity of the pre-damaged finite element model, The yield displacement of the pre-damaged finite element model; The effective maximum displacement of the pre-damaged finite element model, The cumulative hysteresis energy consumption of the pre-damage finite element model, This refers to the limit displacement when a standard finite element model is subjected to monotonically pushed-over loading.

[0080] In this technical solution, the damage coefficient is a dimensionless numerical value used to comprehensively characterize the degree of damage to a component from its intact state to complete failure. The damage model used is based on... Park - Ang The two-parameter damage model framework considers the contribution of the maximum deformation of the component under seismic loading and the cumulative energy consumption during cyclic loading to the overall damage. The maximum deformation reflects the most severe state reached by the component in a single loading, while the cumulative energy consumption characterizes the effect of repeated plastic deformation on material fatigue and damage accumulation. The combination coefficient β is used to balance the weight of these two factors on the total damage index.

[0081] In this technical solution, cyclic loading is applied to steel-concrete composite members with initial out-of-plane geometric defects. The damage evolution exhibits typical characteristics. In the initial stage of loading, the steel plate with initial deformation may experience intensified local buckling under stress. Simultaneously, the fasteners connecting the steel plate and concrete may yield or fail due to stress concentration under repeated interfacial shear and pull-out forces. Fastener failure further weakens the concrete's constraint on the steel plate, allowing the buckling deformation of the steel plate to develop more freely, forming more pronounced buckling ripples. This series of local damages significantly alters the force-displacement relationship of the steel-concrete composite member. On the one hand, it reduces the peak bearing capacity and subsequent bearing capacity retention of the member; on the other hand, it alters the hysteretic behavior of the member, reducing its energy dissipation capacity. This damage model captures the changes in overall deformation capacity caused by buckling and fastener failure through the displacement term, and captures the changes in energy dissipation characteristics caused by the deterioration of hysteretic behavior through the energy term.

[0082] The parameters in the damage model are obtained as follows: Yield bearing capacity and yield displacement The yield strength can be determined through the skeleton curve of the component using the equal energy method or geometric construction method. Generally, 75% of the peak load in the skeleton curve can be used as the yield strength. (Effective maximum displacement) The maximum displacement experienced during cyclic loading can be used, or the displacement corresponding to the bearing capacity decreasing to 85% after exceeding the peak load can be used to determine the coefficient. βIts value can be calculated based on the fitting formula obtained from regression of a large amount of experimental data in relevant research papers according to the component type. This formula usually includes design parameters such as axial compression ratio, shear span ratio, and steel ratio. To simplify the calculation, for steel-concrete composite members, β The value can be set between 0.1 and 0.15.

[0083] For example, a pre-damaged finite element model is subjected to cyclic loading using finite element software, and the yield bearing capacity is... =950.9 kN, corresponding yield displacement =21.7mm, effective maximum displacement =35mm, cumulative hysteresis energy consumption =150000 kN·mm, the ultimate displacement of a standard finite element model under monotonically pushed-over loading. =103.3 mm, combination coefficient β The value is 0.1334, which is the damage coefficient corresponding to this pre-damaged finite element model. DI =(1-0.1334)×0.163+0.1334×1.933=0.399.

[0084] In another technical solution, when the damage coefficient DI ≤ 0.1 When the steel-concrete composite structure is determined to be basically intact, the corresponding repair plan is continued use; when 0.1 < DI ≤ 0.5 When the steel-concrete composite member is determined to have minor damage, the corresponding repair plan is to locally repair the buckling area; when 0.5 < DI ≤ 1.0 When the damage coefficient is determined to be moderate, the corresponding repair solution is overall reinforcement of the component; when the damage coefficient is... DI > 1.0 When the steel-concrete composite member is determined to be severely damaged, the corresponding repair solution is to replace the member.

