A building overall settlement calculation method and system based on multi-source data fusion

Abstract

The application provides a building overall settlement calculation method and system based on multi-source data fusion, relates to the technical field of building safety, and generates natural settlement data of the building under a non-reconstruction working condition through a recurrent neural network based on obtained building settlement data; a first building digital twin is constructed according to the spatial position relationship of the building structure, the settlement state of a key structure of the building is labeled in the first building digital twin according to real-time settlement data and natural settlement data, and a second building digital twin is obtained; the settlement evolution process of the building under a reconstruction working condition is simulated by using the second building digital twin according to the reconstruction construction procedure of the building, and settlement evolution data is obtained; and the overall settlement of the building is calculated based on the real-time settlement data, the natural settlement data and the settlement evolution data, so that high-precision dynamic calculation of the overall settlement of the building and identification of abnormal settlement of key positions are realized.

Description

Technical Field

[0001] This application relates to the field of building safety technology, and in particular to a method and system for calculating the overall settlement of buildings based on multi-source data fusion. Background Technology

[0002] With the increasing number of urban renewal and renovation projects of existing buildings, higher requirements are being placed on the dynamic monitoring of the structural safety status of buildings. In the process of renovation and construction of high-rise buildings, historical buildings, or sensitive areas along subway lines, accurately grasping the overall settlement trend of buildings is not only related to construction safety, but also directly affects the surrounding environment and residents' lives.

[0003] To address these needs, existing research has proposed a settlement analysis method that combines multi-source sensor data fusion with structural simulation models. This method acquires settlement information from different locations within a building using various monitoring devices, employs filtering techniques to perform spatiotemporal alignment and denoising of heterogeneous data, and inputs the processed time-series data into a pre-built physical simulation model. Material parameters are repeatedly adjusted to fit measured deformations, thereby inverting the overall settlement state. However, existing methods still have significant shortcomings. For example, their simulation models rely on pre-defined material properties and boundary conditions, making it difficult to accurately reflect nonlinear settlement behavior under long-term service and complex construction disturbances; data fusion focuses only on numerical consistency, neglecting spatial correlations and collaborative deformation mechanisms between components, resulting in insufficient response to anomalies in key areas; and model parameters require manual calibration and cannot automatically distinguish between natural settlement and construction impacts, limiting their predictive and early warning capabilities during dynamic renovation processes. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for calculating the overall settlement of buildings based on multi-source data fusion, so as to solve the problems of insufficient prediction accuracy and early warning capability in the prior art due to the model's reliance on preset parameters, neglect of spatial collaborative deformation, and inability to automatically separate settlement components.

[0005] To address the aforementioned technical problems, in a first aspect, this application provides a method for calculating the overall settlement of a building based on multi-source data fusion, comprising:

[0006] Acquire building settlement data, which includes historical settlement time series data and real-time settlement data;

[0007] Based on the building settlement data, natural settlement data of the building under no modification conditions is generated by a recurrent neural network.

[0008] Based on the spatial relationship of the building structure, a first building digital twin is constructed. Based on the real-time settlement data and natural settlement data, the settlement status of key building structures is marked in the first building digital twin to obtain a second building digital twin.

[0009] Based on the building renovation and construction procedures, the second building digital twin is used to simulate the settlement evolution process of the building under renovation conditions, and settlement evolution data is obtained.

[0010] Based on the real-time settlement data, natural settlement data, and settlement evolution data, the overall settlement of the building is calculated.

[0011] Optionally, based on the building settlement data, natural settlement data of the building under unmodified conditions is generated using a recurrent neural network, including:

[0012] The historical settlement time series data in the building settlement data is correlated and integrated with the geological and environmental parameters of the area where the building is located to form time-stamped sequence data;

[0013] The real-time settlement data from the building settlement data is added to the time-stamped sequence data to generate network input data;

[0014] The internal processing unit of the recurrent neural network processes the network input data, geological parameters, and environmental parameters to generate natural settlement data of the building under unmodified conditions.

[0015] Optionally, the network input data, geological parameters, and environmental parameters are processed by the internal processing unit of a recurrent neural network to generate natural settlement data of the building under unmodified conditions, including:

[0016] Based on the soil layer depth and the time stamp in the network input data, the geological parameters are split to obtain multiple sets of stratified geological parameters corresponding to different time nodes. At the same time, environmental time series parameters matching each time node are extracted from the environmental parameters.

[0017] Based on the settlement characteristics of key building structures, assign corresponding settlement impact weights to each key structure.

[0018] Based on the spatial distribution of each key structure, determine the spatial correlation parameters between key structures of adjacent buildings;

[0019] The network input data is integrated with the layered geological parameters, environmental time series parameters, subsidence impact weights, and spatial correlation parameters to form an extended input sequence;

[0020] The internal processing state of the recurrent neural network is initialized to obtain the initial processing state.

[0021] According to the time stamp order in the network input data, the sequence segment corresponding to the first time node in the extended input sequence is associated with the initial processing state to obtain the associated information;

[0022] The associated information is processed hierarchically by the internal processing unit of the recurrent neural network until all sequence segments corresponding to all time nodes in the extended input sequence have been processed, generating the final updated processing state. Based on the final updated processing state, the natural settlement data of the building under no modification conditions is generated by the recurrent neural network.

[0023] Secondly, this application provides a method and system for calculating the overall settlement of a building based on multi-source data fusion, including:

[0024] The acquisition module is used to acquire building settlement data, which includes historical settlement time series data and real-time settlement data;

[0025] The generation module is used to generate natural settlement data of the building under no modification conditions based on the building settlement data through a recurrent neural network.

[0026] The construction module is used to construct a first building digital twin based on the spatial relationship of the building structure, and to mark the settlement status of key building structures in the first building digital twin based on the real-time settlement data and natural settlement data, so as to obtain a second building digital twin.

[0027] The simulation module is used to simulate the settlement evolution process of the building under the renovation conditions using the second building digital twin, based on the building renovation construction procedures, and to obtain settlement evolution data.

[0028] The calculation module is used to calculate the overall settlement of the building based on the real-time settlement data, natural settlement data, and settlement evolution data.

[0029] Thirdly, this application provides an electronic device, comprising:

[0030] Memory, used to store computer programs;

[0031] A processor is configured to execute the computer program to implement the steps of a method for calculating overall building settlement based on multi-source data fusion as described in the first aspect above.

[0032] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the building settlement calculation method based on multi-source data fusion as described in the first aspect above.

[0033] This application provides a method for calculating overall building settlement based on multi-source data fusion. By acquiring building settlement data including historical and real-time information, a recurrent neural network is used to learn and generate the natural settlement trend under no-modification conditions, eliminating the interference of construction disturbance on settlement observation. On this basis, a digital twin is constructed by combining the spatial positional relationship of the building structure, and the measured and natural settlement states are mapped into the model to form an enhanced digital twin that can reflect the settlement differences of key components. The twin is driven by the actual modification process to simulate the settlement evolution process under different construction stages. Finally, multi-dimensional settlement information is integrated to realize the dynamic and systematic calculation of the overall building settlement, thereby improving the temporal continuity, spatial correlation and adaptability of settlement assessment.

[0034] Furthermore, by introducing hierarchical geological parameters, environmental temporal parameters, key structural settlement weights, and spatial correlation parameters into the recurrent neural network, an extended input sequence oriented towards spatiotemporal heterogeneity is constructed. Relying on the progressive update mechanism of the network's internal state, a refined modeling of the natural settlement process is achieved. This overcomes the problem that traditional physical simulation models, which rely on fixed material properties and boundary assumptions, are unable to characterize the nonlinear evolution of settlement under complex service environments. At the same time, it avoids the subjective bias caused by manual parameter tuning, enabling the natural settlement benchmark to have stronger environmental adaptability and structural perception capabilities, laying a reliable foundation for the subsequent accurate identification of construction disturbance effects. Attached Figure Description

[0035] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 A flowchart illustrating a method for calculating overall building settlement based on multi-source data fusion, provided for an embodiment of this application;

[0037] Figure 2 This is a schematic diagram illustrating the generation of natural settlement data using a recurrent neural network, as provided in an embodiment of this application.

