A full life cycle intelligent detection and evaluation system for a shock insulation and mitigation structure

Abstract

This invention discloses an intelligent detection and evaluation system for the entire life cycle of seismic isolation and damping structures, belonging to the field of intelligent monitoring technology for structural engineering. It includes: intelligent seismic isolation and damping bearings and multi-source sensing modules, a seismic isolation and damping layer edge gateway module, a full life cycle digital twin platform, and a regional building cluster collaborative monitoring and early warning module. It achieves system-level collaborative monitoring from individual bearings to regional building clusters; it integrates data across the entire chain from design to production, construction, operation and maintenance, and demolition, and uses blockchain technology to achieve tamper-proof traceability; it establishes a multi-physics field coupled damage evolution model, integrating environmental and mechanical parameters to achieve accurate quantitative damage assessment; and it constructs a remaining life prediction model based on historical damage accumulation, improving prediction accuracy. This invention solves the common problems of existing technologies, such as system fragmentation, lack of full life cycle coverage, insufficient multi-physics field coupling analysis, poor long-term sensor reliability, low remaining life prediction accuracy, and disconnect between monitoring and control.

Description

Technical Field

[0001] This invention belongs to the field of intelligent monitoring technology for structural engineering, specifically relating to an intelligent detection and evaluation system for the entire life cycle of seismic isolation and damping structures. Background Technology

[0002] Earthquakes are among the most serious natural disasters threatening human life and property, and the seismic performance of building structures directly affects the extent of damage caused by earthquakes. Seismic isolation and damping technology, as a new generation of seismic-resistant technology, effectively isolates and dissipates seismic energy by setting up seismic isolation layers or energy-dissipating structures between the building foundation and the superstructure, significantly reducing the seismic response of the superstructure.

[0003] The performance of seismic isolation and damping systems directly determines the seismic safety level of buildings. However, seismic isolation and damping components, such as rubber seismic isolation bearings, friction pendulum bearings, and viscous dampers, are subject to temperature changes, humidity erosion, ultraviolet aging, fatigue loads, and seismic forces during long-term service, leading to gradual degradation of their mechanical properties and even internal damage and functional failure. Traditional seismic isolation and damping system testing mainly relies on regular manual inspections and sampling. With the development of sensor technology, Internet of Things (IoT) technology, and artificial intelligence technology, intelligent monitoring technology for seismic isolation and damping structures is gradually developing. However, current related technologies still have at least the following shortcomings: 1) Existing technologies only target individual seismic isolation bearings or energy dissipation and damping components, lacking coordinated monitoring and evaluation of the entire seismic isolation and damping system consisting of the isolation layer, superstructure, and foundation; 2) The testing and evaluation mainly focuses on the service and maintenance phase of the seismic isolation and damping structure, lacking testing and evaluation of the entire life cycle of the structure from design, production and construction; resulting in serious data gaps in the above-mentioned stages, making it impossible to form a complete digital archive; 3) Existing detection and evaluation models are mostly based on single material aging tests, without considering the cumulative damage from historical earthquakes and the coupling effect of multiple factors, resulting in low evaluation accuracy. Summary of the Invention

[0004] This invention provides an intelligent detection and evaluation system for the entire life cycle of seismic isolation and damping structures. It adopts a four-level collaborative modular architecture to achieve system-level collaborative monitoring from individual supports to entire building complexes. It integrates data across the entire chain from design to production, construction, operation and maintenance, and demolition, using blockchain technology to achieve tamper-proof traceability. A multi-physics coupled damage evolution model is established, fusing environmental and mechanical parameters to achieve accurate quantitative damage assessment. A self-calibrating and self-healing sensor network is designed to meet long-term service requirements. A remaining life prediction model based on historical damage accumulation is constructed, improving prediction accuracy. It achieves integrated monitoring, evaluation, and control, and links with active / semi-active devices to complete the transformation from passive monitoring to proactive disaster prevention. This invention solves the common problems of existing technologies, such as system fragmentation, lack of full life cycle coverage, insufficient multi-physics coupled analysis, poor long-term sensor reliability, low remaining life prediction accuracy, and disconnect between monitoring and control.