[0085] In this technical solution, when the damage coefficient does not exceed 0.1, it indicates that the integrity and bearing capacity of the steel-concrete composite member have not significantly degraded after the earthquake, the structure is in an elastic or basically elastic working state, and the member is undamaged or has only extremely small local deformation that does not affect the bearing capacity.

[0086] When the damage coefficient is between 0.1 and 0.5, it indicates that the steel-concrete composite member may experience more local buckling and failure of some fittings, but the core bearing capacity and energy dissipation performance of the steel-concrete composite member are not seriously weakened and the overall deformation is controllable. The corresponding repair scheme is defined as "local repair", such as correcting and reinforcing the buckled steel plate.

[0087] When the damage coefficient is between 0.5 and 1, it indicates that although the steel-concrete composite member has not yet lost its load-bearing capacity and there is still a synergistic effect between the steel plate and the internal concrete, the safety margin has been greatly reduced, and there is a risk of further damage in aftershocks. The corresponding repair plan is "overall reinforcement", which involves cutting the steel plate and strengthening local areas, pressure grouting to repair concrete cracks, wrapping the member with a steel sleeve, and pasting carbon fiber cloth to comprehensively improve the load-bearing capacity and ductility of the member.

[0088] When the damage coefficient is greater than 1, it means that the synergistic effect between the steel plate and the concrete in the steel-concrete composite member is basically eliminated, the outer steel plate buckles over a large area and separates from the concrete, and a large number of embedded parts fail. The corresponding repair solution is to replace the whole structure.

[0089] In another technical solution, based on the aligned pre-earthquake point cloud data and the corrected post-earthquake point cloud data, the displacement deviation Δ in the direction normal to the outer steel plate is selected to be greater than the critical buckling displacement. U a The region is designated as the buckling region, where the critical buckling displacement is... , s For the design spacing of the embedded parts, t The thickness of the outer steel plate. f The range of values ​​for the empirical coefficient 0.15～ 0.25, Specifically, by aligning the two sets of point cloud data, technicians can use the cross-sectional analysis function in existing commercial point cloud processing software (such as CloudCompare and Geomagic Control X) to generate horizontal cross-sections at certain intervals along the height direction of the component. On any cross-section, by comparing the X-axis and Y-axis coordinates of the pre-earthquake and post-earthquake point clouds, discrete points with displacement deviations Δ greater than the critical buckling displacement Ua can be identified. Subsequently, by connecting these discrete points exceeding the limit using the software's built-in contour detection or boundary extraction tools, the buckling region contour on the horizontal cross-section can be clearly defined. By synthesizing the contours of all adjacent horizontal cross-sections, the buckling region can be finally determined.

[0090] buckling area ratio , The total area of ​​the buckling region. This represents the total surface area of ​​the steel plate.

[0091] Embedded component failure rate , This refers to the number of inserts located within the buckling region. This refers to the total number of embedded parts on the steel plate. Specifically, for a steel-concrete composite member that has already been constructed, the position and spacing of the embedded parts inside the steel plate can be obtained from the member design drawings.

[0092] In another technical solution, the buckling parameter also includes the proportion of the most unfavorable weak constraint area. Its calculation methods include:

[0093] The horizontal section of the component with the most failed inserts is identified as the most unfavorable section. Specifically, the purpose of selecting the most unfavorable section is to identify the weakest link in the insert-concrete structure system. Since the number of failed inserts is the largest at this section, it means that the constraint capacity of the outer steel plate on the inner concrete is most severely lost. When rows of inserts fail, the combined effect between the steel plate and the concrete is weakened, and the concrete is more likely to undergo lateral expansion and crushing under pressure, thereby significantly reducing the overall load-bearing capacity and ductility of the component.

