Building monitoring data accurate registration method based on space-time dynamic weight

Abstract

The invention discloses a time-space dynamic weight-based building monitoring data accurate registration method, which comprises the following steps of: establishing a response model considering nonlinear behaviors of materials by analyzing deformation characteristics in multi-source monitoring data; constructing a dynamic weight function based on a structural stress state, and realizing self-adaptive distribution of sensor data reliability weight; geometric deformation correction is carried out by using the material nonlinear response model, and geometric distortion caused by material nonlinear behaviors is corrected; space-time registration of multi-source data is realized by adopting a weight fusion and deformation gradient constraint method; and parameter adaptive optimization is carried out through an online learning mechanism. According to the method, the technical problems of low geometric correction precision of a high-stress area, dynamic change of sensor reliability and the like are solved, and submillimeter-level precision building monitoring data registration is realized.

Description

Technical Field

[0001] The present invention relates to structural health monitoring, and in particular to a method for accurately registering building monitoring data based on spatiotemporal dynamic weights. Background Art

[0002] Accurate registration of building monitoring data is a key link in structural health monitoring and safety assessment. With the development of multi-source monitoring technologies such as satellite remote sensing, drones and ground monitoring equipment, the ability to monitor building deformation has been significantly improved, but it also brings challenges in the temporal and spatial inconsistency of data. Due to differences in acquisition cycle, location and observation angle, multi-source monitoring data produces registration errors that are difficult to eliminate. Especially in building monitoring scenarios with sub-millimeter accuracy requirements, these errors will seriously affect the reliability of analysis results. Accurate registration not only affects the accuracy of structural safety assessment, but is also directly related to post-disaster emergency decision-making, building life prediction and the formulation of maintenance and reinforcement plans, and has important economic and social value.

[0003] The current building monitoring data registration methods are mainly based on three technical routes: feature-based registration methods, region-based registration methods, and model-based registration methods. Feature-based methods achieve registration by extracting and matching key points, but the matching success rate is often low on modern building surfaces with sparse texture features; region-based methods achieve registration by maximizing similarity metrics such as mutual information, but are sensitive to noise and have high computational complexity; model-based methods describe the correspondence between data by establishing a geometric transformation model, but mostly use simplified linear models and have difficulty in handling complex nonlinear deformations. Existing methods generally adopt a static weight strategy, which presets weights based on the inherent accuracy of the sensor and environmental factors, and cannot adapt to changing monitoring conditions. At the same time, traditional geometric correction methods are mostly based on rigid body or linear elastic assumptions, and the accuracy drops significantly when the building structure is subjected to high stress or undergoes plastic deformation.

[0004] These technical routes face a number of technical difficulties in practical applications: First, the existing registration methods ignore the nonlinear behavior of building materials under high stress, resulting in a significant increase in the error of geometric correction in the stress-close-to-yield region; second, the weight distribution of sensor data does not consider the impact of the structural stress state, and the reliability changes of the same sensor under different stress conditions are not fully reflected. These technical difficulties have seriously restricted the further improvement of the registration accuracy of building monitoring data, and innovative solutions are urgently needed. Summary of the invention

[0005] The purpose of the invention is to provide a method for accurate registration of building monitoring data based on spatiotemporal dynamic weights to solve the above-mentioned problems existing in the prior art.

[0006] The technical solution is a method for accurately registering building monitoring data based on spatiotemporal dynamic weights, comprising the following steps:

[0007] Read satellite data, drone data and ground monitoring data, perform preprocessing and standardization, and generate standardized monitoring data sets;

[0008] Analyze the deformation characteristics in the standardized monitoring data set, combine the characteristics of building materials, establish a material model that takes into account nonlinear behavior, and generate a material nonlinear response model and structural stress distribution map;

[0009] According to the structural stress distribution diagram and the pre-stored sensor characteristics, a dynamic weight function is constructed to calculate the reliability weight of each sensor data in different stress areas and output the dynamic sensor weight matrix;

[0010] The geometric deformation of the standardized monitoring data set is corrected by using the material nonlinear response model to obtain a nonlinear correction data set;

[0011] Fusion of nonlinear correction data set and dynamic sensor weight matrix to generate accurate registration results;

[0012] Based on the precise registration results, the parameters are adaptively optimized, possible data anomalies or structural anomalies are identified, and the optimized configuration parameters are generated and fed back for iterative optimization.

[0013] Beneficial effects: The present invention introduces two core factors, namely, material nonlinear behavior and structural stress state, and establishes a deep fusion of data registration and physical model; by modeling the nonlinear materials of the standardized monitoring data set, the geometric correction can accurately reflect the complex deformation behavior of the material under high stress state; by the dynamic weight allocation of structural stress state perception, the reliability assessment problem of different monitoring equipment under changing conditions is solved; the registration accuracy is improved, the adaptability and robustness of the method are enhanced, and more reliable data support is provided for the safety monitoring of building structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A flowchart of the steps of a method for accurate alignment of building monitoring data based on spatiotemporal dynamic weights provided in an embodiment of the present application.

[0015] Figure 2 A flow chart of the steps for generating a material nonlinear response model provided in an embodiment of the present application.

[0016] Figure 3 A flowchart of the steps for generating a structural stress distribution diagram provided in an embodiment of the present application.

[0017] Figure 4 A flowchart of the steps of constructing a dynamic weight function and outputting a dynamic sensor weight matrix provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0019] It should be noted that in order to clearly show the steps of this application, serial numbers are marked for each step in the specification. These serial numbers are only used for the convenience of explanation and do not limit the order of execution of the steps. In actual operation, according to the technical requirements of the specific implementation scenario, the steps can be executed in a different order than that shown in the specification, and in some cases, parallel processing between steps can be achieved.

[0020] like Figure 1 As shown in FIG. 1 , the precise registration method of building monitoring data based on spatiotemporal dynamic weights includes the following steps:

[0021] S1. Read satellite data, drone data and ground monitoring data and perform preprocessing and standardization, including coordinate system unification, noise filtering and data format standardization, to generate standardized monitoring data sets;

[0022] S2. Analyze the deformation characteristics in the standardized monitoring data set, combine the characteristics of building materials, establish a material model that considers nonlinear behavior, and generate a material nonlinear response model and a structural stress distribution map;

[0023] S3, construct a dynamic weight function according to the structural stress distribution diagram and the pre-stored sensor characteristics, calculate the reliability weight of each sensor data in different stress areas, and output the dynamic sensor weight matrix;

[0024] S4, using the material nonlinear response model to perform geometric deformation correction on the standardized monitoring data set, correcting the geometric distortion caused by the nonlinear behavior of the material, and obtaining a nonlinear correction data set;

[0025] S5, integrates the nonlinear correction data set and the dynamic sensor weight matrix to achieve spatiotemporal registration of multi-source data and generate accurate registration results; and outputs a registration accuracy report by comparing the accuracy with the reference points;

[0026] S6. Based on the precise registration results, perform parameter adaptive optimization, identify possible data anomalies or structural anomalies, generate optimized configuration parameters and anomaly detection reports, and feed them back to the previous steps for iterative optimization.

[0027] According to one aspect of the present application, step S1 further comprises:

[0028] S11. Conduct multi-source data collection and metadata extraction: read satellite monitoring data, drone monitoring data and ground monitoring equipment data, extract monitoring metadata such as timestamp, collection location, observation angle, sensor accuracy, etc., perform preliminary verification of data integrity, and generate an initial data quality report.

[0029] S12. Coordinate system unification and time base alignment: Convert different coordinate systems (such as WGS84, local coordinate system, etc.) of satellite monitoring data, UAV monitoring data and ground monitoring equipment data to a unified engineering coordinate system, and align different time bases (such as UTC, local time) to a unified time base, and output a unified coordinate and time data set.

[0030] S13. Reconcile multi-scale data resolutions: Reconcile data of different resolutions in a unified coordinate and time dataset, and convert satellite data (meter level), drone data (centimeter level) and ground monitoring data (millimeter level) into a consistent resolution dataset through an adaptive interpolation algorithm to ensure comparability of spatial dimensions.

[0031] S14. Perform noise identification and signal enhancement: Analyze the noise pattern in the consistent resolution data set, identify and remove environmental noise, sensor noise and random errors through wavelet transform and adaptive filter combination, enhance the effective signal, and obtain a noise suppression data set.

[0032] S15. Data format standardization and quality assessment: Convert the noise suppression dataset into a unified data structure and format, add quality indicators (signal-to-noise ratio, completeness, reliability), and generate the final standardized monitoring dataset and data quality assessment report.

[0033] According to one aspect of the present application, step S2 is further:

[0034] S21. Initialize structural material parameters: Read the building structure design parameters and material property database, combine the initial deformation characteristics in the standardized monitoring data set, initialize the constitutive parameters of building materials such as concrete and steel, and establish the initial material parameter set.

[0035] S22. Analyze multi-phase deformation data: Extract structural deformation data at multiple time points from the standardized monitoring data set, calculate the deformation rate, deformation gradient and cumulative deformation, identify significant deformation areas and deformation patterns, and output deformation feature maps.

[0036] S23. Identify stress-strain nonlinear relationship: Based on the deformation characteristic map and the initial material parameter set, the inverse deduction method is used to calculate the stress state of each key point of the structure. The actual stress-strain curve is identified in combination with the measured deformation data, especially the nonlinear characteristics of the elastic-plastic transition region, to generate a nonlinear stress-strain model.

[0037] S24. Evaluate the material status of each region: Divide the building structure into multiple regions, evaluate the current status of the material in each region (elasticity, elastic-plastic transition, plasticity) based on the nonlinear stress-strain model and deformation characteristic map, and establish a material status zoning map.

[0038] S25. Construct a nonlinear material response model: Integrate the nonlinear stress-strain model and the material state partition diagram to construct a mathematical model that takes into account the nonlinear characteristics of the material. This model can describe the deformation response of the material under different stress levels, and output the material nonlinear response model and a detailed structural stress distribution diagram.

[0039] like Figure 2 As shown, according to one aspect of the present application, step S23 is further:

[0040] S231, based on the standardized monitoring data set and building material properties, the deformation data and the initial material parameter set are extracted, and the initial stress distribution field is calculated by the finite element inversion algorithm;

[0041] S232, extracting multiple time-corresponding force-strain data pairs based on the initial stress distribution field and the initial material parameter set to form a regional stress-strain data set;

[0042] S233, a piecewise nonlinear fitting method is used for the regional stress-strain data set to construct a nonlinear constitutive model and calculate the strain rate influence factor;

[0043] S234, constructing an environmental factor correction function in combination with environmental parameters, integrating the nonlinear constitutive model, strain rate influencing factor and environmental factor correction function, and generating a nonlinear stress-strain model;

[0044] S235. Verify the prediction accuracy of the nonlinear stress-strain model, quantify model uncertainty, and output model reliability assessment report and material nonlinear response model.

