Biaxial coupling anti-seismic property evaluation method and system for flanged shear wall

Abstract

The invention relates to the technical field of shear wall performance evaluation, and discloses a biaxial coupling anti-seismic performance evaluation method and system for a flanged shear wall, and the system comprises an acquisition and preprocessing module which is used for acquiring geometric, material and biaxial coupling load data of the flanged shear wall in real time; after processing, a parameter-performance association database is established in combination with historical data; the model building module performs modeling based on the database and introduces a pre-diagnosis algorithm to identify potential damage risk points; the analysis module inputs real-time data to calculate an anti-seismic performance index, judges a health level, analyzes a degradation rate, predicts a fault probability and generates a pre-diagnosis report; and the management and evaluation module generates a management strategy, outputs an evaluation report, feeds back a data iterative optimization model, and realizes dynamic evaluation of the anti-seismic performance of the shear wall. Scientific basis is provided for safety management of the shear wall, and the anti-seismic safety and operation and maintenance reasonability of the structure are effectively guaranteed.

Description

Technical Field

[0001] This invention relates to the field of shear wall performance evaluation technology, and more specifically, to a biaxially coupled seismic performance evaluation method and system for flanged shear walls. Background Technology

[0002] As a core lateral force resisting component in high-rise buildings and large-span structures, the seismic performance of flanged shear walls directly determines the safety and stability of the overall structure. Performance evaluation under biaxial coupled loads is crucial for engineering design and operation and maintenance.

[0003] Traditional methods for assessing the seismic performance of flanged shear walls are mostly based on uniaxial load modeling, neglecting the biaxial load coupling effect, which leads to a large deviation between the assessment results and the actual stress state. Damage identification relies on manual inspection or simple finite element simulation, making it difficult to accurately locate potential risks in key areas such as flange-web connection points and around openings. The lack of real-time data acquisition and dynamic feedback mechanisms makes it impossible to adjust the assessment model in accordance with the performance degradation patterns during structural service. The timeliness and accuracy of health level determination and fault prediction are insufficient, making it difficult to meet the safety management needs of the entire structural life cycle.

[0004] Therefore, it is necessary to design a biaxial coupling seismic performance evaluation method and system for flanged shear walls to address the problems of insufficient consideration of biaxial coupling effects, low accuracy of damage identification, lack of real-time feedback and dynamic optimization, and lack of systematic performance prediction and health management in traditional evaluation methods. Summary of the Invention

[0005] In view of this, the present invention proposes a biaxial coupling seismic performance evaluation method and system for flanged shear walls, aiming to solve the problems of insufficient consideration of biaxial coupling effect, low accuracy of damage identification, lack of real-time feedback and dynamic optimization, and lack of systematic performance prediction and health management in traditional evaluation methods.

[0006] In one aspect, the present invention proposes a biaxially coupled seismic performance evaluation system for flanged shear walls, comprising: The acquisition and preprocessing module is configured to acquire geometric data, material data and biaxial coupled load data of shear walls with flanges in real time, perform outlier removal and standardization processing on the acquired data, and obtain historical seismic test data, damage records and maintenance files of shear walls to establish a parameter-performance correlation database. The model building module is configured to construct a digital finite element foundation model considering the biaxial load coupling effect based on a parameter-performance correlation database; at the same time, a pre-diagnosis algorithm is introduced to form a biaxial coupling seismic model; the stress-strain distribution, component displacement and corresponding load of the shear wall under biaxial load are simulated through the biaxial coupling seismic model, and potential damage risk points at the connection between the wall flange and web, around the opening and at the bottom of the wall are identified; The analysis module is configured to input preprocessed real-time data into the dual-axis coupled seismic model, calculate seismic performance indicators, preset health status threshold ranges, compare real-time performance indicators with thresholds to determine the current health level, and combine historical performance data to analyze the performance degradation rate using a trend prediction algorithm, predict the probability of potential failures under different load cycles, and generate a pre-diagnosis report. The management and evaluation module is configured to generate health management strategies based on the pre-diagnosis report and performance degradation trend; to conduct a comprehensive evaluation of the biaxial coupling seismic performance, output an evaluation report, and feed the evaluation data back to the parameter-performance correlation database for iterative optimization of the model and pre-diagnosis algorithm, which is used for subsequent dynamic evaluation of the seismic performance of shear walls.

[0007] Furthermore, the acquisition and preprocessing module includes: The geometric parameter acquisition unit is configured to acquire the flange width, flange thickness, web height, web thickness, opening size and reinforcement arrangement parameters of the flanged shear wall through a three-dimensional laser scanning device and a BIM model extraction tool, and generate a structured geometric parameter dataset. The material parameter acquisition unit is configured to acquire the compressive strength, axial tensile strength, yield strength, and elastic modulus of concrete cubes through material mechanics testing equipment, and simultaneously input the material production batch and service life information to form a material performance parameter library. The biaxial coupling load acquisition unit is configured to acquire reciprocating biaxial load data in real time through load sensors and data acquisition instruments, including horizontal X-axis load amplitude, horizontal Y-axis load amplitude, vertical load amplitude and biaxial loading path history, and record the load application rate and holding time. The data preprocessing unit is configured to perform noise reduction on the acquired data using a wavelet filtering algorithm, remove outliers in geometric and material parameters using an isolated forest algorithm, and map load and performance data to the [0,1] interval using a min-max normalization method to generate a standardized dataset and store it in the parameter-performance correlation database.

[0008] Furthermore, the model building module includes: The finite element basic model building unit is configured to be based on a standardized dataset in the parameter-performance correlation database. Shell elements are used to simulate the concrete and steel reinforcement diffusion layer of the shear wall. The concrete constitutive model and the steel reinforcement constitutive model are defined, and biaxial load coupling boundary conditions are set. The pre-diagnosis algorithm unit is configured as a damage feature extraction algorithm based on a convolutional neural network. The damage feature extraction algorithm is trained by damage images and stress-strain data from historical seismic tests and extracts stress concentration features and displacement mutation features in the model simulation process in real time. The damage risk point identification unit is configured to identify potential damage risk points at the junction of the wall flange and web, the upper and lower edges of the opening, and the bottom of the wall limb based on the feature data output by the finite element model and the pre-diagnosis algorithm, combined with a preset risk judgment threshold, and to mark the risk level, which includes low risk, medium risk, and high risk.

[0009] Furthermore, the analysis module includes: The performance index calculation unit is configured to input preprocessed data into the biaxially coupled seismic model and calculate seismic performance indices, including displacement ductility coefficient, bearing capacity attenuation rate, stiffness degradation coefficient, and damage coefficient. The health level determination unit is configured with a preset health status threshold range. It compares the real-time calculated seismic performance index with the preset threshold range and determines the current health level of the flanged shear wall based on the comparison result. The degradation trend analysis and fault prediction unit is configured to use a long short-term memory network algorithm, combined with historical performance data in the parameter-performance correlation database, to fit a performance degradation curve and analyze the performance degradation rate; based on the degradation curve, it predicts the probability of potential fault occurrence under different load cycles and generates a pre-diagnostic report containing fault type, predicted occurrence time, and risk level.

[0010] Furthermore, when the health level determination unit determines the health level by comparing real-time performance indicators with preset thresholds, it includes: The health level determination unit presets four health status threshold ranges, namely, excellent health threshold range, good health threshold range, medium health threshold range and poor health threshold range; The health level determination unit compares the real-time calculated displacement ductility coefficient, bearing capacity attenuation rate, stiffness degradation coefficient, and damage coefficient with the four-level threshold intervals respectively: When all performance indicators are within the health excellent threshold range, the current health level of the flanged shear wall is determined to be excellent. When at least three performance indicators are within the healthy good threshold range and no indicator is below the healthy medium threshold range, the current health level is determined to be good. When at least two performance indicators are within the healthy intermediate threshold range and no indicator is below the healthy poor threshold range, the current health level is determined to be intermediate. When any performance index is within the poor health threshold range, or when the damage evolution index is greater than the boundary between the intermediate health threshold range and the poor health threshold range, the current health level is determined to be poor. After the determination is completed, the health level determination unit records the determination time, the original values ​​of the performance indicators to be compared and the threshold range, and stores them in the parameter-performance association database.

