A method and system for monitoring and analyzing the seismic safety of urban building complexes

Abstract

This invention relates to a method and system for earthquake safety monitoring and analysis of urban building clusters. The method includes the following steps: based on structural design drawings and structural inspection data, acquiring characteristic parameters of the building clusters, classifying the urban building clusters, and selecting representative index buildings in each cluster; establishing an equivalent numerical analysis model, analyzing the structural resilience using elastoplastic analysis based on measured seismic time histories, and determining the performance level of buildings within each cluster in conjunction with the building cluster characteristic parameters; establishing a mapping relationship between the performance level of the index buildings and the performance levels of other buildings within the cluster; extracting the latest seismic time histories recorded by local stations, calculating the performance level of the index buildings, and obtaining the performance level of all buildings within the cluster based on the mapping relationship, thereby completing the earthquake safety monitoring and analysis of urban building clusters. Compared with existing technologies, this invention has advantages such as high-precision monitoring, efficient analysis, and engineering applicability.

Description

Technical Field

[0001] The present invention relates to the field of seismic safety monitoring and analysis of building structures, and in particular to a method and system for seismic safety monitoring and analysis of urban building groups. Background Art

[0002] Seismic safety monitoring and assessment of urban building groups is a key link in the urban disaster prevention and mitigation system, aiming to use structural monitoring data, building attribute information, and ground motion records to timely evaluate the safety status of building groups under seismic action, providing a basis for urban emergency response and post-disaster recovery. At present, most seismic safety assessment methods still take individual buildings as the basic analysis object, and data processing, dynamic characteristic identification, and numerical simulation analysis need to be carried out for each building separately. For example, Chinese Patent CN118898095A discloses a method for seismic risk assessment of urban building groups based on multi-source measured information, which uses a method based on multi-source measured information. By obtaining the basic design information and measured dynamic characteristics of urban building groups, an equivalent numerical analysis model is established, and a seismic risk assessment is carried out in combination with an intelligent prediction model to achieve automatic and accurate assessment. However, for large-scale urban building groups, such methods not only have a huge calculation amount and a long assessment period, but also it is difficult to quickly form a group safety assessment result after an earthquake, limiting their application in real-time monitoring and early warning and rapid response.

[0003] To improve efficiency, some studies have tried to cluster buildings according to similar characteristics and select representative buildings for alternative analysis to reduce the overall calculation amount of the building group. However, the existing clustering methods generally have the following deficiencies: (1) The clustering features often only include limited basic indicators such as building height, number of floors, and construction year, and the clustering results cannot comprehensively reflect the true dynamic behavior characteristics and geographical locations of buildings; (2) The selection of indicator buildings usually adopts simple methods such as "the closest to the cluster center" or manual judgment, without considering the dispersion degree of structural parameters within the cluster, the diversity of text features, and the distribution of buildings in physical space, resulting in insufficient representativeness of candidate buildings; (3) The existing technology lacks a mechanism to automatically determine the number of indicator buildings based on the number of buildings within the cluster and the complexity of features, and cannot adapt to different cluster scales and feature diversities; (4) The performance inference of other buildings within the cluster generally relies on empirical relationships or single models, lacking a mapping mechanism driven by multi-source structural features, limiting the accuracy and universality of the assessment results.

[0004] In the context of large-scale urban building groups, complex structural types, and multi-source monitoring data, the existing seismic safety monitoring and analysis methods have significant deficiencies in the selection of representative buildings, the inference of performance levels within the cluster, and the overall analysis efficiency. There is an urgent need for a seismic safety monitoring and analysis method that can overcome the above defects to achieve accurate and efficient analysis of the seismic safety of urban building groups. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for earthquake safety monitoring and analysis of urban building clusters. By using initial design information and actual service dynamic characteristics, cluster analysis is used to select index buildings, establish finite element models to calculate their performance levels, and quickly infer the performance levels of other buildings based on intra-cluster mapping relationships, thereby achieving efficient and accurate earthquake safety monitoring and analysis.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A method for monitoring and analyzing the seismic safety of urban building complexes includes the following steps:

[0008] S1, Building Cluster Feature Acquisition: Based on structural design drawings and structural inspection data, obtain building cluster feature parameters, which include the initial design information and actual service dynamic characteristics of the urban building cluster.

[0009] S2, Clustering of Building Clusters and Selection of Indicator Buildings: Based on the building cluster characteristic parameters, cluster analysis is used to classify urban building clusters, dividing buildings with similar design features and dynamic characteristics into similar clusters, and selecting representative indicator buildings in each cluster.

[0010] S3, Numerical Model Analysis of Building Clusters: Establish an equivalent numerical analysis model, analyze the structural recovery capacity based on the measured ground motion time history using the elastoplastic analysis method, and determine the performance level of buildings in each cluster of urban building clusters by combining the characteristic parameters of the building clusters.