[0038] Figure 3 This is a schematic diagram of a building settlement calculation system based on multi-source data fusion, provided as an embodiment of this application. Detailed Implementation

[0039] To address the shortcomings of existing settlement analysis methods, such as difficulty in accurately distinguishing between natural settlement and construction disturbance, lack of modeling capability for spatial collaborative deformation between building components, and insufficient dynamic adaptability due to reliance on manual parameter adjustment, this application utilizes a recurrent neural network to generate a natural settlement benchmark under unmodified conditions based on historical and real-time settlement data, thus avoiding dependence on preset physical parameters. Subsequently, a digital twin is established based on the spatial topological relationship of the building structure, mapping the measured and natural settlement states to key locations to form a settlement representation with spatial semantics. Furthermore, the model is driven by actual modification procedures to simulate the settlement evolution process under construction disturbance. Finally, through the collaborative calculation of multi-source settlement information, a dynamic and structurally perceptive assessment of the overall building settlement is achieved.

[0040] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0041] The core of this application is to provide a method for calculating the overall settlement of buildings based on multi-source data fusion, and a flowchart of one specific implementation method is shown below. Figure 1 As shown, the method includes:

[0042] Step 101: Obtain building settlement data, which includes historical settlement time series data and real-time settlement data.

[0043] In this step, historical settlement time series data refers to the settlement-related data recorded continuously in chronological order during the building's past service period. This historical settlement time series data includes settlement values ​​at different time points, data collection time markers, etc.

[0044] Real-time settlement data refers to building settlement data continuously collected in the current or recent time period. This real-time settlement data includes the settlement values ​​at the current moment and within a preset time interval.

[0045] In this embodiment, settlement monitoring equipment deployed at key parts of the building collects real-time settlement data. This data covers the current settlement values ​​of each key structure of the building. The collection frequency of the settlement monitoring equipment can be set according to the building type and renovation construction needs. For example, historical buildings in sensitive areas along subway lines can be collected at a frequency of once per hour. By retrieving the building's previous settlement monitoring files and operation and maintenance records, and organizing the acquired data in chronological order, historical settlement time series data including settlement values ​​at different historical stages and corresponding collection times can be obtained. For example, settlement data of the building's foundation columns collected once a month over the past five years can be retrieved.

[0046] Step 102: Based on the building settlement data, generate natural settlement data of the building under no-modification conditions using a recurrent neural network.

[0047] In this step, the natural settlement data under no modification conditions refers to the building settlement data caused solely by factors such as natural geological consolidation, long-term aging of building materials, and natural environmental changes, excluding the impact of modification construction. This natural settlement data includes predicted settlement values ​​at different future time points.

[0048] Step 103: Based on the spatial relationship of the building structure, construct a first building digital twin. Based on the real-time settlement data and natural settlement data, mark the settlement status of key building structures in the first building digital twin to obtain a second building digital twin.

[0049] In this step, the first building digital twin refers to a three-dimensional model constructed by integrating the building's structural information and geological information of the area where the building is located, based on the spatial relationship of the building's physical structure. Its structural form and geological distribution are consistent with the actual building situation.

[0050] The settlement status of a building's critical structure refers to the settlement of the parts of the building that play a decisive role in the overall structural safety. This settlement status includes the settlement value, the trend of settlement change, and whether the settlement exceeds the safety threshold.

[0051] The second building digital twin refers to a three-dimensional model of the building's key structures annotated with the settlement status of the first building digital twin. It integrates real-time settlement data and natural settlement data, and can intuitively present the current settlement status of the building's key structures.

[0052] Step 104: Based on the building renovation construction process, use the second building digital twin to simulate the settlement evolution process of the building under renovation conditions and obtain settlement evolution data.

[0053] In this step, settlement evolution data refers to the data recorded during the simulated building renovation construction process, showing how building settlement changes as the construction progresses.

[0054] Step 105: Calculate the overall settlement of the building based on the real-time settlement data, natural settlement data, and settlement evolution data.

[0055] In this step, the overall building settlement refers to the data that reflects the degree and trend of the overall building settlement, calculated by integrating real-time settlement data, natural settlement data, and settlement evolution data. It can be used to comprehensively assess the structural safety status of the building during the renovation and construction process.

[0056] In this embodiment, using natural settlement data as a benchmark, the deviation adjustment amount is first calculated based on the difference between real-time settlement data and natural settlement data. Then, this deviation adjustment amount is superimposed on the total construction impact amount reflected by the settlement evolution data onto the natural settlement data to obtain the overall building settlement, providing a direct basis for structural safety monitoring and early warning during the renovation construction process. The total construction impact amount is calculated as follows: the actual settlement value of the same key structure at each time node in the settlement evolution data is subtracted from the natural settlement data of that key structure at the corresponding time node to obtain the settlement impact amount of a single process. Finally, all single-process settlement impact amounts are superimposed in chronological order to obtain the total construction impact amount.

[0057] This application provides a complete temporal basis for separating natural and disturbed settlement by using historical and real-time settlement data; it improves the adaptability of the benchmark settlement; it constructs a digital twin that integrates spatial location relationships and marks the settlement status of key structures, strengthening the expression of coordinated deformation between components; it simulates the settlement evolution process in conjunction with the renovation process, enabling the model to have the ability to perceive the construction stage; and it realizes a systematic and dynamic assessment of building settlement trends under complex renovation scenarios.

[0058] This application provides a specific embodiment. Step 102 involves generating natural settlement data of the building under no-modification conditions using a recurrent neural network based on the building settlement data. This specifically includes the following steps:

[0059] Step 201: Link and integrate the historical settlement time series data in the building settlement data with the geological and environmental parameters of the area where the building is located to form time-stamped sequence data.

[0060] In this step, geological parameters refer to the geological characteristics data of the area where the building is located, including soil type, thickness, density, bearing capacity, etc.

[0061] Environmental parameters refer to data on environmental factors affecting settlement in the area where the building is located. These environmental parameters include annual precipitation, temperature changes, and groundwater level fluctuations.

[0062] Time-stamped sequence data refers to a time series data set formed by binding historical settlement time series data with geological and environmental parameters at corresponding time nodes. This sequence data includes time stamps, settlement values ​​at corresponding times, geological parameter values, and environmental parameter values.

[0063] In this embodiment, the historical settlement time series data is first organized to clarify the time node corresponding to each settlement value. Then, geological and environmental parameters of the building's location are extracted, and their changes are organized along the time dimension. Subsequently, historical settlement values, geological parameters, and environmental parameters at the same time node are linked together to form time-stamped sequence data, ensuring that each time marker is associated with complete settlement, geological, and environmental data.

[0064] Step 202: Supplement the real-time settlement data in the building settlement data into the time-stamped sequence data to generate network input data.

[0065] In this step, network input data refers to the complete dataset formed by supplementing real-time settlement data with time-stamped sequence data.

[0066] In this embodiment, the acquisition time node of the real-time settlement data is first determined, ensuring that the acquisition time node is consistent with the time format of the time-stamped sequence data. Then, the settlement value corresponding to the real-time settlement data and the acquisition time stamp are appended to the end of the time-stamped sequence data to obtain supplemented data, thus maintaining the temporal continuity of the data. Finally, the integrity of the supplemented data is checked to ensure that there is no missing or duplicate data, generating network input data covering historical and real-time multi-source data.