[0011] Furthermore, when the degradation trend analysis and fault prediction unit generates a pre-diagnostic report, it includes: The degradation trend analysis and fault prediction unit presets three levels of fault occurrence probability intervals, namely low-risk probability interval, medium-risk probability interval, and high-risk probability interval. The unit calculates the probability of failure under different load cycles based on the fitted performance degradation curve: When the probability of a fault occurrence falls within the low-risk probability range, the fault risk level is marked as low risk, and the pre-diagnosis report recommends maintaining the regular monitoring frequency. When the probability of failure is in the medium risk range, the failure risk level is marked as medium risk. The pre-diagnosis report recommends increasing the monitoring frequency to twice the original frequency and adding detection of the flange-web connection area. When the probability of a failure is in the high-risk probability range, the failure risk level is marked as high-risk, and the pre-diagnosis report recommends immediate shutdown for inspection and assessment of whether reinforcement is needed. The pre-diagnosis report also includes shear wall foundation information, real-time performance index curves, and degradation trend fitting plots, which are output in the form of text and visual charts and uploaded to the remote monitoring platform simultaneously.

[0012] Furthermore, when generating a health management strategy, the management and evaluation module includes: Pre-set basic management strategies corresponding to four health levels: excellent management strategy, good management strategy, medium management strategy, and poor management strategy. Among them, the excellent management strategy is to maintain routine monthly monitoring, the good management strategy is to add bi-monthly special monitoring of the flange-web connection, the intermediate management strategy is to conduct weekly full-dimensional monitoring plus crack width detection, and the poor management strategy is to conduct daily monitoring plus structural reinforcement assessment preparation. The management and evaluation module extracts the current health level and performance degradation rate from the pre-diagnosis report, and presets a degradation rate threshold. Compare the performance degradation rate with the degradation rate threshold, and adjust the basic management strategy based on the comparison results: When the performance degradation rate is less than or equal to the minimum value of the degradation rate threshold, the basic management strategy for maintaining the corresponding health level is implemented. When the performance degradation rate is greater than the minimum value of the degradation rate threshold but less than or equal to the maximum value of the degradation rate threshold, the monitoring frequency of the basic management strategy will be increased to 1.5 times the original frequency. When the performance degradation rate exceeds the maximum value of the degradation rate threshold, the basic management policy will be upgraded to a higher-level management policy. After the strategy is adjusted, the basis for the adjustment is recorded, a health management plan including monitoring items, monitoring frequency, and responsible persons is generated, and the plan is synchronously stored in the parameter-performance association database.

[0013] Furthermore, when the management and evaluation module feeds the evaluation data back to the parameter-performance correlation database to iteratively optimize the model, it includes: A model optimization trigger threshold is preset, which includes a prediction bias threshold and a data accumulation threshold. The management and evaluation module periodically extracts evaluation data for a preset period from the parameter-performance correlation database and calculates the prediction deviation for each data point. The prediction deviation calculation formula is: prediction deviation equals the absolute value of the model prediction value minus the measured value divided by the measured value multiplied by 100%. The percentage of data with a prediction deviation greater than the prediction deviation threshold, and the total accumulated statistical data: When the data percentage is less than or equal to 10% and the total accumulated data is greater than or equal to the data accumulation threshold, the model is deemed to have met the current accuracy standard and no optimization is required. When the data percentage is greater than 10% or the total accumulated data is less than the data accumulation threshold, model iterative optimization is initiated: The newly accumulated assessment data were used as training samples to retrain the pre-diagnosis algorithm and trend prediction algorithm in the biaxially coupled seismic model, and the constitutive parameters and boundary condition weights of the model were adjusted. After training is complete, the prediction bias of the new model is calculated. When the new prediction bias is less than or equal to the prediction bias threshold, the parameters of the new model are saved and the old model is replaced. If the new prediction deviation is greater than the prediction deviation threshold, retraining is performed. If the new prediction deviation is still greater than the prediction deviation threshold after three consecutive training sessions, the alarm module is triggered to issue a warning and record abnormal information about model optimization.

[0014] Furthermore, when the damage risk point identification unit dynamically reviews potential damage risk points, it includes: The preset interval for reviewing damage risk points is used. The review indicators include the stress change rate and displacement increment at the risk points. The preset thresholds for reviewing the stress change rate and displacement increment are also included. Based on the damage risk point review interval, real-time stress data and displacement data of the risk points are extracted through the biaxially coupled seismic model, and the stress change rate and displacement increment are calculated. The stress change rate is compared with the verification stress change rate threshold, and the displacement increment is compared with the verification displacement increment threshold. The risk level is adjusted based on the comparison results. When the rate of change of stress is less than the minimum value of the threshold for the rate of change of stress and the displacement increment is less than the minimum value of the threshold for the displacement increment, the original risk level is maintained. When the rate of change of stress falls within the threshold range of the rate of change of stress or the displacement increment falls within the threshold range of the displacement increment, the original risk level will be upgraded by one level. When the rate of change of stress is greater than the maximum value of the threshold for the rate of change of stress and the displacement increment is greater than the maximum value of the threshold for the displacement increment, the original risk level is upgraded to the highest risk level, which is an upgraded state of high risk. If the risk level is determined to be extremely high after review, the alarm module will immediately issue an emergency alarm and push the risk point location and review data to the structural safety management platform. All review results are stored in the parameter-performance correlation database for subsequent model optimization.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. The system uses a model building module to construct a digital finite element foundation model that considers the biaxial load coupling effect based on a parameter-performance correlation database, accurately simulating the stress-strain distribution and internal force transmission path under biaxial loads. This design directly addresses the core problem of traditional assessments neglecting the biaxial coupling effect and resulting in large deviations between assessment results and actual forces. By accurately quantifying the coupling effect, it significantly improves the fit and accuracy of seismic performance assessments.

[0016] 2. Relying on pre-diagnostic algorithms and damage risk point identification mechanisms, the system can extract features such as stress concentration and sudden displacement in real time, accurately locating potential damage in key areas such as the flange-web connection, the perimeter of openings, and the bottom of walls. Compared to the crude identification methods of traditional manual inspection or simple simulation, this technology achieves precise locking of damage risks, completely solving the shortcomings of traditional methods such as low accuracy in damage identification and easy omission of key risk points.

[0017] 3. The real-time data acquisition and multi-dimensional data processing capabilities of the acquisition and preprocessing module, combined with the data feedback iteration mechanism of the management and evaluation module, form a closed loop of "acquisition-analysis-feedback-optimization". This design effectively solves the problems of traditional evaluation lacking real-time feedback and being unable to adapt to performance changes during structural service. Through dynamic optimization of models and algorithms, it ensures that the evaluation results always closely match the actual service status of the structure.

[0018] 4. The system integrates multi-dimensional seismic performance index calculation, health level determination, performance degradation rate analysis, and targeted health management strategy generation, constructing a comprehensive and systematic assessment and management system. Addressing the shortcomings of traditional methods, such as insufficient timeliness in performance prediction and lack of systematic health management, this system achieves full-chain coverage from performance assessment to operation and maintenance guidance, providing scientific and continuous support for the safety management of structures throughout their entire lifecycle.

[0019] On the other hand, this application also provides a method for evaluating the seismic performance of biaxially coupled shear walls with flanges, including the following steps: Real-time acquisition of geometric data, material data, and biaxial coupled load data of flanged shear walls; outlier removal and standardization processing of the acquired data; acquisition of historical seismic test data, damage records, and maintenance archives of shear walls; and establishment of a parameter-performance correlation database. Based on a parameter-performance correlation database, a digital finite element foundation model considering biaxial load coupling effect is constructed; at the same time, a pre-diagnosis algorithm is introduced to form a biaxial coupling seismic model; the stress-strain distribution, component displacement and corresponding load of shear wall under biaxial load are simulated through the biaxial coupling seismic model to identify potential damage risk points at the connection between the wall flange and web, around the opening and at the bottom of the wall. The preprocessed real-time data is input into the dual-axis coupled seismic model to calculate the seismic performance index. A health status threshold range is preset, and the real-time performance index is compared with the threshold to determine the current health level. At the same time, combined with historical performance data, a trend prediction algorithm is used to analyze the performance degradation rate, predict the probability of potential failures under different load cycles, and generate a pre-diagnosis report. Based on the pre-diagnosis report and performance degradation trend, a health management strategy is generated; a comprehensive assessment of the biaxial coupling seismic performance is conducted, an assessment report is output, and the assessment data is fed back to the parameter-performance correlation database for iterative optimization of the model and pre-diagnosis algorithm, which is used for subsequent dynamic assessment of the seismic performance of shear walls.