[0011] S4, Establish intra-cluster performance level mapping relationship: Based on the performance level of buildings in each cluster of the urban building complex, establish a mapping relationship between the performance level of the index building and the performance level of other buildings in the cluster;

[0012] S5, Seismic Safety Monitoring and Analysis of Urban Building Clusters: Extract the latest seismic motion time history recorded by local stations, calculate the performance level of index buildings, and obtain the performance level of all buildings within the cluster based on the aforementioned mapping relationship, thereby completing the seismic safety monitoring and analysis of urban building clusters.

[0013] The initial design information for the urban building complex includes:

[0014] Textual features: intended use, structural type, plan shape, site category, foundation type, design code adopted, and seismic fortification intensity;

[0015] Numerical characteristics: building height, building length, building width, number of floors, and year of construction;

[0016] Spatial location characteristics: the latitude and longitude of the building's location and the distance of the building from the nearest seismic station.

[0017] The actual service dynamic characteristics are obtained by acquiring the acceleration response of the building structure under environmental pulsation collected by acceleration sensors deployed on the target building structure as structural inspection data, and performing time-domain and frequency-domain analysis on the acceleration response to obtain the actual service dynamic characteristics, which include modal period and modal damping ratio, both of which are numerical features.

[0018] The clustering analysis methods include K-means clustering, K-prototypes clustering, or feature-based dimensionality reduction clustering analysis methods.

[0019] The method for selecting representative landmark buildings is as follows:

[0020] A unified feature space is constructed based on the numerical and textual features of the building cluster's feature parameters. The textual features are encoded into numerical form using integer mapping. The Euclidean geometric distance of each building relative to the cluster center is calculated to quantify the similarity of buildings in terms of structural attributes and dynamic characteristics. The clustering distance similarity score is then calculated.

[0021] ,

[0022] In the formula, Indicates the first Cluster distance similarity score between each building and the cluster center; Indicates the cluster center is at the th Cluster centroids on each feature dimension; Indicates the first in the cluster The building in The values ​​taken on each feature dimension The sum of the feature dimensions for textual and numerical features;

[0023] Perform statistical distribution analysis on the numerical features in the building cluster feature parameters of the current cluster, and calculate the numerical feature scores;

[0024] ,

[0025] In the formula, For the first Numerical feature scores for each building; For the first Numerical characteristics of a building The value; Numerical features within the cluster The mean; Numerical features within the cluster The maximum value; Feature dimensions for numerical features;

[0026] Frequency statistics are performed on the textual features of each dimension in the building complex feature parameters to determine the dominant type under the corresponding dimension, and buildings belonging to the dominant type are selected. The textual feature score is then calculated.

[0027] ,

[0028] ,

[0029] In the formula, For the first Textual feature scores for each building; It is the first The building features textual characteristics The type on; Intra-cluster textual features The dominant type; For text-type features, the feature dimensions are... For the first Textual features of individual buildings The rating;

[0030] Spatial location score is calculated based on the distance between the building and the nearest seismic station. :

[0031] ,

[0032] In the formula, For the first A score for the spatial location of each building. For the first The distance between a building and its nearest corresponding seismic station; This is the average distance between all buildings within the cluster and their nearest corresponding seismic station; This represents the maximum distance between all buildings within the cluster and their corresponding nearest seismic station.

[0033] The cluster distance similarity score, numerical feature score, textual feature score, and spatial location score are weighted and fused to obtain a comprehensive representativeness score for buildings within the cluster. :

[0034] ,

[0035] in, These are the weighting coefficients for the four categories of ratings, and their sum is 1; This is the building function importance coefficient. For ordinary buildings, its value is 1, and for key function buildings, the value is less than 1.

[0036] Based on comprehensive representativeness score Sort by size from smallest to largest, and take the first few. Each corresponding building is a landmark building. The number of buildings is an indicator.

[0037] The number of index buildings is determined based on the number of buildings within the cluster, the dispersion of numerical features within the cluster, and the diversity of textual features, wherein:

[0038] ,

[0039] In the formula, Based on quantity, , The basic proportional coefficient, For the first The total number of buildings within the cluster; It is a numerical feature dispersion index within a cluster. , This is the dispersion influence coefficient. Numerical features within the cluster standard deviation Numerical features within the cluster The mean; As an index of textual feature diversity within a cluster, , This represents the text feature coverage coefficient. Text-type features Number of types, Text-type features The first in The frequency of each type within a cluster.

[0040] S3 includes the following steps:

[0041] S301, based on the initial design information of urban building complexes and actual service dynamic characteristics, and with reference to the technical manual, determines the performance points of the capability curve, including the yield point, the maximum bearing capacity point and the ultimate deformation point;

[0042] S302, Establish an equivalent numerical analysis model, input the measured ground motion time history into the equivalent numerical model, and perform elastoplastic analysis to obtain the structural restoring force skeleton curve;

[0043] S303 classifies the performance levels of engineering structures into five levels based on the structural restoring force skeleton curve and the determined capacity curve performance points: normal operation, immediate use, life safety, near collapse, and collapse.

