Intelligent building measuring system

Abstract

The invention relates to the field of intelligent structure monitoring, and discloses an intelligent building measurement system, which comprises the steps of collecting structure stress, deformation, micro-seismic response, material attenuation and environmental load curves of a building in different life cycle stages, generating a real-time structure degradation fingerprint vector, and calculating a real-time structure degradation fingerprint vector; the method comprises the following steps: identifying load, vibration, temperature anomaly and basic displacement from an extreme environment and an operation scene library, generating a local response sequence, combining the local response sequence with a real-time structure degradation fingerprint vector, generating a key load transmission topology, selecting a key path and nodes to form a local response event sequence, calculating structure offset and degradation, and generating a key load transmission topology. And generating an adaptive correction structure index set, constructing a long-period degradation track and performing trend prediction, and when a prediction index exceeds a safety threshold, generating a structure safety intervention scheme, a dynamic measurement optimization scheme and a data restoration scheme. The method has the advantage of improving the accuracy and safety of building structure state evaluation.

Description

Technical Field

[0001] This invention relates to the field of intelligent structure monitoring, specifically to an intelligent building measurement system. Background Technology

[0002] With the increasing scale and complexity of modern buildings, the requirements for building structural safety monitoring are becoming more stringent. Existing intelligent building measurement systems primarily rely on real-time collected structural data to monitor stress, deformation, and environmental load conditions. However, these systems suffer from several problems: First, when buildings encounter extreme environmental events such as strong winds, earthquakes, or overload operations, the system can only provide current response information and cannot effectively predict potential degradation risks and remaining service life. Second, in the event of sensor failure or data loss, analysis based on partial observation data can easily lead to overall decision-making biases, reducing the accuracy of structural safety assessments. Third, the system cannot comprehensively analyze and compare real-time observations with historical structural evolution patterns at different lifecycle stages, making it difficult to predict potential weaknesses and degradation trends in key components. Existing systems have significant limitations in long-term trend prediction, anomaly identification, and comprehensive decision-making capabilities. Therefore, it is essential to design an intelligent building measurement system that improves the accuracy and safety of building structural condition assessments. Summary of the Invention

[0003] To address the shortcomings of existing technologies, this invention provides an intelligent building measurement system that improves the accuracy and safety of building structural condition assessment, thus solving the problems mentioned in the background section.

[0004] To achieve the aforementioned goal of improving the accuracy and safety of building structural condition assessment, this invention provides the following technical solution: an intelligent building measurement system, comprising: Data Granularity Module: Collects structural stress, deformation, microseismic response, material attenuation and environmental load curves of buildings at different life stages, and generates real-time structural degradation fingerprint vectors through multi-scale evolution comparison and time continuity verification. Feature encoding module: Based on real-time structural degradation fingerprint vectors, it identifies load, vibration, temperature anomalies and foundation displacement from extreme environment and operation scenario library, detects sensor failure, data loss and noise interference, performs dynamic measurement optimization and data repair, and generates local response sequences; Anomaly identification module: Combines local response sequences with real-time structural degradation fingerprint vectors to analyze key stress channels in the building, select key paths and nodes, and form local response event sequences; Feedback Iteration Module: Based on the local response event sequence and real-time measurement data, it calculates structural offset and degradation, dynamically adjusts sensor weights and key indicator thresholds, and performs reverse calibration when the offset or degradation index exceeds the threshold to generate an adaptive correction set of structural indicators. Global prediction module: Based on continuous multi-round real-time structural degradation fingerprints and adaptive correction of structural index sets, it constructs long-cycle degradation trajectories and performs trend prediction. Based on the changes in degradation trends, it generates structural safety intervention, dynamic measurement optimization and data repair schemes.

[0005] Preferably, the process of generating real-time structural degradation fingerprint vectors is as follows: The stress, deformation, microseismic response, and environmental load data collected by building structure sensors during the construction, use, and maintenance periods are processed into a time series. Perform integrity checks on data for each time period, including sampling interval consistency checks, missing data marking, and outlier identification; Spatial scale decomposition and frequency domain analysis are performed on various structural response data to calculate local stress concentration, microseismic frequency characteristics and dynamic response modes. At the same time, key structural behavior characteristics are extracted by combining material attenuation parameters and environmental load change rate. By using a multi-scale comparison algorithm, the characteristics of the same component at different life cycle stages are compared horizontally and vertically to identify stable behavior patterns and degradation trends. The results are then integrated according to time series to generate a real-time structural degradation fingerprint vector.