[0067] Step 203: The network input data, geological parameters, and environmental parameters are processed by the internal processing unit of the recurrent neural network to generate natural settlement data of the building under no modification conditions.

[0068] In this step, the internal processing unit refers to the core component in the recurrent neural network responsible for data processing, which includes a memory unit and a computation unit.

[0069] Optionally, such as Figure 2 As shown, step 203 involves processing the network input data, geological parameters, and environmental parameters through the internal processing unit of the recurrent neural network to generate natural settlement data of the building under unmodified conditions. This specifically includes the following steps:

[0070] Step 211: Based on the soil depth and the time stamp in the network input data, the geological parameters are split to obtain multiple sets of stratified geological parameters corresponding to different time nodes. At the same time, environmental time series parameters matching each time node are extracted from the environmental parameters.

[0071] In this step, soil depth refers to the vertical distribution depth of the underground soil layers in the area where the building is located, which includes the top and bottom depths of various types of soil layers.

[0072] A timestamp refers to a specific time point in time when data is collected or corresponding to a data point. This timestamp includes historical time points and real-time data time points.

[0073] Layered geological parameters refer to multi-dimensional geological data obtained by splitting geological-related parameters according to soil layer depth and time marker. These layered geological parameters include geological characteristic values ​​corresponding to different time nodes and different soil layer depths.

[0074] Environmental time series parameters refer to environmental data extracted from environmental-related parameters and arranged by time nodes. These environmental time series parameters include the environmental factor values ​​corresponding to each time node.

[0075] In this embodiment, the soil depth classification standard for the building area is first determined to clarify the depth range of each soil layer. Then, using the time stamps in the network input data as a benchmark, the geological parameters corresponding to each time node are split according to the soil depth range to obtain layered geological parameters, ensuring that each time node corresponds to multiple sets of geological data for different soil depths. Simultaneously, corresponding environmental time-series parameters are extracted from the environmental parameters in time stamp order to ensure that each time node has matching environmental data.

[0076] Step 212: Assign corresponding settlement influence weights to each key structure based on its settlement characteristics.

[0077] In this step, settlement characteristics refer to the sensitivity of the building's critical structure to factors affecting settlement, that is, the ease with which the critical structure will experience settlement changes when geological, environmental, and other factors change.

[0078] Settlement impact weight refers to the weight coefficient assigned to each key structure of a building based on its settlement characteristics. The larger the weight coefficient, the more significant the impact of the settlement state of the key structure on the overall safety of the building.

[0079] In this embodiment, the key structures of the building are first identified, such as load-bearing walls, foundation columns, and core tubes. The material properties, dimensions, stress distribution, and geological location of each key structure are analyzed to assess its settlement sensitivity to geological changes and environmental fluctuations. Then, a weighting coefficient is assigned to each key structure based on its settlement sensitivity, ensuring that the weighting is positively correlated with the settlement characteristics of the key structure.

[0080] Step 213: Determine the spatial correlation parameters between key structures of adjacent buildings based on the spatial distribution of each key structure.

[0081] In this step, the spatial distribution location refers to the three-dimensional coordinate position of the key building structure within the building entity, which includes both horizontal and vertical coordinates.

[0082] Spatial correlation parameters refer to parameters that quantify the degree of mutual influence of settlement between key structures of adjacent buildings. These spatial correlation parameters include the distance, location, and settlement transmission coefficient of adjacent structures.

[0083] In this embodiment, the spatial distribution data of each building's key structure is first obtained to determine the pairing relationship between adjacent building key structures; then, the straight-line distance between each pair of adjacent building key structures is calculated, and the positional relationship between the two is analyzed, such as horizontal adjacency or vertical superposition; finally, based on the straight-line distance, a preliminary settlement transmission coefficient is determined, wherein the closer the straight-line distance, the larger the settlement transmission coefficient. The preliminary settlement transmission coefficient is corrected in combination with the positional relationship, and finally, the spatial correlation parameters between each adjacent building key structure are obtained.

[0084] The preliminary calculation formulas for the settlement conduction coefficient and spatial correlation parameters are as follows:

[0085] (1)

[0086] in, α is the preliminary settlement conduction coefficient; a larger value indicates a stronger conduction effect. k is a proportional calibration constant, pre-calibrated according to the building structure type and geological conditions. For example, k can be set to 5.0 for a Class A frame structure in a dense clay layer, and k can be set to 3.5 for a Class B brick-concrete structure in a loose sand layer. The calibration is based on historical monitoring data of settlement conduction of similar buildings. d is the straight-line distance between adjacent key structures, in meters, calculated through the spatial distribution of the key structures, i.e., the Euclidean distance of the coordinate difference between two points. K is a spatial correlation parameter. α is a positional correction coefficient, determined according to the positional relationship of adjacent key structures. For example, when adjacent key structures are horizontally adjacent, α is 1.0; when adjacent key structures are vertically stacked, α is 1.2; when adjacent key structures are obliquely adjacent, α is 0.8. This embodiment does not limit the value of α; it can be set according to the actual situation.

[0087] Step 214: Integrate the network input data with the layered geological parameters, environmental time series parameters, settlement influence weights, and spatial correlation parameters to form an extended input sequence;

[0088] In this step, the extended input sequence refers to the comprehensive data sequence formed by integrating network input data, layered geological parameters, environmental time series parameters, subsidence influence weights, and spatial correlation parameters.

[0089] In this embodiment, the time stamp of the network input data is used as a unified benchmark. Layered geological parameters are correlated with the network input data according to time nodes. Then, environmental time-series parameters, settlement impact weights of each key structure, and spatial correlation parameters of adjacent key structures are sequentially matched to the corresponding time nodes and key building structures. Finally, the data after the above correlation and matching are organized in chronological order to form an extended input sequence, ensuring that each time node and each key structure is associated with complete multi-dimensional data.

[0090] Step 215: Initialize the internal processing state of the recurrent neural network to obtain the initial processing state;

[0091] In this step, the internal processing state refers to the operating state of the internal processing units of the recurrent neural network, which includes the initial values ​​of the memory units, the parameter configuration of the computation units, etc.

[0092] The initial processing state refers to the initial settings made to the internal processing state of the recurrent neural network before it begins to process data. This initial processing state includes setting the memory unit values ​​to initial values ​​and loading the trained computational parameters.

[0093] In this embodiment, the initialization module of the recurrent neural network is first invoked to reset the values ​​of the internal memory units to preset initial values, ensuring that they are not affected by residual data processing from the previous process. Then, pre-trained computational unit parameters are loaded, including data fusion weights and temporal feature extraction coefficients, enabling the processing unit to perform data processing. Finally, the completeness and rationality of the parameters in the internal processing state are checked, and the initial processing state is obtained after confirming that everything is correct.

[0094] Step 216: According to the time stamp order in the network input data, associate the sequence segment corresponding to the first time node in the extended input sequence with the initial processing state to obtain the associated information;

[0095] In this step, the associated information refers to the comprehensive information formed after binding the sequence fragment of the first time node with the initial processing state. This associated information includes the parameters of the initial processing state and the multi-dimensional data of the first time node.

[0096] In this embodiment, the extended input sequence is first sorted according to the time stamp order of the network input data to determine the sequence segment corresponding to the first time node. Then, all data in this sequence segment is extracted, including settlement values, geological parameters of each soil layer, environmental parameters, key structural weights, and spatial correlation parameters. Finally, this data is bound and fused with the memory unit values ​​and calculation parameters of the initial processing state to obtain the correlated information, ensuring that the processing unit can utilize both the initial state and the data from the first time node simultaneously.

[0097] Step 217: The associated information is processed hierarchically by the internal processing unit of the recurrent neural network until all sequence segments corresponding to all time nodes in the extended input sequence are processed, generating the final updated processing state. Based on the final updated processing state, the natural settlement data of the building under no modification conditions is generated by the recurrent neural network.