[0044] The establishment of the equivalent numerical analysis model specifically involves:

[0045] Establish the dynamic equations for an equivalent single-degree-of-freedom elastoplastic model of the building structure:

[0046] ,

[0047] In the formula, For quality; It is the viscous damping coefficient; For restorative power; , and These are the displacement, velocity, and acceleration of the structure relative to the ground, respectively. This refers to the absolute acceleration of the ground.

[0048] The aforementioned dynamic equations are mass-normalized to obtain normalized dynamic equations, which serve as the equivalent numerical analysis model, as shown in the following equation:

[0049] ,

[0050] In the formula, The structural angular frequency; The damping ratio; The restoring force after mass normalization; , , Determine using the following formula:

[0051] ,

[0052] In the formula, The measured modal period.

[0053] S4 includes the following steps:

[0054] S401, Construct the mapping model database;

[0055] Structural characteristic parameters, dynamic characteristic parameters, spatial location characteristic parameters, and performance levels of indicative buildings within a cluster are extracted and used as inputs to a performance mapping model. The performance levels within the cluster are used as outputs to construct a mapping model database.

[0056] S402, Establish a performance level mapping model;

[0057] A performance level mapping model is constructed and trained using a mapping model database. By learning the relationship between structural features and performance response, the performance level of buildings within a cluster can be quickly predicted.

[0058] The performance level mapping model, based on traditional regression or neural network models, introduces a structural similarity weight kernel function in the output stage to enhance the propagation ability of the indicative building performance response to buildings within the cluster. The structural similarity weight kernel function is defined as follows:

[0059] ,

[0060] In the formula, and The first Feature vectors of individual buildings and landmark buildings and The structure is a latitude and longitude vector. and To control the range of weight decay, This is the structural similarity weight kernel function;

[0061] Based on the structural similarity weight kernel function, the refined analysis results of the indicative building are generalized to other buildings within the cluster to obtain the predicted performance values ​​of the buildings:

[0062] ,

[0063] In the formula, As a benchmark for the performance level of buildings, The first prediction of traditional regression models or neural network models The performance level of each building This represents the performance level output by the performance level mapping model.

[0064] An earthquake safety monitoring and analysis system for urban building complexes, used to implement the method, the system comprising:

[0065] Building cluster feature acquisition module: Based on structural design drawings and structural inspection data, acquire building cluster feature parameters, which include the initial design information and actual service dynamic characteristics of the urban building cluster;

[0066] Building Cluster Feature Clustering and Indicator Building Selection Module: Based on the building cluster feature parameters, cluster analysis is used to classify urban building clusters, dividing buildings with similar design features and dynamic characteristics into similar clusters, and selecting representative indicator buildings in each cluster.

[0067] Building Cluster Numerical Model Analysis Module: Establish an equivalent numerical analysis model, analyze the structural resilience based on the measured ground motion time history using elastoplastic analysis, and determine the performance level of buildings within each cluster of the urban building cluster by combining the characteristic parameters of the building cluster.

[0068] Module for establishing intra-cluster performance level mapping relationship: Based on the performance level of buildings in each cluster of the urban building complex, establish a mapping relationship between the performance level of the index building and the performance level of other buildings in the cluster;

[0069] The urban building cluster seismic safety monitoring and analysis module extracts the latest seismic motion time history recorded by local stations, calculates the performance level of index buildings, and obtains the performance level of all buildings within the cluster based on the aforementioned mapping relationship, thereby completing the urban building cluster seismic safety monitoring and analysis.

[0070] Compared with the prior art, the present invention has the following beneficial effects:

[0071] (1) This invention obtains the characteristic parameters of the building complex by simultaneously introducing structural design drawing information and structural inspection data, and combines the initial design state of the building with the actual service state. This can effectively make up for the uncertainty brought about by relying solely on design data or experience parameters, and provide a more reliable data foundation for subsequent cluster analysis and performance calculation.

[0072] (2) This invention divides buildings with similar design features and dynamic characteristics into similar clusters by cluster analysis based on building cluster feature parameters, and selects representative index buildings within the clusters. Under the premise of ensuring the coverage of building cluster structural features, it can significantly reduce the number of buildings that need to be analyzed, thereby reducing the overall calculation scale and improving the efficiency of building cluster analysis.

[0073] (3) By establishing an equivalent numerical analysis model and combining it with measured ground motion time history to carry out elastoplastic analysis, this invention can quantitatively characterize the nonlinear response and recovery capacity of building structures under earthquake action, and on this basis determine the performance level of the building, so that the safety status analysis of building groups is no longer limited to elastic assumptions or empirical judgments, thereby improving the engineering credibility of the performance level judgment results.