[0006] Preferably, the process of identifying load, vibration, temperature anomalies, and foundation displacement from the extreme environment and operating scenario library is as follows: Real-time structural degradation fingerprint vectors are mapped to a database of extreme building environments and a library of historical operating scenarios. Based on multidimensional feature distance measurement and weighted similarity calculation, the types of loads, vibration events, temperature anomalies and foundation displacements that the current structure may bear are identified. Based on the matching results, structural response prediction sequences for various scenarios are generated.

[0007] Preferably, the process of generating the local response sequence is as follows: Perform integrity verification and outlier detection on the collected real-time sensor data; By combining the matching results of various scenarios and the structural response prediction sequence, interpolation, filtering and weighted correction are performed on missing or abnormal data; Based on the corrected data and the predicted sequence, a local structural response sequence is generated according to the event window. Key features are weighted and encoded according to load intensity, frequency response, and environmental sensitivity to form a local response sequence.

[0008] Preferably, the process of analyzing the key stress channels of a building is as follows: The local response sequences are aggregated at multiple scales according to spatial location and time window, and mapped to building structural units and key component locations; By combining the scene response prediction sequence corresponding to each event and the evolution trend of the real-time structural degradation fingerprint vector, the main load transfer paths and nodes are identified; For each stress channel, key indicators are calculated, including stress change rate, local stress concentration, vibration amplitude, material attenuation rate, and environmental sensitivity coefficient.

[0009] Preferably, the process of forming a local response event sequence is as follows: Based on the analysis results of the key indicators of the critical stress channels, the key nodes and load paths of each channel are prioritized. Consider the rate of change of force, local stress concentration, material attenuation rate, and coupling strength between channels; For high-risk and high-degradation regions in the sorting results, extract the corresponding local response events within the time window; Define the triggering conditions, load type, response amplitude, and duration of each event, and integrate them according to time sequence and spatial location to generate a sequence of local structural response events.

[0010] Preferably, the process of calculating structural offset and degradation is as follows: Based on the local response event sequence and real-time measurement data, the event sequence is aligned and matched with the acquired sensor data in time and space; For each critical node, the structural offset, stress concentration variation, and degradation rate are calculated based on the local response amplitude, event duration, and historical degradation trend. By combining multi-scale spatial interpolation algorithms, node degradation indices are mapped to surrounding components to form a continuous local degradation distribution map; Multi-scale analysis methods are used to integrate the local degradation distribution, assess the global degradation status of the overall structure, and generate a structural health parameter matrix.

[0011] Preferably, the process of generating the adaptive correction structure index set is as follows: The calculated structural offsets, degradation rates, and local and global degradation states of each key node are mapped to the structural health parameter matrix. Based on the matrix analysis results, the weights of each sensor and the thresholds of relevant key indicators are adjusted in real time. When the comprehensive degradation index exceeds the preset threshold, the reverse calibration algorithm is invoked to correct the sensor data and local response sequence of the abnormal node, update the structural health parameter matrix, and generate an adaptive correction structural index set.

[0012] Preferably, the process of constructing long-term degradation trajectories and predicting trends is as follows: The adaptive correction set of structural indicators is integrated with the multi-round real-time structural degradation fingerprint vectors according to the time series. Long-term degradation trajectories were constructed using multi-scale regression and trend analysis methods. Based on the weight adjustment of the adaptive correction structural index set and the node degradation mapping, the future structural state evolution trend is predicted, potential weaknesses and degradation rate changes are identified, and quantifiable prediction results are generated.

[0013] Preferably, the process of generating structural safety intervention, dynamic measurement optimization, and data repair solutions based on changes in degradation trends is as follows: By comparing long-term degradation trajectories and quantifiable prediction results with preset safety thresholds and historical degradation benchmarks, out-of-limit components, critical nodes, and potentially high-risk areas can be identified. For the identified areas exceeding the limits, multi-dimensional weighted analysis is used to generate structural safety intervention strategies, including local reinforcement schemes, load adjustment suggestions, and vibration control measures; Based on the predicted degradation trend and sensor coverage, the measurement strategy is dynamically optimized, and the sensor layout, sampling frequency and monitoring priority of key indicators are adjusted. For nodes with missing or abnormal data, a data repair path is planned by combining degradation prediction and the response of neighboring components, and a complete corrected dataset is generated that can be executed by the system. By integrating safety intervention strategies, dynamic measurement optimization schemes, and data repair paths, a structural safety management and maintenance implementation plan is formed.

