Dynamic monitoring and early warning system of construction surrounding environment based on 3D laser point cloud technology

Abstract

The invention discloses a dynamic monitoring and early warning system for the construction surrounding environment based on three-dimensional laser point cloud technology, which relates to the technical field of construction monitoring. The system comprises: a state information acquisition module, which acquires the construction displacement vector of the foundation pit in the preset construction area; an environmental settlement prediction module, which processes the construction displacement vector through a stress distribution state reconstruction model to obtain the predicted settlement vector of the surrounding environment; an environmental surrounding monitoring module, which uses a three-dimensional laser point cloud monitoring device to monitor the surrounding environment and obtain the environmental settlement modeling result; a monitoring settlement extraction module, which extracts the monitoring settlement vector from the surrounding environment settlement modeling result; an abnormal analysis module, which analyzes the monitoring settlement vector according to the predicted settlement vector to obtain the coordinates of the settlement abnormal area; a rendering mark early warning module, which renders the coordinates of the settlement abnormal area in the environmental settlement modeling result, and visually displays the early warning. Thus, the technical effect of improving the accuracy and stability of the early warning is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of construction monitoring, and in particular to a dynamic monitoring and early warning system for a construction surrounding environment based on three-dimensional laser point cloud technology. Background Art

[0002] Traditional construction environment monitoring and settlement warning systems rely on threshold values ​​for warning triggering. These threshold values ​​are usually determined based on experience and expert knowledge, and are highly subjective and unstable. Due to the complexity of the construction process and the increase in environmental requirements, there are technical problems such as low warning accuracy and poor warning stability. Summary of the invention

[0003] The present invention provides a dynamic monitoring and early warning system for the construction surrounding environment based on three-dimensional laser point cloud technology to solve the technical problems of low early warning accuracy and poor early warning stability in the prior art, and achieve the technical effect of improving the accuracy and stability of early warning.

[0004] The construction surrounding environment dynamic monitoring and early warning system based on three-dimensional laser point cloud technology provided by the present invention includes:

[0005] A status information acquisition module is used to obtain the construction status information of a foundation pit in a preset construction area, wherein the construction status information includes a construction displacement vector.

[0006] The environmental settlement prediction module is used to reconstruct the model through the stress distribution state of the preset construction area, process the construction displacement vector, and obtain the surrounding environment predicted settlement vector.

[0007] The environment surrounding monitoring module is used to monitor the surrounding environment through three-dimensional laser point cloud monitoring equipment to obtain surrounding environment settlement modeling results.

[0008] A monitoring settlement extraction module is used to extract the surrounding environment monitoring settlement vector from the surrounding environment settlement modeling results.

[0009] The anomaly analysis module is used to perform settlement anomaly analysis on the settlement vector monitored by the surrounding environment according to the settlement vector predicted by the surrounding environment, so as to obtain the coordinates of the settlement anomaly area.

[0010] A rendering mark warning module is used to render and mark the coordinates of the abnormal settlement area in the settlement modeling results of the surrounding environment, and display and warn through a visual interface.

[0011] The invention discloses a dynamic monitoring and early warning system for the construction surrounding environment based on three-dimensional laser point cloud technology, including: a state information acquisition module, which is used to obtain the foundation pit construction state information of a preset construction area, including a construction displacement vector; an environmental settlement prediction module, which processes the construction displacement vector through a stress distribution state reconstruction model of the preset construction area to obtain a predicted settlement vector for the surrounding environment; an environmental surrounding monitoring module, which performs surrounding environment monitoring through a three-dimensional laser point cloud monitoring device to obtain a surrounding environment settlement modeling result; a monitoring settlement extraction module, which extracts a surrounding environment monitoring settlement vector from the surrounding environment settlement modeling result; an abnormality analysis module, which performs settlement abnormality analysis on the surrounding environment monitoring settlement vector according to the surrounding environment predicted settlement vector to obtain the coordinates of the settlement abnormality area; a rendering mark early warning module, which renders and marks the coordinates of the settlement abnormality area in the surrounding environment settlement modeling result, and displays the early warning through a visual interface. The dynamic monitoring and early warning system for the construction surrounding environment based on three-dimensional laser point cloud technology disclosed by the invention solves the technical problems of low early warning accuracy and poor early warning stability, and achieves the technical effect of improving the accuracy and stability of early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 It is a structural schematic diagram of the construction surrounding environment dynamic monitoring and early warning system based on three-dimensional laser point cloud technology of the present invention;