[0098] In this step, the final update processing state refers to the final running state of the internal processing unit after the recurrent neural network has processed all the sequence segments at all time points. This final update processing state includes the feature memory of all data, the accumulation of calculation results, etc.

[0099] In this embodiment, the temporal feature extraction module of the internal processing unit extracts the settlement change features and geological environment impact features of the first time node from the associated information. Then, the multi-parameter fusion module fuses these two types of features with the settlement impact weight and spatial correlation parameters to obtain the intermediate settlement prediction result for that time node, and updates the internal processing state to obtain the updated processing state for that time node.

[0100] Next, following the time-stamped order, the sequence fragment of the next time node is associated with the updated processing state. This hierarchical processing process is repeated until the sequence fragments of all time nodes have been processed, generating the final updated processing state. Finally, based on the feature memory and cumulative calculation results in the final updated processing state, the settlement values ​​of each key structure at different future time nodes are generated through the output module of the recurrent neural network, obtaining the natural settlement data under the unmodified condition.

[0101] This application's embodiments improve the accuracy of natural settlement prediction by analyzing the temporal changes in geology and environment and the synergistic effects between key structures. It solves the problems of existing methods failing to reflect nonlinear settlement behavior and neglecting spatial correlation, and supports settlement differentiation and safety assessment during dynamic transformation processes.

[0102] This application provides a specific embodiment. Step 103 involves constructing a first digital twin of the building based on the spatial relationship of the building structure, specifically including the following steps:

[0103] Step 301: Collect information on the building's physical structure and the geological information of the area where the building is located. Combine this with the spatial relationship of the building structure to construct a three-dimensional model framework that is consistent with the physical structure of the building.

[0104] In this step, the entity structure information refers to the data set that reflects the structural characteristics of the building entity. This information includes the size, material type, connection method of the building components, and distribution information of key structures.

[0105] Geological information refers to the underground geological distribution data of the area where the building is located. This geological information includes soil layer type, thickness of each soil layer, geological bearing capacity, and coordinates of the layer boundary.

[0106] The spatial relationship of a building structure refers to the relative position and coordinate relationship of each component in three-dimensional space. This spatial relationship includes the horizontal coordinates, vertical elevation, and spacing between the components and adjacent components.

[0107] A three-dimensional model framework refers to a three-dimensional shell model constructed based on the entity's structural information, geological information, and spatial relationships. This three-dimensional model framework includes the three-dimensional outline of the building entity, the spatial occupancy of each component, and the three-dimensional layered outline of the geological region, but does not include detailed parameter information.

[0108] In this embodiment, the dimensions, materials, and connection methods of each building component are collected, and missing structural information is supplemented by combining it with architectural design drawings. Next, the geological survey report of the building's location is retrieved to extract geological information such as soil layer type, thickness, and bearing capacity. The accuracy of key geological data is verified through on-site drilling. Then, using 3D modeling technology, a 3D outline of the building's physical structure is first constructed in a virtual environment based on the spatial relationships of the building structure to determine the spatial coordinates and relative positions of each component. A 3D layered outline of the geological area is then constructed, and these two types of outlines are combined to form a preliminary 3D model framework. Finally, the consistency between this preliminary 3D model framework and the physical structure is checked, and any spatial coordinate deviations are corrected, ultimately forming a 3D model framework consistent with the physical structure of the building.

[0109] Step 302: Classify and integrate the entity structure information according to the spatial position relationship of the building structure to form a building structure information set, and integrate the geological information information layer by layer according to the soil layer depth order to form a geological layer information set.

[0110] In this step, the building structure information set refers to the structured data set formed after classifying and integrating the spatial positional relationships of the building structure. The building structure information set includes subsets divided by floor, area or component type. Each subset includes the size, material, connection method and spatial coordinates of the corresponding component.

[0111] Soil depth sequence refers to the vertical distribution order of underground soil layers from the surface to the deepest underground layers.

[0112] A geological stratification information set refers to a data set formed by integrating geological information in stratification according to soil layer depth. This geological stratification information set includes the type, thickness, bearing capacity, stratification boundary coordinates, and physical property parameters of the soil layer.

[0113] In this embodiment, based on the spatial relationship of the building structure, a primary category is divided by floor, and a secondary category is divided by the functional area of ​​each floor. The component dimensions, materials, and other data in the physical structure information are correspondingly assigned to each category, forming a building structure information set, ensuring that each category is associated with complete component information. Simultaneously, according to the soil depth from shallow to deep, geological information for each soil layer is extracted sequentially, establishing an independent data subset for each soil layer, and marking the layer boundary coordinates of that soil layer, forming a geological layer information set, ensuring that the subset corresponds to the actual soil layer distribution.

[0114] Step 303: Bind the information of each component in the building structure information set to the building structure area of ​​the three-dimensional model frame according to the spatial position relationship of the building structure, and bind the information of each soil layer in the geological stratification information set to the geological area of ​​the three-dimensional model frame according to the soil layer depth order, to form the first building digital twin.

[0115] In this step, the component information refers to the specific component data included in each subset of the building structure information set. This component information includes the component's size, material, connection method, spatial coordinates, and structural function.

[0116] The architectural structure area of ​​the 3D model frame refers to the virtual area corresponding to the building entity structure in the 3D model frame. This architectural structure area includes the 3D outline area of ​​each component, the floor area, and the functional zoning.

[0117] Soil layer information refers to the specific geological data included in each subset of the geological stratification information set. This soil layer information includes soil layer type, thickness, bearing capacity, and physical property parameters.

[0118] The geological region in the 3D model framework refers to the virtual region in the 3D model framework that corresponds to the underground geological distribution. This geological region includes the 3D layered outline region of each soil layer.

[0119] The first building digital twin refers to the complete digital model formed by binding the information of each component and each soil layer to the corresponding area of ​​the three-dimensional model framework. The digital twin includes detailed parameters of the building entity structure and detailed distribution of the geological environment.

[0120] In this embodiment, based on the spatial relationship of the building structure, component outlines corresponding to each subset of the building structure information set are found in the building structure region of the 3D model frame. The component dimensions, materials, connection methods, and other information in each subset are bound to the corresponding outline, ensuring that each building structure region is associated with complete component parameters. Simultaneously, based on soil depth, soil layer outlines corresponding to each subset of the geological stratification information set are found in the geological region of the 3D model frame. Soil layer types, bearing capacity, and other information in each subset are bound to the corresponding outline, ensuring that each geological region is associated with complete soil layer parameters. Finally, the accuracy of the information binding is checked, and parameter mismatches or coordinate deviations are corrected, thereby forming a first building digital twin that includes complete building and geological information.

[0121] For example, taking a historical building along the subway line in area A as an application scenario, the dimensions of the building's load-bearing walls and foundation columns are first collected using a laser rangefinder. Combined with the original architectural design drawings, the material of the components is confirmed to be brick and concrete. Spatial coordinates of each component are obtained through on-site surveying, forming the physical structural information. Then, the geological survey report of the area where the building is located is retrieved to determine that the underground soil layers, from shallow to deep, are miscellaneous fill, clay, and sand. The thickness and bearing capacity of each soil layer are extracted, and the thickness data of the clay layer is verified through on-site drilling, forming geologically relevant information.

[0122] Using 3D modeling technology, a 3D outline of the building is constructed based on the spatial relationships of its components, and a layered outline is constructed based on geological distribution. These two types of outlines are combined to form a 3D model framework. The structural information is divided into first-floor and second-floor subsets, and each floor is further grouped into load-bearing and non-load-bearing zones, forming a set of structural information. Information on miscellaneous fill, clay, and sand is integrated into subsets according to soil depth, forming a set of geological stratification information. Finally, the component information of each floor is bound to the structural area of ​​the 3D model framework, and the soil layer information is bound to the geological area of ​​the 3D model framework, forming a first digital twin of the historical protected building and its geological environment.