[0045] In one embodiment of the present application, the initial stress distribution is reversely calculated. The deformation characteristic map and the initial material parameter set are read, and the finite element inversion algorithm is applied to solve the unknown stress field through the known deformation field. The specific process is: establish a structural discrete finite element model, use the displacement in the deformation characteristic map as the boundary condition, use the conjugate gradient method to iteratively solve the stress distribution equation, and obtain the initial stress distribution field. This calculation takes into account the structural continuity constraint to ensure that the stress distribution satisfies the static equilibrium equation.

[0046] Separate the linear and nonlinear regions of the material. Analyze the initial stress distribution field, combine the material yield strength in the initial material parameter set, and identify the linear deformation area (area where the stress is lower than the yield strength) and the potential nonlinear deformation area (area where the stress is close to or exceeds the yield strength) in the structure. Adopt an adaptive threshold segmentation algorithm, consider stress gradient and stress concentration factors, and generate a material behavior partition map, which divides the structure into multiple regions and marks the deformation behavior characteristics (linear, critical or nonlinear) of each region.

[0047] Extract multi-time stress-strain data. Extract stress-strain data pairs at multiple time points from the standardized monitoring data set and the initial stress distribution field. For each region (especially the nonlinear region marked in the material behavior partition diagram), establish a stress-strain data matrix, where each row represents a time point, each column represents a spatial point, and the matrix elements are the corresponding stress-strain value pairs. Apply a noise-robust data cleaning algorithm to remove outliers and obtain a regional stress-strain data set.

[0048] Perform iterative fitting of nonlinear constitutive relations. For each region in the regional stress-strain data set, a piecewise nonlinear fitting method is used to construct a stress-strain relationship model. First, the elastic modulus and yield point of the material are determined using piecewise linear fitting. Then, for data exceeding the yield point, a nonlinear optimization algorithm (such as the Levenberg-Marquardt algorithm) is used to fit a multi-parameter nonlinear model (such as the Ramberg-Osgood model or the modified Johnson-Cook model). During the iterative optimization process, the cross-validation method is used to evaluate the prediction ability of the model under different load conditions, and the model parameters are dynamically adjusted until the residual reaches the minimum value, and the regional nonlinear constitutive model is output.

[0049] Calculate the strain rate influence factor. Analyze the dynamic deformation data in the regional stress-strain data set, calculate the strain rate in different regions, and evaluate the influence of strain rate on the nonlinear behavior of the material. The specific method is: calculate the ratio of the strain increment to the time interval between consecutive time points to obtain the strain rate field; divide the strain rate into multiple levels, fit the constitutive model to the stress-strain data at each level separately, compare the changes in model parameters under different strain rates, establish the relationship function between parameters and strain rate, and output the strain rate influence factor.

[0050] Correct temperature and environmental factors. Read environmental parameters (temperature, humidity, etc.) in the monitoring metadata and analyze the impact of these factors on the nonlinear behavior of the material. Use multivariate regression analysis to establish the mapping relationship between environmental parameters and constitutive model parameters. For temperature effects, special consideration is given to the coupling of thermal expansion effects and material softening effects; for concrete materials, the impact of humidity on Young's modulus is also considered. Based on the analysis results, an environmental factor correction function is generated to adjust the constitutive model parameters under different environmental conditions.

[0051] Construct a comprehensive nonlinear stress-strain model. Integrate the regional nonlinear constitutive model, strain rate influencing factors and environmental factor correction functions to construct a comprehensive nonlinear stress-strain model that takes into account the influence of multiple factors. The model adopts a layered structure: the base layer is a static nonlinear constitutive relationship, the middle layer considers the influence of strain rate, and the top layer applies environmental correction. The model is expressed by tensors and can describe the nonlinear behavior of anisotropic materials under three-dimensional stress states. The model parameters are stored in a spatially distributed form, covering the entire monitoring area, and finally output a nonlinear stress-strain model, which will be used in subsequent steps to accurately predict deformation behavior.

[0052] Perform model validation and uncertainty quantification. Use a portion of the regional stress-strain data set (about 20%) as a validation set to evaluate the prediction accuracy of the nonlinear stress-strain model. Calculate the relative error between the predicted stress and the measured stress, and generate error distribution statistics. At the same time, use the Monte Carlo method to perform random sampling in the model parameter space to evaluate the impact of parameter uncertainty on the model output and quantify the reliability interval of the model. Based on the validation results and uncertainty analysis, generate a model reliability assessment report, which contains the reliability indicators of the model at different stress levels and in different regions, and will serve as an important reference for subsequent weight allocation.

[0053] This embodiment constructs a comprehensive nonlinear stress-strain model to overcome the limitations of traditional linear material models. The model can accurately describe the nonlinear behavior of materials under different stress states, especially the deformation characteristics close to the yield point and plastic region. Compared with the traditional linear material model, the prediction accuracy of the nonlinear model in high stress areas is improved by more than 3 times, and the material behavior prediction error is reduced from more than 30% to less than 10%. This high-precision material behavior prediction directly improves the accuracy of geometric correction, enabling the registration method to handle complex nonlinear deformation fields, which is particularly suitable for monitoring building structures that undergo large deformation or partial plastic deformation, and improves the reliability of monitoring data under extreme conditions (such as after an earthquake and during construction).

[0054] like Figure 3 As shown, according to one aspect of the present application, step S24 is further:

[0055] S241, read the building structure design parameters and standardized monitoring data set, and divide the structural function zoning map based on the structural function characteristics and geometric form;

[0056] S242, analyzing deformation characteristics in the standardized monitoring data set, identifying areas with similar deformation characteristics, and generating a deformation behavior partition map;

[0057] S243, reading the initial stress distribution field and the nonlinear stress-strain model, calculating the stress state index, and generating a stress state classification diagram;

[0058] S244, determining the elastic-plastic state of the material in each region based on the material yield criterion, taking into account the differences in material properties, and outputting a material elastic-plastic state diagram;

[0059] S245, analyzing the historical load data in the standardized monitoring data set, calculating the cumulative plastic strain and fatigue damage, and generating a cumulative damage distribution map;

[0060] S246, integrate the structural function zoning diagram, deformation behavior zoning diagram, stress state classification diagram, material elastic-plastic state diagram and cumulative damage distribution diagram, and generate the material state zoning diagram and structural stress distribution diagram through multi-index evaluation.

[0061] In one embodiment of the present application, the structural functional areas are divided. The building structure design parameters and standardized monitoring data sets are read, and based on the functional characteristics and geometric morphology of the structure, a hierarchical clustering algorithm is used to divide the entire structure into functionally similar areas. The algorithm first performs an initial classification based on material type and component function, and then considers geometric continuity and stress transfer path to merge or subdivide the areas, and finally outputs a structural functional zoning map. The zoning map contains area boundaries, area numbers, and area functional attributes, providing a basic spatial reference framework for subsequent material status assessment.

[0062] Analyze deformation feature clustering. Read the deformation feature map, apply the K-means++ clustering algorithm to the displacement, velocity and acceleration data of each spatial point, and identify areas with similar deformation features. Before clustering, use principal component analysis (PCA) to reduce the dimension and retain the principal components that explain more than 90% of the variance; during the clustering process, the silhouette coefficient is used to automatically determine the optimal number of clusters. Apply spatial continuity constraints to the clustering results to ensure that the same cluster area is relatively continuous in space, and output the deformation behavior partition map.

[0063] Perform detailed analysis of stress state. Read the initial stress distribution field and nonlinear stress-strain model, and calculate the stress state indicators in each area, including: principal stress distribution, shear stress distribution, stress triaxiality and stress intensity. Use stress invariant theory to calculate the von Mises stress and maximum principal stress at each point, and compare them with the material yield strength to determine the stress utilization rate at each point. Based on these indicators, use fuzzy logic algorithm to classify the stress state and output the stress state classification diagram.

[0064] Determine the elastic-plastic state of the material. Read the stress state classification diagram and nonlinear stress-strain model, and determine the elastic-plastic state of the material in each region based on the material's yield criterion (such as the von Mises criterion for metal materials and the Drucker-Prager criterion for concrete). For each region, calculate the probability distribution of the three states of elasticity, elastic-plastic transition and plasticity, consider the uncertainty information in the model reliability assessment report, and use the Bayesian inference method to determine the final state classification. For concrete materials, special consideration is given to the tension-compression asymmetry and cracking behavior, and the material elastic-plastic state diagram is output.

[0065] Evaluate cumulative damage. Analyze historical load data in the standardized monitoring data set and calculate the cumulative plastic strain and fatigue damage experienced by the materials in each region. For metal materials, the Palmgren-Miner linear cumulative damage theory is used in combination with the Rainflow counting method to handle variable amplitude loads; for concrete materials, the cracking damage model is considered to evaluate the development of microcracks. By combining the low-cycle fatigue and high-cycle fatigue models, the regional damage index is calculated and the cumulative damage distribution map is output.

[0066] Conduct a comprehensive assessment of the material state. Integrate the structural function zoning diagram, deformation behavior zoning diagram, stress state classification diagram, material elastic-plastic state diagram and cumulative damage distribution diagram to establish a multi-index evaluation system. Use the analytic hierarchy process (AHP) to determine the weight of each indicator, and calculate the comprehensive state index by weighted summation. According to the index value, the material state is divided into five levels: fully elastic, elastic-dominated, elastic-plastic transition, plastic-dominated and near-failure, and a comprehensive material state assessment diagram is generated.

[0067] Refine the state boundary. Read the material comprehensive state assessment diagram, detect the boundary position of different state areas, and use the adaptive grid refinement algorithm to improve the assessment accuracy of the boundary area. Increase the density of calculation points in the area near the boundary, re-execute the state assessment, and ensure the accurate description of the state change area. Apply the boundary smoothing algorithm to reduce unnecessary boundary complexity, while retaining the real state change characteristics, and output a refined material state boundary diagram.

[0068] Conduct time evolution analysis of material states. Compare the material comprehensive state assessment diagrams at different time points to analyze the time evolution trend of material states. Use the state transition matrix to describe the transition probability between states and identify areas of rapid state change and stable areas. Based on the historical evolution law, use time series prediction models (such as ARIMA or LSTM networks) to predict the state change trend in the short term, output the state evolution prediction report and the final material state partition diagram, which will serve as an important basis for subsequent nonlinear correction and weight allocation.