[0013] Figure 2 It is a flow chart of the execution steps of obtaining the predicted settlement vector of the surrounding environment in the dynamic monitoring and early warning system of the construction surrounding environment based on the three-dimensional laser point cloud technology of the present invention.

[0014] Explanation of the reference numerals: state information acquisition module 11 , environmental settlement prediction module 12 , environmental surrounding monitoring module 13 , monitoring settlement extraction module 14 , abnormality analysis module 15 , rendering identification warning module 16 . DETAILED DESCRIPTION

[0015] The technical solution provided in the embodiments of the present invention is to solve the technical problems of low warning accuracy and poor warning stability in the prior art. The overall idea adopted is as follows:

[0016] First, the status information acquisition module obtains the construction status information of the foundation pit in the preset construction area, wherein the construction status information includes the construction displacement vector. Then, the environmental settlement prediction module reconstructs the model through the stress distribution state of the preset construction area, processes the construction displacement vector, and obtains the predicted settlement vector of the surrounding environment. Then, the environmental surrounding monitoring module monitors the surrounding environment through the three-dimensional laser point cloud monitoring equipment to obtain the surrounding environment settlement modeling results. Next, the monitoring settlement extraction module extracts the surrounding environment monitoring settlement vector from the surrounding environment settlement modeling results. Furthermore, the anomaly analysis module performs settlement anomaly analysis on the surrounding environment monitoring settlement vector according to the surrounding environment predicted settlement vector, and obtains the coordinates of the settlement abnormal area. Finally, the rendering mark warning module renders and marks the coordinates of the settlement abnormal area in the surrounding environment settlement modeling results, and displays the warning through a visual interface.

[0017] The above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods of the specification to better understand the above technical solution. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments used only to explain the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, it should be noted that, for the convenience of description, only the parts related to the present invention are shown in the drawings, rather than all of them. Example

[0018] Figure 1 The structure diagram of the dynamic monitoring and early warning system for the construction surrounding environment based on three-dimensional laser point cloud technology of the present invention is shown in FIG. 1 , wherein the system includes:

[0019] The status information acquisition module 11 is used to obtain the construction status information of the foundation pit in the preset construction area, wherein the construction status information includes the construction displacement vector.

[0020] Specifically, the foundation pit information of the preset construction area is obtained and parsed to obtain the construction displacement vector. The construction displacement vector is determined based on the current foundation pit state and the ideal state of the next construction node in the engineering design data, and can be understood as a displacement vector quantitative representation of the construction target. By obtaining the construction displacement vector, it is helpful to understand the current state of the foundation pit and determine the displacement or adjustment required to reach the next construction node.

[0021] Specifically, first, obtain the real-time status information of the foundation pit through field measurement or using monitoring equipment (such as displacement sensors); then, obtain the ideal state of the next construction node through design drawings, construction plans, etc., including the ideal geometry, size, and position of the foundation pit; then, calculate the construction displacement vector based on the current state of the foundation pit and the ideal state of the next construction node. The construction displacement vector represents the displacement or adjustment required to reach the next construction node from the current state.

[0022] That is, the starting point of the construction displacement vector corresponds to the current state of the foundation pit construction, the end point of the construction displacement vector corresponds to the ideal state of the next construction node, the direction of the construction displacement vector represents the direction of the change or adjustment required to reach the next construction node from the current state, and the length (or size) of the construction displacement vector represents the degree of change or adjustment required. The longer the length, the greater the change or adjustment required.