[0020] It is understandable that the above-mentioned biaxially coupled seismic performance evaluation method and system for flanged shear walls have the same beneficial effects, and will not be elaborated further here. Attached Figure Description

[0021] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This is a functional block diagram of a biaxially coupled seismic performance evaluation system for flanged shear walls provided in an embodiment of the present invention. Figure 2 This is a flowchart illustrating the workflow of a biaxially coupled seismic performance evaluation system for flanged shear walls provided in an embodiment of the present invention. Figure 3 This is a flowchart of a biaxially coupled seismic performance evaluation method for flanged shear walls provided in an embodiment of the present invention. Detailed Implementation

[0022] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0023] Reference Figure 1 and Figure 2 In some embodiments of this application, a biaxially coupled seismic performance evaluation system for flanged shear walls includes: an acquisition and preprocessing module, a model building module, an analysis module, and a management and evaluation module.

[0024] Specifically, the acquisition and preprocessing module is configured to acquire geometric data, material data and biaxial coupled load data of the flanged shear wall in real time, perform outlier removal and standardization processing on the acquired data, and obtain historical seismic test data, damage records and maintenance files of the shear wall to establish a parameter-performance correlation database. The model building module is configured to construct a digital finite element foundation model that considers the biaxial load coupling effect based on a parameter-performance correlation database; at the same time, a pre-diagnosis algorithm is introduced to form a biaxial coupling seismic model; the biaxial coupling seismic model is used to simulate the stress-strain distribution, component displacement and corresponding load of shear walls under biaxial loads, and to identify potential damage risk points at the connection between the wall flange and web, around openings and at the bottom of the wall. The analysis module is configured to input preprocessed real-time data into a dual-axis coupled seismic model, calculate seismic performance indicators, preset health status threshold ranges, compare real-time performance indicators with thresholds to determine the current health level, and combine historical performance data to use trend prediction algorithms to analyze the performance degradation rate, predict the probability of potential failures under different load cycles, and generate a pre-diagnosis report. The management and evaluation module is configured to generate health management strategies based on the pre-diagnosis report and performance degradation trend; to conduct a comprehensive evaluation of the biaxial coupling seismic performance, output an evaluation report, and feed the evaluation data back to the parameter-performance correlation database for iterative optimization of the model and pre-diagnosis algorithm, which is used for subsequent dynamic evaluation of the seismic performance of shear walls.

[0025] Specifically, the acquisition and preprocessing module includes a geometric parameter acquisition unit, which is configured to acquire the flange width, flange thickness, web height, web thickness, opening size and reinforcement arrangement parameters of the flanged shear wall through a three-dimensional laser scanning device and a BIM model extraction tool, and generate a structured geometric parameter dataset. The material parameter acquisition unit is configured to acquire the compressive strength, axial tensile strength, yield strength, and elastic modulus of concrete cubes through material mechanics testing equipment, and simultaneously input the material production batch and service life information to form a material performance parameter library. The biaxial coupling load acquisition unit is configured to acquire reciprocating biaxial load data in real time through load sensors and data acquisition instruments, including horizontal X-axis load amplitude, horizontal Y-axis load amplitude, vertical load amplitude and biaxial loading path history, and record the load application rate and holding time. The data preprocessing unit is configured to perform noise reduction on the acquired data using a wavelet filtering algorithm, remove outliers in geometric and material parameters using an isolated forest algorithm, and map load and performance data to the [0,1] interval using a min-max normalization method to generate a standardized dataset and store it in the parameter-performance correlation database.

[0026] Specifically, the model building module includes: finite element basic model building unit, which is configured to be based on a standardized dataset in the parameter-performance correlation database, using shell elements to simulate the concrete and steel reinforcement diffuse layer of the shear wall, defining the concrete constitutive model and the steel reinforcement constitutive model, and setting biaxial load coupling boundary conditions; The pre-diagnosis algorithm unit is configured as a damage feature extraction algorithm based on a convolutional neural network. The damage feature extraction algorithm is trained by damage images and stress-strain data from historical seismic tests and extracts stress concentration features and displacement change features in real time during the model simulation process. The damage risk point identification unit is configured to identify potential damage risk points at the junction of the wall flange and web, the upper and lower edges of openings, and the bottom of the wall limbs based on the feature data output by the finite element model and the pre-diagnosis algorithm, combined with a preset risk judgment threshold, and to mark the risk level, which includes low risk, medium risk, and high risk.

[0027] Specifically, the analysis module includes a performance index calculation unit, which is configured to input preprocessed data into a biaxially coupled seismic model to calculate seismic performance indices, including displacement ductility coefficient, bearing capacity attenuation rate, stiffness degradation coefficient, and damage coefficient. The health level determination unit is configured with a preset health status threshold range. It compares the real-time calculated seismic performance index with the preset threshold range and determines the current health level of the flanged shear wall based on the comparison result. The degradation trend analysis and fault prediction unit is configured to use a long short-term memory network algorithm, combined with historical performance data in the parameter-performance correlation database, to fit a performance degradation curve and analyze the performance degradation rate; based on the degradation curve, it predicts the probability of potential fault occurrence under different load cycles and generates a pre-diagnostic report containing fault type, predicted occurrence time, and risk level.

[0028] In a specific embodiment of this application, the above steps are implemented as follows: In practical applications, the biaxially coupled seismic performance evaluation system for flanged shear walls completes the overall evaluation process through multi-device collaborative acquisition and multi-module linkage. The geometric parameter acquisition unit uses laser scanning equipment to accurately scan and capture the physical dimensions of the flanged shear wall, such as flange width, flange thickness, web height, and web thickness. At the same time, it uses BIM model extraction tools to export structural parameters such as opening dimensions and reinforcement layout from the constructed building information model, jointly generating a standardized geometric parameter dataset. The material parameter acquisition unit utilizes professional materials mechanics testing equipment to test the cubic compressive strength and axial tensile strength of the concrete used in the shear wall, and to detect the yield strength and elastic modulus of the reinforcing steel. Simultaneously, it inputs relevant information such as material production batches and service years into the system, forming a complete material performance parameter database. The biaxial coupling load acquisition unit uses load sensors and data acquisition instruments deployed in the shear wall test area or actual service scenario to capture the load amplitude in the horizontal X-axis, horizontal Y-axis, and vertical directions in real time, recording key load information such as the number of loading cycles, load application rate, and holding time. The data preprocessing unit performs unified processing on the various types of raw data collected above. It uses wavelet filtering algorithms to filter out environmental interference noise in the data, identifies and removes abnormal data points in geometric and material parameters using the isolated forest algorithm, and then uses the min-max standardization method to uniformly map the load data and subsequently generated performance data to the [0,1] interval, generating a standardized dataset. Simultaneously, the system retrieves past seismic test data, historical damage records, and maintenance archives of the shear wall, integrating them to establish a comprehensive parameter-performance correlation database. The model building module is based on the standardized dataset in the parameter-performance correlation database. It uses shell elements to numerically simulate the concrete and steel reinforcement materials of the shear wall, defines a concrete constitutive model and a steel reinforcement constitutive model that conform to the actual material properties, and reasonably sets the boundary conditions under biaxial load coupling to construct a digital finite element foundation model. The pre-diagnosis algorithm unit introduces a mature damage feature extraction algorithm based on convolutional neural networks, trained with damage images and stress-strain data from a large number of historical seismic tests, into the finite element foundation model to form a biaxially coupled seismic model with real-time feature extraction capabilities. This model can dynamically simulate the stress-strain distribution, nodal displacement changes, and internal force transmission paths inside the shear wall under biaxial load. The damage risk point identification unit, based on the stress concentration features and displacement mutation features output by the model, combined with the preset risk judgment criteria, accurately locates potential risk points that are prone to damage, such as the junction of the wall flange and web, the upper and lower edges of openings, the bottom of the wall, and densely reinforced areas, and marks each risk point as low-risk, medium-risk, or high-risk.The analysis module receives standardized real-time data output from the data preprocessing unit and inputs it into the biaxially coupled seismic model. The performance index calculation unit automatically calculates core seismic performance indicators such as displacement ductility coefficient, bearing capacity attenuation rate, stiffness degradation coefficient, and damage evolution index. The degradation trend analysis and fault prediction unit calls historical performance data stored in the parameter-performance correlation database and performs in-depth mining on these data through a long short-term memory network algorithm to fit the performance degradation curve of the shear wall, intuitively presenting the performance degradation rate. At the same time, based on this curve, it predicts the probability of potential faults that may occur under different load cycles, and finally generates a pre-diagnostic report containing key information such as fault type, predicted occurrence time, and risk level. After receiving the pre-diagnosis report, the management and evaluation module formulates targeted health management strategies based on the performance degradation trend. Simultaneously, it conducts a comprehensive evaluation of the biaxially coupled seismic performance of the flanged shear wall, generating a detailed evaluation report and outputting it to relevant terminals. The system also fully feeds back all data generated during this evaluation process to the parameter-performance correlation database, providing sufficient data support for parameter optimization of the biaxially coupled seismic model and iterative upgrades of the pre-diagnosis algorithm. This ensures that subsequent dynamic evaluations of the seismic performance of the flanged shear wall and similar structures have higher accuracy and reliability.