[0069] This embodiment achieves a detailed assessment of the state of building structure materials. Compared with the traditional single standard zoning method, this embodiment enhances the adaptability to complex structures and improves the zoning accuracy by more than 35%. The detailed material state assessment enables the system to adopt differentiated processing strategies for different state areas, especially the detailed identification of elastic-plastic transition areas and damaged areas, which makes subsequent high-precision registration possible. In addition, by considering the cumulative damage factor, this embodiment also improves the monitoring capability of long-term service structures and provides more reliable data support for structural life assessment.

[0070] Existing weight assignment methods are mainly based on the inherent accuracy of sensors and environmental factors, ignoring the stress state of the building structure itself. The reliability of the same sensor will vary significantly under different stress distribution conditions, and the structural stress state is not taken into account. Existing static weight schemes and dynamic calibration techniques use predefined weights or simple environmental parameter adjustments, which cannot respond to changes in stress distribution in the structure. This leads to a significant reduction in registration accuracy under complex loading conditions, especially in areas of non-uniform deformation. Therefore, Figure 4 As shown, according to one aspect of the present application, step S3 is further:

[0071] S31. Analyze the accuracy characteristics of various sensors (satellite SAR, UAV optical / laser, ground displacement / tilt / strain sensors, etc.) under different working conditions, establish a sensor accuracy characteristic model, and quantify the relationship between influencing factors and measurement errors;

[0072] S32, read the structural stress distribution diagram and the sensor accuracy characteristic model, analyze the reliability variation law of various sensors under different stress levels, establish a quantitative correlation function between stress level and sensor reliability, and output a stress-reliability mapping table;

[0073] S33. Based on the structural stress distribution map and engineering safety assessment standards, identify high stress areas, stress concentration areas, areas with large stress gradients, and key load-bearing parts in the structure, set monitoring priorities for different areas, and generate a structural area priority map;

[0074] S34, integrating the stress-reliability mapping table and the structural area priority map, constructing a dynamic weight calculation function, which can adaptively adjust the weight coefficient according to the regional stress state, sensor type and regional priority, and generate a weight calculation function (dynamic weight function, the same below);

[0075] S35. Apply the weight calculation function to perform weight calculation on each data point in the standardized monitoring data set, taking into account the spatial position of the data point, the acquisition time, the stress state of the area and the sensor characteristics, and generate a dynamic sensor weight matrix covering the entire monitoring area and time period.

[0076] This embodiment overcomes the limitations of the traditional static weight method. Compared with the traditional static weight method, the dynamic weight mechanism improves the rationality of weight distribution under changing load conditions, and the data fusion accuracy is improved by 25-40%. Especially in complex structures with uneven stress distribution, it can accurately reflect the monitoring needs of local areas and improve the ability to identify potential risk areas. In addition, this embodiment also improves the system's adaptability to sensor anomalies. When the performance of some sensors decreases, the system can automatically reduce their weights to maintain the overall monitoring quality, thereby enhancing the robustness and fault tolerance of the monitoring system.

[0077] According to one aspect of the present application, step S34 is further:

[0078] S341, reading the stress-reliability mapping table, the structural area priority map and the sensor accuracy characteristic model, establishing a multidimensional factor evaluation system affecting weight distribution, including stress state index, regional importance index and sensor characteristic index, and obtaining a weight influencing factor evaluation system;

[0079] S342. Based on the weight influencing factor evaluation system, a weight optimization objective function is constructed, and the maximization of deformation monitoring accuracy, the minimization of spatial discontinuity of weight distribution, and the maximization of monitoring reliability of key areas are considered at the same time to obtain a multi-objective weight optimization function;

[0080] S343, reading the structural stress distribution diagram and the stress-reliability mapping table, designing a stress response function for each sensor type, and outputting a stress response function set;

[0081] S344, constructing constraint conditions to ensure spatial continuity of weight distribution, constructing a gradient energy functional of the weight field, and outputting a spatial continuity constraint function;

[0082] S345, analyzing the deformation rate data in the standardized monitoring data set, constructing a time response mechanism for dynamic weight adjustment, including a fast response part and a slow adaptation part, and outputting a time response adjustment function;

[0083] S346, constructing a robust processing mechanism for abnormal data using a weight function, reducing the impact of abnormal values ​​on weight calculation, and outputting an abnormal data processing function;

[0084] S347. Integrate multi-objective weight optimization function, stress response function set, spatial continuity constraint function, time response adjustment function and abnormal data processing function to build a complete dynamic weight calculation framework and generate weight calculation function.

[0085] This embodiment achieves optimal weight allocation under complex conditions. Compared with the traditional single-factor weight method, this embodiment improves the rationality and adaptability of weight allocation, and improves the fusion accuracy by more than 30% in complex monitoring scenarios. In particular, the introduction of the time responsiveness mechanism enables the system to respond quickly to emergencies, such as instantaneous deformation caused by earthquakes or explosion loads, while gradually adapting to slow processes such as creep or seasonal changes of buildings through long-term learning. The robust processing mechanism for abnormal data improves the reliability of the system in harsh environments or partial failure of sensors, extending the effective working time of the monitoring system by 35%.

[0086] According to one aspect of the present application, the steps of constructing a dynamic weight function and outputting a dynamic sensor weight matrix include:

[0087] Read the structural stress distribution diagram and stress-reliability mapping table, and design the stress response function for each sensor type in the form of a piecewise function: in the elastic region, use a slowly varying S-shaped curve to describe the change in sensor reliability; in the elastic-plastic transition region, increase the function slope; in the plastic region, introduce an inflection point or saturation characteristic;

[0088] The pre-stored measured data are fitted by nonlinear least square method to determine the stress response function parameters of various sensors and output the stress response function set.

[0089] This embodiment achieves accurate mapping of structural stress state and sensor reliability. Compared with simple linear mapping, this embodiment improves the accuracy of reliability estimation by 45%. Accurate stress-reliability mapping enables the system to select the most reliable data source for fusion under complex stress distribution conditions, avoiding the error amplification problem caused by reduced sensor reliability in high stress areas. In addition, this embodiment can also adapt to the differences in characteristics of different sensors, such as the reduced accuracy of SAR interferometric sensors in stress-induced large deformation areas, and the lack of sensitivity of optical related technologies in small deformation areas, thereby maximizing the advantages of various sensors and improving overall monitoring efficiency.

[0090] In one embodiment of the present application, the weight influencing factors are analyzed quantitatively. The stress-reliability mapping table, the structural area priority map and the sensor accuracy characteristic model are read to establish a multidimensional factor space that affects the weight distribution. The principal component analysis method is used to identify the main influencing factors and their relative importance. For each factor, a quantitative evaluation index is established, such as stress state index (stress level, stress gradient), regional importance index (structural safety factor, functional criticality) and sensor characteristic index (baseline accuracy, random error, systematic error), and the weight influencing factor evaluation system is output.

[0091] Construct a multi-objective weight optimization function. Based on the weight influencing factor evaluation system, construct a weight optimization objective function. This objective function contains three sub-objectives: maximizing deformation monitoring accuracy, minimizing spatial discontinuity of weight distribution, and maximizing monitoring reliability in key areas. The Pareto optimization method is used to deal with multi-objective conflicts, and a comprehensive objective function is constructed through convex optimization technology. Different combinations of objective weights are designed for different application scenarios (such as conventional monitoring, post-disaster assessment, and construction monitoring), and a multi-objective weight optimization function is output.

[0092] Construct stress state response function. Read the structural stress distribution diagram and stress-reliability mapping table, and design a stress response function for each sensor type. This function describes the law of sensor reliability changing with stress level, and adopts a piecewise function form: in the low stress area (elastic range), the function is a slowly changing S-shaped curve; in the transition area (elastic-plastic transition area), the function slope increases; in the high stress area (plastic area), the function may have an inflection point or saturation phenomenon. The function parameters are determined by fitting the measured data with the nonlinear least squares method, and the stress response function set for each type of sensor is output.

[0093] Construct spatial continuity constraints. Design constraints to ensure spatial continuity of weight distribution. Use the variational principle to construct the gradient energy functional of the weight field and quantify the weight space discontinuity into energy values. Set strong continuity constraints in regions with similar material states; allow appropriate discontinuities at material state boundaries. The constraint strength parameters are determined through numerical experiments on typical scenarios, so that the weight changes are smooth and reflect the real sensor reliability changes, and output the spatial continuity constraint function.

[0094] Construct a time-responsive mechanism. Analyze the deformation rate data in the deformation feature map and design a time-responsive mechanism for dynamic weight adjustment. The mechanism consists of two parts: a fast response part and a slow adaptation part. The fast response part is implemented through an exponential filter to quickly adjust the weights for sudden deformation events (such as impact loads); the slow adaptation part uses a long short-term memory (LSTM) network to learn long-term deformation trends and predict weight adjustment needs. The results of the two parts are combined through an adaptive mixer to output a time response adjustment function.

[0095] Build a robust outlier processing mechanism. Design a robust processing mechanism for abnormal data in the weight function. Use the Huber loss function instead of the traditional mean square error to reduce the impact of outliers on weight calculation. Combine the random sampling consensus (RANSAC) algorithm to detect and process data anomalies to prevent abnormal data from causing unreasonable weight distribution. For temporary sensor failure or sudden change in accuracy, design a fast weight reduction strategy and backup sensor activation mechanism, and output an abnormal data processing function.

[0096] Perform weight normalization and smoothing strategies. Design a weight normalization strategy to ensure the comparability of weights of different types of sensors and the rationality of the overall weight distribution. Use nonlinear normalization methods (such as variants of the softmax function) to maintain the relative difference of weights while controlling the ratio of the maximum and minimum weights. To avoid sudden changes in weights over time, apply a time smoothing filter to limit the maximum amplitude of weight changes per unit time, and output a weight normalization and smoothing function.

[0097] Integrated comprehensive dynamic weight function. Integrate multi-objective weight optimization function, stress response function set, spatial continuity constraint function, time response adjustment function, abnormal data processing function and weight normalization and smoothing function to build a complete dynamic weight calculation framework. Adopt a hierarchical calculation strategy: first calculate the basic weight based on stress state and sensor characteristics, then apply spatial continuity constraints, then make time adjustments, and finally handle anomalies and normalize. The framework is implemented in the form of a function library, which can be flexibly configured according to different monitoring needs, and finally outputs a weight calculation function, which will be used to generate a dynamic weight matrix covering the entire monitoring area.

[0098] According to one aspect of the present application, step S4 is further:

[0099] S41. Analysis of the error of the traditional geometric transformation model: Compare the standardized monitoring data set with the traditional geometric transformation results based on the linear elastic assumption, analyze the error distribution under different stress states, quantify the error size and distribution characteristics caused by the linear assumption, and output the linear model error map.