[0014] Compared with existing technologies, the present invention provides an intelligent building measurement system, which has the following advantages: This invention collects structural stress, deformation, microseismic response, material attenuation, and environmental load curves at different stages of a building's life cycle. Combined with multi-scale evolution comparison and temporal continuity verification, it generates a real-time structural degradation fingerprint vector, providing a reliable foundation for subsequent analysis. Based on this fingerprint vector, the system can identify load, vibration, temperature anomalies, and foundation displacement from extreme environments and operational scenario libraries. Simultaneously, it detects sensor failures, data loss, and noise interference, performs dynamic measurement optimization and data repair, and forms a local response sequence. Combining this local response sequence with the degradation fingerprint allows analysis of key stress channels in the building, generating a key load transfer topology, and selecting key paths and nodes to form a local response event sequence. This enables accurate anomaly identification in key areas, based on the event sequence and real-time measurement data. The system dynamically calculates structural offset and degradation, adjusts sensor weights and key indicator thresholds, and performs reverse calibration when offset or degradation exceeds the threshold, generating an adaptive correction set of structural indicators to continuously optimize the monitoring closed loop. Ultimately, it constructs a long-cycle degradation trajectory and performs trend prediction through multiple rounds of degradation fingerprints and adaptive indicator sets. When the predicted indicators exceed the safety threshold, the system can promptly generate structural safety intervention plans, dynamic measurement optimization, and data repair strategies, achieving continuous closed-loop monitoring from local anomaly identification to global trend prediction. Its innovation lies in granularizing multi-lifecycle data to generate degradation fingerprints, deeply coupling local response event sequences with key load topology, and an adaptive feedback iterative closed-loop mechanism, effectively improving the accuracy, timeliness, and operability of structural health monitoring. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the structure of the present invention. Detailed Implementation

[0016] 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.

[0017] Example 1: Please refer to Figure 1 As shown in the embodiment of the present invention, the intelligent building measurement system includes: Data Granularity Module: Collects structural stress, deformation, microseismic response, material attenuation and environmental load curves of buildings at different life stages, and generates real-time structural degradation fingerprint vectors through multi-scale evolution comparison and time continuity verification. The process of generating real-time structural degradation fingerprint vectors in the data granularity module is as follows: The stress, deformation, microseismic response, and environmental load data collected by building structure sensors during the construction, use, and maintenance periods are processed into a time series. The stress, deformation, microseismic response, and environmental load data collected by sensors deployed during the construction, use, and maintenance phases of the building structure are sorted according to the collection timestamp, a unified timeline is established, different types of sensor data are converted into a unified data format and unit, the completeness of data records is checked, the sampling interval is ensured to be consistent, and the processed data is stored in a database or local buffer to provide a unified interface for subsequent processing.

[0018] Perform integrity checks on data for each time period, including sampling interval consistency checks, missing data marking, and outlier identification; Perform a sampling interval consistency check on the processed time series data, check whether consecutive sampling points meet the preset sampling frequency, mark missing points, and identify outliers using sliding window analysis or statistical methods.

[0019] Spatial scale decomposition and frequency domain analysis are performed on various structural response data to calculate local stress concentration, microseismic frequency characteristics and dynamic response modes. At the same time, key structural behavior characteristics are extracted by combining material attenuation parameters and environmental load change rate. The structural response data is spatially divided according to components or nodes. Using mesh generation or node mapping methods, stress and deformation data are mapped to the surface or internal elements of the components. Fourier transform or wavelet transform is performed on the microseismic response signal to calculate the frequency characteristics and response amplitude distribution. At the same time, the material attenuation coefficient is calculated by combining the material type and historical experimental parameters. The environmental load data is converted into gradient change rate to form the local structural feature vector of each node or component.

[0020] Using a multi-scale comparison algorithm, the characteristics of the same component at different life stages are compared horizontally and vertically to identify stable behavior patterns and degradation trends. The results are then integrated according to time series to generate a real-time structural degradation fingerprint vector. The local response characteristics of the same component at different life cycle stages are compared horizontally and vertically. A multi-scale comparison algorithm is used to calculate the response mode stability index of nodes or components, extract local stress concentration changes, microseismic frequency evolution and material performance degradation trends, integrate the characteristics of each stage in chronological order to generate a continuously traceable structural behavior feature table, and vectorize the integrated time series structural behavior features, including local stress, microseismic frequency, deformation and environmental load evolution information, and arrange them in chronological order to form a continuous real-time structural degradation fingerprint vector.