[0094] At the most unfavorable section, the intersection of the effective embedded part and the outer steel plate, and the intersection of adjacent plates of the outer steel plate are considered as constraint points. Two adjacent constraint points are considered as one calculation unit. The formula for calculating the area of ​​the weakly constrained concrete region corresponding to each calculation unit is as follows: ,in θ 0 For the boundary curve tangent, b k For the first k The distance between constraint points on both sides of each computational unit, and the proportion of the area with the most unfavorable weak constraint. ,in The area of ​​the most unfavorable section is specifically defined as follows: for a horizontal section containing embedded fasteners, the connection points between the fasteners that can function normally and the steel plate, as well as the intersections of adjacent plates of the steel plate itself, can all be considered as points where the steel plate exerts its constraint on the internal concrete. Between two constraint points, the constraint force of the steel plate on the concrete gradually decreases from the two constraint points towards the middle. Therefore, the area of ​​the weak constraint region can be calculated based on the existing theory of strong-weak constraint zones in steel-concrete composite structures. On a section, the ratio between the sum of the areas of all weak constraint regions and the section size is the proportion of the most unfavorable weak constraint area.

[0095] In this technical solution, the overall mechanical properties of steel-concrete composite members are strongly correlated with their weakest link. The buckling area ratio and the failure ratio of the embedded fasteners are used to characterize the damage to the steel plates of the steel-concrete composite structure, while the ratio of the weakest constraint area is used to characterize the weakening of the constraint on the concrete after an earthquake.

[0096] For example, consider a steel-concrete composite member using tie bolts as fasteners, wherein the cross-sectional length is... B =2000mm, cross-sectional width L =200mm, the embedded part spacing is 600mm, the outermost embedded part is 100mm from the edge line, according to relevant research in the field of square steel tube concrete technology, the boundary curve chamfer. θWe can uniformly take 0.21 rad.

[0097] like Figure 3 As shown in (a), under condition (a), all fasteners are intact and the outer steel plate does not buckle. There are a total of 12 calculation units on the most unfavorable section. Three of these are located in the middle along the long side of the outer steel plate, with a constraint point spacing of 600mm. Two are located at the edge, with a constraint point spacing of 100mm. Since there are no fasteners on the side of the steel-concrete composite member, the intersection of the two plates is considered as the constraint point, with a constraint point spacing of 200mm. The areas of the weakly constrained regions corresponding to all calculation units on this most unfavorable section are summarized and compared with the section area to obtain the final result. =0.207.

[0098] like Figure 3 As shown in (b), in condition (b), one embedding component fails. There are a total of 10 calculation units on the most unfavorable cross-section. Since one embedding component fails and can no longer exert tensile force on the outer steel plate to constrain the concrete, one long side of the outer steel plate changes to one calculation unit located in the middle with a constraint point spacing of 1200mm, one calculation unit located in the middle with a constraint point spacing of 600mm, and two calculation units located at the edge with a constraint point spacing of 100mm. The areas of the weakly constrained regions corresponding to all calculation units on this most unfavorable cross-section are summarized and compared with the cross-sectional area to obtain the final result. =0.327.

[0099] like Figure 3 As shown in (c), in condition (c), two embedded fasteners fail. There are a total of 8 calculation units on the most unfavorable cross-section. The outer steel plate has one calculation unit located in the middle with a constraint point spacing of 1800mm, and two calculation units located at the edge with a constraint point spacing of 100mm. The areas of the weakly constrained regions corresponding to all calculation units on this most unfavorable cross-section are summarized and compared with the cross-sectional area to obtain the final result. =0.581.

[0100] When the two middle embedded parts fail;

[0101] In another technical solution, the damage coefficient assessment model employs a BP neural network model comprising an input layer, a hidden layer, and an output layer. The input layer of the BP neural network model includes multiple input neuron nodes, and the hidden layer has multiple hidden neuron nodes. Each input neuron node is connected to each of the hidden neuron nodes, and all hidden neuron nodes are connected to the output layer. The buckling parameters include the buckling area ratio, the failure ratio of the embedded fasteners, and the ratio of the area of ​​the most unfavorable weak constraint. The design parameters include the outer steel plate thickness, the cross-sectional width, cross-sectional length, and height of the steel-concrete composite member, the embedded fastener type code, the embedded fastener design spacing, the concrete strength, and the steel plate yield strength. The buckling parameters and design parameters are used as input values ​​for the input neuron nodes, and the output results are compared with the corresponding damage coefficients. The algorithm optimizes the neural network model to obtain the damage coefficient assessment model.