[0029] The above scenarios are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0030] In the above embodiments, the system constructs a digital finite element foundation model considering biaxial load coupling effects based on a parameter-performance correlation database through a model building module, accurately simulating the stress-strain distribution and internal force transmission path under biaxial loads. This design directly addresses the core problem of traditional assessments neglecting biaxial coupling effects and resulting in large deviations between assessment results and actual stress. By accurately quantifying the coupling effect, it significantly improves the fit and accuracy of seismic performance assessments. Relying on pre-diagnosis algorithms and damage risk point identification mechanisms, the system can extract features such as stress concentration and displacement abrupt changes in real time, accurately locating potential damage in key areas such as flange-web connections and the perimeter of openings. Compared to the crude identification methods of traditional manual inspection or simple simulation, this technology achieves precise locking of damage risks, completely solving the shortcomings of traditional methods such as low accuracy in damage identification and easy omission of key risk points. The real-time data acquisition and multi-dimensional data processing capabilities of the acquisition and preprocessing module, combined with the data feedback iteration mechanism of the management and assessment module, form a closed loop of "acquisition-analysis-feedback-optimization". This design effectively addresses the shortcomings of traditional assessments, such as lack of real-time feedback and inability to adapt to performance changes during structural service. Through dynamic optimization of models and algorithms, it ensures that assessment results always closely reflect the actual service status of the structure. The system integrates multi-dimensional seismic performance index calculation, health level determination, performance degradation rate analysis, and the generation of targeted health management strategies, constructing a comprehensive and systematic assessment and management framework. Addressing the shortcomings of traditional methods, such as insufficient timeliness in performance prediction and lack of systematic health management, this system achieves full-chain coverage from performance assessment to operation and maintenance guidance, providing scientific and continuous support for the safety management of the entire structural lifecycle.

[0031] Specifically, when the health level determination unit determines the health level by comparing real-time performance indicators with preset thresholds, it includes: the health level determination unit presets four health status threshold ranges, namely, the health excellent threshold range, the health good threshold range, the health intermediate threshold range, and the health poor threshold range. The health level assessment unit compares the real-time calculated displacement ductility coefficient, bearing capacity attenuation rate, stiffness degradation coefficient, and damage coefficient with the four-level threshold intervals: When all performance indicators are within the health excellent threshold range, the current health level of the flanged shear wall is determined to be excellent. When at least three performance indicators are within the healthy good threshold range and no indicator is below the healthy medium threshold range, the current health level is determined to be good. When at least two performance indicators are within the healthy intermediate threshold range and no indicator is below the healthy poor threshold range, the current health level is determined to be intermediate. When any performance indicator is within the poor health threshold range, or when the damage coefficient is greater than the boundary between the intermediate health threshold range and the poor health threshold range, the current health level is determined to be poor. After the determination is completed, the health level determination unit records the determination time, the original values ​​of the performance indicators to be compared and the threshold range, and stores them in the parameter-performance association database.

[0032] Specifically, the health level assessment unit presets four health status threshold ranges. Each range is set based on the seismic design code for flanged shear walls and engineering practice data. Specifically, they are: excellent health threshold range (displacement ductility coefficient μ≥4.5, bearing capacity attenuation rate λ≤5%, stiffness degradation coefficient η≤0.2, damage coefficient D≤0.1), good health threshold range (3.5≤μ<4.5, 5%<λ≤10%, 0.2<η≤0.35, 0.1<D≤0.25), medium health threshold range (2.5≤μ<3.5, 10%<λ≤20%, 0.35<η≤0.5, 0.25<D≤0.4), and poor health threshold range (μ<2.5, λ>20%, η>0.5, D>0.4). The dividing values ​​between the medium and poor health threshold ranges are μ=2.5, λ=20%, η=0.5, and D=0.4, respectively. The health level determination unit first normalizes the real-time calculated displacement ductility coefficient, bearing capacity attenuation rate, stiffness degradation coefficient, and damage evolution index (retaining two decimal places), and then compares them one by one with the four threshold intervals: when all four performance indicators are completely within the excellent health threshold interval, the current health level of the flanged shear wall is determined to be excellent; when at least three performance indicators are within the good health threshold interval, and the remaining indicators are all within the medium health threshold interval (with no indicator below the lower limit of the medium health threshold interval), the current health level is determined to be good; when at least two performance indicators are within the medium health threshold interval, and the remaining indicators are all within the good or medium health threshold interval (with no indicator below the lower limit of the poor health threshold interval), the current health level is determined to be medium; when any performance indicator is within the poor health threshold interval, or the damage evolution index D is greater than the boundary value of 0.4 between medium and poor health, the current health level is determined to be poor. After the judgment is completed, the health level judgment unit automatically records the judgment timestamp accurate to the second, the original measurement values ​​of the four performance indicators, the upper and lower limits of the comparison threshold range for each indicator, the final health level and the core judgment basis (such as "3 indicators are in the good range, 1 indicator is in the medium range, and the judgment is good"), and stores it in the parameter-performance association database in a structured data format for subsequent degradation trend analysis and model optimization.

[0033] The above embodiments, by pre-setting a four-level health status threshold range that conforms to the seismic design code and engineering practice of flanged shear walls, conduct a comprehensive comparison around four core seismic performance indicators: displacement ductility coefficient, bearing capacity attenuation rate, stiffness degradation coefficient, and damage evolution index. With clear quantitative constraints and interval limitation rules (such as all indicators being excellent for a grade of excellent, at least three indicators being good and no indicators below the intermediate level for a grade of good), the health level is accurately defined. This avoids the one-sidedness of single-indicator evaluation and reduces errors caused by human intervention through standardized judgment logic. Furthermore, after judgment, the judgment time, original values ​​of performance indicators, and threshold ranges are fully recorded and stored in the database. This provides traceable basic data support for subsequent performance degradation trend analysis and model iteration optimization, and also provides an objective and reliable basis for the health management and maintenance strategy formulation of flanged shear walls, effectively ensuring the scientific and practical nature of seismic performance assessment.

[0034] Specifically, when the degradation trend analysis and fault prediction unit generates a pre-diagnosis report, it includes: the degradation trend analysis and fault prediction unit presets three levels of fault occurrence probability intervals, namely low-risk probability interval, medium-risk probability interval, and high-risk probability interval. The unit calculates the probability of failure under different load cycles based on the fitted performance degradation curve: When the probability of a fault occurrence falls within the low-risk probability range, the fault risk level is marked as low risk, and the pre-diagnosis report recommends maintaining the regular monitoring frequency. When the probability of failure is in the medium risk range, the failure risk level is marked as medium risk. The pre-diagnosis report recommends increasing the monitoring frequency to twice the original frequency and adding detection of the flange-web connection area. When the probability of a failure is in the high-risk probability range, the failure risk level is marked as high-risk, and the pre-diagnosis report recommends immediate shutdown for inspection and assessment of whether reinforcement is needed. The pre-diagnosis report also includes shear wall foundation information, real-time performance index curves, and degradation trend fitting plots, which are output in the form of text and visual charts and uploaded to the remote monitoring platform simultaneously.