[0021] Feature encoding module: Based on real-time structural degradation fingerprint vectors, it identifies load, vibration, temperature anomalies and foundation displacement from extreme environment and operation scenario library, detects sensor failure, data loss and noise interference, performs dynamic measurement optimization and data repair, and generates local response sequences; The process of identifying load, vibration, temperature anomalies, and foundation displacement from the extreme environment and operating scenario library in the feature encoding module is as follows: Real-time structural degradation fingerprint vectors are mapped to a database of extreme building environments and a library of historical operating scenarios. The generated real-time structural degradation fingerprint vectors are mapped to the building extreme environment database and historical operation scenario database according to the correspondence between nodes and components. A corresponding index is established between the fingerprint vectors and various environmental and operation condition features in the database to ensure that the stress, deformation, microseismic response and environmental load information of each node are consistent with the same data structure in the database.

[0022] Based on multidimensional feature distance measurement and weighted similarity calculation, the types of loads, vibration events, temperature anomalies and foundation displacements that the current structure may bear are identified. Multidimensional feature distance calculation is performed between the mapped fingerprint vector and the scene features in the database. Methods such as Euclidean distance, Manhattan distance or weighted distance are used. Combined with key indicators such as stress amplitude, micro-seismic frequency, deformation and environmental change rate, weighted similarity is calculated. Thresholds are used to determine the load type, possible vibration events, temperature anomaly range and foundation displacement of the current structure, and scene matching results at the node or component level are generated.

[0023] Generate structural response prediction sequences for various scenarios based on the matching results; Based on the matched extreme environments or historical operating scenarios, the structural response patterns under various loads, vibrations, and temperature anomalies are extracted from the database, and node and component-level structural response prediction sequences are generated according to the time step, including stress changes, deformation amplitudes, and microseismic frequency evolution information. At the same time, each sequence is weighted according to load intensity, frequency response, and environmental sensitivity to form a structural response prediction sequence.

[0024] The process of generating local response sequences in the feature encoding module is as follows: Perform integrity verification and outlier detection on the collected real-time sensor data; The stress, deformation, microseismic response, and environmental load data collected by building structure sensors during the construction, use, and maintenance periods are processed into a time series according to a preset sampling frequency. The data for each time period are subjected to integrity verification, including checking whether the sampling interval is consistent, marking missing data, removing duplicate data, and identifying outliers. Outlier identification is performed by calculating the mean and standard deviation of data at nearby times to detect deviations, and data that deviate from the threshold are labeled.

[0025] By combining the matching results of various scenarios and the structural response prediction sequence, interpolation, filtering and weighted correction are performed on missing or abnormal data; The raw data from each sensor is aligned with the predicted structural response sequences under matched extreme environments and historical operating scenarios. Missing data is filled using linear interpolation or local polynomial interpolation based on time windows. Abnormal data is corrected using low-pass filters or adaptive weighted filters. The filter weights are allocated according to the historical stability and environmental sensitivity of the sensors to generate a continuous and smooth signal sequence.

[0026] Based on the corrected data and the predicted sequence, a local structural response sequence is generated according to the event window. The corrected sensor data is synchronized with the structural response prediction sequence and divided according to the time window defined by the key events. Each window covers possible load changes or microseismic events. The data in each window is spatially aggregated according to the node location and component distribution to form local response units. The responses of different components and nodes are integrated into the same event window to generate a continuous local structural response sequence.

[0027] Key features are weighted and encoded according to load intensity, frequency response, and environmental sensitivity to form a local response sequence; For each local response unit's key features, such as peak stress, deformation amplitude, vibration frequency, and microseismic response mode, weights are assigned according to load intensity, frequency response characteristics, and sensitivity to environmental factors. The weighted features are then encoded into vectors or matrices in a uniform format and arranged in chronological order of events to generate a local response sequence.

[0028] Anomaly identification module: Combines local response sequences with real-time structural degradation fingerprint vectors to analyze key stress channels in the building, generate key load transfer topologies, select key paths and nodes, and form local response event sequences; The process of analyzing key stress channels in a building in the anomaly identification module is as follows: The local response sequences are aggregated at multiple scales according to spatial location and time window, and mapped to building structural units and key component locations; The generated local response sequences are grouped according to the location of the acquisition sensors and the time window. First, the responses of adjacent nodes are spatially weighted and averaged within the micro-area to calculate the comprehensive response value within each structural unit. Then, the responses of each micro-area are hierarchically aggregated at a larger spatial scale to form multi-scale response data covering the entire component or structural unit.