[0123] This application embodiment restores the spatial relationship between the building's physical structure and the geological environment, avoiding the problem of existing models being disconnected from the actual structure and geology. It provides an accurate digital carrier for subsequent annotation of settlement status, supports the spatial accuracy of settlement simulation under subsequent renovation conditions, and improves the overall reliability of building settlement analysis.

[0124] This application provides a specific embodiment. Step 103 involves marking the settlement status of key building structures in the first building digital twin based on the real-time settlement data and natural settlement data, thereby obtaining a second building digital twin. This specifically includes the following steps:

[0125] Step 311: Extract the spatial coordinates and structural identification information of the key structures of the building from the first building digital twin.

[0126] In this step, the spatial coordinates of the building's critical structures refer to the three-dimensional coordinate data of the parts in the first building digital twin that play a decisive role in the overall safety of the building in virtual space.

[0127] Structural identification information refers to the unique identification information assigned to each key structure in the first building digital twin. This structural identification information includes the key structure type, the floor to which it belongs, and its functional attributes.

[0128] The first building digital twin refers to a three-dimensional digital model that integrates the building's structural information with geological stratification information.

[0129] In this embodiment, key building structures are identified in the first building digital twin, and non-critical components, such as non-load-bearing partitions and decorative components, are excluded using model filtering functions. Then, the spatial coordinates of each key building structure are extracted from the model database. Subsequently, the attribute information of each key building structure is extracted, and unique structural identifier information is generated for each key building structure according to the rule of structure type-floor-serial number.

[0130] Step 312: Associate the real-time settlement data with the structural identification information to obtain first settlement association data, and associate the natural settlement data with the structural identification information to obtain second settlement association data.

[0131] In this step, the first settlement correlation data refers to the data set formed by binding real-time settlement data with corresponding structural identification information. This first settlement correlation data includes building structure identification, real-time settlement value, and data acquisition time. The second settlement correlation data refers to the data set formed by binding natural settlement data with corresponding structural identification information, including building structure identification, natural settlement value, and prediction time node.

[0132] In this embodiment, the objects from which the real-time settlement data is collected are first analyzed to identify the actual key structures corresponding to each settlement value. Then, based on the description of the structure type-location in the structure identification information, the real-time settlement values ​​are bound to the building structure identifications of the corresponding actual key structures, and the integrity of the data is checked to ensure that no structure identifications are omitted or data mismatches are made, thus forming the first settlement association data. Following the same logic, the predicted settlement values ​​of each key building structure in the natural settlement data are bound to the corresponding building structure identifications to form the second settlement association data.

[0133] Step 313: Based on the presentation requirements of the building's key structures, the first settlement correlation data, and the second settlement correlation data, as well as the visualization characteristics of the first building digital twin, determine the labeling rules for the settlement status.

[0134] In this step, the presentation requirements refer to the information display requirements when showing the key structure of the building, the first settlement correlation data, and the second settlement correlation data. These presentation requirements include whether to display settlement values, whether to distinguish settlement types, and whether to mark settlement trends.

[0135] The visualization features refer to the information display capabilities supported by the first building digital twin. These visualization features include functions such as color differentiation, icon labeling, numerical overlay, and trend curve drawing.

[0136] The labeling rules for settlement status refer to the specific specifications for labeling settlement information in the first building digital twin, which are formulated in combination with presentation requirements and visualization characteristics. The labeling rules include the correspondence between data types and labeling forms, the meaning of colors / icons, and the numerical display format.

[0137] In this embodiment of the application, the requirements are first clearly presented based on the core objectives of building settlement monitoring and safety assessment, specifically including: in order to accurately distinguish settlement data from different sources, it is necessary to classify real-time and natural settlement types; in order to intuitively grasp the degree of settlement, it is necessary to display specific settlement values; in order to identify risks in a timely manner, it is necessary to indicate whether the settlement exceeds the safety threshold.

[0138] Based on the clearly defined presentation requirements, we further analyzed the visualization characteristics of the first building digital twin. Specifically, we determined which characteristics could meet the requirements and ultimately confirmed that its supported color gradient annotations could be used to distinguish settlement types, numerical floating displays could be used to present specific values, and icon warning functions could be used to indicate that safety thresholds were exceeded, thus providing functional support for subsequent rule formulation.

[0139] Subsequently, the labeling rules were detailed based on the matching of needs and characteristics: In order to distinguish between real-time and natural settlement, and in combination with the color gradient characteristics, real-time settlement data was labeled with blue and natural settlement data with green, and both types of data were displayed by floating numerical values ​​to meet the need to display specific values; In order to label safety risks, and in combination with the icon warning characteristics, a red warning icon was set to be superimposed when the settlement exceeds the safety threshold; At the same time, settlement trend labeling was added, and the line drawing function supported by the model was used to show the rising or falling trend of settlement through short broken lines.

[0140] Finally, the adaptability was verified with the core question of whether the rules could be implemented: the visualization features corresponding to each rule were checked one by one to ensure that all annotation forms could be stably implemented in the first building digital twin without any functional conflicts or inability to be presented, thus forming a complete and feasible settlement status annotation rule.

[0141] Step 324: According to the labeling rules of the settlement state, the first settlement association data and the second settlement association data are respectively labeled to the model positions in the first building digital twin corresponding to the spatial position coordinates, to form the second building digital twin.

[0142] In this step, the model location refers to the virtual structural part in the first building digital twin that corresponds to the spatial location coordinates.

[0143] The second building digital twin refers to a three-dimensional digital model based on the first building digital twin, annotated with the first and second settlement correlation data.

[0144] In this embodiment, the spatial coordinates of the corresponding key structure are first found in the first building digital twin based on the structural identification information in the first settlement association data, and the model position is determined; then, according to the settlement status labeling rules, blue labels and real-time settlement values ​​are superimposed on the model position.

[0145] Following the same logic, the corresponding model location is found based on the second settlement correlation data, and green labels and natural settlement values ​​are superimposed. If the settlement of a key structure exceeds the safety threshold, a red warning icon is superimposed on the corresponding model location. Finally, the accuracy and clarity of all labels are checked to ensure that there are no misaligned labels or duplicate information, and finally integrated to form the second building digital twin.

[0146] This application's embodiments achieve precise binding of settlement data with key structures, enabling the differentiation between real-time and natural settlement states. It solves the problem of insufficient display of settlement response in key parts by existing methods, providing a digital carrier with a benchmark settlement state for subsequent settlement simulation of modification conditions, and supporting intuitive comparison and safety assessment of settlement changes during the modification process.

[0147] This application provides a specific embodiment. Step 104 involves simulating the settlement evolution process of the building under renovation conditions using the second building digital twin, based on the building's renovation construction sequence, to obtain settlement evolution data. This specifically includes the following steps:

[0148] Step 401: Determine the construction influencing factors corresponding to the building renovation construction procedures, and extract the basic data for settlement simulation from the second building digital twin.

[0149] In this step, construction influencing factors refer to various factors that may cause changes in building settlement during the renovation construction. These factors include construction type, magnitude of force, scope of work, and type of material replacement.

[0150] The basic data for settlement simulation refers to the core data extracted from the second building digital twin for settlement simulation. This basic data includes spatial correlation parameters of key building structures, settlement influence weights, component material properties, geological stratification parameters, etc.

[0151] In this embodiment, a complete list of construction procedures for building renovation is first integrated to clarify the construction content of each procedure. Then, factors that may affect settlement are analyzed for each procedure to determine the corresponding construction influencing factors. Subsequently, the database of the second building digital twin is accessed to extract basic data required for settlement simulation, such as spatial correlation parameters, settlement influence weights, component material properties, and geological stratification parameters.

