Large-scale public building health monitoring method based on three-dimensional laser scanning and Beidou positioning

Abstract

The invention discloses a large-scale public building health monitoring method based on three-dimensional laser scanning and Beidou positioning. The method comprises the following steps: acquiring three-dimensional point cloud data and Beidou positioning data; based on the three-dimensional point cloud data and the Beidou positioning data, a topological structure between the buildings is obtained, a finite element analysis method is adopted to analyze the topological structure, and an analysis result is obtained; extracting feature information of the building group by adopting a multi-scale analysis method, and obtaining a health state index of the building group in combination with an analysis result; and obtaining to-be-monitored building group data, inputting the to-be-monitored building group data and the building group health state index into the building group abnormity monitoring model, and obtaining a monitoring result. According to the invention, by combining the three-dimensional point cloud data and the Beidou positioning data, the precision and efficiency of building group safety monitoring can be effectively improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of information monitoring, and in particular relates to a large-scale public building health monitoring method based on three-dimensional laser scanning and Beidou positioning. Background Art

[0002] Health monitoring of large public building complexes is a complex system engineering project, involving the structural safety, functional use and overall coordination of the building complex. Traditional monitoring methods usually target individual buildings and are difficult to fully assess the overall condition of the complex. 3D laser scanning technology can be used to obtain a detailed 3D model of the complex, but how to associate massive 3D point cloud data with the actual health status of the complex is a key technical issue that needs to be solved urgently.

[0003] The Beidou positioning system can provide accurate spatiotemporal location information, but in complex building environments, the positioning accuracy is easily affected by factors such as signal shielding and multipath effects, making it difficult to meet the high-precision positioning requirements for building health monitoring. At the same time, there are differences in the spatiotemporal scale between Beidou positioning data and 3D laser scanning data. How to effectively integrate the two types of data and build a unified spatiotemporal benchmark is also a technical difficulty. Therefore, there is an urgent need for a large-scale public building health monitoring method based on 3D laser scanning and Beidou positioning. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a large-scale public building health monitoring method based on three-dimensional laser scanning and Beidou positioning, which can effectively improve the accuracy and efficiency of building complex safety monitoring.

[0005] The present invention provides a large-scale public building health monitoring method based on three-dimensional laser scanning and Beidou positioning, comprising:

[0006] Obtain 3D point cloud data and Beidou positioning data;

[0007] Based on the three-dimensional point cloud data and Beidou positioning data, a topological structure between buildings is obtained, and a finite element analysis method is used to analyze the topological structure to obtain an analysis result;

[0008] A multi-scale analysis method is used to extract characteristic information of the building complex, and the health status index of the building complex is obtained by combining the analysis results;

[0009] The data of the building complex to be monitored are obtained, and the data of the building complex to be monitored and the building complex health status indicators are input into the building complex abnormality monitoring model to obtain the monitoring results.

[0010] Optionally, obtaining the three-dimensional point cloud data includes:

[0011] Obtain initial 3D point cloud data of the building complex;

[0012] A statistical filtering algorithm is used to remove noise points and outliers in the initial three-dimensional point cloud data, and the processed point cloud data is unified in a spatial coordinate system to obtain the three-dimensional point cloud data.

[0013] Optionally, based on the three-dimensional point cloud data and Beidou positioning data, obtaining the topological structure between buildings includes:

[0014] Aligning the three-dimensional point cloud data with Beidou positioning data to establish a unified space-time reference;

[0015] Based on the unified spatiotemporal reference, the topological structure between buildings is obtained.

[0016] Optionally, registering the three-dimensional point cloud data with Beidou positioning data to establish a unified space-time reference includes:

[0017] The coordinate conversion algorithm is used to convert the position information of the Beidou positioning data into a target coordinate system to obtain the position information in a unified coordinate system;

[0018] According to the position information in the unified coordinate system, corresponding three-dimensional point cloud data is acquired, and the position information is registered with the three-dimensional point cloud data using a feature matching algorithm to obtain the registered position information and three-dimensional point cloud data;

[0019] Performing time synchronization on the registered position information and the three-dimensional point cloud data, unifying the registered position information and the three-dimensional model data to the same time reference;

[0020] According to the same time reference, a spatial transformation algorithm is used to coordinate the registered position information and three-dimensional point cloud data to establish a unified time-space reference.