[0035] Specifically, when generating a pre-diagnostic report, the degradation trend analysis and fault prediction unit includes: The unit presets three levels of fault occurrence probability intervals, based on the design service life of the flanged shear wall, material fatigue characteristics, and engineering risk control standards. These are: low-risk probability interval (0% ≤ P < 30%), medium-risk probability interval (30% ≤ P < 70%), and high-risk probability interval (P ≥ 70%), with probability calculations retained to two decimal places. Based on a nonlinear performance degradation curve fitted by a long short-term memory network algorithm, the unit selects the current load cycle number as the starting point and calculates the fault occurrence probability of each node within the range up to the design limit load cycle number (typically set to 2000 cycles) at intervals of 50 cycles / step. A 95% confidence interval is also marked to reflect the prediction reliability. When the probability of failure falls within the low-risk range, the failure risk level is marked as low-risk. The pre-diagnosis report recommends maintaining the monthly routine monitoring frequency, with monitoring items including external crack observation and stress monitoring of key components. When the probability of failure falls within the medium-risk range, the failure risk level is marked as medium-risk. It is recommended to increase the monitoring frequency to twice the original frequency (i.e., once every half month), and add ultrasonic testing of the flange-web connection and strain gauge data acquisition around the opening. When the probability of failure falls within the high-risk range, the failure risk level is marked as high-risk. It is recommended to shut down the machine for inspection within 24 hours, using the rebound method to test the concrete strength and the electromagnetic induction method to test the degree of steel corrosion. Simultaneously, a structural engineer should be organized to assess whether steel plate reinforcement or carbon fiber reinforcement is required. The pre-diagnosis report also includes shear wall foundation information (project name, shear wall number, building and floor, core dimensions of flange and web, concrete strength grade, steel reinforcement type and reinforcement ratio), real-time time-series curves of four performance indicators (horizontal axis is the monitoring timestamp, vertical axis is the indicator value, with reference lines for threshold intervals marked), and degradation trend fitting graph (horizontal axis is the number of load cycles, vertical axis is the performance indicator value, with historical data points and predicted curves superimposed). It is output in PDF format with high-definition PNG visualization charts. The report number is compiled according to the rule of "project number-shear wall number-generation date (YYYYMMDD)-version number" and is uploaded to the "pre-diagnosis report management" module of the remote monitoring platform within 10 seconds after generation. It is associated with the unique identification code of the corresponding shear wall and supports hierarchical viewing (administrators can download and edit, and maintenance personnel can only view). It also records the upload time, reception status and operation log.

[0036] The above embodiments, by pre-setting a three-level failure probability range that fits the actual needs of flanged shear wall projects and risk control requirements, and combining the fitted performance degradation curves, accurately calculate the failure probability under different load cycles. This achieves scientific definition of risk levels and differentiated responses. Targeted recommendations corresponding to low, medium, and high risk levels (maintaining routine monitoring, increasing monitoring frequency and adding special tests, and immediate shutdown for inspection, evaluation, and reinforcement) not only avoid the waste of resources caused by over-maintenance but also effectively prevent the expansion of risks. At the same time, the pre-diagnosis report integrates shear wall foundation information, real-time performance index curves, and degradation trend fitting graphs, outputting them in an intuitive form of text and visual charts and uploading them simultaneously to the remote monitoring platform. This not only provides convenience for relevant personnel to quickly grasp the performance status of the shear wall and trace data but also supports cross-scenario collaborative management and decision-making, helping to predict potential failures in advance and optimize maintenance strategies. It provides reliable technical support for the seismic safety assurance, full life cycle health management, and service life extension of flanged shear walls.

[0037] Specifically, when the management and assessment module generates health management strategies, it includes: pre-setting basic management strategies corresponding to four levels of health, namely, excellent management strategy, good management strategy, intermediate management strategy, and poor management strategy; Among them, the excellent management strategy is to maintain routine monthly monitoring, the good management strategy is to add bi-monthly special monitoring of the flange-web connection, the intermediate management strategy is to conduct weekly full-dimensional monitoring plus crack width detection, and the poor management strategy is to conduct daily monitoring plus structural reinforcement assessment preparation. The management and evaluation module extracts the current health level and performance degradation rate from the pre-diagnosis report and presets a degradation rate threshold. Compare the performance degradation rate with the degradation rate threshold, and adjust the basic management strategy based on the comparison results: When the performance degradation rate is less than or equal to the minimum value of the degradation rate threshold, the basic management strategy for maintaining the corresponding health level is implemented. When the performance degradation rate is greater than the minimum value of the degradation rate threshold but less than or equal to the maximum value of the degradation rate threshold, the monitoring frequency of the basic management strategy will be increased to 1.5 times the original frequency. When the performance degradation rate exceeds the maximum value of the degradation rate threshold, the basic management policy will be upgraded to a higher-level management policy. After the strategy is adjusted, the basis for the adjustment is recorded, a health management plan including monitoring items, monitoring frequency, and responsible persons is generated, and the plan is synchronously stored in the parameter-performance association database.

[0038] Specifically, four basic management strategies are pre-defined for each of the four health levels. The superior management strategy involves maintaining routine monthly monitoring (conducting visual inspections and collecting key seismic performance data on the 15th of each month, with data accuracy retained to three decimal places). The good management strategy adds bi-weekly specialized monitoring of the flange-web connection (using ultrasonic testing equipment to detect flaws in the connection every 15 and 30 days, recording the sound wave propagation speed and reflected signal amplitude). The intermediate management strategy includes weekly comprehensive monitoring plus crack width detection (conducting comprehensive monitoring of 12 indicators including appearance, stress, and displacement every Friday, simultaneously using a crack width meter to detect the area around the opening and...). Cracks at the connection points (accuracy 0.01mm, recording changes in crack length and width); the differential management strategy involves daily monitoring and structural reinforcement assessment preparation (collecting core indicator data at 10:00 AM daily, completing preliminary discussions of the reinforcement plan within 3 working days, and submitting a detailed assessment report within 7 working days); simultaneously, a performance degradation rate threshold is preset (using the monthly change rate of the damage evolution index as the core indicator, with a minimum threshold of 0.005 / month and a maximum of 0.01 / month; the performance degradation rate is calculated as the average monthly change rate of the damage degradation index over the past 3 months). This module extracts the current health level from the pre-diagnosis report and the calculated performance degradation rate, and then... Its comparison with thresholds: When the performance degradation rate is ≤0.005 / month, maintain the basic management strategy corresponding to the health level; when 0.005 / month < performance degradation rate ≤0.01 / month, increase the monitoring frequency of the basic management strategy to 1.5 times the original frequency (Excellent to once every 20 days, Good to once every 10 days, Medium to once every 4.7 days, Poor to once every 16 hours); when the performance degradation rate is >0.01 / month, upgrade the basic management strategy to a higher-level management strategy (Excellent to Good, Good to Medium, Medium to Poor, Poor to 24-hour monitoring + emergency hardening review). After slight adjustments, detailed records of the adjustment basis (including the current health level, specific values ​​of performance degradation rate, threshold range, and comparison results) are generated. A health management plan is generated, which includes specific monitoring items (clearly defining testing equipment, operating procedures, and data accuracy requirements), precise monitoring frequency (accurate to specific dates or time periods), and responsible persons (on-site maintenance personnel are responsible for excellent levels, and structural engineers are responsible for good levels and above, with two dedicated testing personnel). The plan is named in the format of "project number-shear wall number-strategy generation date" and synchronously stored in the parameter-performance association database in the form of a structured data table, linking the corresponding pre-diagnosis report and health level determination record.

[0039] The above embodiments, by pre-setting a four-level basic management strategy tailored to different health states of flanged shear walls, and combining it with degradation rate thresholds (0.005 / month, 0.01 / month) based on the monthly change rate of the damage evolution index, dynamically adjust the management strategy (maintain the basic strategy, increase the monitoring frequency by 1.5 times, or upgrade to a higher-level strategy) according to the comparison results of the current health level and performance degradation rate. This achieves precise matching between the management strategy and the actual health status and degradation trend of the shear wall, avoiding the problems of over-maintenance or under-maintenance caused by a single fixed strategy. Furthermore, by generating a standardized health management plan that includes specific monitoring items, precise monitoring frequencies, and clearly defined responsible persons, the feasibility and implementation efficiency of the strategy are improved. At the same time, the plan and adjustment basis are synchronously stored in the parameter-performance correlation database, providing reliable support for subsequent strategy optimization, historical data tracing, and full life cycle health management. This effectively ensures the continuous stability of the seismic performance of flanged shear walls, extends their service life, and reduces safety risks.