[0005] To achieve the above-mentioned technical objectives, the present invention is implemented through the following technical solution: A smart detection and evaluation system for the entire life cycle of seismic isolation and damping structures, comprising: Intelligent vibration isolation bearings and multi-source sensing modules, vibration isolation layer edge gateway modules, full life cycle digital twin platform, and regional building cluster collaborative monitoring and early warning modules; The intelligent vibration isolation and damping support and the multi-source sensing module include the intelligent vibration isolation and damping support, which is equipped with a multi-source sensor array for collecting multi-source data; The edge computing unit has a built-in processor and a memory. The memory stores computer-executable programs for preprocessing, feature extraction, and preliminary damage identification of the collected raw multi-source data. The edge gateway module of the vibration isolation layer, based on the processor and memory provided in the edge computing unit, executes a computer-executable program to complete the synchronous acquisition of multi-source data; performs preprocessing such as denoising, missing value completion, and standardization on the acquired multi-source data; performs preliminary damage assessment of the vibration isolation layer based on edge computing; and links with the vibration isolation device to achieve real-time control. The full lifecycle digital twin platform includes a full lifecycle data management unit for recording and managing data throughout the entire lifecycle, from the design, production, construction, operation and maintenance, and demolition phases. It assigns a unique blockchain identity to each set of seismic isolation bearings based on blockchain technology. The full lifecycle data is stored on the blockchain to prevent tampering. It enables cross-platform data sharing and traceability. The digital twin modeling unit is used for multi-dimensional modeling processing, including geometric modeling, mechanical modeling, and multi-scale modeling; and it updates the parameters of the digital twin in real time based on on-site monitoring data using data assimilation technology. The intelligent assessment and decision-making unit constructs a multi-physics coupled damage evolution model to quantitatively assess and locate damage to seismic isolation and damping structures; predicts the remaining service life of seismic isolation and damping structures based on historical damage accumulation; and completes post-earthquake damage assessment and repair strategy recommendations. The regional building cluster collaborative monitoring and early warning module is used to realize the sharing of monitoring data of multiple isolation and damping buildings in the region and to assess the overall seismic toughness level of isolation and damping buildings in the region.

[0006] Preferably, the multi-source sensor array includes: Distributed fiber optic strain sensors are used to monitor the strain distribution in rubber layers and steel plates, with high accuracy. ; A triaxial MEMS accelerometer for monitoring horizontal and vertical acceleration; Laser displacement sensor, used to monitor three-dimensional displacement; Piezoelectric ceramic sensors are used to monitor internal damage and interface delamination. Temperature and humidity sensors are used to monitor temperature and humidity data inside the vibration isolation bearings; Ultraviolet (UV) monitors are used to monitor the intensity of UV radiation in the environment and assess the degree of rubber aging.

[0007] Preferably, the system sensor network is arranged based on the sensor array in the following pattern: Superstructure monitoring: Accelerometers and displacement sensors are installed at key floor locations to monitor the seismic response of the superstructure; Foundation monitoring: Settlement sensors and inclinometers are installed at the bottom of the foundation to monitor uneven settlement of the foundation; Environmental monitoring: Temperature, humidity, ultraviolet radiation, and chloride ion concentration sensors are installed in the isolation layer to monitor the impact of environmental parameters on the isolation system.

[0008] Preferably, the multi-source sensor array performs self-calibration in the following manner: Each set of vibration isolation and damping structures has 3 built-in redundant sensors, which achieve in-situ self-calibration through mutual calibration. When a sensor fails, it automatically switches to the redundant sensor to ensure continuous system operation. A calibration procedure is performed once a month, with the calibration accuracy set to ±2%.

[0009] Preferably, the multi-source data is denoised using wavelet transform and Kalman filtering; missing values ​​are filled in using interpolation; and the data is standardized using the Min-Max standardization method.

[0010] Preferably, the preliminary damage assessment method for the damping layer is as follows: 1) Real-time threshold pre-screening: A dynamic threshold method is used with a sliding window, a window size of 10 seconds, and a step size of 1 second. The statistical thresholds of each parameter are calculated to detect sudden anomalies. 2) Lightweight Machine Learning for Damage Recognition: Extracting lightweight, damage-sensitive features from pre-processed multi-source data. Time-domain characteristics: peak value, root mean square value, kurtosis, margin factor, waveform factor Frequency domain characteristics: dominant frequency offset, 1st to 3rd octave amplitude ratio, and bandwidth energy proportion. Simple time-frequency characteristics: low-frequency energy from 3-layer wavelet packet decomposition; A pre-trained Lightweight Gradient Boosting Decision Tree (LightGBM) model is used to output the damage probability and the initial damage type, which includes: rubber aging, interface peeling, steel plate yielding, and connection loosening. 3) Multi-support, multi-source evidence fusion An improved DS evidence theory was used to integrate the evaluation results of all smart bearings within the same seismic isolation layer; An evidence weighting coefficient is introduced to dynamically adjust the credibility of each support's evidence based on the sensor's self-calibration results; Output the overall damage status and initial damage area of ​​the vibration isolation and damping layer.