[0029] By combining the scene response prediction sequence corresponding to each event and the evolution trend of the real-time structural degradation fingerprint vector, the main load transfer paths and nodes are identified; By combining the scene response prediction sequence corresponding to each event, the temporal evolution characteristics of the local response sequence are compared with the prediction sequence to identify the possible load types and vibration modes. At the same time, the real-time structural degradation fingerprint vector is mapped to each node according to the time sequence, the node degradation trend is calculated, and the main load transmission paths and key node locations are identified by using node response intensity, load direction, connectivity and degradation trend information through graph structure analysis or path search methods, forming the preliminary identification results of the building's key stress channels.

[0030] For each stress channel, key indicators are calculated, including stress change rate, local stress concentration, vibration amplitude, material attenuation rate, and environmental sensitivity coefficient. For each identified stress channel, the response characteristics of each node on the channel are extracted, including the rate of change of stress amplitude, the degree of local stress concentration, vibration amplitude, material attenuation rate and sensitivity coefficient under corresponding environmental conditions. The characteristics in the channel are integrated by spatial and temporal weighting to generate a key index matrix for each stress channel.

[0031] The process of forming a local response event sequence in the anomaly detection module is as follows: Based on the analysis results of the key indicators of the critical stress channels, the key nodes and load paths of each channel are prioritized. Based on the key indicators of the critical stress channels obtained from the previous analysis, each node and load path within the channel is quantitatively scored according to the rate of change of stress, the degree of local stress concentration, the material attenuation rate, and the channel coupling strength, generating a multi-dimensional feature vector.

[0032] Consider the rate of change of force, local stress concentration, material attenuation rate, and coupling strength between channels; In the priority ranking results, a threshold is set to filter out nodes and paths that experience drastic stress changes, high stress concentration, large degradation rate, and strong coupling with other channels. For these selected areas, micro-regions are divided according to spatial coordinates and structural units to ensure that the selected areas cover potential critical stress-bearing parts in the structure.

[0033] For high-risk and high-degradation regions in the sorting results, extract the corresponding local response events within the time window; For the selected high-risk and high-degradation areas, response events are extracted from the local response sequence by time window. Each event is recorded by node or component, including the start and end time, response amplitude, load type, frequency characteristics, and environmental conditions. Combined with the time evolution characteristics of the local response sequence, a preliminary dataset of structural local response events is formed.

[0034] Define the triggering conditions, load type, response amplitude, and duration of each event, and integrate them according to time series and spatial location to generate a sequence of local structural response events; The extracted local response events are arranged in a time sequence and mapped to spatial locations, so that each event has clear triggering conditions, load type, response amplitude and duration information. The events are then encoded to form a unified format of structural local response event sequence.

[0035] Feedback Iteration Module: Based on the local response event sequence and real-time measurement data, it calculates structural offset and degradation, dynamically adjusts sensor weights and key indicator thresholds, and performs reverse calibration when the offset or degradation index exceeds the threshold to generate an adaptive correction set of structural indicators. The process of calculating structural offset and degradation in the feedback iteration module is as follows: Based on the local response event sequence and real-time measurement data, the event sequence is aligned and matched with the acquired sensor data in time and space; Based on the established sequence of local response events, each event is matched temporally with the timestamp of the sensor data, and spatially associated with the corresponding structural node or component location. The sensor signals are interpolated and resampled to ensure that the start and end times of the events correspond to continuous and synchronous sensor data points.

[0036] For each critical node, the structural offset, stress concentration variation, and degradation rate are calculated based on the local response amplitude, event duration, and historical degradation trend. For each critical node, the instantaneous structural offset and stress concentration changes of the node are calculated by utilizing the event response amplitude, duration, and historical degradation trend of the node. The degradation rate is estimated by numerical differentiation method, and the cumulative effect of continuous events is processed by time weighting and exponential smoothing method, so that the degradation evolution of the node is quantified into a numerical index that can be used for matrix construction.

[0037] By combining multi-scale spatial interpolation algorithms, node degradation indices are mapped to surrounding components to form a continuous local degradation distribution map; By combining multi-scale spatial interpolation algorithms, the degradation index of each key node is mapped to surrounding components and adjacent nodes according to its spatial location in the structure, generating a continuous local degradation distribution map. In the interpolation process, the geometric information of structural units and the topological relationship of node neighborhoods are used to extend the discrete degradation index to the surface of components and adjacent nodes, forming a continuous distribution to reflect the local degradation state.