[0040] Specifically, when the management and evaluation module feeds evaluation data back to the parameter-performance correlation database to iteratively optimize the model, it includes: presetting model optimization trigger thresholds, which include prediction bias thresholds and data accumulation thresholds; The management and evaluation module periodically extracts evaluation data for a preset period from the parameter-performance correlation database and calculates the prediction deviation for each data point. The prediction deviation is calculated as follows: the prediction deviation equals the absolute value of the model prediction value minus the measured value divided by the measured value multiplied by 100%. The percentage of data with a prediction deviation greater than the prediction deviation threshold, and the total accumulated statistical data: When the data percentage is less than or equal to 10% and the total accumulated data is greater than or equal to the data accumulation threshold, the model is deemed to have met the current accuracy standard and no optimization is required. When the data percentage is greater than 10% or the total accumulated data is less than the data accumulation threshold, model iterative optimization is initiated: The newly accumulated assessment data were used as training samples to retrain the pre-diagnosis algorithm and trend prediction algorithm in the biaxially coupled seismic model, and the constitutive parameters and boundary condition weights of the model were adjusted. After training is complete, the prediction bias of the new model is calculated. When the new prediction bias is less than or equal to the prediction bias threshold, the parameters of the new model are saved and the old model is replaced. If the new prediction deviation is greater than the prediction deviation threshold, retraining is performed. If the new prediction deviation is still greater than the prediction deviation threshold after three consecutive training sessions, the alarm module is triggered to issue a warning and record abnormal information about model optimization.

[0041] Specifically, a pre-set model optimization trigger threshold is established, with the prediction deviation threshold set at 8% (i.e., the upper limit of the relative deviation between the model's predicted value and the measured value), and the data accumulation threshold set at 500 records (each evaluation data record includes real-time parameters, model prediction performance indicators, measured performance indicators, and a judgment timestamp). This module periodically extracts all evaluation data for each preset 3-month period from the parameter-performance correlation database. The prediction deviation for each data record is calculated using the formula "Prediction Deviation = |Model Predicted Value - Measured Value| / Measured Value × 100%" (if the measured value is 0, then...). The prediction bias is corrected using the formula "Prediction bias = |Model predicted value - Measured value| / (Measured value + 0.001) × 100%" to avoid a denominator of zero. Then, the percentage of data with a prediction bias greater than 8% is calculated, along with the total accumulated data for that period. When the percentage of data with a prediction bias exceeding the threshold is ≤10% and the total accumulated data is ≥500 records, the model's current accuracy is considered satisfactory, and no optimization is needed. When the percentage of data with a prediction bias exceeding the threshold is >10% or the total accumulated data is <500 records, model iteration optimization is immediately initiated: the newly accumulated evaluation data is divided into training and validation sets in a 7:2:1 ratio. The test set was used as training samples to retrain the convolutional neural network pre-diagnosis algorithm and the long short-term memory network trend prediction algorithm in the biaxially coupled seismic model. The elastic modulus correction coefficient of the concrete constitutive model, the yield strength adjustment weight of the reinforced steel constitutive model, and the load distribution coefficient of the biaxial load coupling boundary conditions were adjusted. The training process consisted of 50 training rounds with an initial learning rate of 0.001. After each training round, the model accuracy was verified using a validation set. After training, the average prediction deviation of the new model was calculated using the test set. When the average prediction deviation of the new model was ≤8%, the new model parameters were automatically saved and overwritten. Replace the old model and synchronously record the model update time, parameter adjustment details, and accuracy comparison data before and after optimization. When the average prediction deviation of the new model is >8%, restart training (the initial learning rate is decayed by 80% of the previous one each time it is retrained). If the average prediction deviation of the new model is still >8% after 3 consecutive training sessions, trigger the alarm module to issue an audible and visual warning. At the same time, upload the abnormal model optimization information (including optimization start time, number of training sessions, prediction deviation of each training session, total amount of data accumulated and its proportion) to the remote monitoring platform and store it in the parameter-performance correlation database in the form of structured logs.

[0042] The above embodiments, by presetting a clear prediction deviation threshold (8%) and a data accumulation threshold (500 data points), periodically extract evaluation data for a preset time period and accurately calculate the prediction deviation according to a standardized formula. Combining the data proportion and the total accumulation amount, the accuracy of the model is determined. This avoids the waste of resources caused by blind optimization and ensures the targeted nature of model optimization. When optimization is initiated, the newly accumulated data is divided into training, validation, and test sets according to a scientific ratio. The pre-diagnosis algorithm and trend prediction algorithm are retrained, and the model constitutive parameters and boundary condition weights are adjusted. With the help of a multi-round training attenuation mechanism and validation process, the prediction accuracy and adaptability of the model for the biaxial coupling seismic performance of flanged shear walls are effectively improved. At the same time, the mechanism of triggering an alarm after three consecutive training failures ensures the reliability and safety of model optimization. The evaluation data feedback and model iteration form a closed loop. The continuously optimized model provides more accurate technical support for subsequent dynamic seismic performance evaluation, further improving the scientificity and effectiveness of health management of flanged shear walls.

[0043] Specifically, when the damage risk point identification unit dynamically reviews potential damage risk points, it includes: setting a preset damage risk point review interval, review indicators including stress change rate and displacement increment at the risk point, and setting a preset review stress change rate threshold and a review displacement increment threshold. Based on the damage risk point review interval, real-time stress data and displacement data of the risk points are extracted through the biaxially coupled seismic model, and the stress change rate and displacement increment are calculated. The stress change rate is compared with the verification stress change rate threshold, and the displacement increment is compared with the verification displacement increment threshold. The risk level is adjusted based on the comparison results. When the rate of change of stress is less than the minimum value of the threshold for the rate of change of stress and the displacement increment is less than the minimum value of the threshold for the displacement increment, the original risk level is maintained. When the rate of change of stress falls within the threshold range of the rate of change of stress or the displacement increment falls within the threshold range of the displacement increment, the original risk level will be upgraded by one level. When the rate of change of stress exceeds the maximum value of the threshold for the rate of change of stress and the displacement increment exceeds the maximum value of the threshold for the displacement increment, the original risk level is upgraded to the highest risk level, which is an upgraded state of high risk. If the risk level is determined to be extremely high after review, the alarm module will immediately issue an emergency alarm and push the risk point location and review data to the structural safety management platform. All review results are stored in the parameter-performance correlation database for subsequent model optimization.

[0044] Specifically, when the damage risk point identification unit dynamically reviews potential damage risk points, the preset review interval is 15 days (calculated from the date of initial risk point identification, with automatic reviews triggered on the 1st and 16th of each month). Review indicators include the stress change rate (unit: % / day) and displacement increment (unit: mm) at the risk point. Preset thresholds for the stress change rate (minimum 0.05% / day, maximum 0.1% / day) and displacement increment (minimum 0.02mm, maximum 0.05mm) are also provided. The stress change rate is calculated using the formula: "(real-time stress value - previous review stress value) / previous review stress value × 10". The calculation of "0% / review interval days" (results are rounded to four decimal places) is used. The displacement increment is the difference between the real-time displacement value and the previous review displacement value (accurate to 0.001 mm). Real-time stress data (in MPa units, rounded to two decimal places) and displacement data (three-dimensional coordinate form, accurate to 0.001 mm) are extracted from each risk point using a biaxially coupled seismic model at 15-day review intervals. The stress change rate and displacement increment are then calculated using the formulas described above. The calculation results are compared with preset thresholds: when the stress change rate is ≤0.05% / day and the displacement increment is <0.02 mm, the original risk level (low, medium, high risk) is maintained; when 0.05% / day < stress change rate, the original risk level is maintained. When the rate of change is ≤0.1% / day or the displacement increment is ≤0.05mm and ≤0.02mm, the original risk level will be upgraded by one level (low risk to medium risk, medium risk to high risk, high risk to extreme risk). When the stress change rate is >0.1% / day and the displacement increment is >0.05mm, the original risk level will be directly upgraded to extreme risk (extreme risk is an upgraded state of high risk, corresponding to an emergency state where the structure is on the verge of damage and instability). If the risk level is extreme risk after review, the alarm module will be immediately triggered to issue an audible and visual emergency alarm (the alarm lasts for 5 minutes, cycling every 1 minute), and at the same time, the three-dimensional coordinates of the risk point (accurate to 0.01m) will be pushed through the 4G module. The specific values ​​of stress change rate, displacement increment, review timestamp (accurate to the second), and corresponding load cycle number are sent to the structural safety management platform (the platform provides real-time pop-up notifications and sends SMS notifications to structural engineers and project managers). All review results (including risk point ID, original risk level, adjusted risk level, real-time stress / displacement data, calculation process, comparison threshold, and review time) are named in the format of "Review Date (YYYYMMDD) - Risk Point ID" and stored in the parameter-performance association database in the form of a structured data table. This provides measured data support for the subsequent optimization of risk judgment threshold and training of damage feature extraction algorithm in the biaxially coupled seismic model.