[0102] In this technical solution, the neural network model is a BP neural network model, and the method for training and optimizing the neural network model includes the following steps;

[0103] B1. Taking buckling parameters, design parameters, and other factors as influencing factors, the total number of parameter types is used as the number of input neuron nodes m, and the predicted damage coefficient is used as the sole output value, i.e., the number of nodes in the output layer. c =1, number of hidden neurons in the hidden layer c 1 for , a A random constant between 1 and 10;

[0104] B2. Normalize the sample set data; its mathematical expression is: ,in Sample data representing influencing factors. , These are the minimum and maximum values ​​in the sample data, respectively. The data on influencing factors are after dimensionless processing;

[0105] B3. Initialize the mapping relationship between the population particles and the weights and thresholds of the BP neural network, including particle dimension, initial velocity, population size, learning factor, and inertia weight;

[0106] B4. Input the normalized input and output variables into the neural network model, calculate the particle's fitness function value, and obtain the particle's historical best fitness and global fitness. The particle's fitness function value is the mean square error of the calculation result, and its function expression is: ,in, Represented as the first i The predicted value for each sample, For the firstj The true value of each sample n This represents the total number of calculation results from the neural network model.

[0107] B5. Iteratively calculate the particle fitness, update the historical best fitness and the global fitness according to the preset update conditions, until the preset iteration end condition is met;

[0108] B6. Update the weights and thresholds of the neural network model to obtain the damage coefficient evaluation model.

[0109] It should be noted that although the steps are described in a specific order above, this does not mean that they must be performed in that order. In fact, some of these steps can be executed concurrently, or even in a different order, as long as the required functionality is achieved. The number of devices and processing scale described herein are for simplification of the invention; applications, modifications, and variations of this invention will be readily apparent to those skilled in the art.