[0011] Preferably, after the initial damage assessment, it is linked with the vibration isolation and damping device, adopting a phased linkage control strategy, as follows: Normal operating conditions: Predictive performance optimization, the edge gateway performs a health assessment of the damping layer every hour, and fine-tunes the basic damping parameters of the semi-active damping device based on ambient temperature, humidity and long-term performance degradation trends; the cloud platform verifies and optimizes the edge-side control strategy once a day and updates the control parameter library. For example, when the ambient temperature is below 5℃, the stiffness of the rubber bearing will increase by 15%~20%, and the system will automatically reduce the current of the magnetorheological damper to keep the overall equivalent damping ratio of the vibration isolation layer at the design value. Early warning level: Local damage suppression. When one or a few supports show early signs of damage, the edge gateway immediately increases the damping force of the damper near the damaged support, transferring the energy of the damaged area to the healthy area. At the same time, the sampling frequency of the damaged support and its adjacent supports was increased from 10Hz to 100Hz, and the abnormal data was uploaded to the cloud for high-precision analysis. The cloud platform generates a detailed damage assessment report within 5 minutes and provides further control recommendations; Emergency Level: Full System Coordinated Control. When significant damage occurs to the isolation and damping layer, such as multiple support strains exceeding yield strain or displacements exceeding 80% of the design value, the edge gateway immediately activates the full isolation layer coordinated control algorithm. For semi-active seismic isolation systems: Magnetorheological damper: The current is instantly adjusted to its maximum value to provide maximum damping force. Variable friction damper: Increases normal force and improves friction energy dissipation capacity. Active tuned mass damper (TMD): Adjusts the frequency to the dominant frequency of the isolation layer to maximize energy absorption effect; For active seismic isolation systems: Active control forces are activated to counteract the inertial forces of the superstructure. The model predictive control (MPC) algorithm is used to optimize the magnitude and direction of the control force in real time. Simultaneously, a local audible and visual alarm is triggered, and an emergency signal is sent to the cloud platform and the building management center. Hazard level: Emergency safety protection. When the damping layer is severely damaged, such as when the horizontal displacement of the support exceeds the limit or the connecting bolts break, the edge gateway will immediately activate emergency braking control. Lock the horizontal displacement of the seismic isolation layer to prevent excessive swaying or even overturning of the superstructure; cut off unnecessary power supply in the building, activate emergency lighting and evacuation broadcasts; send distress signals and building location information to fire, emergency and other departments.

[0012] Preferably, the multiphysics coupled damage evolution model includes the following three-level model structure: First layer: Microscopic material-level multiphysics coupled damage constitutive model For rubber materials, the aging-fatigue-stress coupled damage model adopts the improved Arrhenius-fatigue coupled damage constitutive model. Aging damage considers the coupling effect of temperature, humidity, ultraviolet radiation and static stress; fatigue damage considers the improved Miner rule that takes into account the influence of mean stress and temperature. Steel plate material: The corrosion-fatigue-stress coupled damage model adopts the electrochemical-mechanical coupled damage model; Corrosion damage: The effects of humidity, chloride ion concentration and stress are considered; Interface layer: The bond-peel-environment coupled damage model adopts the cohesion-damage coupling model, which considers the influence of temperature and humidity on the interfacial bond strength; Second layer: Mesoscopic component-level performance degradation model The refined reduced-order mechanical model of the vibration isolation and damping bearings, based on the POD-Galerkin method, reduces the order of the three-dimensional finite element model, decreasing the degrees of freedom from 10^5 to 10^2, thus improving computational efficiency by more than 1000 times. Model input: Microscopic material damage parameters (D_rubber, D_steel, D_interface) Model outputs: horizontal stiffness K_h, vertical stiffness K_v, equivalent damping ratio ξ, and ultimate bearing capacity F_u of the support; Third layer: Macro-system level damage evolution and assessment layer The overall finite element model of the seismic isolation system uses a lumped mass model or a rod system model for the superstructure; the seismic isolation and damping bearings use a second-layer mesoscopic reduced-order model; the foundation uses a rigid foundation or an elastic half-space model; the model updates the component performance parameters every 24 hours based on monitoring data. A multi-support collaborative damage evolution model considers the interactions between various supports within the damping layer: When damage to a support leads to a decrease in stiffness, the load on adjacent supports will increase, accelerating the damage to the adjacent supports. The model uses a damage transfer matrix to describe this synergistic effect and can predict the development trend of damage to the seismic isolation layer and possible failure modes.

[0013] Preferably, the application and implementation process of the multi-physics coupled damage evolution model is as follows: 1) Quantitative assessment and localization of damage Damage is identified by inverting the model-measured response residual minimization method; Input: Displacement, acceleration, and strain monitoring data for each support; Output: Damage degree (0~1) and damage type for each support; Positioning accuracy: ±1 support; damage quantitative error <8%; 2) Remaining useful life prediction a. Model calibration based on historical accumulated damage to obtain the current damage state; Input: All monitoring data from the structure's construction to the current moment, including temperature, humidity, displacement, acceleration, strain, etc.; Run the multiphysics coupled damage evolution model and use the Bayesian inference algorithm to invert and update the model parameters; Output: The accurate damage state vector of each seismic isolation bearing at the current moment; This step is the core link based on historical accumulated damage, ensuring that the prediction starts from the actual state of the structure, rather than from the ideal design state; b. Generate a random load sample set for the next 30 years, based on local meteorological statistics for the past 30 years, and generate time series samples of temperature, humidity, and ultraviolet radiation. Based on the results of local seismic hazard analysis, a sample of possible ground motions over the next 30 years is generated, including the time of occurrence, magnitude, and spectral characteristics. Based on the building's usage function, time series samples of usage load and wind load are generated; The Monte Carlo method was used to generate 10,000 independent future load samples, covering all possible extreme cases; c. Starting from the current damage state, simulate the future damage evolution. For each set of future load samples, use D_current as the initial condition. Run a multiphysics coupled damage evolution model to simulate damage development over the next 30 years over time. When any damage index reaches the failure threshold, such as D_rubber≥0.8, D_interface≥0.9, or the ultimate bearing capacity drops to 70% of the design value, the time is recorded as the failure time. d. Statistical analysis was performed to obtain the remaining lifetime at different confidence levels. Statistical analysis was conducted on the failure times obtained from 10,000 sets of simulations to obtain the probability distribution of failure times. Calculate the remaining useful life at different confidence levels based on the probability distribution: 90% confidence level: The time during which the structure will not fail with a 90% probability. 95% confidence level: The time during which the structure will not fail with a 95% probability. 99% confidence level: The time during which the structure will not fail with a 99% probability; 3) Recommendations for post-earthquake damage assessment and repair strategies Immediately after the earthquake, the model is updated with seismic records and monitoring data; the post-earthquake damage level and remaining bearing capacity of each bearing are calculated; the effects of different repair schemes, such as replacing bearings, reinforcing bearings, and adding dampers, are simulated; the optimal repair strategy is given by comprehensively considering safety, economy, and construction difficulty; and detailed repair schemes and cost estimates are generated.