[0045] The above embodiments, by pre-setting a 15-day scientific review interval, use stress change rate (threshold 0.05% / day, 0.1% / day) and displacement increment (threshold 0.02mm, 0.05mm) as core review indicators. They accurately calculate real-time changes in risk points using standardized formulas and compare these changes with the thresholds, enabling dynamic adjustment of risk levels (maintaining the original level, upgrading by one level, or directly upgrading to the highest risk level). This avoids omissions and misjudgments caused by the solidification of risk point states. Furthermore, the mechanism of emergency alarms for highest risk levels and real-time push of risk point locations and review data to the structural safety management platform provides valuable time for emergency risk handling. Simultaneously, all review results are stored in a parameter-performance correlation database, providing measured data support for subsequent optimization of the biaxially coupled seismic model. This effectively improves the timeliness, accuracy, and reliability of damage risk identification for flanged shear walls, building a dynamic protective barrier for structural seismic safety.

[0046] Reference Figure 3 In some embodiments of this application, a method for evaluating the seismic performance of a flanged shear wall using biaxial coupling includes the following steps: Step S100: Real-time acquisition of geometric data, material data and biaxial coupled load data of the flanged shear wall; outlier removal and standardization processing of the acquired data; acquisition of historical seismic test data, damage records and maintenance files of the shear wall; and establishment of a parameter-performance correlation database. Step S200: Based on the parameter-performance correlation database, construct a digital finite element foundation model that considers the biaxial load coupling effect; at the same time, introduce a pre-diagnosis algorithm to form a biaxial coupling seismic model; simulate the stress-strain distribution, component displacement and corresponding load of the shear wall under biaxial load through the biaxial coupling seismic model, and identify potential damage risk points at the connection between the wall flange and web, around the opening and at the bottom of the wall. Step S300: Input the preprocessed real-time data into the dual-axis coupled seismic model, calculate the seismic performance index, preset the health status threshold range, compare the real-time performance index with the threshold to determine the current health level; at the same time, combine historical performance data, use a trend prediction algorithm to analyze the performance degradation rate, predict the probability of potential failures under different load cycles, and generate a pre-diagnosis report. Step S400: Generate a health management strategy based on the pre-diagnosis report and performance degradation trend; conduct a comprehensive evaluation of the biaxial coupling seismic performance, output an evaluation report, and feed the evaluation data back to the parameter-performance correlation database for iterative optimization of the model and pre-diagnosis algorithm, which will be used for subsequent dynamic evaluation of the seismic performance of the shear wall.

[0047] It is understandable that the above-mentioned biaxially coupled seismic performance evaluation method and system for flanged shear walls have the same beneficial effects, and will not be elaborated further here.

[0048] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0049] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0050] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0051] These computer program instructions can also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0052] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A biaxially coupled seismic performance evaluation system for flanged shear walls, characterized in that, include: The acquisition and preprocessing module is configured to acquire geometric data, material data and biaxial coupled load data of shear walls with flanges in real time, perform outlier removal and standardization processing on the acquired data, and obtain historical seismic test data, damage records and maintenance files of shear walls to establish a parameter-performance correlation database. The model building module is configured to construct a digital finite element foundation model considering the biaxial load coupling effect based on a parameter-performance correlation database; at the same time, a pre-diagnosis algorithm is introduced to form a biaxial coupling seismic model; the stress-strain distribution, component displacement and corresponding load of the shear wall under biaxial load are simulated through the biaxial coupling seismic model, and potential damage risk points at the connection between the wall flange and web, around the opening and at the bottom of the wall are identified; The analysis module is configured to input preprocessed real-time data into the dual-axis coupled seismic model, calculate seismic performance indicators, preset health status threshold ranges, compare real-time performance indicators with thresholds to determine the current health level, and combine historical performance data to analyze the performance degradation rate using a trend prediction algorithm, predict the probability of potential failures under different load cycles, and generate a pre-diagnosis report. The management and assessment module is configured to generate health management strategies based on pre-diagnosis reports and performance degradation trends; A comprehensive assessment of the biaxially coupled seismic performance is conducted, an assessment report is output, and the assessment data is fed back to the parameter-performance correlation database for iterative optimization of the model and pre-diagnostic algorithm, which is then used for the dynamic assessment of the seismic performance of the shear wall in the future.

2. The biaxially coupled seismic performance evaluation system for flanged shear walls according to claim 1, characterized in that, The acquisition and preprocessing module includes: The geometric parameter acquisition unit is configured to acquire the flange width, flange thickness, web height, web thickness, opening size and reinforcement arrangement parameters of the flanged shear wall through a three-dimensional laser scanning device and a BIM model extraction tool, and generate a structured geometric parameter dataset. The material parameter acquisition unit is configured to acquire the compressive strength, axial tensile strength, yield strength, and elastic modulus of concrete cubes through material mechanics testing equipment, and simultaneously input the material production batch and service life information to form a material performance parameter library. The biaxial coupling load acquisition unit is configured to acquire reciprocating biaxial load data in real time through load sensors and data acquisition instruments, including horizontal X-axis load amplitude, horizontal Y-axis load amplitude, vertical load amplitude and biaxial loading path history, and record the load application rate and holding time. The data preprocessing unit is configured to perform noise reduction on the acquired data using a wavelet filtering algorithm, remove outliers in geometric and material parameters using an isolated forest algorithm, and map load and performance data to the [0,1] interval using a min-max normalization method to generate a standardized dataset and store it in the parameter-performance correlation database.

3. The biaxially coupled seismic performance evaluation system for flanged shear walls according to claim 1, characterized in that, The model building module includes: The finite element basic model building unit is configured to be based on a standardized dataset in the parameter-performance correlation database. Shell elements are used to simulate the concrete and steel reinforcement diffusion layer of the shear wall. The concrete constitutive model and the steel reinforcement constitutive model are defined, and biaxial load coupling boundary conditions are set. The pre-diagnosis algorithm unit is configured as a damage feature extraction algorithm based on a convolutional neural network. The damage feature extraction algorithm is trained by damage images and stress-strain data from historical seismic tests and extracts stress concentration features and displacement mutation features in the model simulation process in real time. The damage risk point identification unit is configured to identify potential damage risk points at the junction of the wall flange and web, the upper and lower edges of the opening, and the bottom of the wall limb based on the feature data output by the finite element model and the pre-diagnosis algorithm, combined with a preset risk judgment threshold, and to mark the risk level, which includes low risk, medium risk, and high risk.

4. The biaxially coupled seismic performance evaluation system for flanged shear walls according to claim 1, characterized in that, The analysis module includes: The performance index calculation unit is configured to input preprocessed data into the biaxially coupled seismic model and calculate seismic performance indices, including displacement ductility coefficient, bearing capacity attenuation rate, stiffness degradation coefficient, and damage coefficient. The health level determination unit is configured with a preset health status threshold range. It compares the real-time calculated seismic performance index with the preset threshold range and determines the current health level of the flanged shear wall based on the comparison result. The degradation trend analysis and fault prediction unit is configured to use a long short-term memory network algorithm, combined with historical performance data in the parameter-performance correlation database, to fit a performance degradation curve and analyze the performance degradation rate; based on the degradation curve, it predicts the probability of potential fault occurrence under different load cycles and generates a pre-diagnostic report containing fault type, predicted occurrence time, and risk level.