[0110] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

Claims

1. An intelligent evaluation method for a seismic damage steel reinforced concrete structure, characterized by, The method comprises the following steps: S1, a plurality of monitoring points are arranged on the outer steel plate surface of the steel reinforced concrete component in the building, and pre-earthquake point cloud data containing all the monitoring points are obtained through three-dimensional laser scanning; S2, after the earthquake, three-dimensional laser scanning is performed on the same steel reinforced concrete component to obtain post-earthquake point cloud data of the steel reinforced concrete component; S3, based on the post-earthquake point cloud data, the interlayer displacement index of the steel reinforced concrete component is calculated, and when the interlayer displacement index exceeds a limited threshold value, the steel reinforced concrete component is determined to be a macroscopic damage component; S4, for the steel reinforced concrete component not determined to be macroscopic damage, the pre-earthquake point cloud data and the post-earthquake point cloud data are aligned in coordinates based on the monitoring points, the horizontal displacement of the post-earthquake point cloud data compared with the pre-earthquake point cloud data is compared, and the buckling parameters of the steel reinforced concrete component are obtained, the buckling parameters including a buckling area ratio and a fastener failure ratio; Buckling area ratio , Total area of buckling region, Total surface area of outer steel plate; Proportion of embedded fasteners that fail , Number of embedded fasteners located in the bending region, Total number of embedded fasteners on the outer steel sheet; S5, a damage coefficient evaluation model based on a neural network is established, the buckling parameters and the design parameters of the component are input into the damage coefficient evaluation model, and the damage coefficient of the steel reinforced concrete component is calculated; S6, according to the value of the damage coefficient, the corresponding damage state of the steel reinforced concrete component is determined, and the corresponding component repair scheme is determined. 2.The intelligent evaluation method of the seismic damage steel-concrete composite structure according to claim 1, wherein, In step S3, representative segments with height of H / 100~H / 150 are selected at the bottom and top of the steel reinforced concrete member respectively, and the average of all post-earthquake point cloud coordinates in the bottom segment is calculated as the bottom reference coordinate P b (x b、 y b、 z b ) The average of all post-earthquake point cloud coordinates in the top segment is calculated as the top reference coordinate P t (x t、 y t、 z t ) Wherein, taking the height direction as the Z axis, the first inter-story drift angle and the second inter-story drift angle of the steel reinforced concrete member are calculated, if the first inter-story drift angle or the second inter-story drift angle , the steel reinforced concrete member is determined as a macroscopic failure member. 3.The intelligent evaluation method of the seismic damage steel-concrete composite structure according to claim 2, wherein, Step S4 comprises the following steps: S41, for any one point in the post-earthquake point cloud data P i (x i、 y i、 z i ) carrying out elastic deformation correction, , , form the post-earthquake correction coordinates, and all the corrected point positions form the corrected post-earthquake point cloud data; S42. Establish a pre-earthquake coordinate system based on pre-earthquake point cloud data, define the coordinates of all monitoring points, and form a pre-earthquake coordinate set. And calculate the pre-earthquake distance between any two monitoring points. This forms a set of pre-earthquake distances. A post-earthquake coordinate system was established based on the post-earthquake corrected point cloud data, defining the coordinates of all monitoring points and forming a set of post-earthquake coordinates. Calculate the post-earthquake distance between any two monitoring points. Forming a set of post-earthquake distances ,in, N To monitor the number of monitoring points, For the first l Pre-earthquake coordinates of the monitoring points. For the first m Pre-earthquake coordinates of the monitoring points. For the first n Pre-earthquake coordinates of the monitoring points. For the first l Post-earthquake corrected coordinates of monitoring points. For the first p Post-earthquake corrected coordinates of monitoring points. For the first q Post-earthquake corrected coordinates of monitoring points; S43, Multi-point combination that iterates through all monitoring points From the pre-earthquake distance set Extract the corresponding pre-earthquake distance From the post-earthquake distance set Extract the corresponding post-earthquake distance Calculate the global distance error of this multi-point combination. Select the ones that make The smallest combination of multiple points is taken as the optimal reference point group, and the center point of the optimal reference point group is set as the alignment point. M The number of monitoring points within a multi-point combination. a 1 、a 2 、 a M For monitoring point index, a i and a j This is an index of monitoring points within a multi-point combination; S44, the center points of the pre-earthquake point cloud data and the corrected post-earthquake point cloud data are aligned, the buckling area ratio is calculated by identifying the buckling area and calculating the buckling area ratio through the coordinate deviation of the points at the same height in the pre-earthquake point cloud data and the corrected post-earthquake point cloud data in the normal direction of the outer steel plate, and the fastener failure ratio is calculated by defining the fastener in the buckling area as invalid. 