[0074] (4) By establishing a mapping relationship between the performance level of an index building and the performance level of other buildings in the cluster, this invention can extend the detailed analysis results of a small number of index buildings to the remaining buildings in the cluster, realize the rapid inference of the performance level of the building group, avoid repeated modeling and analysis of each building, effectively balance computational efficiency and result accuracy, and improve the feasibility of the analysis method in large-scale urban applications.

[0075] (5) This invention utilizes the latest seismic motion time history recorded by local seismic stations to calculate the performance level of index buildings and obtains the overall performance distribution of the building group by combining the performance level mapping relationship within the cluster. This enables the analysis of the seismic safety status of urban building groups based on real-time monitoring data.

[0076] (6) The present invention realizes rapid monitoring and analysis of earthquake safety of building clusters. Compared with the existing technology, the present invention can take into account the analysis accuracy, coverage and real-time performance in large-scale building cluster scenarios, and provides an efficient, reliable and engineering-promotable technical solution for the safety assessment of urban building clusters. Attached Figure Description

[0077] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0078] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. These embodiments are based on the technical solution of the present invention and provide detailed implementation methods and specific operating procedures. However, the scope of protection of the present invention is not limited to the following embodiments.

[0079] This embodiment provides a method for monitoring and analyzing the seismic safety of urban building complexes, such as... Figure 1 As shown, it includes the following steps:

[0080] S1, Building Complex Feature Acquisition: Based on structural design drawings and structural inspection data, obtain building complex feature parameters.

[0081] Specifically, prior to an earthquake, the initial design information and actual service dynamic characteristics of urban building complexes are obtained as characteristic parameters of the building complexes.

[0082] (1) The initial design information of the urban building complex includes:

[0083] Textual features include multiple dimensions such as intended use, structural type, plan shape, site category, foundation type, applicable design codes, and seismic fortification intensity. Each dimension has corresponding feature types. For example, intended use includes residential, commercial, industrial, school, hospital, and public spaces; structural type includes shear wall, frame, frame-shear wall, masonry, and steel structures; plan shape includes I-shaped, cross-shaped, and rectangular; site category can be classified as Class A, B, C, and D according to national standards; foundation type includes raft foundation, box foundation, isolated foundation, strip foundation, and pile foundation; applicable design codes include various relevant codes; and seismic fortification intensity is classified as six, seven, eight, and nine degrees according to national standards.

[0084] Numerical features include multiple dimensions such as building height, building length, building width, number of floors, and construction year, with each dimension corresponding to a feature value;

[0085] Spatial location characteristics: the latitude and longitude of the building's location and the distance of the building from the nearest seismic station.

[0086] (2) Actual service dynamic characteristics

[0087] In this embodiment, two orthogonal acceleration sensors are deployed near the centroid of the top layer of the target building structure to obtain the acceleration response of the building structure under environmental pulsation collected by the acceleration sensors as structural inspection data. The random subspace method is used to perform time-domain and frequency-domain analysis on the acceleration response to obtain the true service dynamic characteristics. The true service dynamic characteristics include modal period and modal damping ratio, which are all numerical features.

[0088] S2, Clustering of Building Clusters and Selection of Indicator Buildings: Based on the building cluster characteristic parameters, cluster analysis is used to classify urban building clusters, dividing buildings with similar design features and dynamic characteristics into similar clusters, and selecting representative indicator buildings in each cluster.

[0089] S201, Cluster Analysis.

[0090] In this step, the clustering analysis methods that can be used include, but are not limited to, K-means clustering, K-prototypes clustering, or feature-based dimensionality reduction clustering analysis methods.

[0091] The following is the specific implementation process:

[0092] First, statistical analysis was conducted on the sample characteristics of the buildings: descriptive statistics were used for numerical characteristics to provide the mean, standard deviation and interval distribution; frequency statistics were performed on textual characteristics to identify the dominant type.

[0093] Next, the features are screened: for feature combinations that are conceptually and theoretically strongly correlated, Pearson correlation analysis is used to remove redundant features; for multicollinear features (such as height and number of layers), those with stronger physical meaning are retained; categorical variables that occur too infrequently (such as sample size <5%) are merged or removed, and only the regions where the feature distribution is more important are considered.

[0094] In this embodiment, the K-prototypes clustering method is used, which comprehensively considers textual and numerical features, and the elbow method is used to determine the optimal number of clusters.

[0095] Before performing feature clustering, the features need to be standardized to eliminate differences in the units of different features and to make numerical and textual features comparable in the same metric space.

[0096] For numerical features, standardization is performed based on their mean and standard deviation; for textual features, different processing methods can be adopted depending on the clustering method, and no processing is required when using the K-prototypes method.

[0097] The specific methods of clustering are common practices for those skilled in the art, and this embodiment will not elaborate on them in detail.

[0098] S202, select representative landmark buildings.