[0021] Optionally, the topological structure is analyzed by using a finite element analysis method, and the analysis results obtained include:

[0022] Using finite element analysis method, the stress and deformation of buildings in the topological structure under load are calculated to obtain the load transfer process between buildings;

[0023] According to the load transfer process, the vibration characteristic parameters of the building are obtained, and according to the vibration characteristic parameters, the resonance effect of the building complex under the earthquake dynamic load is analyzed to obtain the vibration mode and frequency of the building complex;

[0024] According to the load transfer process between the buildings and the vibration mode and frequency of the building complex, the overall performance and safety of the building complex are evaluated to obtain the building complex performance analysis results.

[0025] Optionally, using a multi-scale analysis method to extract characteristic information of the building complex includes:

[0026] A multi-scale analysis method is used to extract key characteristic information of the building complex from three spatial scales: single building, local and overall building complex, and three temporal scales: short-term, medium-term and long-term.

[0027] Optionally, the building complex anomaly monitoring model is trained using a training set, which is historical monitoring data including sensor data, environmental data and historical detection records.

[0028] Optionally, the monitoring results include:

[0029] If the monitoring data of the building complex deviates from the building complex health status index, abnormal alarm information is output, and the abnormal status level of the building complex is determined based on the abnormal alarm information.

[0030] Compared with the prior art, the present invention has the following advantages and technical effects:

[0031] The present invention discloses a building complex health monitoring method based on multi-source data fusion. The method obtains high-precision point cloud data of the building complex through three-dimensional laser scanning, combines the spatiotemporal information of the Beidou positioning system for registration, and constructs a three-dimensional model of the building complex under a unified spatiotemporal reference. On this basis, a topological structure model between buildings is established, and finite element analysis is used to simulate load transfer and resonance effects. Key features are extracted through multi-scale analysis and fused to obtain comprehensive health status indicators. The health status indicators can be used to monitor abnormalities in the building complex data, effectively improving the accuracy and efficiency of building complex safety monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0033] Figure 1 It is a flow chart of a large public building health monitoring method based on three-dimensional laser scanning and Beidou positioning according to an embodiment of the present invention. DETAILED DESCRIPTION

[0034] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0035] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0036] The present invention proposes a large-scale public building health monitoring method based on three-dimensional laser scanning and Beidou positioning. Figure 1 As shown, the specific steps include:

[0037] Obtain 3D point cloud data and Beidou positioning data;

[0038] Based on 3D point cloud data and Beidou positioning data, the topological structure between buildings is obtained, and the finite element analysis method is used to analyze the topological structure and obtain the analysis results;

[0039] The multi-scale analysis method is used to extract the characteristic information of the building complex, and the health status index of the building complex is obtained by combining the analysis results;

[0040] Obtain the data of the building complex to be monitored, input the data of the building complex to be monitored and the building complex health status indicators into the building complex abnormality monitoring model, and obtain the monitoring results.

[0041] Furthermore, obtaining three-dimensional point cloud data includes:

[0042] Obtain initial 3D point cloud data of the building complex;

[0043] A statistical filtering algorithm is used to remove noise points and outliers in the initial three-dimensional point cloud data, and the processed point cloud data is unified in the spatial coordinate system to obtain three-dimensional point cloud data.

[0044] Specifically, three-dimensional point cloud data of the building complex is obtained, and the three-dimensional point cloud data includes noise points and outliers;

[0045] The statistical filtering algorithm is used to process the three-dimensional point cloud data, remove noise points and outliers, and obtain the point cloud data after preliminary processing;

[0046] Through the registration and splicing of point cloud data, the spatial coordinate system of the initially processed point cloud data is unified to obtain a complete three-dimensional point cloud model of the building complex;

[0047] According to the point cloud density and building complex characteristics of the three-dimensional point cloud model of the building complex, the main structure of the building is extracted using the regional growing algorithm to obtain the building skeleton point cloud data;

[0048] Based on the building skeleton point cloud data, a regular 3D building model is reconstructed through plane fitting and edge extraction, and model simplification and texture mapping are performed;

[0049] If there are occluded or missing areas in the 3D point cloud model of the building complex, a point cloud completion algorithm based on deep learning is used to repair the occluded or missing areas to obtain a complete 3D model of the building complex;

[0050] According to the application requirements of the three-dimensional model of the building complex, mesh simplification and multi-level of detail technology are used to generate three-dimensional model data with different accuracy and detail levels.