[0152] Step 402: Establish the correspondence between the renovation construction process and each model area in the second building digital twin. Based on the correspondence, determine the target model area corresponding to the work scope of each renovation construction process.

[0153] In this step, the corresponding association relationship refers to the mapping relationship that binds the renovation construction process to a specific model area in the second building digital twin.

[0154] In this embodiment, the scope of work for each renovation construction process is analyzed to identify the building components, floors, or geological areas to which it applies. Then, a model region with a spatial location consistent with the scope of work is located within the second building digital twin, and its identifier is determined. Subsequently, a correspondence between the renovation construction process and the model region identifier is established to ensure that each process is accurately mapped to one or more unique model regions. Finally, based on this correspondence, the model regions corresponding to the scope of work for each renovation construction process are selected as target model regions.

[0155] Step 403: Based on the construction influencing factors, assign working condition simulation parameters to each modification construction process. The working condition simulation parameters include the force application method, material property parameter adjustment rules, and simulation time span.

[0156] In this step, the working condition simulation parameters refer to the set of parameters used to simulate construction working conditions in the second building digital twin.

[0157] The method of applying force refers to the type of force applied to building components or geological areas during construction.

[0158] The material property parameter adjustment rules refer to the adjustment standards for material parameters in the second building digital twin when materials are replaced or modified during construction.

[0159] The simulated time span refers to the simulated time length corresponding to a single renovation and construction process.

[0160] In this embodiment of the application, the construction influencing factors of each modification construction process are analyzed to determine the corresponding force application method. For example, if the construction influencing factors of the structural reinforcement process are pressure application and high-strength material replacement, then the corresponding force application method is to apply uniform pressure to the reinforcement component.

[0161] Next, based on the type of material replacement, rules for adjusting material property parameters are formulated. For example, the original concrete strength parameters are increased to the strength parameters of the reinforcing material. Then, referring to the estimated operation time of this modification construction procedure in the construction plan, the simulation time span is determined. For example, if the actual construction takes 7 days, the simulation time span is set to 7 days. Finally, the force application method, material property parameter adjustment rules, and simulation time span are integrated into the working condition simulation parameters for this procedure.

[0162] Step 404: Based on the working condition simulation parameters and the basic data of the settlement simulation, use the second building digital twin to simulate the settlement evolution process of the building under the renovation working condition in each target model area, and record the actual settlement values ​​of each building key structure and the corresponding time nodes after the simulation of each renovation construction procedure is completed, forming stage settlement data.

[0163] In this step, the "modification working condition" refers to the mechanical environment and construction conditions of the building during the simulated modification and construction process.

[0164] The settlement evolution process refers to the dynamic process of settlement gradually changing with the construction progress under the condition of building renovation. This process includes the generation, transmission and accumulation of settlement in key structures.

[0165] Phased settlement data refers to a set of data that records the actual settlement values ​​of each key structure of a building after the completion of a single renovation construction process and the corresponding time nodes. This phased settlement data includes the process name, time node, key structure identifier, and actual settlement value.

[0166] Step 405: Integrate all the phased settlement data to form settlement evolution data.

[0167] In this embodiment, the phased settlement data of all renovation construction processes are sorted according to the time nodes corresponding to each renovation construction process. Then, the actual settlement values ​​of all key structures at the same time node are integrated, and the actual settlement values ​​of the same key structure at different time nodes are arranged in order to obtain the sorted data. Finally, the consistency of the sorted data is verified, and after the consistency verification is passed, a settlement evolution data covering the entire renovation process and with complete time sequence is formed.

[0168] Optionally, step 404 involves using the second building digital twin to simulate the settlement evolution process of the building under the renovation conditions in each target model area based on the simulation parameters and the basic data of the settlement simulation, and recording the actual settlement values ​​and corresponding time points of each key structure of the building after the simulation of each renovation construction procedure, thus forming staged settlement data. This specifically includes the following steps:

[0169] Step 411: Based on the construction type in the construction influencing factors and the basic data of the settlement simulation, generate the adaptation parameters of the construction type and the geological stratification response parameters.

[0170] In this step, the adaptation parameters for the construction type refer to the simulation parameters optimized for a specific construction type. These adaptation parameters include the construction force coefficient, material adaptation coefficient, etc.

[0171] Geological stratification response parameters refer to the response coefficients of different soil layers to construction actions. These parameters include soil layer settlement sensitivity and mechanical transmission coefficient.

[0172] In this embodiment, the construction type of each modification construction procedure is extracted from the construction influencing factors. The construction type includes structural reinforcement, component demolition, etc. Then, based on the basic data of settlement simulation, the required parameters for each construction type are analyzed to generate the appropriate parameters for that construction type. For example, the structural reinforcement type requires the generation of material strength enhancement coefficient and pressure adaptation coefficient, while the component demolition type requires the generation of unloading coefficient and component removal adaptation coefficient.

[0173] The material strength enhancement coefficient is obtained by dividing the strength parameters of the new material by the strength parameters of the original material. The strength parameters of the new material refer to the strength parameters corresponding to the new material used in the reinforcement construction, which are obtained from the construction plan or material specifications. The strength parameters of the original material refer to the material strength parameters of the original building components, which are obtained from the component material properties in the basic data of settlement simulation. The pressure adaptation coefficient is obtained by dividing the applied pressure value in the construction plan by the ultimate bearing capacity of the component or the geological conditions. The applied pressure value in the construction plan refers to the magnitude of the pressure to be applied to the target area during construction, which is obtained from the magnitude of the forces in the construction influencing factors. The ultimate bearing capacity of the component or the geological conditions refers to the maximum bearing value of the component or the geological conditions in the target model area, which is obtained from the component bearing parameters or geological bearing capacity parameters in the basic data of settlement simulation.

[0174] The unloading coefficient is obtained by dividing the load value borne by the demolished component by the total load of the original structure in the area. The load value borne by the demolished component refers to the size of the demolished component itself and the additional load it bears, which comes from the load action attribute in the basic data of settlement simulation. The total load of the original structure in the area refers to the sum of the overall structural loads in the area where the demolished component is located, which comes from the load action attribute in the basic data of settlement simulation. The component removal adaptation coefficient is obtained by dividing the stiffness parameter of the demolished component by the total stiffness of the structure in the area. The stiffness parameter of the demolished component refers to the bending and compressive stiffness values ​​of the demolished component itself, which comes from the component material properties in the basic data of settlement simulation. The total stiffness of the structure in the area refers to the sum of the stiffness of all structural components in the area where the demolished component is located, which comes from the component material properties and spatial correlation parameters in the basic data of settlement simulation.

[0175] Based on the geological stratification parameters, the geological stratification response parameters corresponding to each soil layer are calculated. The specific calculation formula is as follows:

[0176] (2)

[0177] in, For the first The geological stratification response parameter of the soil layer indicates that the soil layer responds more significantly to the current construction type. β is the geological influence coefficient corresponding to the construction type, which is determined according to the construction type and calibrated based on construction influencing factors and engineering experience. For example, the geological influence coefficient is 1.2 when the construction type is pressure construction, 0.8 when the construction type is unloading construction, and 1.5 when the construction type is foundation treatment. This embodiment does not limit the size of the geological influence coefficient, and it can be set according to the actual situation. For the first The compressibility modulus of the soil layer reflects the soil layer's ability to resist compressive deformation and comes from the geological parameter attributes in the basic data of the settlement simulation; μ is the geological correction coefficient, adjusted according to the soil layer homogeneity and comes from the geological stratification parameters in the basic data of the settlement simulation.