[0099] (1) Based on the numerical and textual features in the building cluster feature parameters of the current cluster, a unified feature space is constructed. The textual features are encoded into numerical form using integer mapping. The Euclidean geometric distance of each building relative to the cluster center is calculated to quantify the similarity of buildings in the dimensions of structural attributes and dynamic characteristics. The clustering distance similarity score is calculated:

[0100] ,

[0101] In the formula, Indicates the first Cluster distance similarity score between each building and the cluster center; Indicates the cluster center is at the th Cluster centroids on each feature dimension; Indicates the first in the cluster The building in The values ​​taken on each feature dimension The sum of the feature dimensions of textual and numerical features. , For numerical features, the feature dimension is... The feature dimension for text-type features.

[0102] (2) Statistical distribution analysis of numerical features such as height, number of floors, and construction year in the current cluster of building features, and calculation of numerical feature scores to ensure the representativeness of candidate buildings in terms of numerical features;

[0103] ,

[0104] In the formula, For the first Numerical feature scores for each building; For the first Numerical characteristics of a building The value; Numerical features within the cluster The mean; Numerical features within the cluster The maximum value.

[0105] (3) Perform frequency statistics on the textual features of each dimension (e.g., structural type, site category, etc.) in the building cluster feature parameters, determine the dominant type under the corresponding dimension, filter buildings belonging to the dominant type, and calculate the textual feature score:

[0106] ,

[0107] ,

[0108] In the formula, For the first Textual feature scores for each building; It is the first The building features textual characteristics The type on; Intra-cluster textual features The dominant type; For the first Textual features of individual buildings The rating.

[0109] For example, regarding structural types, including shear wall structures, frame structures, and masonry structures, if the dominant type is determined to be a shear wall structure, then for the first... If a building's structural type is a shear wall structure, then The value is 0, which results in lower scores for textual features corresponding to buildings that are closer to the dominant type.

[0110] (4) Calculate spatial location score based on the distance between the building and the nearest seismic station :

[0111] ,

[0112] In the formula, For the first A score for the spatial location of each building. For the first The distance between a building and its nearest corresponding seismic station; This is the average distance between all buildings within the cluster and their nearest corresponding seismic station; This represents the maximum distance between all buildings within the cluster and their corresponding nearest seismic station.

[0113] (5) The cluster distance similarity score, numerical feature score, textual feature score and spatial location score are weighted and fused to obtain the comprehensive representativeness score of buildings within the cluster. :

[0114] ,

[0115] in, These are the weighting coefficients for the four categories of ratings, and their sum is 1; This is the building function importance coefficient. For ordinary buildings, its value is 1, while for key functional buildings such as hospitals, schools, and government departments, the value is less than 1.

[0116] Based on the above settings, buildings with lower scores are more representative.

[0117] (6) Scoring based on comprehensive representativeness Sort by size from smallest to largest, and take the first few. Each corresponding building is a landmark building. The number of buildings is an indicator.

[0118] In this embodiment, the number of index buildings is determined based on the number of buildings within the cluster, the dispersion of numerical features within the cluster, and the diversity of textual features, wherein:

[0119] ,

[0120] In the formula, Based on quantity, , The basic proportionality coefficient can be taken as 0.1. For the first The total number of buildings within the cluster; It is a numerical feature dispersion index within a cluster. , The dispersion influence coefficient can be set to 0.5. If it is necessary to cover as many numerical features as possible for the index buildings, this value can be appropriately increased, but the maximum value should not exceed 1. Numerical features within the cluster standard deviation Numerical features within the cluster The mean; As an index of textual feature diversity within a cluster, , This is the text feature coverage coefficient, which can be set to 0.5. If it is necessary to cover as many text features as possible for the indicative buildings, this value can be appropriately increased, but the maximum value should not exceed 1. Text-type features Number of types, Text-type features The first in The frequency of occurrence of each type within a cluster. For example, for structural types, including shear wall structures, frame structures, and masonry structures, The cluster contains 10 buildings, with corresponding structural types including 4 shear wall structures, 3 frame structures, and 3 masonry structures. What is the frequency corresponding to the shear wall structure? The frequencies corresponding to frame structures and masonry structures .

[0121] The selection principle for indicative buildings in this embodiment not only comprehensively considers the numerical and textual features of buildings and their distribution characteristics within the cluster, but also introduces an adaptive method for determining the number of indicative buildings based on feature dispersion and textual diversity. Furthermore, it combines physical spatial distribution and functional importance coefficients to construct a comprehensive scoring system, which significantly improves the accuracy of feature coverage and performance level mapping within the cluster.

[0122] S3, Numerical Model Analysis of Building Clusters: An equivalent numerical analysis model is established. Based on the measured ground motion time history, the structural recovery capacity is analyzed using the elastoplastic analysis method. Combined with the characteristic parameters of the building clusters, the performance level of buildings in each cluster of the urban building clusters is determined.