[0051] Furthermore, based on the 3D point cloud data and Beidou positioning data, the topological structure between buildings is obtained, including:

[0052] Align 3D point cloud data with Beidou positioning data to establish a unified space-time benchmark;

[0053] Based on a unified spatiotemporal reference, the topological structure between buildings is obtained.

[0054] Furthermore, the 3D point cloud data and Beidou positioning data are registered to establish a unified space-time benchmark including:

[0055] The coordinate conversion algorithm is used to convert the location information of Beidou positioning data into the target coordinate system to obtain the location information in a unified coordinate system;

[0056] According to the position information in the unified coordinate system, the corresponding three-dimensional point cloud data is obtained, and the position information and the three-dimensional point cloud data are registered using a feature matching algorithm to obtain the registered position information and three-dimensional point cloud data;

[0057] Time synchronization is performed on the registered position information and 3D point cloud data, and the registered position information and 3D model data are unified to the same time reference;

[0058] Based on the same time reference, the space transformation algorithm is used to coordinate the registered position information and three-dimensional point cloud data to establish a unified space-time reference.

[0059] Specifically, the latitude, longitude, elevation and time information of the current location are first obtained through the Beidou positioning system, with an accuracy of centimeters. Then, the seven-parameter coordinate transformation model is used to transform the location information in the geodetic coordinate system into the local coordinate system where the three-dimensional model of the building complex is located, and the conversion error is controlled within 5 centimeters. Then, SIFT feature extraction and RANSAC algorithm are used to extract the feature points of the location point cloud and the building model, and the optimal transformation matrix is ​​estimated to achieve accurate registration of the two, with a registration error of less than 10 centimeters. On this basis, the timing service of the Beidou system is used to obtain high-precision time information, and the timestamps of the location point cloud and the building model are unified to Beijing time, with a time synchronization error of less than 1 millisecond. Then, the quaternion interpolation algorithm is used to perform coordinate transformation and time interpolation on the location and model under a unified space-time reference to obtain a seamless and dynamically updated three-dimensional scene. According to the spatial relationship between the current location point and the building model, the octree index and ray tracing algorithm are used to determine the location information of the floor, room, etc. in real time, with a positioning accuracy better than 1 meter. Finally, the Unity 3D engine is used to fuse the positioning results with the building's three-dimensional model for rendering, and combined with AR, VR and other interactive technologies to provide users with an immersive navigation and visualization experience, with a stable frame rate of more than 30 frames per second.

[0060] Furthermore, the finite element analysis method is used to analyze the topological structure, and the analysis results obtained include:

[0061] Finite element analysis is used to calculate the stress and deformation of buildings in the topological structure under load, and the load transfer process between buildings is obtained;

[0062] According to the load transfer process, the vibration characteristic parameters of the building are obtained. Based on the vibration characteristic parameters, the resonance effect of the building complex under the earthquake dynamic load is analyzed to obtain the vibration mode and frequency of the building complex.

[0063] According to the load transfer process between buildings and the vibration mode and frequency of the building complex, the overall performance and safety of the building complex are evaluated and the performance analysis results of the building complex are obtained.