[0014] The beneficial effects of this invention are: The technical solution provided by this invention realizes multi-level collaborative monitoring and evaluation from a single support to the entire seismic isolation system and then to a regional building complex; it establishes a complete digital archive of data from the entire life cycle of the seismic isolation and damping structure, including design, production, construction, operation and maintenance and demolition; the multi-physics field coupled damage evolution model realizes accurate location and quantitative assessment of damage; in addition to real-time monitoring health data of the seismic isolation and damping structure, it also predicts the remaining service life based on historical damage accumulation through the above model, significantly improving the prediction accuracy. Attached Figure Description

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 This is a system module structure block diagram of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Example

[0018] A smart detection and evaluation system for the entire life cycle of seismic isolation and damping structures, comprising: The intelligent vibration isolation bearing and multi-source sensing module include the intelligent vibration isolation bearing, which is equipped with a multi-source sensor array for collecting multi-source data. The multi-source sensor array configured in this embodiment is as follows: Distributed fiber optic strain sensors are used to monitor the strain distribution of the rubber layer and steel plate in vibration isolation structures, achieving high accuracy. ; A triaxial MEMS accelerometer is used to monitor the horizontal and vertical acceleration of a vibration isolation and damping structure when it is subjected to vibration disturbance. Laser displacement sensors are used to monitor the three-dimensional displacement of vibration isolation and damping structures. Piezoelectric ceramic sensors are used to monitor internal damage and interfacial delamination in vibration isolation and damping structures. Temperature and humidity sensors are used to monitor temperature and humidity data inside the vibration isolation and damping structure; Ultraviolet (UV) monitors are used to monitor the UV intensity data of the environment and assess the degree of aging of the rubber in the vibration damping structure. In this embodiment, each set of vibration isolation and damping structures has three redundant sensors built in. In-situ self-calibration is achieved through mutual calibration. When the current sensor fails, it automatically switches to the redundant sensor to collect data, ensuring the continuous operation of the system. In addition, the sensors are calibrated once a month with a calibration accuracy of ±2% to ensure the accuracy of data collection. The sensor arrays described above are arranged into a system sensor network according to the following pattern: Superstructure monitoring: Accelerometers and displacement sensors are installed at key floor locations to monitor the seismic response of the superstructure; Foundation monitoring: Settlement sensors and inclinometers are installed at the bottom of the foundation to monitor uneven settlement of the foundation; Environmental monitoring: Temperature, humidity, ultraviolet radiation, and chloride ion concentration sensors are installed in the isolation layer to monitor the impact of environmental parameters on the isolation system; The edge computing unit has a built-in processor and memory. The memory stores computer-executable programs for preprocessing, feature extraction, and preliminary damage identification of the acquired raw multi-source data. The edge gateway module of the vibration isolation and damping layer is based on the processor and memory provided in the edge computing unit. The memory stores computer-executable programs to complete the synchronous acquisition of multi-source data. The acquired multi-source data is denoised by wavelet transform and Kalman filtering, missing values ​​are filled by interpolation, and the data is standardized by Min-Max. A preliminary damage assessment of the damping and isolation layer is performed based on edge computing. The specific assessment method is as follows: 1) Real-time threshold pre-screening: A dynamic threshold method is used with a sliding window. The window size is set to 10 seconds and the step size is set to 1 second. The statistical thresholds of each parameter are calculated to detect sudden anomalies. 2) Lightweight Machine Learning for Damage Recognition: Extracting lightweight, damage-sensitive features from pre-processed multi-source data. Time-domain characteristics: peak value, root mean square value, kurtosis, margin factor, waveform factor Frequency domain characteristics: dominant frequency offset, 1st-3rd octave amplitude ratio, and bandwidth energy proportion. Simple time-frequency characteristics: low-frequency energy from 3-level wavelet packet decomposition; A pre-trained Lightweight Gradient Boosting Decision Tree (LightGBM) model is used to output the damage probability and the initial damage type, which includes: rubber aging, interface peeling, steel plate yielding, and connection loosening. 3) Multi-support, multi-source evidence fusion An improved DS evidence theory was used to integrate the evaluation results of all smart bearings within the same seismic isolation layer; An evidence weighting coefficient is introduced to dynamically adjust the credibility of each support's evidence based on the sensor's self-calibration results; Output the overall damage status and initial damage area of ​​the vibration damping layer. After initial damage assessment, the system is linked with the vibration isolation and damping device, employing a phased linkage control strategy, as detailed below: Normal operating conditions: Predictive performance optimization, the edge gateway performs a health assessment of the damping layer every hour, and fine-tunes the basic damping parameters of the semi-active damping device based on ambient temperature, humidity and long-term performance degradation trends; the cloud platform verifies and optimizes the edge-side control strategy once a day and updates the control parameter library. For example, when the ambient temperature is below 5℃, the stiffness of the rubber bearing will increase by 15%~20%, and the system will automatically reduce the current of the magnetorheological damper to keep the overall equivalent damping ratio of the vibration isolation layer at the design value. Early warning level: Local damage suppression. When one or a few supports show early signs of damage, the edge gateway immediately increases the damping force of the damper near the damaged support, transferring the energy of the damaged area to the healthy area. At the same time, the sampling frequency of the damaged support and its adjacent supports was increased from 10Hz to 100Hz, and the abnormal data was uploaded to the cloud for high-precision analysis. The cloud platform generates a detailed damage assessment report within 5 minutes and provides further control recommendations; Emergency Level: Full System Coordinated Control. When significant damage occurs to the isolation and damping layer, such as multiple support strains exceeding yield strain or displacements exceeding 80% of the design value, the edge gateway immediately activates the full isolation layer coordinated control algorithm. For semi-active seismic isolation systems: Magnetorheological damper: The current is instantly adjusted to its maximum value to provide maximum damping force. Variable friction damper: Increases normal force and improves friction energy dissipation capacity. Active tuned mass damper (TMD): Adjusts the frequency to the dominant frequency of the isolation layer to maximize energy absorption effect; For active seismic isolation systems: Active control forces are activated to counteract the inertial forces of the superstructure. The model predictive control (MPC) algorithm is used to optimize the magnitude and direction of the control force in real time. Simultaneously, a local audible and visual alarm is triggered, and an emergency signal is sent to the cloud platform and the building management center. Hazard level: Emergency safety protection. When the damping layer is severely damaged, such as when the horizontal displacement of the support exceeds the limit or the connecting bolts break, the edge gateway will immediately activate emergency braking control. Lock the horizontal displacement of the seismic isolation layer to prevent excessive swaying or even overturning of the superstructure; cut off unnecessary power supply in the building, activate emergency lighting and evacuation broadcasts; send distress signals and building location information to fire, emergency and other departments.