5. The biaxially coupled seismic performance evaluation system for flanged shear walls according to claim 4, characterized in that, When the health level determination unit determines the health level by comparing real-time performance indicators with preset thresholds, it includes: The health level determination unit presets four health status threshold ranges, namely, excellent health threshold range, good health threshold range, medium health threshold range and poor health threshold range; The health level determination unit compares the real-time calculated displacement ductility coefficient, bearing capacity attenuation rate, stiffness degradation coefficient, and damage coefficient with the four-level threshold intervals respectively: When all performance indicators are within the health excellent threshold range, the current health level of the flanged shear wall is determined to be excellent. When at least three performance indicators are within the healthy good threshold range and no indicator is below the healthy medium threshold range, the current health level is determined to be good. When at least two performance indicators are within the healthy intermediate threshold range and no indicator is below the healthy poor threshold range, the current health level is determined to be intermediate. When any performance index is within the poor health threshold range, or when the damage evolution index is greater than the boundary between the intermediate health threshold range and the poor health threshold range, the current health level is determined to be poor. After the determination is completed, the health level determination unit records the determination time, the original values ​​of the performance indicators to be compared and the threshold range, and stores them in the parameter-performance association database.

6. The biaxially coupled seismic performance evaluation system for flanged shear walls according to claim 4, characterized in that, When the degradation trend analysis and fault prediction unit generates a pre-diagnostic report, it includes: The degradation trend analysis and fault prediction unit presets three levels of fault occurrence probability intervals, namely low-risk probability interval, medium-risk probability interval, and high-risk probability interval. The unit calculates the probability of failure under different load cycles based on the fitted performance degradation curve: When the probability of a fault occurrence falls within the low-risk probability range, the fault risk level is marked as low risk, and the pre-diagnosis report recommends maintaining the regular monitoring frequency. When the probability of failure is in the medium risk range, the failure risk level is marked as medium risk. The pre-diagnosis report recommends increasing the monitoring frequency to twice the original frequency and adding detection of the flange-web connection area. When the probability of a failure is in the high-risk probability range, the failure risk level is marked as high-risk, and the pre-diagnosis report recommends immediate shutdown for inspection and assessment of whether reinforcement is needed. The pre-diagnosis report also includes shear wall foundation information, real-time performance index curves, and degradation trend fitting plots, which are output in the form of text and visual charts and uploaded to the remote monitoring platform simultaneously.

7. The biaxially coupled seismic performance evaluation system for flanged shear walls according to claim 1, characterized in that, When the management and assessment module generates a health management strategy, it includes: Pre-set basic management strategies corresponding to four health levels: excellent management strategy, good management strategy, medium management strategy, and poor management strategy. Among them, the excellent management strategy is to maintain routine monthly monitoring, the good management strategy is to add bi-monthly special monitoring of the flange-web connection, the intermediate management strategy is to conduct weekly full-dimensional monitoring plus crack width detection, and the poor management strategy is to conduct daily monitoring plus structural reinforcement assessment preparation. The management and evaluation module extracts the current health level and performance degradation rate from the pre-diagnosis report, and presets a degradation rate threshold. Compare the performance degradation rate with the degradation rate threshold, and adjust the basic management strategy based on the comparison results: When the performance degradation rate is less than or equal to the minimum value of the degradation rate threshold, the basic management strategy for maintaining the corresponding health level is implemented. When the performance degradation rate is greater than the minimum value of the degradation rate threshold but less than or equal to the maximum value of the degradation rate threshold, the monitoring frequency of the basic management strategy will be increased to 1.5 times the original frequency. When the performance degradation rate exceeds the maximum value of the degradation rate threshold, the basic management policy will be upgraded to a higher-level management policy. After the strategy is adjusted, the basis for the adjustment is recorded, a health management plan including monitoring items, monitoring frequency, and responsible persons is generated, and the plan is synchronously stored in the parameter-performance association database.

8. The biaxially coupled seismic performance evaluation system for flanged shear walls according to claim 1, characterized in that, When the management and evaluation module feeds evaluation data back to the parameter-performance correlation database to iteratively optimize the model, it includes: A model optimization trigger threshold is preset, which includes a prediction bias threshold and a data accumulation threshold. The management and evaluation module periodically extracts evaluation data for a preset period from the parameter-performance correlation database and calculates the prediction deviation for each data point. The prediction deviation calculation formula is: prediction deviation equals the absolute value of the model prediction value minus the measured value divided by the measured value multiplied by 100%. The percentage of data with a prediction deviation greater than the prediction deviation threshold, and the total accumulated statistical data: When the data percentage is less than or equal to 10% and the total accumulated data is greater than or equal to the data accumulation threshold, the model is deemed to have met the current accuracy standard and no optimization is required. When the data percentage is greater than 10% or the total accumulated data is less than the data accumulation threshold, model iterative optimization is initiated: The newly accumulated assessment data were used as training samples to retrain the pre-diagnosis algorithm and trend prediction algorithm in the biaxially coupled seismic model, and the constitutive parameters and boundary condition weights of the model were adjusted. After training is complete, the prediction bias of the new model is calculated. When the new prediction bias is less than or equal to the prediction bias threshold, the parameters of the new model are saved and the old model is replaced. If the new prediction deviation is greater than the prediction deviation threshold, retraining is performed. If the new prediction deviation is still greater than the prediction deviation threshold after three consecutive training sessions, the alarm module is triggered to issue a warning and record abnormal information about model optimization.

9. The biaxially coupled seismic performance evaluation system for flanged shear walls according to claim 3, characterized in that, When the damage risk point identification unit dynamically reviews potential damage risk points, it includes: The preset interval for reviewing damage risk points is used. The review indicators include the stress change rate and displacement increment at the risk points. The preset thresholds for reviewing the stress change rate and displacement increment are also included. Based on the damage risk point review interval, real-time stress data and displacement data of the risk points are extracted through the biaxially coupled seismic model, and the stress change rate and displacement increment are calculated. The stress change rate is compared with the verification stress change rate threshold, and the displacement increment is compared with the verification displacement increment threshold. The risk level is adjusted based on the comparison results. When the rate of change of stress is less than the minimum value of the threshold for the rate of change of stress and the displacement increment is less than the minimum value of the threshold for the displacement increment, the original risk level is maintained. When the rate of change of stress falls within the threshold range of the rate of change of stress or the displacement increment falls within the threshold range of the displacement increment, the original risk level will be upgraded by one level. When the rate of change of stress is greater than the maximum value of the threshold for the rate of change of stress and the displacement increment is greater than the maximum value of the threshold for the displacement increment, the original risk level is upgraded to the highest risk level, which is an upgraded state of high risk. If the risk level is determined to be extremely high after review, the alarm module will immediately issue an emergency alarm and push the risk point location and review data to the structural safety management platform. All review results are stored in the parameter-performance correlation database for subsequent model optimization.

10. A method for evaluating the seismic performance of a biaxially coupled shear wall with flanges, characterized in that, The biaxially coupled seismic performance evaluation system for flanged shear walls as described in any one of claims 1-9 includes: Real-time acquisition of geometric data, material data, and biaxial coupled load data of flanged shear walls; outlier removal and standardization processing of the acquired data; acquisition of historical seismic test data, damage records, and maintenance archives of shear walls; and establishment of a parameter-performance correlation database. Based on a parameter-performance correlation database, a digital finite element foundation model considering biaxial load coupling effect is constructed; at the same time, a pre-diagnosis algorithm is introduced to form a biaxial coupling seismic model; the stress-strain distribution, component displacement and corresponding load of shear wall under biaxial load are simulated through the biaxial coupling seismic model to identify potential damage risk points at the connection between the wall flange and web, around the opening and at the bottom of the wall. The preprocessed real-time data is input into the dual-axis coupled seismic model to calculate the seismic performance index. A health status threshold range is preset, and the real-time performance index is compared with the threshold to determine the current health level. At the same time, combined with historical performance data, a trend prediction algorithm is used to analyze the performance degradation rate, predict the probability of potential failures under different load cycles, and generate a pre-diagnosis report. Based on the pre-diagnosis report and performance degradation trend, a health management strategy is generated; a comprehensive assessment of the biaxial coupling seismic performance is conducted, an assessment report is output, and the assessment data is fed back to the parameter-performance correlation database for iterative optimization of the model and pre-diagnosis algorithm, which is used for subsequent dynamic assessment of the seismic performance of shear walls.

Images