[0038] Multi-scale analysis methods are used to integrate the local degradation distribution, assess the global degradation status of the overall structure, and generate a structural health parameter matrix. Using a multi-scale analysis method, the local degradation distribution map is weighted and integrated at the component scale, unit scale, and overall structural scale. Local and global degradation state indices are calculated, and the offsets, stress concentration changes, and degradation rates of each node and component are arranged according to structural location to construct a structural health parameter matrix.

[0039] The process of generating the adaptive correction structure index set in the feedback iteration module is as follows: The calculated structural offsets, degradation rates, and local and global degradation states of each key node are mapped to the structural health parameter matrix. The structural offsets, degradation rates, and local and global degradation states of each key node are mapped to a structural health parameter matrix according to the spatial location of the node in the building structure. The rows and columns of the matrix correspond to the node number and component number. Each element records the node's offset, stress concentration, degradation rate, and environmental load response. Interpolation is performed using the geometric topological relationship between the node and the component during the mapping process.

[0040] Based on the matrix analysis results, the weights of each sensor and the thresholds of relevant key indicators are adjusted in real time. Based on the structural health parameter matrix, the importance of each sensor node or component is quantified. Initial weights are set according to offset, degradation rate, and environmental sensitivity. The offset and degradation distribution in the matrix are analyzed in real time. The sensor weights and key indicator thresholds are adjusted linearly or nonlinearly to make the weights proportional to the node degradation state, load response, and data reliability. When the comprehensive degradation index exceeds the preset threshold, the reverse calibration algorithm is invoked to correct the sensor data and local response sequence of the abnormal node, update the structural health parameter matrix, and generate an adaptive correction structural index set. When the comprehensive degradation index of a node or region in the structural health parameter matrix exceeds the preset threshold, the reverse calibration algorithm is invoked to correct the sensor data of the corresponding node, including interpolating missing data, filtering out abnormal signals, and correcting the local response sequence according to weights. After calibration, the node offset, degradation rate, and local and global degradation state are recalculated, and the structural health parameter matrix is ​​updated. An adaptive correction set of structural indicators is generated through matrix iterative updates.

[0041] Global prediction module: Based on continuous multi-round real-time structural degradation fingerprints and adaptive correction of structural index sets, it constructs long-cycle degradation trajectories and performs trend prediction. Based on the changes in degradation trends, it generates structural safety intervention, dynamic measurement optimization and data repair schemes.

[0042] The process of constructing long-term degradation trajectories and performing trend prediction in the global prediction module is as follows: The adaptive correction set of structural indicators is integrated with the multi-round real-time structural degradation fingerprint vectors according to the time series. The offsets, degradation rates, and corrected local and global degradation states of each key node in the adaptive correction structural index set are aligned with the real-time structural degradation fingerprint vectors collected in multiple rounds in chronological order to form a unified time series dataset. Data from different collection periods are interpolated or time-aligned to ensure the data integrity of each node at each time point and to maintain the consistency of the matrix structure. Each row of the integrated sequence represents the state of a node or component at different time points, and each column represents key parameters such as structural offset, stress concentration, and degradation rate.

[0043] Long-term degradation trajectories were constructed using multi-scale regression and trend analysis methods. Using a multi-scale regression algorithm, the integrated time series data is analyzed hierarchically, including node-level, component-level, and region-level degradation indices. For the data at each scale, linear regression or weighted nonlinear regression is used to fit the historical trend, and the long-term offset change rate, cumulative degradation rate, and stress concentration evolution are calculated. The results of each scale are fused to generate a complete long-period degradation trajectory matrix, and each trajectory records the degradation state changes of nodes or components over time.

[0044] Based on the weight adjustment and node degradation mapping of the adaptive correction structural index set, the future structural state evolution trend is predicted, potential weaknesses and degradation rate changes are identified, and quantifiable prediction results are generated. Based on the node weight information in the long-cycle degradation trajectory and adaptive correction structural index set, historical trends are mapped to future time steps. The future offset, degradation rate and local and global degradation state of each node or component are calculated iteratively step by step. A multi-scale prediction method is adopted, which combines the coupling relationship between nodes and the environmental sensitivity coefficient to generate the future structural state evolution trend. During the prediction process, potential weak nodes and abnormal changes in degradation rate are marked, forming a prediction result matrix that can be quantified by node, component and time dimensions.