[0064] Specifically, firstly, a 3D laser scanner is used to obtain the point cloud data of the building complex, and the point cloud density is not less than 100 points per square meter. Then, the RANSAC algorithm and the regional growing algorithm are used to extract the geometric features such as planes and edges in the point cloud, and the 3D solid model of the building is reconstructed with an accuracy better than 10 cm. Then, the spatial position coordinates, length, width and height and other parameter information of the building are extracted from the 3D solid model. According to the relative position relationship of the buildings, the topological structure model between the buildings is constructed using the Delaunay triangulation algorithm. The topological structure model is converted into a finite element mesh model with a mesh size of no more than 1 meter, and the elastic modulus, Poisson's ratio, density and other attribute parameters of the building materials, as well as the load size, direction and constraint conditions are defined. The finite element analysis software ANSYS is used to calculate the stress cloud map and deformation vector map of the building under the action of self-weight, wind load, seismic load, etc., and analyze the stress transfer path and deformation trend. The environmental vibration test system is used to obtain the first 10 vibration mode frequencies and vibration mode coefficients of the building, and the modal participation mass ratio is greater than 90%. Apply earthquake waves consistent with the actual measurements to the finite element model, analyze the dynamic response of the building complex, and determine whether there is a resonance effect and the risk of progressive collapse. Combined with the results of static and dynamic analysis, the seismic performance, wind resistance and overall stability of the building complex are evaluated to determine the weak points and potential risks. The evaluation results are visualized in the form of stress cloud maps, deformation vector maps, animations, etc., to intuitively reflect the stress state and destruction process of the building complex. Finally, the evaluation results are compared with national standards such as GB50011 and GB50009 to calculate the seismic margin, displacement angle, inter-story displacement ratio and other indicators of the building complex. If the requirements of the standards are not met, optimization plans are proposed from the aspects of seismic isolation, support addition, and quality adjustment to ensure the safety and applicability of the building complex.

[0065] Furthermore, the multi-scale analysis method is used to extract the characteristic information of the building complex, including:

[0066] A multi-scale analysis method is used to extract key characteristic information of the building complex from three spatial scales: single building, local and overall building complex, and three temporal scales: short-term, medium-term and long-term.

[0067] Specifically, the spatial layout information of the building complex is obtained, and a multi-level topological structure model including three levels of building units, local building complexes and overall building complexes is constructed; the monitoring data of the building complex at three different time scales of short-term, medium-term and long-term are obtained, and the short-term monitoring data includes vibration data and strain data, the medium-term monitoring data includes settlement data and displacement data, and the long-term monitoring data includes material aging data and environmental impact data; the monitoring data is preprocessed to remove outliers and noise, and key characteristic parameters are extracted; the key characteristic parameters are input into the preset finite element analysis model to obtain the stress-strain distribution and dynamic response of the building complex under different working conditions; a multi-scale feature fusion algorithm is used to perform weighted combination of the characteristic parameters of the building units, local building complexes and overall building complexes to obtain a comprehensive health status index; the comprehensive health status index is input into the preset machine learning model to establish a mapping relationship between the comprehensive health status index and the performance degradation of the building complex, and the prediction results of the future performance of the building complex are obtained; according to the prediction results, the weak parts of the building complex are identified, the potential risks are evaluated, and the maintenance and reinforcement plan of the building complex is generated.

[0068] Furthermore, the building complex anomaly monitoring model is trained through a training set, which is historical monitoring data including sensor data, environmental data and historical detection records.

[0069] Furthermore, obtaining monitoring results includes:

[0070] If the monitoring data of the building complex deviates from the building complex health status index, an abnormal alarm message is output, and the abnormal status level of the building complex is determined based on the abnormal alarm message.

[0071] Specifically, sensor monitoring data, environmental monitoring data, and historical detection record-related data of the building complex are obtained to obtain building complex monitoring data;

[0072] Preprocessing the building complex monitoring data, including data cleaning and feature extraction of the building complex monitoring data, to obtain preprocessed building complex monitoring data;

[0073] The unsupervised learning algorithm is used to train the pre-processed building complex monitoring data to obtain a deep learning model that characterizes the distribution characteristics of various monitoring data under the normal health state of the building complex.

[0074] Obtain the current monitoring data of the building complex, input the current monitoring data into the deep learning model, reconstruct errors or generate errors through the deep learning model, and determine whether the current monitoring data deviates from the normal data distribution of the building complex;

[0075] If the current monitoring data deviates from the normal state data distribution of the building complex by more than the preset threshold, the building complex is judged to be in an abnormal state and an abnormal alarm information is output;

[0076] Obtain historical anomaly detection records of the building complex, fine-tune the deep learning model based on the historical anomaly detection records, and obtain an optimized deep learning anomaly detection model;

[0077] The optimized deep learning anomaly detection model is used to detect the abnormal state of the building complex and obtain the anomaly detection results of the building complex;

[0078] Obtain monitoring data of buildings in abnormal conditions, analyze the degree of deviation of various health indicators of buildings from normal conditions, and determine the comprehensive health status level of buildings;

[0079] Visualize the anomaly detection results and comprehensive health status assessment results of the building complex.