[0019] The full lifecycle digital twin platform includes a full lifecycle data management unit for recording and managing data throughout the entire lifecycle, from the design, production, construction, operation and maintenance, and demolition phases. It assigns a unique blockchain identity to each set of seismic isolation bearings based on blockchain technology. The full lifecycle data is stored on the blockchain to prevent tampering. It enables cross-platform data sharing and traceability. The digital twin modeling unit is used for multi-dimensional modeling processing, including geometric modeling, mechanical modeling, and multi-scale modeling; and it updates the parameters of the digital twin in real time based on on-site monitoring data using data assimilation technology. The intelligent assessment and decision-making unit constructs a multi-physics coupled damage evolution model to quantitatively assess and locate damage to seismic isolation and damping structures; predicts the remaining service life of seismic isolation and damping structures based on historical damage accumulation; and completes post-earthquake damage assessment and repair strategy recommendations. The multiphysics coupled damage evolution model in this embodiment includes the following three-level model structure: First layer: Microscopic material-level multiphysics coupled damage constitutive model For rubber materials, the aging-fatigue-stress coupled damage model adopts the improved Arrhenius-fatigue coupled damage constitutive model. Aging damage considers the coupling effect of temperature, humidity, ultraviolet radiation and static stress; fatigue damage considers the improved Miner rule that takes into account the influence of mean stress and temperature. Steel plate material: The corrosion-fatigue-stress coupled damage model adopts the electrochemical-mechanical coupled damage model; Corrosion damage: The effects of humidity, chloride ion concentration and stress are considered; Interface layer: The bond-peel-environment coupled damage model adopts the cohesion-damage coupling model, which considers the influence of temperature and humidity on the interfacial bond strength; Second layer: Mesoscopic component-level performance degradation model The refined reduced-order mechanical model of the seismic isolation bearings, based on the POD-Galerkin method, reduces the order of the three-dimensional finite element model, decreasing the degrees of freedom from 10^5 to 10^2, thus improving computational efficiency by more than 1000 times. Model input: Microscopic material damage parameters (D_rubber, D_steel, D_interface) Model outputs: horizontal stiffness K_h, vertical stiffness K_v, equivalent damping ratio ξ, and ultimate bearing capacity F_u of the support; Third layer: Macro-system level damage evolution and assessment layer The overall finite element model of the seismic isolation system uses a lumped mass model or a rod system model for the superstructure; the seismic isolation bearings use the aforementioned mesoscopic reduced-order model; the foundation uses a rigid foundation or an elastic half-space model; the model updates the component performance parameters every 24 hours based on monitoring data. A multi-support collaborative damage evolution model considers the interactions between various supports within the damping layer: When damage to a support leads to a decrease in stiffness, the load on adjacent supports will increase, accelerating the damage to the adjacent supports. The model uses a damage transfer matrix to describe this synergistic effect and can predict the development trend of damage to the isolation layer and possible failure modes. Based on the above multiphysics coupled damage evolution model, the following method is used for 1) quantitative damage assessment and localization: Damage is identified by inverting the model-measured response residual minimization method; Input: Displacement, acceleration, and strain monitoring data for each support; Output: Damage degree (0~1) and damage type for each support; Positioning accuracy: ±1 support; damage quantitative error <8%; 2) The method for predicting remaining useful life is as follows: a. Model calibration based on historical accumulated damage to obtain the current damage state; Input: All monitoring data from the structure's construction to the current moment, including temperature, humidity, displacement, acceleration, strain, etc.; Run the multiphysics coupled damage evolution model and use the Bayesian inference algorithm to invert and update the model parameters; Output: The accurate damage state vector of each seismic isolation bearing at the current moment; This step is the core link based on historical accumulated damage, ensuring that the prediction starts from the actual state of the structure, rather than from the ideal design state; b. Generate a random load sample set for the next 30 years, based on local meteorological statistics for the past 30 years, and generate time series samples of temperature, humidity, and ultraviolet radiation. Based on the results of local seismic hazard analysis, a sample of possible ground motions over the next 30 years is generated, including the time of occurrence, magnitude, and spectral characteristics. Based on the building's usage function, time series samples of usage load and wind load are generated; The Monte Carlo method was used to generate 10,000 independent future load samples, covering all possible extreme cases; c. Starting from the current damage state, simulate the future damage evolution. For each set of future load samples, use D_current as the initial condition. Run a multiphysics coupled damage evolution model to simulate damage development over the next 30 years step by step; When any damage index reaches the failure threshold, such as D_rubber≥0.8, D_interface≥0.9, or the ultimate bearing capacity drops to 70% of the design value, the time is recorded as the failure time. d. Statistical analysis was performed to obtain the remaining lifetime at different confidence levels. Statistical analysis was conducted on the failure times obtained from 10,000 sets of simulations to obtain the probability distribution of failure times. Calculate the remaining useful life at different confidence levels based on the probability distribution: 90% confidence level: The time during which the structure will not fail with a 90% probability. 95% confidence level: The time during which the structure will not fail with a 95% probability. 99% confidence level: The time during which the structure will not fail with a 99% probability; 3) Recommendations for post-earthquake damage assessment and repair strategies Immediately after the earthquake, the model is updated with seismic records and monitoring data; the post-earthquake damage level and remaining bearing capacity of each bearing are calculated; the effects of different repair schemes, such as replacing bearings, reinforcing bearings, and adding dampers, are simulated; the optimal repair strategy is given by comprehensively considering safety, economy, and construction difficulty; and detailed repair schemes and cost estimates are generated.