[0100] S42. Decomposition of nonlinear deformation modes: Decompose the deformation field in the standardized monitoring data set into linear and nonlinear parts. Through principal component analysis and modal decomposition technology, extract the main nonlinear deformation modes to form a nonlinear deformation mode library.

[0101] S43. Perform deformation prediction based on local material state perception: Combine the material nonlinear response model and the material state partition diagram to establish a local nonlinear deformation prediction model for each area of ​​the structure. The model can predict the deformation behavior based on the known stress state and material properties and output a set of local deformation prediction functions.

[0102] S44. Construct a hierarchical geometric transformation matrix: Based on the nonlinear deformation pattern library and the local deformation prediction function set, construct a multi-level geometric transformation matrix, including a global linear transformation layer, a regional nonlinear transformation layer and a local detail transformation layer, and generate a hierarchical geometric transformation operator.

[0103] S45, perform nonlinear correction and residual optimization; apply the hierarchical geometric transformation operator to perform geometric correction on the standardized monitoring data set, and then further optimize the correction result through the iterative minimization residual method, and finally output the nonlinear correction data set and correction residual evaluation report.

[0104] The current geometric correction method assumes that the building structure follows linear elastic deformation, ignoring the nonlinear behavior of the material when it approaches the limit state, resulting in registration errors in high stress areas. When the stress approaches the yield point, the error of the linear geometric transformation model increases by more than 300%. Existing real-time optimization methods are all based on rigid body or linear elastic assumptions for geometric transformation, and cannot accurately express nonlinear deformation behavior. In particular, these methods perform particularly poorly in the crack development stage of concrete structures or in the plastic deformation area of ​​steel structures. Therefore, according to one aspect of the present application, step S43 is further:

[0105] S431, read the material nonlinear response model and material state partition diagram, combine with structural mechanics theory, build a library containing various typical deformation modes, and form a physics-based deformation mode library;

[0106] S432, decomposing the complex deformation field in the standardized monitoring data set into a series of linear combinations of orthogonal modes, identifying the dominant mode and the secondary mode and combining with the deformation mode library, outputting the deformation mode decomposition result;

[0107] S433, reading the material state partition diagram, parametrically expressing the material state of each area, and generating a material state parameter field;

[0108] S434, combining the deformation mode decomposition result and the material state parameter field, constructing a local deformation response function for each region, and generating a regional deformation response function set;

[0109] S435, analyzing interface conditions between adjacent regions, ensuring the continuity and compatibility of the deformation field at the region boundary, and generating an interface compatibility condition set;

[0110] S436. Based on the regional deformation response function set and the interface compatibility condition set, a global load-deformation mapping matrix is ​​constructed to describe the relationship between the external load and the structural deformation;

[0111] S437, integrate the regional deformation response function set, interface compatibility condition set and global load-deformation mapping matrix, develop an integrated deformation prediction engine, and output the local deformation prediction function set for subsequent geometric correction;

[0112] S438. Perform geometric correction on the standardized monitoring data set based on the local deformation prediction function set to obtain a nonlinear correction data set.

[0113] This embodiment establishes an accurate mapping from material properties to deformation behavior. Compared with the traditional homogeneous material assumption method, the deformation prediction accuracy of this embodiment in complex structures is improved by 55%, especially in the elastic-plastic transition area and the damaged area, the prediction error is reduced from 30-40% to less than 10%. Accurate deformation prediction directly improves the accuracy of geometric correction and makes the registration results more consistent with physical laws. In addition, this embodiment can also handle material discontinuities and interface conditions, and is suitable for composite materials structures and buildings with complex geometric shapes. It expands the application scope of the registration method and provides technical support for the accurate monitoring of various types of buildings.

[0114] According to one aspect of the present application, the steps of constructing a local deformation response function for each region and generating a set of regional deformation response functions include:

[0115] Combining the deformation mode decomposition results and the material state parameter field, special deformation response functions are designed for different areas;

[0116] For the elastic region, the linear hyperelastic theory is used; for the elastic-plastic transition region, the flow theory combined with the hardening rule is used; for the plastic region, the large deformation theory and plastic flow effect are considered;

[0117] Each response function is implemented using the extended finite element method, which enables it to handle material discontinuities and damage effects, and outputs a set of regional deformation response functions.

[0118] This embodiment achieves an accurate description of complex deformation behaviors. Compared with the unified model method, this embodiment improves the adaptability and accuracy of deformation prediction, and reduces the prediction error by more than 40% under complex load conditions. This embodiment is particularly suitable for processing heterogeneous materials and irregular structures, such as complex building forms such as multi-layer composite walls and steel-concrete composite structures. By accurately simulating various nonlinear effects such as strain hardening, material softening, and local buckling, this embodiment improves the monitoring capability under extreme conditions, provides reliable data support for the safety assessment of important buildings after extreme events, and has important engineering practical value.

[0119] In one embodiment of the present application, a physics-based deformation pattern library is constructed. The nonlinear stress-strain model and the material state partition diagram are read, and a library containing various typical deformation patterns is constructed in combination with the classical structural mechanics theory. For different types of structural elements (beams, columns, plates, shells, etc.), the deformation responses under various boundary conditions and load conditions are pre-calculated. The influence of the nonlinear behavior of the material on the deformation mode is specially considered, and the deformation difference between the linear region and the nonlinear region is calculated to form a physics-based deformation pattern library.

[0120] Decompose measured deformation data. Read the deformation field data in the standardized monitoring data set, and apply orthogonal decomposition technology (such as POD, Proper Orthogonal Decomposition) to decompose the complex deformation field into a linear combination of a series of orthogonal modes. For each mode, analyze its spatial distribution characteristics and time evolution law, and identify the dominant mode and secondary mode. Compare the decomposition results with the theoretical mode in the physics-based deformation mode library, establish a mapping relationship between the measured mode and the theoretical mode, and output the deformation mode decomposition results.

[0121] Parameterize the local material state. Read the material state partition map and express the material state of each area in a parameterized way. The parameters include: elastic modulus distribution, yield stress distribution, plastic hardening coefficient, damage coefficient, etc. For the elastic-plastic transition area, a state transition function is specially constructed to describe how the parameters smoothly transition from elastic values ​​to plastic values. For concrete materials, the influence of cracking degree and cracking direction is also considered. The spline interpolation method is used to ensure the continuous distribution of parameters in space and output the material state parameter field.

[0122] Construct regional deformation response functions. Combine the deformation modal decomposition results and the material state parameter field to construct a local deformation response function for each region. This function describes the expected deformation of each point in the region under a given stress state. For the elastic region, the linear hyperelastic theory is used; for the elastic-plastic transition region, the flow theory (Flow Theory) combined with the hardening rule is used; for the plastic region, the large deformation theory and plastic flow effect are considered. Each response function is implemented using the extended finite element method (XFEM), which can handle material discontinuities and damage effects, and output a set of regional deformation response functions.

[0123] Process interface compatibility conditions. Analyze the interface conditions between adjacent regions to ensure the continuity and compatibility of the deformation field at the region boundary. Use the Lagrange multiplier method to add interface continuity as a constraint condition to the deformation prediction model. For interfaces with significant differences in material states, establish a transition layer model to avoid non-physical mutations in the deformation field. Consider possible slip and separation phenomena at the interface, use the contact mechanics model to describe the interface behavior, and output the interface compatibility condition set.

[0124] Construct a load-deformation mapping matrix. Based on the set of regional deformation response functions and the set of interface compatibility conditions, a global load-deformation mapping matrix is ​​constructed. This matrix maps external loads and boundary conditions to the structural deformation field, using a block matrix structure, with each block corresponding to a material state region. The matrix elements are calculated using the finite element method or boundary element method, taking into account material nonlinearity and geometric nonlinearity effects. To improve computational efficiency, the model reduction technique is used to retain the most critical degrees of freedom for deformation prediction and output the load-deformation mapping matrix.

[0125] Quantify deformation uncertainty. Evaluate sources of uncertainty in deformation prediction, including material parameter uncertainty, boundary condition uncertainty, model error, and measurement noise. Use Monte Carlo simulation to generate a large number of deformation field samples by random sampling in parameter space. Analyze the statistical distribution of these samples and calculate the mean, variance, and confidence interval of deformation prediction at each point. For computationally complex areas, use multi-fidelity models and surrogate models to accelerate uncertainty analysis and output deformation prediction uncertainty maps.

[0126] Develop an integrated deformation prediction engine. Integrate the regional deformation response function set, interface compatibility condition set, load-deformation mapping matrix and deformation prediction uncertainty map to develop an integrated deformation prediction engine. The engine adopts a layered architecture: the bottom layer is the material state and response function library, the middle layer is the regional deformation calculation module, and the top layer is the global coordination and optimization module. The engine supports two working modes: forward mode (predicting deformation given load) and reverse mode (inferring the complete deformation field given partial deformation). The engine has a built-in adaptive solver that can automatically select the most appropriate numerical method according to the complexity of the problem, and finally output an efficient and reliable set of local deformation prediction functions, which will be used for subsequent geometric transformation matrix construction.

[0127] Since it is difficult for existing multi-scale feature matching algorithms to balance matching efficiency and accuracy when processing multi-resolution data from satellites, drones, and ground equipment, according to one aspect of the present application, step S44 is further as follows:

[0128] S441, read the reference point coordinates in the standardized monitoring data set, calculate the global rigid body transformation parameters using the least squares method, and output the global rigid body transformation matrix set;

[0129] S442, reading the material state partition map, calculating the local affine transformation matrix for each region, and outputting the regional affine transformation matrix set and the affine transformation residual map;

[0130] S443, analyzing the affine transformation residual image, extracting the nonlinear deformation component, performing multi-scale decomposition through wavelet transformation, identifying the deformation mode at different spatial scales, and outputting the nonlinear residual decomposition result;

[0131] S444. Based on the nonlinear residual decomposition result, a radial basis function network is designed to express the nonlinear geometric transformation, a suitable basis function type is selected for different types of nonlinear deformation, and a radial basis function network structure is output;

[0132] S445, reading the material nonlinear response model and the local deformation prediction function set, taking the deformation behavior predicted by the physical model as a constraint condition, optimizing the radial basis function network parameters, and outputting the radial basis function parameter set of the physical constraint;

[0133] S446, constructing a transformation hierarchy integration strategy, combining the global rigid body transformation matrix set, the regional affine transformation matrix set, and the radial basis function parameter set of physical constraints into a complete transformation expression, and outputting a hierarchical geometric transformation operator;

[0134] S447. Use a layered geometric transformation operator to perform geometric correction on the standardized monitoring data set to obtain a nonlinear correction data set.