[0045] The process of generating structural safety intervention, dynamic measurement optimization, and data repair schemes based on changes in degradation trends in the global prediction module is as follows: By comparing long-term degradation trajectories and quantifiable prediction results with preset safety thresholds and historical degradation benchmarks, out-of-limit components, critical nodes, and potentially high-risk areas can be identified. The long-term degradation trajectory and future structural state prediction results are compared and analyzed node by node with the preset safety threshold and historical degradation benchmark. The degradation index and threshold difference are calculated for each component, and the over-limit state is marked according to the spatial location and importance level of the node or component. Through load transfer relationship between components and node coupling analysis, a list of potential high-risk areas is generated.

[0046] For the identified areas exceeding the limits, multi-dimensional weighted analysis is used to generate structural safety intervention strategies, including local reinforcement schemes, load adjustment suggestions, and vibration control measures; For identified oversized components and high-risk areas, intervention plans are generated using weighted analysis methods, taking into account load type, local stress concentration, and vibration amplitude data.

[0047] Based on the predicted degradation trend and sensor coverage, the measurement strategy is dynamically optimized, and the sensor layout, sampling frequency and monitoring priority of key indicators are adjusted. Based on the predicted degradation trend and the current sensor coverage, the sensor deployment is analyzed to determine the monitoring priority of key nodes and high-risk areas, and the sampling frequency and key indicator monitoring plan are adjusted.

[0048] For nodes with missing or abnormal data, a data repair path is planned by combining degradation prediction and the response of neighboring components, and a complete corrected dataset is generated that can be executed by the system. For nodes with missing or abnormal data, interpolation and path planning are performed using predicted degradation trends and response patterns of neighboring components. Data repair sequences are generated according to the spatial location and coupling relationship of nodes, including repair order, algorithm selection, weighting parameters and time windows. The repaired data structure is then merged with the original measurement data to form a complete dataset.

[0049] Integrate safety intervention strategies, dynamic measurement optimization schemes, and data repair paths to form a structural safety management and maintenance execution plan; The intervention strategy, dynamic measurement optimization plan, and data repair path are integrated according to nodes, components, and time sequence to form a unified execution plan data structure, including the execution conditions, target nodes, parameter settings, and execution time of each measure.

[0050] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0051] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An intelligent building measurement system, characterized in that, include: Data Granularity Module: Collects structural stress, deformation, microseismic response, material attenuation and environmental load curves of buildings at different life stages, and generates real-time structural degradation fingerprint vectors through multi-scale evolution comparison and time continuity verification. Feature encoding module: Based on real-time structural degradation fingerprint vectors, it identifies load, vibration, temperature anomalies and foundation displacement from extreme environment and operation scenario library, detects sensor failure, data loss and noise interference, performs dynamic measurement optimization and data repair, and generates local response sequences; Anomaly identification module: Combines local response sequences with real-time structural degradation fingerprint vectors to analyze key stress channels in the building, select key paths and nodes, and form local response event sequences; Feedback Iteration Module: Based on the local response event sequence and real-time measurement data, it calculates structural offset and degradation, dynamically adjusts sensor weights and key indicator thresholds, and performs reverse calibration when the offset or degradation index exceeds the threshold to generate an adaptive correction set of structural indicators. Global prediction module: Based on continuous multi-round real-time structural degradation fingerprints and adaptive correction of structural index sets, it constructs long-cycle degradation trajectories and performs trend prediction. Based on the changes in degradation trends, it generates structural safety intervention, dynamic measurement optimization and data repair schemes.

2. The intelligent building measurement system according to claim 1, characterized in that, The process of generating real-time structural degradation fingerprint vectors is as follows: The stress, deformation, microseismic response, and environmental load data collected by building structure sensors during the construction, use, and maintenance periods are processed into a time series. Perform integrity checks on data for each time period, including sampling interval consistency checks, missing data marking, and outlier identification; Spatial scale decomposition and frequency domain analysis are performed on various structural response data to calculate local stress concentration, microseismic frequency characteristics and dynamic response modes. At the same time, key structural behavior characteristics are extracted by combining material attenuation parameters and environmental load change rate. By using a multi-scale comparison algorithm, the characteristics of the same component at different life cycle stages are compared horizontally and vertically to identify stable behavior patterns and degradation trends. The results are then integrated according to time series to generate a real-time structural degradation fingerprint vector.

3. The intelligent building measurement system according to claim 2, characterized in that, The process of identifying load, vibration, temperature anomalies, and foundation displacement from a database of extreme environments and operational scenarios is as follows: Real-time structural degradation fingerprint vectors are mapped to a database of extreme building environments and a library of historical operating scenarios. Based on multidimensional feature distance measurement and weighted similarity calculation, the types of loads, vibration events, temperature anomalies and foundation displacements that the current structure may bear are identified. Based on the matching results, structural response prediction sequences for various scenarios are generated.