4.The intelligent evaluation method of the seismic damage steel-concrete composite structure according to claim 1, wherein, The damage coefficient evaluation model established in step S5 comprises the following steps: A1, in the finite element calculation software, a plurality of standard finite element models of steel reinforced concrete components with different design parameters are made, random initial out-of-plane geometric defects are introduced on the outer steel plate surface of the standard finite element model to make a pre-damage finite element model, and the pre-damage finite element model is simulated and loaded with horizontal reciprocating load to obtain complete hysteresis response data and related mechanical characteristic parameters; A2, based on the mechanical characteristic parameter, calculating the damage coefficient corresponding to each pre-damage finite element model through a preset damage model DI wherein the damage coefficient of the standard finite element model in the elastic stage DI The calculated value is zero. A3, for each pre-damage finite element model, the corresponding buckling parameters and design parameters before simulation and loading are extracted to form a training sample set of the neural network model; A4, a neural network model with the buckling parameters and the design parameters as the input layer and the damage coefficient as the output layer is constructed, the training sample set is input into the neural network model for supervised learning, the network weight and bias are adjusted through the error back propagation algorithm until the prediction error of the neural network model converges, and a trained damage coefficient evaluation model is obtained. 5.The intelligent evaluation method for seismic damage of steel reinforced concrete structure according to claim 4, wherein, The damage model is: ； wherein, DI is a damage coefficient, β is a combination coefficient, with a value ranging from 0.1 to 0.15; the yield load of the pre-damage finite element model under simulated loading, the yield displacement of the pre-damage finite element model under simulated loading; the effective maximum displacement of the pre-damage finite element model under simulated loading, the cumulative hysteretic energy of the pre-damage finite element model under simulated loading, the ultimate displacement when monotonic pushover loading is applied to the standard finite element model. 6.The intelligent evaluation method of the seismic damage steel-concrete composite structure according to claim 5, characterized in that, When the damage coefficient is DI ≤ 0.1 , the steel reinforced concrete member is determined to be basically intact, and the corresponding repair scheme is to continue to use; when 0.1 < DI ≤ 0.5 , the steel reinforced concrete member is determined to be slightly damaged, and the corresponding repair scheme is to locally repair the buckling part; when 0.5 < DI ≤ 1.0 , the steel reinforced concrete member is determined to be moderately damaged, and the corresponding repair scheme is to reinforce the whole member. When the damage coefficient DI > 1.0 is greater than 1.5, the steel-concrete composite member is determined to be severely damaged, and the corresponding repair scheme is to replace the member. 7.The intelligent evaluation method of the seismic steel-concrete composite structure according to claim 3, wherein, Based on the aligned pre-earthquake point cloud data and the corrected post-earthquake point cloud data, a region with a displacement deviation Δ in the normal direction of the outer steel plate greater than a critical buckling displacement U a is selected as a buckling region, where the critical buckling displacement s is the design spacing of the embedded part, t is the thickness of the outer steel plate, f is an empirical coefficient with a value range of 0.15～0.25 .​ 8.The intelligent evaluation method of the seismic steel-concrete composite structure according to claim 1, wherein, The buckling parameter further comprises a least favorable weak constraint area ratio The calculation method comprises: The component horizontal section with the largest number of fastener failures is determined as the most unfavorable section; In the most unfavorable cross section, the intersection position of the effective embedded part and the outer steel plate is regarded as a constraint point, and two adjacent constraint points are regarded as a calculation unit. The calculation formula of the corresponding concrete weak constraint area of each calculation unit is: Wherein θ 0 is the tangent angle of the boundary curve, b k is the distance between the constraint points on both sides of the i th calculation unit, and the most unfavorable weak constraint area ratio k Wherein is the cross-sectional area of the most unfavorable cross section.​ 9.The intelligent evaluation method of the seismic steel-concrete composite structure according to claim 4, wherein, The damage coefficient evaluation model adopts a BP neural network model including an input layer, a hidden layer and an output layer, the input layer of the BP neural network model includes a plurality of input neuron nodes, a plurality of hidden neuron nodes are arranged on the hidden layer, each input neuron node is connected with each hidden neuron node respectively, the hidden neuron nodes are connected with the output layer, the buckling parameters include a buckling area ratio, a fastener failure ratio and a most unfavorable weak constraint area ratio, the design parameters include an outer steel plate thickness, a cross section width of a steel reinforced concrete member, a cross section length, a height, a fastener type code, a fastener design spacing, a concrete strength and a steel plate yield strength, the buckling parameters and the design parameters are taken as input neuron node input values, an output result is compared with a corresponding damage coefficient, an algorithm optimized neural network model is obtained, and thus the damage coefficient evaluation model is obtained.

Images