[0023] The environmental settlement prediction module 12 is used to reconstruct the model through the stress distribution state of the preset construction area, process the construction displacement vector, and obtain the surrounding environment predicted settlement vector.

[0024] Specifically, the stress distribution state reconstruction model of the preset construction area is used to describe the stress changes during the foundation pit excavation process and its impact on the surrounding environment. The stress distribution state reconstruction model is constructed based on the geological data of the preset construction area, the foundation pit design parameters and the construction plan, and is calibrated through on-site monitoring data and historical construction data to improve the accuracy of the model.

[0025] Specifically, the construction displacement vector is input into the stress distribution state reconstruction model, and the stress changes at various positions during the foundation pit excavation process are calculated to obtain the impact data of the foundation pit excavation on the surrounding environment, and the predicted settlement vector is calculated and obtained.

[0026] In some embodiments, Figure 2 As shown, by reconstructing the stress distribution state model of the preset construction area, processing the construction displacement vector, and obtaining the surrounding environment predicted settlement vector, the execution steps include:

[0027] Collect basic information of the preset construction area, wherein the basic information of the preset construction area includes groundwater level, geological type, and annual average soil moisture; configure the stress distribution state reconstruction model based on the graph neural network according to the groundwater level, the geological type, and the annual average soil moisture; fit the construction displacement vector through the stress distribution state reconstruction model to obtain the stress distribution state prediction result; perform surrounding environment settlement simulation based on the stress distribution state prediction result to obtain the surrounding environment predicted settlement vector.

[0028] Specifically, first, basic information of the preset construction area is obtained, which includes the groundwater level, geological type information, and annual average soil moisture of the construction area. The geological type information includes the distribution and properties of rock and soil layers, which are determined through drilling, sampling analysis, and historical geological data.

[0029] Specifically, the collected basic information such as groundwater level, geological type and annual average soil moisture is preprocessed, including data cleaning and standardization, to form standard data input. Then, according to the spatial distribution and geological information of the preset construction area, a graph structure reflecting the geological and environmental characteristics is constructed. Then, the graph neural network (GNN) algorithm is used to construct a training stress distribution state reconstruction model. Exemplarily, node information (such as groundwater level, geological type, soil moisture) and edge information (such as the relationship between different geological units) are input to initialize the configuration stress distribution state reconstruction model.

[0030] Specifically, the stress distribution state and stress action mode of the construction area with different basic information are not the same. In other words, the construction area with different soil and geological conditions responds differently to the same construction displacement vector. Therefore, based on the groundwater level, geological type, and annual average soil humidity, the model parameters of the stress distribution state reconstruction model are initialized.

[0031] Specifically, the monitored construction displacement vector is input into the configured stress distribution state reconstruction model, and the construction displacement vector is fitted and calculated to obtain the prediction result of the stress distribution state during the foundation pit construction process.

[0032] Specifically, according to the prediction results of the stress distribution state, a settlement simulation model of the surrounding environment is constructed. First, a mathematical model is constructed to describe the physical properties and geological conditions of the soil and structures (such as buildings and roads). Usually, it includes the mechanical properties of the soil (such as density, shear strength, etc.), the weight and position of the structure, the groundwater level and other information. Then, the weight of the structure (including buildings, engineering structures, roads, etc.) is applied to the model as a load. The load will cause the stress distribution in the soil to change. Then, based on the finite element analysis method and tools (such as ANSYS, FLAC3D, etc.), the stress distribution in the soil under the load is calculated. Finally, based on the calculated stress distribution, the settlement of the soil is simulated. It is expressed as a predicted settlement vector, and the direction and magnitude of the predicted settlement vector represent the predicted settlement direction and degree, respectively.