4. The intelligent building measurement system according to claim 3, characterized in that, The process of generating a local response sequence is as follows: Perform integrity verification and outlier detection on the collected real-time sensor data; By combining the matching results of various scenarios and the structural response prediction sequence, interpolation, filtering and weighted correction are performed on missing or abnormal data; Based on the corrected data and the predicted sequence, a local structural response sequence is generated according to the event window. Key features are weighted and encoded according to load intensity, frequency response, and environmental sensitivity to form a local response sequence.

5. The intelligent building measurement system according to claim 4, characterized in that, The process of analyzing the key stress channels of a building is as follows: The local response sequences are aggregated at multiple scales according to spatial location and time window, and mapped to building structural units and key component locations; By combining the scene response prediction sequence corresponding to each event and the evolution trend of the real-time structural degradation fingerprint vector, the main load transfer paths and nodes are identified; For each stress channel, key indicators are calculated, including stress change rate, local stress concentration, vibration amplitude, material attenuation rate, and environmental sensitivity coefficient.

6. The intelligent building measurement system according to claim 5, characterized in that, The process of forming a local response event sequence is as follows: Based on the analysis results of the key indicators of the critical stress channels, the key nodes and load paths of each channel are prioritized. Consider the rate of change of force, local stress concentration, material attenuation rate, and coupling strength between channels; For high-risk and high-degradation regions in the sorting results, extract the corresponding local response events within the time window; Define the triggering conditions, load type, response amplitude, and duration of each event, and integrate them according to time sequence and spatial location to generate a sequence of local structural response events.

7. The intelligent building measurement system according to claim 6, characterized in that, The process of calculating structural offset and degradation is as follows: Based on the local response event sequence and real-time measurement data, the event sequence is aligned and matched with the acquired sensor data in time and space; For each critical node, the structural offset, stress concentration variation, and degradation rate are calculated based on the local response amplitude, event duration, and historical degradation trend. By combining multi-scale spatial interpolation algorithms, node degradation indices are mapped to surrounding components to form a continuous local degradation distribution map; Multi-scale analysis methods are used to integrate the local degradation distribution, assess the global degradation status of the overall structure, and generate a structural health parameter matrix.

8. The intelligent building measurement system according to claim 7, characterized in that, The process of generating the adaptive correction structure index set is as follows: The calculated structural offsets, degradation rates, and local and global degradation states of each key node are mapped to the structural health parameter matrix. Based on the matrix analysis results, the weights of each sensor and the thresholds of relevant key indicators are adjusted in real time. When the comprehensive degradation index exceeds the preset threshold, the reverse calibration algorithm is invoked to correct the sensor data and local response sequence of the abnormal node, update the structural health parameter matrix, and generate an adaptive correction structural index set.

9. The intelligent building measurement system according to claim 8, characterized in that, The process of constructing long-term degradation trajectories and predicting trends is as follows: The adaptive correction set of structural indicators is integrated with the multi-round real-time structural degradation fingerprint vectors according to the time series. Long-term degradation trajectories were constructed using multi-scale regression and trend analysis methods. Based on the weight adjustment of the adaptive correction structural index set and the node degradation mapping, the future structural state evolution trend is predicted, potential weaknesses and degradation rate changes are identified, and quantifiable prediction results are generated.

10. The intelligent building measurement system according to claim 9, characterized in that, The process of generating structural safety intervention, dynamic measurement optimization, and data repair solutions based on changes in degradation trends is as follows: By comparing long-term degradation trajectories and quantifiable prediction results with preset safety thresholds and historical degradation benchmarks, out-of-limit components, critical nodes, and potentially high-risk areas can be identified. For the identified areas exceeding the limits, multi-dimensional weighted analysis is used to generate structural safety intervention strategies, including local reinforcement schemes, load adjustment suggestions, and vibration control measures; Based on the predicted degradation trend and sensor coverage, the measurement strategy is dynamically optimized, and the sensor layout, sampling frequency and monitoring priority of key indicators are adjusted. For nodes with missing or abnormal data, a data repair path is planned by combining degradation prediction and the response of neighboring components, and a complete corrected dataset is generated that can be executed by the system. By integrating safety intervention strategies, dynamic measurement optimization schemes, and data repair paths, a structural safety management and maintenance implementation plan is formed.

Images