[0123] S301, based on the initial design information of urban building complexes and actual service dynamic characteristics, and with reference to the Hazus technical manual, determines the performance points of the capability curve, including the yield point, the maximum bearing capacity point and the ultimate deformation point;

[0124] S302, Establish an equivalent numerical analysis model, input the measured ground motion time history into the equivalent numerical model, and perform elastoplastic analysis to obtain the structural restoring force skeleton curve;

[0125] Establish the dynamic equations for an equivalent single-degree-of-freedom elastoplastic model of the building structure:

[0126] ,

[0127] In the formula, For quality; It is the viscous damping coefficient; For restorative power; , and These are the displacement, velocity, and acceleration of the structure relative to the ground, respectively. This refers to the absolute acceleration of the ground.

[0128] The aforementioned dynamic equations are mass-normalized to obtain normalized dynamic equations, which serve as the equivalent numerical analysis model, as shown in the following equation:

[0129] ,

[0130] In the formula, The structural angular frequency; The damping ratio; The restoring force after mass normalization; , , Determine using the following formula:

[0131] ,

[0132] In the formula, The measured modal period.

[0133] S303 classifies the performance levels of engineering structures into five levels based on the structural restoring force skeleton curve and the determined capacity curve performance points: normal operation, immediate use, life safety, near collapse, and collapse.

[0134] S4, Establish intra-cluster performance level mapping relationship: Based on the performance level of buildings in each cluster of the urban building complex, establish a mapping relationship between the performance level of the index building and the performance level of other buildings in the cluster.

[0135] S401, Construct the mapping model database;

[0136] The cluster features (including structural, dynamic, and spatial characteristics) and performance levels of indicative buildings are extracted and used as inputs to the performance mapping model. The cluster performance levels are then used as outputs to construct the mapping model database.

[0137] S402, Establish a performance level mapping model;

[0138] A performance level mapping model is constructed and trained using a mapping model database. By learning the relationship between structural features and performance response, the performance level of buildings within a cluster can be predicted quickly.

[0139] The performance level mapping model, based on traditional regression or neural network models, introduces a structural similarity weight kernel function in the output stage to enhance the propagation ability of the indicative building performance response to buildings within the cluster. The structural similarity weight kernel function is defined as follows:

[0140] ,

[0141] In the formula, and The first Feature vectors of individual buildings and landmark buildings and The structure is a latitude and longitude vector. and To control the range of weight decay, This is the structural similarity weight kernel function.

[0142] Based on the structural similarity weight kernel function, the refined analysis results of the indicative building are generalized to other buildings within the cluster to obtain the predicted performance values ​​of the buildings:

[0143] ,

[0144] In the formula, As a benchmark for the performance level of buildings, The first prediction of traditional regression models or neural network models The performance level of each building This represents the performance level output by the performance level mapping model.

[0145] The performance level mapping model can quantify the structural and spatial differences between buildings into similarity weights, enabling the refined response of the benchmark building to propagate to buildings within the cluster according to the degree of similarity, thus achieving controllable diffusion and rapid inference of performance levels.

[0146] S5, Seismic Safety Monitoring and Analysis of Urban Building Clusters: Extract the latest seismic motion time history recorded by local stations, calculate the performance level of index buildings, and obtain the performance level of all buildings within the cluster based on the aforementioned mapping relationship, thereby completing the seismic safety monitoring and analysis of urban building clusters.

[0147] The above is an introduction to the method embodiments. The following system embodiments will further illustrate the solution of the present invention.

[0148] An earthquake safety monitoring and analysis system for urban building complexes, used to implement the method, the system comprising:

[0149] Building cluster feature acquisition module: Based on structural design drawings and structural inspection data, acquire building cluster feature parameters, which include the initial design information and actual service dynamic characteristics of the urban building cluster;

[0150] Building Cluster Feature Clustering and Indicator Building Selection Module: Based on the building cluster feature parameters, cluster analysis is used to classify urban building clusters, dividing buildings with similar design features and dynamic characteristics into similar clusters, and selecting representative indicator buildings in each cluster.

[0151] Building Cluster Numerical Model Analysis Module: Establish an equivalent numerical analysis model, analyze the structural resilience based on the measured ground motion time history using elastoplastic analysis, and determine the performance level of buildings within each cluster of the urban building cluster by combining the characteristic parameters of the building cluster.

[0152] Module for establishing intra-cluster performance level mapping relationship: Based on the performance level of buildings in each cluster of the urban building complex, establish a mapping relationship between the performance level of the index building and the performance level of other buildings in the cluster;

[0153] The urban building cluster seismic safety monitoring and analysis module extracts the latest seismic motion time history recorded by local stations, calculates the performance level of index buildings, and obtains the performance level of all buildings within the cluster based on the aforementioned mapping relationship, thereby completing the urban building cluster seismic safety monitoring and analysis.