[0020] The regional building cluster collaborative monitoring and early warning module is used to realize the sharing of monitoring data of multiple isolation and damping buildings in the region and to assess the overall seismic toughness level of isolation and damping buildings in the region.

Claims

1. A smart detection and evaluation system for the entire life cycle of seismic isolation and damping structures, characterized in that, include: The intelligent vibration isolation and damping support and the multi-source sensing module include the intelligent vibration isolation and damping support, which is equipped with a multi-source sensor array for collecting multi-source data; The edge computing unit has a built-in processor and a memory. The memory stores computer-executable programs for preprocessing, feature extraction, and preliminary damage identification of the collected raw multi-source data. The edge gateway module of the vibration isolation layer, based on the processor and memory provided in the edge computing unit, executes a computer-executable program to complete the synchronous acquisition of multi-source data; performs preprocessing such as denoising, missing value completion, and standardization on the acquired multi-source data; performs preliminary damage assessment of the vibration isolation layer based on edge computing; and links with the vibration isolation device to achieve real-time control. The full lifecycle digital twin platform includes a full lifecycle data management unit for recording and managing data throughout the entire lifecycle, from the design, production, construction, operation and maintenance, and demolition phases. It assigns a unique blockchain identity to each set of seismic isolation bearings based on blockchain technology. The full lifecycle data is stored on the blockchain to prevent tampering. It enables cross-platform data sharing and traceability. The digital twin modeling unit is used for multi-dimensional modeling processes, including geometric modeling, mechanical modeling, and multi-scale modeling. Based on on-site monitoring data, the parameters of the digital twin are updated in real time using data assimilation technology; The intelligent assessment and decision-making unit constructs a multi-physics coupled damage evolution model to quantitatively assess and locate damage to the vibration isolation and damping structure; and predicts the remaining service life of the vibration isolation and damping structure based on historical damage accumulation. And complete the post-earthquake structural damage assessment and repair strategy recommendations; The regional building cluster collaborative monitoring and early warning module is used to realize the sharing of monitoring data of multiple isolation and damping buildings in the region and to assess the overall seismic toughness level of isolation and damping buildings in the region.