[0080] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A large-scale public building health monitoring method based on three-dimensional laser scanning and Beidou positioning, characterized in that: include: Obtain 3D point cloud data and Beidou positioning data; Based on the three-dimensional point cloud data and Beidou positioning data, a topological structure between buildings is obtained, and a finite element analysis method is used to analyze the topological structure to obtain an analysis result; A multi-scale analysis method is used to extract characteristic information of the building complex, and the health status index of the building complex is obtained by combining the analysis results; The data of the building complex to be monitored are obtained, and the data of the building complex to be monitored and the building complex health status indicators are input into the building complex abnormality monitoring model to obtain the monitoring results.

2. According to claim 1, a large-scale public building health monitoring method based on three-dimensional laser scanning and Beidou positioning is characterized in that: Acquiring the three-dimensional point cloud data includes: Obtain initial 3D point cloud data of the building complex; A statistical filtering algorithm is used to remove noise points and outliers in the initial three-dimensional point cloud data, and the processed point cloud data is unified in a spatial coordinate system to obtain the three-dimensional point cloud data.

3. The method for health monitoring of large public buildings based on three-dimensional laser scanning and Beidou positioning according to claim 1 is characterized in that: Based on the three-dimensional point cloud data and Beidou positioning data, obtaining the topological structure between buildings includes: Aligning the three-dimensional point cloud data with Beidou positioning data to establish a unified space-time reference; Based on the unified spatiotemporal reference, the topological structure between buildings is obtained.

4. The method for health monitoring of large public buildings based on three-dimensional laser scanning and Beidou positioning according to claim 3 is characterized in that: Aligning the three-dimensional point cloud data with Beidou positioning data to establish a unified space-time reference includes: The coordinate conversion algorithm is used to convert the position information of the Beidou positioning data into a target coordinate system to obtain the position information in a unified coordinate system; According to the position information in the unified coordinate system, corresponding three-dimensional point cloud data is acquired, and the position information is registered with the three-dimensional point cloud data using a feature matching algorithm to obtain the registered position information and three-dimensional point cloud data; Performing time synchronization on the registered position information and the three-dimensional point cloud data, unifying the registered position information and the three-dimensional model data to the same time reference; According to the same time reference, a spatial transformation algorithm is used to coordinate the registered position information and three-dimensional point cloud data to establish a unified time-space reference.

5. The method for health monitoring of large public buildings based on three-dimensional laser scanning and Beidou positioning according to claim 1 is characterized in that: The topological structure is analyzed by using the finite element analysis method, and the analysis results include: Using finite element analysis method, the stress and deformation of buildings in the topological structure under load are calculated to obtain the load transfer process between buildings; According to the load transfer process, the vibration characteristic parameters of the building are obtained, and according to the vibration characteristic parameters, the resonance effect of the building complex under the earthquake dynamic load is analyzed to obtain the vibration mode and frequency of the building complex; According to the load transfer process between the buildings and the vibration mode and frequency of the building complex, the overall performance and safety of the building complex are evaluated to obtain the building complex performance analysis results.

6. The method for health monitoring of large public buildings based on three-dimensional laser scanning and Beidou positioning according to claim 1 is characterized in that: The multi-scale analysis method is used to extract the characteristic information of the building complex, including: A multi-scale analysis method is used to extract key characteristic information of the building complex from three spatial scales: single building, local and overall building complex, and three temporal scales: short-term, medium-term and long-term.

7. The method for health monitoring of large public buildings based on three-dimensional laser scanning and Beidou positioning according to claim 1 is characterized in that: The building complex anomaly monitoring model is trained through a training set, and the training set is historical monitoring data including sensor data, environmental data and historical detection records.

8. The method for health monitoring of large public buildings based on three-dimensional laser scanning and Beidou positioning according to claim 1 is characterized in that: The monitoring results include: If the monitoring data of the building complex deviates from the building complex health status index, abnormal alarm information is output, and the abnormal status level of the building complex is determined based on the abnormal alarm information.

Images