[0154] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the described module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0155] The preferred embodiments of the present invention have been described in detail above. It should be understood that those skilled in the art can make numerous modifications and variations based on the concept of the present invention without creative effort. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the present invention through logical analysis, reasoning, or limited experimentation on the basis of existing technology should be within the scope of protection defined by the claims.

Claims

1. A method for monitoring and analyzing the seismic safety of urban building complexes, characterized in that, Includes the following steps: S1, Building Cluster Feature Acquisition: Based on structural design drawings and structural inspection data, obtain building cluster feature parameters, which include the initial design information and actual service dynamic characteristics of the urban building cluster. S2, Clustering of Building Clusters and Selection of Indicator Buildings: Based on the building cluster characteristic parameters, cluster analysis is used to classify urban building clusters, dividing buildings with similar design features and dynamic characteristics into similar clusters, and selecting representative indicator buildings in each cluster. S3, Numerical Model Analysis of Building Clusters: Establish an equivalent numerical analysis model, analyze the structural recovery capacity based on the measured ground motion time history using the elastoplastic analysis method, and determine the performance level of buildings in each cluster of urban building clusters by combining the characteristic parameters of the building clusters. S4, Establish intra-cluster performance level mapping relationship: Based on the performance level of buildings in each cluster of the urban building complex, establish a mapping relationship between the performance level of the index building and the performance level of other buildings in the cluster; S5, Seismic safety monitoring and analysis of urban building clusters: Extract the latest seismic motion time history recorded by local stations, calculate the performance level of index buildings, and obtain the performance level of all buildings in the cluster according to the mapping relationship, thereby completing the seismic safety monitoring and analysis of urban building clusters. The method for selecting representative landmark buildings is as follows: A unified feature space is constructed based on the numerical and textual features of the building cluster's feature parameters. The textual features are encoded into numerical form using integer mapping. The Euclidean geometric distance of each building relative to the cluster center is calculated to quantify the similarity of buildings in terms of structural attributes and dynamic characteristics. The clustering distance similarity score is then calculated. ， In the formula, Indicates the first Cluster distance similarity score between each building and the cluster center; Indicates the cluster center at the th Cluster centroids on each feature dimension; Indicates the first in the cluster The building in The values ​​taken on each feature dimension The sum of the feature dimensions for textual and numerical features; Perform statistical distribution analysis on the numerical features in the building cluster feature parameters of the current cluster, and calculate the numerical feature scores; ， In the formula, For the first Numerical feature scores for each building; For the first Numerical characteristics of a building The value; Numerical features within the cluster The mean; Numerical features within the cluster The maximum value; Feature dimensions for numerical features; Frequency statistics are performed on the textual features of each dimension in the building complex feature parameters to determine the dominant type under the corresponding dimension, and buildings belonging to the dominant type are selected. The textual feature score is then calculated. ， ， In the formula, For the first Textual feature scores for each building; It is the first The building features textual characteristics The type on; Intra-cluster textual features The dominant type; For text-type features, the feature dimensions are... For the first Textual features of individual buildings The rating; Spatial location score is calculated based on the distance between the building and the nearest seismic station. : ， In the formula, For the first A score for the spatial location of each building. For the first The distance between a building and its nearest corresponding seismic station; This is the average distance between all buildings within the cluster and their nearest corresponding seismic station; This represents the maximum distance between all buildings within the cluster and their corresponding nearest seismic station. The cluster distance similarity score, numerical feature score, textual feature score, and spatial location score are weighted and fused to obtain a comprehensive representativeness score for buildings within the cluster. : ， in, These are the weighting coefficients for the four categories of ratings, and their sum is 1; This is the building function importance coefficient. For ordinary buildings, its value is 1, and for key function buildings, the value is less than 1. Based on comprehensive representativeness score Sort by size from smallest to largest, and take the first few. Each corresponding building is a landmark building. As a benchmark number of buildings; S4 includes the following steps: S401, Build the mapping model database; Structural characteristic parameters, dynamic characteristic parameters, spatial location characteristic parameters, and performance levels of indicative buildings within a cluster are extracted and used as inputs to a performance mapping model. The performance levels within the cluster are used as outputs to construct a mapping model database. S402, Establish a performance level mapping model; A performance level mapping model is constructed and trained using a mapping model database. By learning the relationship between structural features and performance response, the performance level of buildings within a cluster can be quickly predicted. The performance level mapping model, based on traditional regression or neural network models, introduces a structural similarity weight kernel function in the output stage to enhance the propagation ability of the indicative building performance response to buildings within the cluster. The structural similarity weight kernel function is defined as follows: ， In the formula, and The first Feature vectors of individual buildings and landmark buildings and The structure is a latitude and longitude vector. and To control the range of weight decay, This is the structural similarity weight kernel function; Based on the structural similarity weight kernel function, the refined analysis results of the indicative building are generalized to other buildings within the cluster to obtain the predicted performance values ​​of the buildings: ， In the formula, As a benchmark for the performance level of buildings, The first prediction of traditional regression models or neural network models The performance level of each building This represents the performance level output by the performance level mapping model.