[0033] In some implementations, the stress distribution state reconstruction model is configured based on a graph neural network according to the groundwater level, the geological type, and the annual average soil moisture, and the execution steps include:

[0034] Based on the preset initial three-dimensional model of the construction area and the graph neural network, a topological structure of a soil stress distribution reconstruction model is constructed, wherein the topological structure of the soil stress distribution reconstruction model is the same as the topological structure of the initial three-dimensional model of the construction area, and each position has stress distribution monitoring data before construction; with the groundwater level, the geological type, the annual average soil moisture and the initial three-dimensional model of the construction area as limits, a construction displacement vector record data set and a soil stress distribution state monitoring data set are collected; according to the soil stress distribution state monitoring data set and the construction displacement vector record data set, the topological structure of the soil stress distribution reconstruction model is trained using a random gradient descent algorithm to generate the stress distribution state reconstruction model.

[0035] Specifically, first, based on the geological survey data and design plan of the preset construction area, an initial three-dimensional model of the area is constructed. The model reflects the geological and environmental conditions before construction. Then, the topological structure of the soil stress distribution reconstruction model is constructed using graph neural network (GNN) technology. This topological structure is the same as the topological structure of the initial three-dimensional model, ensuring that each location has corresponding stress distribution monitoring data before construction. In other words, the initial three-dimensional model of the preset construction area can be represented as a graph, in which each node represents a location and each edge represents the relationship between locations (for example, physical distance or geological connectivity). Each node can have one or more attributes, such as stress distribution monitoring data before construction.

[0036] Specifically, the construction displacement vector record data set and soil stress distribution state monitoring data set at different time points during the construction process are collected with the groundwater level, geological type and annual average soil moisture information as the screening constraints. The obtained construction displacement vector record data set and soil stress distribution state monitoring data set contain the response relationship between the construction displacement vector and the soil stress distribution under the preset basic information conditions of the construction area.

[0037] Specifically, the soil stress distribution state monitoring dataset and the construction displacement vector record dataset are input into the graph neural network model, the model is trained using the stochastic gradient descent (SGD) algorithm, the model parameters are adjusted to minimize the prediction error, and the performance of the model is evaluated using the validation set and the test set to ensure that the model can accurately predict the stress distribution state. Specifically, the graph neural network can acquire the ability to predict stress distribution by learning the complex mapping pattern between the construction displacement vector record dataset and the soil stress distribution state monitoring dataset. The above-mentioned graph neural network-based method can effectively utilize the spatial relationship of soil stress distribution, thereby improving the accuracy of prediction.

[0038] The surrounding environment monitoring module 13 is used to monitor the surrounding environment through three-dimensional laser point cloud monitoring equipment to obtain surrounding environment settlement modeling results.

[0039] In some embodiments, the surrounding environment is monitored by a three-dimensional laser point cloud monitoring device to obtain the surrounding environment settlement modeling result, and the execution steps include:

[0040] The three-dimensional laser point cloud monitoring device includes a three-dimensional laser point cloud collector, a point cloud locator and a splicing software; the surrounding environment is monitored by the three-dimensional laser point cloud collector and the point cloud locator to obtain the surrounding environment point cloud information, wherein the surrounding environment point cloud information has point cloud coordinate information; the surrounding environment point cloud information is spliced ​​based on the point cloud coordinate information by the splicing software to obtain the surrounding environment point cloud modeling result;

[0041] Based on the surrounding environment point cloud modeling result, the initial modeling result of the surrounding environment is marked with a settlement vector to obtain the surrounding environment settlement modeling result.

[0042] Specifically, the three-dimensional laser point cloud monitoring equipment includes a three-dimensional laser point cloud collector for collecting three-dimensional point cloud data of the surrounding environment; a point cloud locator for determining the geographical location coordinates of the point cloud data to ensure the spatial accuracy of the point cloud data; and splicing software for splicing multiple point cloud data to generate a complete three-dimensional environmental model.

[0043] Optionally, the 3D laser point cloud collector can be portable or mounted on a vehicle or aircraft. Portable devices are suitable for small-scale or inaccessible places, while vehicle-mounted or airborne devices are suitable for large-scale or open places.