2. The intelligent detection and evaluation system for the entire life cycle of seismic isolation and damping structures according to claim 1, characterized in that, The multi-source sensor array includes: Distributed fiber optic strain sensors are used to monitor the strain distribution in rubber layers and steel plates, with high accuracy. ; A triaxial MEMS accelerometer for monitoring horizontal and vertical acceleration; Laser displacement sensor, used to monitor three-dimensional displacement; Piezoelectric ceramic sensors are used to monitor internal damage and interface delamination. Temperature and humidity sensors are used to monitor temperature and humidity data inside the vibration isolation bearings; Ultraviolet (UV) monitors are used to monitor the intensity of UV radiation in the environment and assess the degree of rubber aging.

3. The intelligent detection and evaluation system for the entire life cycle of seismic isolation and damping structures according to claim 2, characterized in that, The system sensor network is arranged based on the sensor array in the following pattern: Superstructure monitoring: Accelerometers and displacement sensors are installed at key floor locations to monitor the seismic response of the superstructure; Foundation monitoring: Settlement sensors and inclinometers are installed at the bottom of the foundation to monitor uneven settlement of the foundation; Environmental monitoring: Temperature, humidity, ultraviolet radiation, and chloride ion concentration sensors are installed in the isolation layer to monitor the impact of environmental parameters on the isolation system.

4. The intelligent detection and evaluation system for the entire life cycle of seismic isolation and damping structures according to claim 1, characterized in that, The multi-source sensor array performs self-calibration in the following manner: Each set of vibration isolation and damping structures has 3 built-in redundant sensors, which achieve in-situ self-calibration through mutual calibration. When a sensor fails, it automatically switches to the redundant sensor to ensure continuous system operation. A calibration procedure is performed once a month, with the calibration accuracy set to ±2%.

5. The intelligent detection and evaluation system for the entire life cycle of seismic isolation and damping structures according to claim 1, characterized in that, The multi-source data were denoised using wavelet transform and Kalman filtering; missing values ​​were filled in using interpolation; and the data were standardized using the Min-Max standardization method.

6. The intelligent detection and evaluation system for the entire life cycle of seismic isolation and damping structures according to claim 1, characterized in that, The preliminary damage assessment method for the damping layer is as follows: 1) Real-time threshold pre-screening: The dynamic threshold method is adopted, and a sliding window is used to calculate the statistical threshold of each parameter to detect sudden anomalies; 2) Lightweight Machine Learning for Damage Recognition: Extracting lightweight, damage-sensitive features from pre-processed multi-source data. Time-domain characteristics: peak value, root mean square value, kurtosis, margin factor, waveform factor Frequency domain characteristics: dominant frequency offset, 1st to 3rd octave amplitude ratio, and bandwidth energy proportion. Simple time-frequency characteristics: low-frequency energy from 3-layer wavelet packet decomposition; A pre-trained lightweight gradient boosting decision tree model is used to output the damage probability and the initial damage type, which includes: rubber aging, interface peeling, steel plate yielding, and connection loosening. 3) Multi-support, multi-source evidence fusion An improved DS evidence theory was used to integrate the evaluation results of all smart bearings within the same seismic isolation layer; An evidence weighting coefficient is introduced to dynamically adjust the credibility of each support's evidence based on the sensor's self-calibration results; Output the overall damage status and initial damage area of ​​the vibration isolation and damping layer.

7. The intelligent detection and evaluation system for the entire life cycle of seismic isolation and damping structures according to claim 1, characterized in that, Following the initial damage assessment, the system is linked with the vibration isolation and damping device, employing a phased linkage control strategy, as detailed below: Normal operating conditions: Predictive performance optimization, the edge gateway performs a health assessment of the damping layer every hour, and fine-tunes the basic damping parameters of the semi-active damping device based on ambient temperature, humidity and long-term performance degradation trends; the cloud platform verifies and optimizes the edge-side control strategy once a day and updates the control parameter library. Early warning level: Local damage suppression. When one or a few supports show early signs of damage, the edge gateway immediately increases the damping force of the damper near the damaged support, transferring the energy of the damaged area to the healthy area. At the same time, the sampling frequency of the damaged support and its adjacent supports was increased from 10Hz to 100Hz, and the abnormal data was uploaded to the cloud for high-precision analysis. The cloud platform generates a detailed damage assessment report and provides further control recommendations. Emergency Level: Full system coordinated control. When significant damage occurs to the damping layer, the edge gateway immediately activates the full damping layer coordinated control algorithm. For semi-active seismic isolation systems: Magnetorheological damper: The current is instantly adjusted to its maximum value to provide maximum damping force. Variable friction damper: Increases normal force and improves friction energy dissipation capacity. Active tuning mass damper: Adjusts the frequency to the dominant frequency of the isolation layer to maximize energy absorption effect; For active seismic isolation systems: Active control forces are activated to counteract the inertial forces of the superstructure. Model predictive control algorithms are used to optimize the magnitude and direction of control force in real time; Simultaneously, a local audible and visual alarm is triggered, and an emergency signal is sent to the cloud platform and the building management center. Hazard Level: Emergency Safety Protection. When the damping layer is severely damaged, the edge gateway will immediately activate emergency braking control. Lock the horizontal displacement of the seismic isolation layer to prevent excessive swaying or even overturning of the superstructure; cut off unnecessary power supply in the building, activate emergency lighting and evacuation broadcasts; send distress signals and building location information to fire, emergency and other departments.