[0135] This embodiment realizes accurate geometric correction from global to local. Compared with the traditional single transformation method, this embodiment improves the registration accuracy by more than 50%, especially in areas with complex deformation patterns, the accuracy improvement is more significant. This embodiment can adapt to deformations of different scales and characteristics: the rigid body layer processes sensor position differences, the affine layer processes uniform strain, and the RBF layer processes local nonlinear deformation, so that the system can fully and accurately describe the geometric changes of building structures under complex stress states. In addition, the layered processing strategy also improves the computational efficiency of the algorithm. Compared with directly processing complex nonlinear transformations, the calculation time is reduced by more than 60%, achieving the unity of high precision and high efficiency.

[0136] According to one aspect of the present application, the steps of optimizing radial basis function network parameters and outputting radial basis function parameter sets include:

[0137] Read the material nonlinear response model and local deformation prediction function set to construct the optimization objective function under physical constraints; the optimization objective function includes data fitting terms, physical consistency terms and smoothing regularization terms, and ensures the rationality of parameter optimization by balancing the influence of data drive and physical laws;

[0138] The alternating direction multiplier method is used to solve the optimization problem, with special attention paid to the nonlinear region of the material to ensure that the radial basis function transformation is consistent with the material behavior; the radial basis function parameter set optimized by physical constraints is output for subsequent geometric transformation matrix construction.

[0139] This embodiment solves the problem of the disconnection between geometric transformation and physical laws. Compared with the pure data-driven method, this embodiment improves the physical rationality of the transformation and reduces more than 80% of non-physical distortion, especially in data sparse areas, where the improvement is more significant. This embodiment solves the overfitting and underfitting problems of the traditional RBF method when dealing with complex nonlinear deformations, ensuring the smoothness and accuracy of the transformation in the entire monitoring area. In addition, by introducing physical constraints, this embodiment also improves the robustness to abnormal data, and can maintain the rationality of the transformation even when the quality of some sensor data decreases, thereby enhancing the reliability and adaptability of the monitoring system in harsh environments.

[0140] In one embodiment of the present application, a rigid body transformation basic matrix is ​​constructed. The reference point coordinates in the standardized monitoring data set are read, and the least squares method is used to calculate the global rigid body transformation parameters, including the translation vector and the rotation matrix. In order to improve stability, an area with a lower stress level is selected as a reference area to reduce the impact of non-rigid body deformation on parameter estimation. For different sensor data (satellites, drones, ground equipment), the corresponding rigid body transformation matrices are calculated respectively, the preliminary correspondence between the data sources is established, and the global rigid body transformation matrix set is output.

[0141] Calculate the regional affine transformation matrix. Read the material state partition map and calculate the local affine transformation matrix for each region (especially the region with less deformation). Use weighted least squares to fit the data points in the region, with the weights based on the reliability of the points and the distance from the center of the region. The affine transformation includes a linear part (scaling, shearing) and a translation part, which can handle uniform deformation within the region. Generate the corresponding transformation matrix for each region, record the fitting residual, and output the regional affine transformation matrix set and affine transformation residual map.

[0142] Decompose the nonlinear deformation field. Analyze the affine transformation residual map and extract the nonlinear deformation components that cannot be expressed by affine transformation. Use wavelet transform to perform multi-scale decomposition of the residual field and identify deformation patterns at different spatial scales. For each scale, apply principal component analysis to extract the main deformation pattern and match it with the template in the nonlinear deformation pattern library to identify the physical characteristics of the deformation. Based on the decomposition results, determine the required nonlinear transformation complexity and output the nonlinear residual decomposition results.

[0143] Construct a radial basis function network. Based on the results of nonlinear residual decomposition, a radial basis function (RBF) network is designed to express nonlinear geometric transformations. For different types of nonlinear deformations (such as local bending, torsion, and wrinkling), select appropriate basis function types (such as Gaussian function, multi-quadratic function, and thin plate spline). Determine the optimal basis function parameters (such as width and smoothness) through a grid search method. The network node layout adopts an adaptive strategy: increase the node density in areas with significant deformation, reduce the number of nodes in areas with gentle deformation, and output the radial basis function network structure.

[0144] Perform physical constraint RBF parameter optimization. Read the nonlinear stress-strain model and the local deformation prediction function set, use the deformation behavior predicted by the physical model as a constraint, and optimize the RBF network parameters. Establish the optimization objective function under physical constraints, including data fitting terms, physical consistency terms, and smoothing regularization terms. Use the alternating direction multiplier method (ADMM) to solve the optimization problem and balance the influence of data drive and physical laws. Pay special attention to the nonlinear region of the material, ensure that the RBF transformation is consistent with the material behavior, and output the physical constraint RBF parameter set.

[0145] Construct adaptive local detail transforms. Construct adaptive local transforms for small-scale local deformation details (such as deformation around cracks and local buckling). Use layered B-spline technology to refine the grid in areas that require fine description. Local transforms are represented by hierarchical sparse grids, and detail information is only stored in complex deformation areas, greatly reducing computational complexity. Local transform parameters are optimized by combining the iterative closest point (ICP) algorithm with local bundle adjustment to output an adaptive local detail transform set.

[0146] Integrate multi-scale transformation hierarchies. Design a transformation hierarchical integration strategy to combine the global rigid body transformation matrix set, the regional affine transformation matrix set, the physically constrained RBF parameter set, and the adaptive local detail transformation set into a complete transformation expression. Adopt a hierarchical design: first apply the global rigid body transformation to establish the initial correspondence, then apply the regional affine transformation to correct the linear deformation, then use the RBF network to handle the mid-scale nonlinear deformation, and finally apply the local detail transformation for fine adjustment. Each layer of transformation is connected by a smooth weight function to ensure the continuous transition of the transformation in space, and output the hierarchical transformation combination strategy.

[0147] Encapsulate and optimize the geometric transformation operator. Encapsulate the hierarchical transformation combination strategy into an efficiently computable geometric transformation operator. Use computational graph optimization technology to analyze transformation dependencies and maximize the potential for parallel computing. For computationally intensive areas, use an adaptive meshing strategy to reduce calculation points while ensuring accuracy. Accelerate key computing links and improve processing speed through: Single Instruction Multiple Data (SIMD) instruction set and graphics processing unit (GPU). Design a cache mechanism to store intermediate results to avoid repeated calculations, especially for iterative optimization processes. Finally, output a high-performance hierarchical geometric transformation operator that can accurately and efficiently correct geometric distortions caused by nonlinear material behavior, providing a reliable foundation for subsequent registration steps.

[0148] According to one aspect of the present application, step S5 is further:

[0149] S51. Perform time series interpolation and synchronization: Analyze the time discontinuity in the nonlinear correction data set, and synchronize data from different sources and different acquisition frequencies to a unified time point through a time series interpolation algorithm that considers the nonlinear characteristics of the material, and output a time-synchronized data set.

[0150] S52. Perform weighted fusion spatial registration: Integrate the time-synchronized dataset and the dynamic sensor weight matrix, perform weighted-based spatial registration, give priority to high-reliability sensor data in high-stress areas, and generate preliminary registration results.

[0151] S53. Perform deformation gradient constrained registration optimization: introduce deformation gradient continuity constraints, optimize the preliminary registration results, ensure smooth transition of the deformation field in space, especially in the area where the material state changes, and output the optimized registration results.

[0152] S54. Verify reference points and evaluate accuracy: Use pre-deployed high-precision reference points (such as GNSS control points, high-precision leveling points, etc.) to verify the optimized registration results, calculate the registration accuracy indicators of each area, including the average error, maximum error and error standard deviation, and generate a registration accuracy report.

[0153] S55, Multi-scale accuracy visualization: Based on the registration accuracy report, generate a multi-scale accuracy visualization map to intuitively display the registration accuracy distribution in different regions and under different stress states, and output the accuracy distribution map and the final accurate registration result.

[0154] According to one aspect of the present application, step S52 is further:

[0155] S521, read the nonlinear correction data set, extract spatial features from satellite data, UAV data and ground monitoring data respectively, identify key points and descriptors, and output a multi-source spatial feature set;

[0156] S522, read the dynamic sensor weight matrix and the multi-source spatial feature set, assign an initial weight to each feature point, and output the initial weight set of the feature points;

[0157] S523, constructing a feature matching framework based on a graph model, transforming the feature matching problem into a graph matching problem, optimizing the matching quality with the initial weight set of feature points as a guide, and outputting a feature point matching pair set;

[0158] S524, analyzing the matching quality of the feature point matching pair set, adaptively adjusting the feature point initial weight set, and outputting the adjusted feature weight set;

[0159] S525, based on the feature point matching pair set and the adjusted feature weight set, using a hierarchical spatial transformation model to estimate the spatial correspondence between different data sources, and output a multi-level spatial transformation model;

[0160] S526. Apply a multi-level spatial transformation model to transform the monitoring data from different sources into a unified reference coordinate system, perform weighted fusion according to the weights in the dynamic sensor weight matrix, and output a preliminary alignment result.

[0161] This embodiment achieves accurate spatiotemporal alignment of multi-source monitoring data. Compared with the traditional equal-weight registration method, this embodiment improves the accuracy of multi-source data fusion, and the registration error is reduced by more than 45%, especially in scenarios with large differences in sensor performance, the improvement is more significant. The feature matching framework based on the graph model improves the success rate of feature point matching, and even on the surfaces of modern buildings with sparse texture features, the matching success rate is improved by more than 30%. The weight adaptive adjustment mechanism ensures the continuous optimization of weight distribution during the registration process, enabling the system to dynamically adjust the strategy according to the matching quality, thereby improving the robustness and adaptability of the algorithm. In addition, the application of the hierarchical spatial transformation model enables the system to simultaneously process spatial transformations of different scales and characteristics, laying a technical foundation for the overall monitoring of large and complex buildings.

[0162] According to one aspect of the present application, the optimization of the preliminary registration result is further performed as follows:

[0163] Perform fine registration processing on the high stress areas marked in the material state partition map; read the material nonlinear response model, predict the impact of material nonlinear behavior on geometric deformation, and modify the registration strategy; in high stress areas, adjust the tolerance parameters of feature matching to allow larger local deformation; at the same time, increase the density of registration points to improve spatial resolution;

[0164] A method combining local weighted regression and radial basis function is used to accurately describe the spatial correspondence of nonlinear deformation areas and output fine registration results of high stress areas. The preliminary registration results and fine registration results of high stress areas are integrated to output further optimized registration results.

[0165] This embodiment solves the problem of insufficient accuracy of traditional registration methods in stress concentration areas. Compared with the unified parameter registration method, the registration accuracy of this embodiment in high stress areas is improved by 65%, reducing the error from millimeter level to sub-millimeter level, meeting the requirements of high-precision structural monitoring. The refined registration strategy enables the system to accurately capture local deformations in stress concentration areas, such as small changes in key locations such as the deformation field around cracks and stress concentration areas near supports, providing the possibility for early identification of structural abnormalities. In addition, by integrating the registration results of ordinary areas and high stress areas, this embodiment also ensures the continuity and rationality of the overall deformation field, avoids the boundary discontinuity problem caused by regional processing, and improves the overall quality and availability of monitoring data.