[0178] Step 412: Based on the adaptation parameters of the construction type, the geological stratification response parameters, and the working condition simulation parameters, adjust the parameter attributes corresponding to each target model area in the second building digital twin to obtain the adjusted parameter attributes, wherein the parameter attributes include component material attributes, geological parameter attributes, and load action attributes.

[0179] In this step, parameter attributes refer to the core data in the second building digital twin used to describe the building structure and geological characteristics, and are key parameters for settlement simulation.

[0180] Component material properties refer to the material characteristics data of building components, including strength, stiffness, elastic modulus, etc.

[0181] Geological parameter attributes refer to the characteristic data of a geological region, including soil density, bearing capacity, compression modulus, etc.

[0182] Load action attributes refer to the load data borne by a building, including self-weight, construction load, and external environmental load.

[0183] The adjusted parameter attributes refer to the new set of parameters obtained by modifying the original parameter attributes based on the adaptation parameters of the construction type, the geological stratification response parameters, and the working condition simulation parameters.

[0184] In this embodiment, the original parameter attributes of each target model region in the second building digital twin are first located, including component material attributes, geological parameter attributes, and load attributes. Then, the corresponding parameter attributes are adjusted dimensionally based on the adaptation parameters of the construction type: For component material attributes, the reinforcement process increases the strength parameters of the original component to the new material strength standard according to the material strength enhancement coefficient; the demolition process removes the material parameters of the demolished component according to the component removal adaptation coefficient, such as resetting stiffness and strength values ​​to zero. For load attributes, the demolition process reduces the load value corresponding to the demolished component according to the unloading coefficient. For example, the component's self-weight load is removed, and the reinforcement process applying pressure superimposes the corresponding pressure load on the target component according to the pressure adaptation coefficient.

[0185] Simultaneously, the geological parameter attributes of the target model area are adjusted according to the geological stratification response parameters. For example, when the construction type is foundation treatment, the bearing capacity parameter of the clay layer is increased according to the response coefficient, and when the construction type is pressure reinforcement, the compression modulus parameter of the sand layer is adjusted according to the response coefficient. Finally, referring to the force application method and material property parameter adjustment rules in the working condition simulation parameters, the above-mentioned multi-dimensional adjusted parameter attributes are finally calibrated to ensure that the adjustment of each parameter attribute conforms to the actual construction logic and obtains the adjusted parameter attributes that accurately match the transformation working conditions.

[0186] Step 413: Based on the adjusted parameter attributes and the geological stratification response parameters, calculate the settlement change of each key building structure;

[0187] In this step, the settlement change refers to the settlement variation value of each key building structure caused by the adjustment of the target model area parameters under the modification conditions.

[0188] In this embodiment, the initial settlement values ​​of each key building structure are obtained from the natural settlement data of the second building digital twin. Then, based on the adjusted parameter attributes and geological stratification response parameters, a structural mechanics analysis method is used to simulate the mechanical deformation process of the key structures under construction, and the simulated settlement values ​​of each key structure are calculated. The simulated settlement values ​​can be obtained by calculating the product of the additional stress, the geological stratification response parameters, and the target soil layer thickness, and the quotient of the corresponding soil layer compression modulus. The additional stress comes from the force application method in the simulation parameters, the corresponding soil layer compression modulus comes from the geological parameter attributes in the adjusted parameter attributes, and the target soil layer thickness comes from the geological stratification parameters in the settlement simulation base data. Finally, the initial settlement values ​​are subtracted from the simulated settlement values ​​to obtain the settlement change of each key building structure under the modification conditions.

[0189] Step 414: Based on the spatial correlation parameters in the basic data of the settlement simulation, calculate the transmitted settlement of each key building structure to the key building structures in the adjacent model area.

[0190] In this step, the critical building structure in the adjacent model area refers to the critical building structure located within the adjacent model area, whose settlement will be indirectly affected by the construction of the target model area.

[0191] Transmitted settlement refers to the settlement change of each building's key structure that is transmitted to the key structures of adjacent model areas through spatial correlation.

[0192] In this embodiment, spatial correlation parameters between each key building structure and the key building structures in adjacent model areas are extracted from the basic data of settlement simulation. Then, for the settlement change of each key building structure, the settlement change is multiplied by the spatial correlation parameters to obtain the transmitted settlement contribution value of that key building structure to each adjacent key building structure. Finally, all transmitted settlement contribution values ​​of the key building structures in each adjacent model area are summed to obtain the final transmitted settlement amount of that key building structure.

[0193] Step 415: Based on the settlement influence weight in the basic data of the settlement simulation, the settlement change and the transmitted settlement are weighted and integrated to obtain the actual settlement values ​​of each key building structure after the simulation of each renovation construction procedure is completed.

[0194] In this embodiment, the weighted integration rule can be: Actual settlement value = (settlement change × settlement influence weight) + (transferred settlement × (1 - settlement influence weight)], where the higher the settlement influence weight, the greater the impact of direct settlement change on the final settlement. Then, by substituting the settlement change, transferred settlement, and settlement influence weight of each key structure of the building, the actual settlement value of each key structure is calculated one by one.

[0195] Step 416: Integrate the time nodes corresponding to the completion of each renovation construction process simulation and the actual settlement values ​​of each building's key structures to form phased settlement data.

[0196] In this embodiment, the time node after the simulation of each renovation construction process is completed is first determined to ensure that the time node is consistent with the simulation time span; then, the building structure identifier of each key building structure and the corresponding actual settlement value are bound to the time node, and the process name information is supplemented to form a complete data record; finally, after checking that the complete data record has no duplication or missing data, the stage settlement data corresponding to the renovation construction process is obtained.

[0197] The embodiments of this application can restore the dynamic changes of settlement throughout the entire renovation and construction process, taking into account both the direct impact of construction and the indirect transmission impact of adjacent areas. It solves the problems of existing methods that are difficult to reflect nonlinear settlement under construction disturbance and ignore spatial correlation, providing accurate construction impact data for the overall settlement calculation of the building, and supporting settlement prediction and safety early warning during the renovation process.

[0198] Figure 3 This is a schematic diagram illustrating a specific implementation of a building settlement calculation system based on multi-source data fusion, as provided in this application. (Refer to...) Figure 3 The system may include:

[0199] The acquisition module 21 is used to acquire building settlement data, which includes historical settlement time series data and real-time settlement data;

[0200] The generation module 22 is used to generate natural settlement data of the building under no-modification conditions by means of a recurrent neural network based on the building settlement data.

[0201] The construction module 23 is used to construct a first building digital twin based on the spatial relationship of the building structure, and to mark the settlement status of key building structures in the first building digital twin based on the real-time settlement data and natural settlement data, so as to obtain a second building digital twin.

[0202] The simulation module 24 is used to simulate the settlement evolution process of the building under the renovation conditions using the second building digital twin according to the building renovation construction procedure, and obtain settlement evolution data.

[0203] The calculation module 25 is used to calculate the overall settlement of the building based on the real-time settlement data, natural settlement data and settlement evolution data.

[0204] This application provides a building settlement calculation system based on multi-source data fusion to implement the aforementioned building settlement calculation method based on multi-source data fusion. Therefore, the specific implementation of the building settlement calculation system based on multi-source data fusion can be found in the embodiment section of the building settlement calculation method based on multi-source data fusion described above. The specific implementation can be referred to the description of the corresponding embodiments, which will not be repeated here.

[0205] This application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of the above-described method for calculating the overall settlement of a building based on multi-source data fusion.

[0206] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of any of the above-described methods for calculating overall building settlement based on multi-source data fusion.

[0207] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0208] The embodiments of this application also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above embodiments of the building settlement calculation method based on multi-source data fusion.