8. The intelligent detection and evaluation system for the entire life cycle of seismic isolation and damping structures according to claim 1, characterized in that, The multiphysics coupled damage evolution model includes the following three-level model structure: First layer: Microscopic material-level multiphysics coupled damage constitutive model For rubber materials, the aging-fatigue-stress coupled damage model adopts the improved Arrhenius-fatigue coupled damage constitutive model. Aging damage considers the coupling effect of temperature, humidity, ultraviolet radiation and static stress; fatigue damage considers the improved Miner rule that takes into account the influence of mean stress and temperature. Steel plate material: The corrosion-fatigue-stress coupled damage model adopts the electrochemical-mechanical coupled damage model; Corrosion damage: The effects of humidity, chloride ion concentration and stress are considered; Interface layer: The bond-peel-environment coupled damage model adopts the cohesion-damage coupling model, which considers the influence of temperature and humidity on the interfacial bond strength; Second layer: Mesoscopic component-level performance degradation model The refined reduced-order mechanical model of the vibration isolation and damping bearings, based on the POD-Galerkin method, reduces the order of the three-dimensional finite element model, decreasing the degrees of freedom from 10^5 to 10^2, thus improving computational efficiency by more than 1000 times. Model input: Micromaterial damage parameters Model outputs: horizontal stiffness K_h, vertical stiffness K_v, equivalent damping ratio ξ, and ultimate bearing capacity F_u of the support; Third layer: Macro-system level damage evolution and assessment layer The overall finite element model of the seismic isolation system uses a lumped mass model or a rod system model for the superstructure; the seismic isolation and damping bearings use a second-layer mesoscopic reduced-order model; the foundation uses a rigid foundation or an elastic half-space model; the model updates the component performance parameters every 24 hours based on monitoring data. A multi-support collaborative damage evolution model considers the interactions between various supports within the damping layer: When damage to a support leads to a decrease in stiffness, the load on adjacent supports will increase, accelerating the damage to those supports; the model uses a damage transfer matrix to describe this synergistic effect.

9. The intelligent detection and evaluation system for the entire life cycle of seismic isolation and damping structures according to claim 1, characterized in that, The application and implementation process of the multiphysics coupled damage evolution model is as follows: 1) Quantitative assessment and localization of damage Damage is identified by inverting the model-measured response residual minimization method; Input: Displacement, acceleration, and strain monitoring data for each support; Output: Damage extent and damage type for each support; 2) Remaining useful life prediction a. Based on historical cumulative damage, perform model calibration to obtain the current damage state; Input: All monitoring data from the structure's construction to the current moment; Run the multiphysics coupled damage evolution model and use the Bayesian inference algorithm to invert and update the model parameters; Output: The precise damage state vector of each seismic isolation bearing at the current moment; b. Generate a random load sample set for the next 30 years, based on local meteorological statistics for the past 30 years, and generate time series samples of temperature, humidity, and ultraviolet radiation. Based on the results of local seismic hazard analysis, a sample of possible ground motions over the next 30 years is generated, including the time of occurrence, magnitude, and spectral characteristics. Based on the building's usage function, time series samples of usage load and wind load are generated; The Monte Carlo method was used to generate 10,000 independent future load samples, covering all possible extreme cases; c. Starting from the current damage state, simulate the future damage evolution. For each set of future load samples, use D_current as the initial condition. Run a multiphysics coupled damage evolution model to simulate damage development over the next 30 years over time. When any damage indicator reaches the failure threshold, record that moment as the failure time. d. Statistical analysis was performed to obtain the remaining lifetime at different confidence levels. Statistical analysis was conducted on the failure times obtained from 10,000 sets of simulations to obtain the probability distribution of failure times. Calculate the remaining useful life at different confidence levels based on the probability distribution: 90% confidence level: The time during which the structure will not fail with a 90% probability. 95% confidence level: The time during which the structure will not fail with a 95% probability. 99% confidence level: The time during which the structure will not fail with a 99% probability; 3) Recommendations for post-earthquake damage assessment and repair strategies Immediately after the earthquake, the model is updated with seismic records and monitoring data; the post-earthquake damage level and remaining bearing capacity of each bearing are calculated; the effects of different repair schemes, such as replacing bearings, reinforcing bearings, and adding dampers, are simulated; the optimal repair strategy is given by comprehensively considering safety, economy, and construction difficulty; and detailed repair schemes and cost estimates are generated.

Images