[0166] In one embodiment of the present application, spatial features of multi-source data are extracted. Time-synchronized data sets are read, and feature extraction is performed on satellite data, drone data, and ground monitoring data, respectively. A method combining scale-invariant feature transform (SIFT), speeded-up robust features (SURF), and deep learning-based feature extractors (such as LoFTR) is used to identify key points and descriptors in each data source. In view of the sparse texture features of building surfaces, edge features, corner features, and structural element features are added to improve the distinctiveness and robustness of features. For each feature point, feature reliability indicators are calculated, including feature response strength, feature uniqueness, and feature stability, and a multi-source spatial feature set is output.

[0167] Perform initial feature weight assignment. Read the dynamic sensor weight matrix and multi-source spatial feature set, and assign initial weights to each feature point. Weight calculation takes into account three factors: the basic weight of the sensor at that location (obtained from the dynamic sensor weight matrix), the reliability index of the feature point itself, and the material state of the area where the feature point is located (obtained from the material state partition map). For feature points in high stress areas, special consideration is given to the degree to which they are affected by the nonlinear behavior of the material, and the weights are adjusted appropriately. Normalization is used to ensure a reasonable weight distribution, and the initial weight set of the feature points is output.

[0168] Feature matching based on graph model. Construct a feature matching framework based on graph model to transform the feature matching problem into a graph matching problem. The nodes of the graph are feature points, and the edges represent the spatial relationship between feature points. The matching process considers node similarity (feature descriptor distance) and structural similarity (degree of spatial relationship preservation). The spectral graph matching algorithm is used in combination with RANSAC (random sampling consensus) to remove outliers. In the matching optimization, the initial weight set of feature points is used as a guide, the matching quality of high-weight feature points is given priority, and the feature point matching pair set is output.

[0169] Perform adaptive weight adjustment. Analyze the matching quality of the feature point matching pair set and adaptively adjust the initial weight set of the feature points. Matching quality indicators include: feature descriptor distance, spatial transformation consistency, and local deformation rationality. For points with low matching quality, reduce their weights; for points with high matching quality and in line with physical expectations, increase their weights. For areas where there is ambiguity in the matching relationship, introduce spatial smoothing constraints to avoid non-physical jumps in weights, and output the adjusted feature weight set.

[0170] Estimate hierarchical spatial transformation. Based on the feature point matching pairs and the adjusted feature weight set, a hierarchical spatial transformation model is used to estimate the spatial correspondence between different data sources. First, the global rigid body transformation is estimated using the weighted least squares method; second, based on the rigid body transformation, the affine transformation of each region is estimated; then, non-parametric transformations (such as thin plate spline transformation, TPS) are introduced to handle nonlinear deformations; finally, a local weighted regression model is used for local details. At each level, the adjusted feature weight set is fully utilized to guide the optimization of transformation parameters, biased towards accurate matching of high-weight features, and output a multi-level spatial transformation model.

[0171] Perform weight-guided spatial data fusion. Apply a multi-level spatial transformation model to transform monitoring data from different sources into a unified reference coordinate system. For each spatial location, a weight-based data fusion strategy is adopted: multi-source data at the same location are weighted averaged according to the weights in the dynamic sensor weight matrix. The complementarity of data sources is considered in the fusion process. Satellite data provides large-scale deformation information, drone data provides medium-precision regional information, and ground monitoring data provides high-precision local information. By adjusting the fusion weights, optimal data utilization is achieved in areas with different accuracy requirements, and preliminary fused data is output.

[0172] Perform fine registration of high stress areas. Perform fine registration processing for high stress areas (such as elastic-plastic transition zone and plastic zone) marked in the material state partition diagram. Read the material nonlinear response model, predict the impact of material nonlinear behavior on geometric deformation, and correct the registration strategy. In these areas, adjust the tolerance parameters of feature matching to allow larger local deformation; at the same time, increase the density of registration points to improve spatial resolution. A method combining local weighted regression and radial basis function is used to accurately describe the spatial correspondence of nonlinear deformation areas and output fine registration results of high stress areas.

[0173] Perform registration quality assessment and optimization. Perform quality assessment on the preliminary fusion data and the fine registration results of high stress areas. Evaluation indicators include: residual distribution, spatial continuity, and consistency with physical model predictions. Identify areas with poor registration quality and analyze the reasons (such as sparse features, extreme deformation, and sensor blind spots). For problem areas, adopt corresponding optimization strategies: add auxiliary feature points, adjust local transformation models, and introduce additional physical constraints. Through the iterative optimization process, continuously improve the registration quality, and finally output high-quality preliminary registration results, which provide a basis for the subsequent optimization of deformation gradient constraints.

[0174] Since the deformation gradient of the material state boundary region (such as the junction of the elastic region and the elastic-plastic region) may change rapidly, the existing registration method lacks a processing strategy for these special regions. Therefore, according to one aspect of the present application, step S53 is further:

[0175] S531, reading the preliminary registration result, using the finite difference method to calculate the deformation gradient field in the three-dimensional space, and outputting the initial deformation gradient field;

[0176] S532, read the material nonlinear response model and the structural stress distribution diagram, predict the deformation gradient distribution under ideal conditions based on the theory of continuous medium mechanics, and output the physical prediction gradient field;

[0177] S533, comparing the initial deformation gradient field with the physical prediction gradient field, calculating the difference distribution between the two, and generating a gradient difference distribution map and a difference area classification table;

[0178] S534. Based on the elastic-plastic mechanics theory, a deformation continuity constraint function is constructed, the function includes a gradient continuity term, a physical consistency term and a data fidelity term, and the deformation continuity constraint function is output;

[0179] S535. Design a multi-scale optimization strategy, gradually optimize the deformation field from coarse to fine, apply the variational method to minimize the objective function, and output the multi-scale optimized deformation field;

[0180] S536, integrating the multi-scale optimized deformation field and the above-mentioned processing results, generating a final deformation field registration solution, and outputting an accurate registration result.

[0181] This embodiment ensures that the registration results comply with the principle of physical continuity. Compared with the registration method based solely on data, this embodiment improves the physical rationality of the registration results by more than 70%, especially in areas where the data is sparse or of poor quality. The multi-scale optimization strategy improves the computational efficiency and stability of the algorithm, enabling the system to process complex large-scale monitoring data, reducing the calculation time by 55% while maintaining high accuracy. In addition, this embodiment can also effectively identify real structural anomalies and data anomalies, distinguish physical deformation from measurement errors, improve the reliability and practicality of the monitoring system, and provide more accurate data support for safety assessment and early warning of building structures.

[0182] According to one aspect of the present application, the precise registration result is also improved, specifically:

[0183] A special registration optimization strategy is designed for the state boundary area in the material state partition diagram; in the state boundary area, the deformation gradient may change rapidly but still maintain physical continuity; an adaptive mesh refinement technique is used to increase the density of computational nodes near the boundary; a transition layer model is introduced to smoothly connect the deformation behaviors of different state areas through an elastic-plastic transformation function; special consideration is given to the influence of nonlinear effects such as material softening and local buckling on the boundary area, and the optimization results of the boundary area are output; the optimization results of the boundary area and the aforementioned processing results are integrated to further improve the registration accuracy and perfect the precise registration results.

[0184] This embodiment solves the technical problem of difficult registration of state transition areas. Compared with the unified processing method, the registration accuracy of the state boundary area in this embodiment is improved by 60%, and the error is reduced from millimeter level to 0.2-0.3 mm, meeting the sub-millimeter accuracy requirements. The fine boundary processing enables the system to accurately capture the complex deformation behavior of the material state transition process, such as key physical phenomena such as yield extension and crack development, and provides high-quality data for structural behavior analysis. In addition, this embodiment also improves the overall continuity of the registration results, eliminates the artificial boundary effect between different state areas, improves the credibility and practicality of the monitoring data, and provides technical support for the fine health assessment of complex structures.

[0185] In one embodiment of the present application, a deformation gradient field is calculated. The preliminary registration result is read, and the deformation gradient field in three-dimensional space is calculated using the finite difference method. Specifically, a regular grid is constructed, the displacement difference between grid points is calculated, and the deformation rate per unit distance is normalized. The deformation characteristics in different directions are considered, and the main deformation direction and the main deformation amount are calculated. Special attention is paid to the continuity and physical rationality of the deformation gradient, and numerical noise is eliminated through spatial filtering technology to obtain a smooth initial deformation gradient field.

[0186] Construct a physical model to predict the gradient. Read the material nonlinear response model and the structural stress distribution diagram, and predict the deformation gradient distribution under ideal conditions based on the theory of continuum mechanics. The prediction process takes into account the nonlinear behavior of the material, especially in the elastic-plastic transition region and the plastic region, where the deformation gradient is no longer linearly related to the stress. For different material state regions, the corresponding constitutive equation is used to calculate the theoretical deformation response and output the physical prediction gradient field.

[0187] Analyze deformation gradient differences. Compare the initial deformation gradient field with the physically predicted gradient field and calculate the difference distribution between the two. Analyze the spatial pattern and statistical characteristics of the differences to identify abnormal areas (areas with significant differences) and consistent areas (areas with small differences). For abnormal areas, further classify them into possible physical anomalies (such as cracks, local buckling) and possible registration errors. Based on the analysis results, generate a gradient difference distribution map and a difference area classification table.

[0188] Construct a deformation continuity constraint function. Based on the theory of elastic-plastic mechanics, a deformation continuity constraint function is constructed. This function quantifies the degree to which the deformation gradient field deviates from physical expectations and contains three components: a gradient continuity term (ensuring smooth changes in the gradient field), a physical consistency term (ensuring consistency with material behavior), and a data fidelity term (ensuring consistency with observed data). The constraint strength is dynamically adjusted according to the material state of the region: in the elastic region, continuity and physical consistency are emphasized; in the elastic-plastic region, the continuity requirements are appropriately relaxed but physical rationality is maintained; in the plastic region or near the crack, the constraints are further relaxed to accommodate local mutations, and the deformation continuity constraint function is output.

[0189] Optimize multi-scale deformation constraints. Design a multi-scale optimization strategy to gradually optimize the deformation field from coarse to fine. In the coarse-scale stage, focus on large-scale deformation trends and use simplified models for rapid optimization; in the medium-scale stage, handle deformation transitions between regions to ensure continuity; in the fine-scale stage, fine-tune local details to meet high-precision requirements. At each scale, apply the variational method to minimize the objective function (including data terms and constraint terms), and iterate the solution through gradient optimization algorithms such as L-BFGS. During the optimization process, adjust the data item weights according to the dynamic sensor weight matrix to ensure the dominant role of high-reliability data and output a multi-scale optimized deformation field.