2. The method for monitoring and analyzing the seismic safety of urban building complexes according to claim 1, characterized in that, The initial design information for the urban building complex includes: Textual features: intended use, structural type, plan shape, site category, foundation type, design code adopted, and seismic fortification intensity; Numerical characteristics: building height, building length, building width, number of floors, and year of construction; Spatial location characteristics: the latitude and longitude of the building's location and the distance of the building from the nearest seismic station.

3. The method for monitoring and analyzing the seismic safety of urban building complexes according to claim 1, characterized in that, The actual service dynamic characteristics are obtained by acquiring the acceleration response of the building structure under environmental pulsation collected by acceleration sensors deployed on the target building structure as structural inspection data, and performing time-domain and frequency-domain analysis on the acceleration response to obtain the actual service dynamic characteristics, which include modal period and modal damping ratio, both of which are numerical features.

4. The method for monitoring and analyzing the seismic safety of urban building complexes according to claim 1, characterized in that, The clustering analysis methods include K-means clustering, K-prototypes clustering, or feature-based dimensionality reduction clustering analysis methods.

5. The method for monitoring and analyzing the seismic safety of urban building complexes according to claim 1, characterized in that, The number of index buildings is determined based on the number of buildings within the cluster, the dispersion of numerical features within the cluster, and the diversity of textual features, wherein: ， In the formula, Based on quantity, , The basic proportional coefficient, For the first The total number of buildings within the cluster; It is a numerical feature dispersion index within a cluster. , This is the dispersion influence coefficient. Numerical features within the cluster standard deviation Numerical features within the cluster The mean; As an index of textual feature diversity within a cluster, , This represents the text feature coverage coefficient. Text-type features Number of types, Text-type features The first in The frequency of each type within a cluster.

6. The method for monitoring and analyzing the seismic safety of urban building complexes according to claim 1, characterized in that, S3 includes the following steps: S301, based on the initial design information of urban building complexes and actual service dynamic characteristics, and with reference to the technical manual, determines the performance points of the capability curve, including the yield point, the maximum bearing capacity point and the ultimate deformation point; S302, Establish an equivalent numerical analysis model, input the measured ground motion time history into the equivalent numerical model, and perform elastoplastic analysis to obtain the structural restoring force skeleton curve; S303 classifies the performance levels of engineering structures into five levels based on the structural restoring force skeleton curve and the determined capacity curve performance points: normal operation, immediate use, life safety, near collapse, and collapse.

7. The method for monitoring and analyzing the seismic safety of urban building complexes according to claim 1, characterized in that, The establishment of the equivalent numerical analysis model specifically involves: Establish the dynamic equations for an equivalent single-degree-of-freedom elastoplastic model of the building structure: ， In the formula, For quality; It is the viscous damping coefficient; For restorative power; , and These are the displacement, velocity, and acceleration of the structure relative to the ground, respectively. This refers to the absolute acceleration of the ground. The aforementioned dynamic equations are mass-normalized to obtain normalized dynamic equations, which serve as the equivalent numerical analysis model, as shown in the following equation: ， In the formula, The structural angular frequency; The damping ratio; The restoring force after mass normalization; , , Determine by the following formula: ， In the formula, The measured modal period.

8. A seismic safety monitoring and analysis system for urban building complexes, characterized in that, For implementing the method as described in any one of claims 1-7, the system comprises: Building cluster feature acquisition module: Based on structural design drawings and structural inspection data, acquire building cluster feature parameters, which include the initial design information and actual service dynamic characteristics of the urban building cluster; Building Cluster Feature Clustering and Indicator Building Selection Module: Based on the building cluster feature parameters, cluster analysis is used to classify urban building clusters, dividing buildings with similar design features and dynamic characteristics into similar clusters, and selecting representative indicator buildings in each cluster. Building Cluster Numerical Model Analysis Module: Establish an equivalent numerical analysis model, analyze the structural resilience based on the measured ground motion time history using elastoplastic analysis, and determine the performance level of buildings within each cluster of the urban building cluster by combining the characteristic parameters of the building cluster. Module for establishing intra-cluster performance level mapping relationship: Based on the performance level of buildings in each cluster of the urban building complex, establish a mapping relationship between the performance level of the index building and the performance level of other buildings in the cluster; The urban building cluster seismic safety monitoring and analysis module extracts the latest seismic motion time history recorded by local stations, calculates the performance level of index buildings, and obtains the performance level of all buildings within the cluster based on the aforementioned mapping relationship, thereby completing the urban building cluster seismic safety monitoring and analysis.

Images