[0209] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0210] The foregoing has provided a detailed description of a method and system for calculating overall building settlement based on multi-source data fusion, as provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A building overall settlement calculation method based on multi-source data fusion, characterized in that, The method comprises the following steps: obtaining building settlement data, the building settlement data comprising historical settlement time series data and real-time settlement data; generating natural settlement data of the building under a non-reconstruction working condition by a recurrent neural network according to the building settlement data; constructing a first building digital twin according to the spatial positional relationship of the building structure, labeling the settlement state of the key structure of the building in the first building digital twin according to the real-time settlement data and the natural settlement data, and obtaining a second building digital twin; simulating the settlement evolution process of the building under a reconstruction working condition by using the second building digital twin according to the reconstruction construction procedure of the building, and obtaining settlement evolution data; calculating the overall settlement of the building based on the real-time settlement data, the natural settlement data and the settlement evolution data, wherein, taking the natural settlement data as a benchmark, a deviation adjustment amount is calculated according to the difference between the real-time settlement data and the natural settlement data, the deviation adjustment amount and the total construction influence amount reflected by the settlement evolution data are superimposed on the natural settlement data to obtain the overall settlement of the building.

2. The building overall settlement calculation method based on multi-source data fusion according to claim 1, characterized in that, The method for generating natural settlement data of the building under a non-reconstruction working condition by a recurrent neural network according to the building settlement data comprises the following steps: integrating the historical settlement time series data in the building settlement data with geological and environmental related parameters of the area where the building is located to form time-labeled sequence data; supplementing the real-time settlement data in the building settlement data to the time-labeled sequence data to generate network input data; processing the network input data, the geological and environmental related parameters by an internal processing unit of the recurrent neural network to generate natural settlement data of the building under a non-reconstruction working condition.

3. The method according to claim 2, wherein, The method for processing the network input data, the geological and environmental related parameters by an internal processing unit of the recurrent neural network to generate natural settlement data of the building under a non-reconstruction working condition comprises the following steps: splitting the geological related parameters according to the soil depth and the time label in the network input data to obtain multiple groups of layered geological parameters corresponding to different time nodes, and extracting environmental time sequence parameters matching each time node from the environmental related parameters; allocating corresponding settlement influence weights to each key structure according to the settlement characteristics of the key structure; determining the spatial correlation parameters between adjacent key structures according to the spatial distribution position of each key structure; integrating the network input data with the layered geological parameters, the environmental time sequence parameters, the settlement influence weights and the spatial correlation parameters to form an extended input sequence; initializing the internal processing state of the recurrent neural network to obtain an initial processing state; associating the sequence segment corresponding to the first time node in the extended input sequence with the initial processing state in the order of the time label in the network input data to obtain associated information; The associated information is processed by an internal processing unit of the recurrent neural network until all time nodes in the extended input sequence are processed, and a final updated processing state is generated. Based on the final updated processing state, the natural settlement data of the building under the unmodified working condition is generated by the recurrent neural network.

4. The building overall settlement calculation method based on multi-source data fusion according to claim 1, characterized in that, According to the reconstruction construction process of the building, the settlement evolution data of the building under the reconstruction working condition is obtained by simulating the settlement evolution process of the building under the reconstruction working condition by using the second building digital twin. Determine the construction influence factors corresponding to the reconstruction construction process of the building, and extract the basic data of the settlement simulation from the second building digital twin. Establish the corresponding association relationship between the reconstruction construction process and each model region in the second building digital twin, and determine the target model region corresponding to the operation range of each reconstruction construction process based on the corresponding association relationship. According to the construction influence factors, the working condition simulation parameters are allocated to each reconstruction construction process, including the force application mode, material characteristic parameter adjustment rule and simulation time span. According to the working condition simulation parameters and the basic data of the settlement simulation, the settlement evolution process of the building under the reconstruction working condition is simulated in each target model region by using the second building digital twin, and the actual settlement value and the corresponding time node of each building key structure after the simulation of each reconstruction construction process are recorded to form the stage settlement data. All stage settlement data are integrated to form the settlement evolution data.

5. The building overall settlement calculation method based on multi-source data fusion according to claim 4, characterized in that, According to the working condition simulation parameters and the basic data of the settlement simulation, the settlement evolution process of the building under the reconstruction working condition is simulated in each target model region by using the second building digital twin, and the actual settlement value and the corresponding time node of each building key structure after the simulation of each reconstruction construction process are recorded to form the stage settlement data, including: Based on the construction type in the construction influence factors and the basic data of the settlement simulation, the adaptive parameters of the construction type and the geological layer response parameters are generated; Based on the adaptive parameters of the construction type, the geological layer response parameters and the working condition simulation parameters, the parameter attributes of each target model region in the second building digital twin are adjusted to obtain the adjusted parameter attributes, wherein the parameter attributes include component material attributes, geological parameter attributes and load action attributes; Based on the adjusted parameter attributes and the geological layer response parameters, the settlement change amount of each building key structure is calculated; According to the spatial association parameters in the basic data of the settlement simulation, the conduction settlement amount of the settlement change amount of each building key structure to the building key structure of the adjacent model region is calculated; According to the settlement influence weight in the basic data of the settlement simulation, the settlement change amount and the conduction settlement amount are weighted and integrated to obtain the actual settlement value of each building key structure after the simulation of each reconstruction construction process; The time node and the actual settlement value of each building key structure after the simulation of each reconstruction construction process are integrated to form the stage settlement data.

6. The method according to claim 1, wherein, According to the spatial position relationship of the building structure, a first building digital twin is constructed, comprising: Collecting entity structure information of the building and geological related information of the area where the building is located, combining the spatial position relationship of the building structure, a three-dimensional model framework consistent with the entity structure form of the building is constructed; According to the spatial position relationship of the building structure, the entity structure information is classified and integrated to form a building structure information set, and the geological related information is layered and integrated according to the soil layer depth sequence to form a geological layered information set; According to the spatial position relationship of the building structure, each component information in the building structure information set is bound to the building structure area of the three-dimensional model framework, and each soil layer information in the geological layered information set is bound to the geological area of the three-dimensional model framework according to the soil layer depth sequence, forming a first building digital twin.

7. The method according to claim 1, wherein, According to the real-time settlement data and the natural settlement data, the settlement state of the key structure of the building is marked in the first building digital twin to obtain a second building digital twin, comprising: Extracting the spatial position coordinates and structure identification information of the key structure of the building from the first building digital twin; Correlating the real-time settlement data with the structure identification information to obtain first settlement correlation data, and correlating the natural settlement data with the structure identification information to obtain second settlement correlation data; According to the presentation requirements of the key structure of the building, the first settlement correlation data and the second settlement correlation data, and the visualization characteristics of the first building digital twin, a marking rule of the settlement state is determined; According to the marking rule of the settlement state, the first settlement correlation data and the second settlement correlation data are marked to the model position corresponding to the spatial position coordinates in the first building digital twin respectively to form a second building digital twin.

8. A building overall settlement calculation system based on multi-source data fusion, characterized in that, Comprising: An acquisition module for acquiring building settlement data, the building settlement data including historical settlement time series data and real-time settlement data; A generation module for generating natural settlement data of the building under the condition of no reconstruction through a recurrent neural network according to the building settlement data; A construction module for constructing a first building digital twin according to the spatial position relationship of the building structure, and marking the settlement state of the key structure of the building in the first building digital twin according to the real-time settlement data and the natural settlement data to obtain a second building digital twin; A simulation module for simulating the settlement evolution process of the building under the condition of reconstruction according to the reconstruction construction procedure of the building by using the second building digital twin to obtain settlement evolution data; A calculation module for calculating the overall settlement of the building based on the real-time settlement data, the natural settlement data and the settlement evolution data.

9. A computing device, comprising: Comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to realize a building overall settlement calculation method based on multi-source data fusion according to any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The computer program is stored and is executed by a computer to realize the building overall settlement calculation method based on multi-source data fusion according to any one of claims 1 to 7.

Images