[0190] Special processing is performed on the material state boundary area. Special registration optimization strategies are designed for the state boundary areas in the material state partition diagram (such as the junction of the elastic area and the elastoplastic area). In these areas, the deformation gradient may change rapidly but still maintain physical continuity. Adaptive mesh refinement technology is used to increase the density of computational nodes near the boundary; a transition layer model is introduced to smoothly connect the deformation behaviors of different state areas through the elastoplastic transformation function. Special consideration is given to the influence of nonlinear effects such as material softening and local buckling on the boundary area, and the optimization results of the boundary area are output.

[0191] Construct strain energy minimization constraints. Introduce the principle of strain energy minimization as an additional physical constraint. According to the theory of elastic mechanics, under the action of external forces, the structure will spontaneously adjust to the state of minimum strain energy. Construct a strain energy functional that takes into account material nonlinearity and add it to the optimization objective as a regularization term. Solve through the variational method to find the deformation field configuration with the minimum strain energy under the premise of satisfying the observation data constraints. This constraint is particularly helpful for processing deformation inference in areas with sparse data, improving physical rationality, and outputting strain energy constraint optimization results.

[0192] Integrate and control the registration results. Integrate the multi-scale optimized deformation field, boundary area optimization results, and strain energy constraint optimization results to generate the final deformation field registration solution. Adopt a region-based weighted fusion strategy: in data-rich areas, the observation data is dominant; in data-sparse areas, the influence of physical constraints is increased; in special areas (such as high stress areas, material state boundaries), specially optimized results are used. Set quality control indicators, including residual statistics, physical consistency scores, and spatial continuity metrics, to ensure that the final registration results meet both observation accuracy and physical rationality requirements, and output high-quality optimized registration results.

[0193] According to one aspect of the present application, step S6 is further:

[0194] S61. Perform alignment quality assessment and problem area identification: Analyze the alignment accuracy report and correction residual assessment report, identify areas with poor alignment quality, and compare them with the structural stress distribution map and material state partition map to determine the cause of the problem and output the problem area report.

[0195] S62. Analyze parameter sensitivity: Perform sensitivity analysis on key parameters in the registration process, including material model parameters, weight calculation function parameters, and geometric transformation parameters, determine the parameters that have the greatest impact on the registration results, and generate a parameter sensitivity report.

[0196] S63. Perform adaptive parameter optimization: Based on the problem area report and parameter sensitivity report, adjust key parameters in a targeted manner, design an iterative optimization strategy, automatically update the parameters in the registration algorithm, and output the optimized configuration parameters.

[0197] S64. Perform structural anomaly and data anomaly detection: By comparing the precise alignment results with the expected structural behavior patterns and combining historical monitoring data, potential structural anomalies (such as abnormal deformation, crack development) and data anomalies (such as sensor failure, external interference) are identified and an anomaly detection report is generated.

[0198] S65. Dynamically adjust the monitoring strategy: According to the anomaly detection report and the registration accuracy report, dynamically adjust the monitoring strategy, including suggesting to increase the monitoring density in a specific area, adjust the monitoring frequency, or change the sensor type, etc., output monitoring optimization suggestions, and feed back the optimized configuration parameters to steps S2-S5 for iterative optimization.

[0199] Since the registration parameter optimization lacks an adaptive mechanism and cannot be dynamically adjusted according to the monitoring data characteristics and structural state, it is difficult to cope with complex and changeable monitoring scenarios. Therefore, according to one aspect of the present application, step S63 is further:

[0200] S631, quantify the sensitivity of each parameter in the registration algorithm, evaluate the influence of parameter changes on registration accuracy, and form a parameter sensitivity quantification table;

[0201] S632, expressing the parameter optimization problem as a multi-objective optimization problem, wherein the objective function includes registration accuracy, physical rationality, spatial continuity and computational efficiency, and outputting the parameter optimization problem definition;

[0202] S633, analyzing the correlation and interaction effects between optimization parameters, designing a parameter conversion scheme, converting the original parameter space into an orthogonal parameter space, and outputting a parameter correlation analysis report and a parameter conversion scheme;

[0203] S634. Design an adaptive gradient descent optimization strategy, use small step sizes to carefully adjust highly sensitive parameters, allow larger step sizes to accelerate convergence for less sensitive parameters, and output the gradient optimization strategy;

[0204] S635. Design a hybrid optimization strategy that combines global exploration with local optimization. Latin hypercube sampling and particle swarm optimization algorithms are used in the global exploration phase, and the quasi-Newton method is used in the local optimization phase. The hybrid optimization strategy is output.

[0205] S636: Automatically generate optimized configuration parameters for current monitoring data characteristics, and generate optimized configuration parameters.

[0206] This embodiment realizes automatic tuning and performance optimization of the registration algorithm. Compared with the fixed parameter method, this embodiment improves the average performance of the system in a variety of environments by more than 50%, especially under extreme conditions, such as strong noise environment, extreme weather or rapid structural changes, the improvement is more significant. The hybrid strategy combining global exploration with local optimization effectively avoids the problem of optimization falling into local optimality, and the global optimal solution achievement rate is increased by 65%, ensuring that the system can find the best configuration under various conditions. The automatic parameter generation function based on the current monitoring data characteristics lowers the threshold for the use of the system, allowing non-professionals to obtain professional-level monitoring results, and promoting the popularization and application of advanced monitoring technologies. In addition, the computational efficiency of this embodiment has also been improved, and the average time consumption of the optimization process has been reduced by 70%, enabling the system to complete parameter adjustment in a short time to meet the needs of quasi-real-time monitoring.

[0207] According to one aspect of the present application, before generating the optimized configuration parameters, it also includes constructing an online learning parameter optimizer, specifically:

[0208] Read the precise registration results, as well as the problem area report and parameter sensitivity report generated during the registration process; establish an online learning mechanism for parameter optimization so that the system can learn from historical optimization experience and continuously improve; create a parameter optimization knowledge base to store problem features, optimization trajectories and final results in historical optimization tasks; use case-based reasoning methods to find the most similar historical cases for new optimization tasks and extract valuable experience; as the system runs, continuously update the knowledge base to gradually improve optimization efficiency and success rate;

[0209] For new problems, an exploration mechanism is established to actively acquire knowledge and output an online learning parameter optimizer. The online learning parameter optimizer is used to generate optimized configuration parameters and feed them back to the aforementioned steps for iterative optimization.

[0210] This embodiment achieves continuous improvement and knowledge accumulation in the parameter optimization process. Compared with the static parameter optimization method, this embodiment improves the optimization efficiency and success rate, the parameter convergence speed is increased by 45%, and the optimization success rate is increased by 35%. The longer the system runs, the richer the accumulated knowledge and experience, and the more significant the performance improvement, which reflects the true "intelligent learning" feature. This embodiment is particularly suitable for long-term monitoring projects. With the accumulation of monitoring data, the system performance will continue to improve, without frequent manual intervention, reducing operation and maintenance costs and technical barriers. In addition, the establishment of a knowledge base also realizes the accumulation and sharing of experience, enabling different monitoring projects to learn from each other and promoting the improvement of the overall monitoring technology level.

[0211] In one embodiment of the present application, the sensitivity of key parameters is quantified. The problem area report and parameter sensitivity report are read, and a systematic method is established to quantify the sensitivity of each parameter in the registration algorithm. The coefficient of variation method and local sensitivity analysis method are used to evaluate the degree of influence of parameter changes on the registration accuracy. For each key parameter, a perturbation experiment is designed, a small amplitude perturbation is performed in the parameter space, and the change in the registration result is observed. The sensitivity index, including the relative influence coefficient and the sensitivity ranking, is calculated to form a parameter sensitivity quantification table.

[0212] Construct a multi-objective optimization problem. Formulate the parameter optimization problem as a multi-objective optimization problem. The objective functions include: registration accuracy (match with reference points), physical rationality (consistency with theoretical models), spatial continuity (smoothness of deformation field), and computational efficiency (algorithm running time). Set the objective weights for different application scenarios: monitoring accuracy priority mode, real-time response priority mode, and balanced mode. Construct constraints to ensure that the optimization parameters are within the valid range and output the parameter optimization problem definition.

[0213] Analyze parameter correlation. Analyze the correlation and interaction effects between optimization parameters. Use principal component analysis and partial correlation analysis to identify strongly correlated parameter groups and independent parameters. For strongly correlated parameters, design parameter conversion schemes to convert the original parameter space into an orthogonal parameter space to simplify the optimization problem. For parameter pairs with significant interaction effects, establish interaction models to describe the impact of joint parameter changes on the optimization target, and output parameter correlation analysis reports and parameter conversion schemes.

[0214] Construct a gradient descent optimization strategy. Based on the parameter sensitivity quantification table and parameter correlation analysis report, design an adaptive gradient descent optimization strategy. For highly sensitive parameters, use small step sizes for careful adjustment; for low-sensitivity parameters, allow larger step sizes to accelerate convergence. Gradient estimation uses numerical difference method to calculate the partial derivatives of the objective function with respect to each parameter. Step size adjustment uses an adaptive strategy: increase the step size after a successful iteration and decrease the step size after a failed iteration. For non-convex optimization problems, introduce momentum terms and random perturbations to avoid falling into local optimality, and output a gradient optimization strategy.

[0215] Combine global exploration with local optimization. Design a hybrid optimization strategy that combines global exploration with local optimization. In the global exploration phase, Latin hypercube sampling and particle swarm optimization algorithms are used to search extensively in the parameter space and identify potential advantageous areas. In the local optimization phase, quasi-Newton methods (such as the BFGS algorithm) are used to perform detailed searches in advantageous areas. The two phases are performed alternately, with global exploration providing good initial values ​​and local optimization improving convergence accuracy. During the optimization process, the resource allocation ratio of global and local searches is dynamically adjusted, and adaptive adjustments are made according to the progress of the search to output a hybrid optimization strategy.

[0216] Build an online learning mechanism. Establish an online learning mechanism for parameter optimization so that the system can learn from historical optimization experience and continuously improve. Create a parameter optimization knowledge base to store problem characteristics, optimization trajectories, and final results in historical optimization tasks. Use case-based reasoning methods to find the most similar historical cases for new optimization tasks and extract valuable experience. As the system runs, continuously update the knowledge base to gradually improve optimization efficiency and success rate. Pay special attention to new types of problems (such as material state combinations or load conditions that have never been encountered), establish an exploration mechanism to actively acquire knowledge, and output an online learning parameter optimizer.

[0217] Verify the adaptability of multi-scenario parameters. Design a variety of typical scenarios to verify the adaptability of parameter optimization. The scenarios include: normal monitoring scenarios (slow and uniform deformation), extreme event scenarios (rapid and non-uniform deformation), scenarios with significant material nonlinearity (large-scale plastic deformation), and sensor abnormality scenarios (partial data loss or reduced accuracy). For each scenario, run the optimization process and evaluate the results to verify the parameter adjustment capability and alignment quality. Analyze the optimization characteristics under different scenarios, identify common and special problems, and output a multi-scenario adaptability evaluation report.

[0218] Automatically generate and verify the optimized configuration. Based on the analysis results of the previous steps, automatically generate the optimized configuration parameters for the current monitoring data characteristics. The configuration includes: material model parameters, weight calculation function parameters, geometric transformation parameters and iterative control parameters. The generation process takes into account data characteristics (such as noise level, spatial distribution), structural state (such as stress level, deformation rate) and environmental conditions (such as temperature, humidity). Through the cross-validation method, evaluate the effectiveness and robustness of the generated parameters and make fine adjustments when necessary. Finally, output the optimized configuration parameters, and this parameter set will be fed back to steps S2-S5 for iterative optimization to continuously improve the registration quality and system performance.

[0219] The present invention establishes a complete set of multi-source monitoring data precision registration process through six organically connected key steps. Compared with the existing registration methods, the present invention introduces two core factors: material nonlinear behavior and structural stress state, and establishes a deep fusion of data registration and physical model. By modeling the nonlinear materials of the standardized monitoring data set, the geometric correction can accurately reflect the complex deformation behavior of the material under high stress; through the dynamic weight distribution of structural stress state perception, the reliability assessment problem of different monitoring equipment under changing conditions is solved; through the registration optimization of deformation gradient constraints, it is ensured that the registration results meet the physical continuity requirements. This method that combines physical model drive with data drive improves the registration accuracy, especially in areas where the nonlinear behavior of the material is significant, the registration accuracy is improved by 40%-60%. At the same time, the complete adaptive optimization mechanism enables the system to cope with complex and changeable monitoring scenarios, enhances the adaptability and robustness of the method, and provides more reliable data support for the safety monitoring of building structures.

[0220] The preferred embodiments of the present invention are described in detail above; however, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all belong to the protection scope of the present invention.

Claims

1. A precise registration method for building monitoring data based on spatiotemporal dynamic weights, characterized in that: The following steps are involved: Read satellite data, drone data and ground monitoring data, perform preprocessing and standardization, and generate standardized monitoring data sets; Analyze the deformation characteristics in the standardized monitoring data set, combine the characteristics of building materials, establish a material model that takes into account nonlinear behavior, and generate a material nonlinear response model and structural stress distribution map; According to the structural stress distribution diagram and the pre-stored sensor characteristics, a dynamic weight function is constructed to calculate the reliability weight of each sensor data in different stress areas and output the dynamic sensor weight matrix; The geometric deformation of the standardized monitoring data set is corrected by using the material nonlinear response model to obtain a nonlinear correction data set; Fusion of nonlinear correction data set and dynamic sensor weight matrix to generate accurate registration results; Based on the precise registration results, the parameters are adaptively optimized, possible data anomalies or structural anomalies are identified, and the optimized configuration parameters are generated and fed back for iterative optimization.

2. The method according to claim 1, characterized in that: The steps to establish a material model that takes nonlinear behavior into account and generate a material nonlinear response model include: Based on the standardized monitoring data set and building material characteristics, deformation data and initial material parameter sets are extracted, and the initial stress distribution field is calculated using the finite element inversion algorithm; Based on the initial stress distribution field and the standardized monitoring data set, multiple time-corresponding force-strain data pairs are extracted to form a regional stress-strain data set; Based on the regional stress-strain data set, a nonlinear constitutive model is constructed and the strain rate influencing factors are calculated; An environmental factor correction function is constructed by combining pre-stored environmental parameters, integrating the nonlinear constitutive model, strain rate influencing factor and environmental factor correction function, and generating a nonlinear stress-strain model; Verify the prediction accuracy of nonlinear stress-strain models, quantify model uncertainty, and output model reliability assessment reports and material nonlinear response models.

3. The method according to claim 2, characterized in that The steps to generate a structural stress distribution map include: Read the building structure design parameters and standardized monitoring data sets, divide the structural functional areas based on the structural functional characteristics and geometric forms, and obtain the structural functional zoning map; Analyze deformation characteristics in standardized monitoring data sets, identify areas with similar deformation characteristics, and generate deformation behavior zoning maps; Read the initial stress distribution field and nonlinear stress-strain model, calculate the stress state index, and generate a stress state classification diagram; determine the elastic-plastic state of the material in each region based on the material yield criterion, consider the differences in material properties, and output the material elastic-plastic state diagram; Analyze historical load data in standardized monitoring data sets, calculate cumulative plastic strain and fatigue damage, and generate cumulative damage distribution maps; The structural function zoning diagram, deformation behavior zoning diagram, stress state classification diagram, material elastic-plastic state diagram and cumulative damage distribution diagram are integrated, and the material state zoning diagram and structural stress distribution diagram are generated through multi-indicator evaluation.

4. The method according to claim 1, characterized in that: The steps of constructing a dynamic weight function and outputting a dynamic sensor weight matrix include: Analyze the accuracy characteristics of the sensor under different working conditions and build a sensor accuracy characteristic model; Read the structural stress distribution diagram and sensor accuracy characteristic model, analyze the reliability variation law of various sensors under different stress levels, establish a quantitative correlation function between stress level and sensor reliability, and output a stress-reliability mapping table; Based on the structural stress distribution map and the preset engineering safety assessment standards, identify the high stress areas and key load-bearing parts in the structure, set the monitoring priorities of different areas, and generate the structural area priority map; Integrate stress-reliability mapping table and structural area priority map to construct dynamic weight function; The dynamic weight function is applied to calculate the weight of each data point in the standardized monitoring data set to generate a dynamic sensor weight matrix covering the entire monitoring area and time period.

5. The method according to claim 4, characterized in that The steps to construct a dynamic weight function include: Read the stress-reliability mapping table, the structural area priority map and the sensor accuracy characteristic model, establish a multidimensional factor space that affects weight distribution, and obtain a weight influencing factor evaluation system; construct a weight optimization objective function to obtain a multi-objective weight optimization function; Read the structural stress distribution map and stress-reliability mapping table, construct a stress response function for each sensor type, and output a stress response function set; Construct constraints to ensure the spatial continuity of weight distribution and generate spatial continuity constraint functions; Analyze the deformation rate data in the standardized monitoring data set, build a time response mechanism for dynamic weight adjustment, and output a time response adjustment function; Construct a robust processing mechanism for abnormal data using weight functions and generate abnormal data processing functions; Integrate multi-objective weight optimization function, stress response function set, spatial continuity constraint function, time response adjustment function and abnormal data processing function to build a complete dynamic weight calculation framework and generate a dynamic weight function.

6. The method according to claim 5, characterized in that The steps of constructing a stress response function for each sensor type and outputting a set of stress response functions include: Read the structural stress distribution diagram and stress-reliability mapping table, and construct the stress response function for each sensor type in the form of piecewise function: in the elastic region, use a slowly varying S-shaped curve; in the elastic-plastic transition region, increase the function slope; in the plastic region, introduce an inflection point or saturation characteristic; The pre-stored measured data are fitted by nonlinear least square method to determine the stress response function parameters of various sensors and output the stress response function set.

7. The method according to claim 3, characterized in that The steps of performing geometric deformation correction to obtain a nonlinear correction data set include: Read the material nonlinear response model and material state partition diagram, and combine with structural mechanics theory to form a physics-based deformation mode library; Decompose the complex deformation field in the standardized monitoring data set into linear combinations and generate deformation mode decomposition results by combining with the deformation mode library; Read the material state partition diagram, parametrically express the material state of each area, and generate the material state parameter field; Combining the deformation mode decomposition results and the material state parameter field, a local deformation response function is constructed for each region to generate a set of regional deformation response functions; Analyze the interface conditions between adjacent regions, ensure the continuity and compatibility of the deformation field at the region boundary, and generate a set of interface compatibility conditions; Based on the set of regional deformation response functions and the set of interface compatibility conditions, a global load-deformation mapping matrix is ​​constructed; Integrate the regional deformation response function set, interface compatibility condition set and global load-deformation mapping matrix, build an integrated deformation prediction engine, and output the local deformation prediction function set; The standardized monitoring data set is geometrically corrected based on the local deformation prediction function set to obtain a nonlinear corrected data set.

8. The method according to claim 7, characterized in that The steps of constructing a local deformation response function for each region and generating a set of regional deformation response functions include: Combining the deformation modal decomposition results and the material state parameter field, special deformation response functions are constructed for different regions: for the elastic region, linear hyperelastic theory is used; for the elastic-plastic transition region, flow theory combined with hardening rules is used; for the plastic region, large deformation theory and plastic flow effect are considered; The extended finite element method is used to implement each dedicated deformation response function and output a set of regional deformation response functions.

9. The method according to claim 7, characterized in that: The step of correcting the geometric deformation of the standardized monitoring data set also includes: constructing a hierarchical geometric transformation operator, specifically: Read the reference point coordinates in the standardized monitoring data set, calculate the global rigid body transformation parameters, and output the global rigid body transformation matrix set; Read the material state partition map, calculate the local affine transformation matrix for each region, and output the regional affine transformation matrix set and the affine transformation residual map; Analyze the affine transformation residual graph, extract the nonlinear deformation components, output the nonlinear residual decomposition results and construct the radial basis function network; Read the material nonlinear response model and the local deformation prediction function set, generate the deformation behavior predicted by the physical model and use it as a constraint to optimize the radial basis function network parameters, and output the radial basis function parameter set; Construct a transformation hierarchy integration strategy to combine the global rigid body transformation matrix set, the regional affine transformation matrix set and the radial basis function parameter set into a complete transformation expression and output a hierarchical geometric transformation operator; The standardized monitoring data set is geometrically corrected using a hierarchical geometric transformation operator to obtain a nonlinear corrected data set.

10. The method according to claim 9, characterized in that The steps of optimizing the radial basis function network parameters and outputting the radial basis function parameter set include: Read the material nonlinear response model and local deformation prediction function set, and construct the optimization objective function under physical constraints, including data fitting terms, physical consistency terms, and smoothing regularization terms; The alternating direction multiplier method is used to solve the optimization problem of the optimization objective function, ensure that the radial basis function transformation is consistent with the material behavior, and output the radial basis function parameter set optimized by physical constraints.

Images