[0044] Optionally, the point cloud locator includes a spatial position sensor and a global navigation satellite system (GNSS) positioning system, wherein the spatial position sensor is used to measure the direction and attitude of the device, and the GNSS positioning system is used to measure the global position of the device. The spatial position of each point cloud data can be accurately determined by combining the two.

[0045] Optionally, the stitching software stitches the collected scattered point cloud data into a continuous 3D model based on a point cloud registration algorithm. Exemplary point cloud registration algorithms include SIFT, ‌SURF, ‌ORB, 4PCS, etc.

[0046] Specifically, based on the collection performance of the 3D laser point cloud collector (collection accuracy, collection range, collection angle, etc.), multiple predetermined collection positions are determined, and then the 3D laser point cloud collector and the point cloud locator are set at the predetermined collection positions, and the 3D laser point cloud collector is activated to scan the surrounding environment to obtain the point cloud information of the surrounding environment; then, the point cloud locator is used to calibrate the geographic location of the collected surrounding environment point cloud information to ensure that the point cloud data has accurate coordinate information.

[0047] Optionally, the point cloud locator is accurately positioned using technologies such as RTK-GPS or total station.

[0048] Furthermore, point cloud processing software (such as Cyclone, CloudCompare, etc.) is used to stitch point cloud data. First, the collected surrounding environment point cloud information is imported into the stitching software, and the surrounding environment point cloud information is aligned based on the point cloud coordinate information. The aligned point cloud data is stitched to generate a complete three-dimensional environment model, which is stored as the surrounding environment point cloud modeling result.

[0049] Specifically, the initial modeling results of the surrounding environment before construction are imported, the point cloud modeling results of the surrounding environment are compared with the initial modeling results of the surrounding environment, the settlement of each monitoring point is analyzed, and then the settlement vector of each monitoring point is calculated according to the change of the point cloud data, the settlement position and settlement amount are marked, and the output is the settlement modeling result of the surrounding environment. The settlement modeling result of the surrounding environment represents the state of the surrounding environment after construction and the impact of construction on the environment.

[0050] In some implementations, the surrounding environment is monitored by the three-dimensional laser point cloud collector and the point cloud locator to obtain point cloud information of the surrounding environment, wherein the point cloud information of the surrounding environment has point cloud coordinate information, and the execution steps include:

[0051] The three-dimensional laser point cloud collector and the point cloud locator are used to monitor the surrounding roads and obtain the surrounding road point cloud information, wherein the surrounding road point cloud information has the surrounding road point cloud coordinate information; the three-dimensional laser point cloud collector and the point cloud locator are used to monitor the surrounding buildings and obtain the surrounding building point cloud information, wherein the surrounding building point cloud information has the surrounding building point cloud coordinate information; the three-dimensional laser point cloud collector and the point cloud locator are used to monitor the surrounding roads and obtain the surrounding pipeline point cloud information, wherein the surrounding pipeline point cloud information has the surrounding pipeline point cloud coordinate information; the surrounding road point cloud information, the surrounding building point cloud information and the surrounding pipeline point cloud information are added into the surrounding environment point cloud information; the surrounding road point cloud coordinate information, the surrounding road point cloud coordinate information and the surrounding pipeline point cloud coordinate information are added into the point cloud coordinate information.

[0052] Specifically, the surrounding environment is monitored to obtain the point cloud information of the surrounding environment, including: monitoring the surrounding roads through a 3D laser point cloud collector and a point cloud locator to obtain the point cloud information of the surrounding roads. This information includes the shape, location and other attributes of the roads. Similarly, the surrounding buildings are monitored to obtain the point cloud information of the surrounding buildings. This information includes the shape, location and other attributes of the buildings. Similarly, the surrounding pipelines are monitored to obtain the point cloud information of the surrounding pipelines. This information includes the shape, location and other attributes of the pipelines.

[0053] Specifically, the collected point cloud information of surrounding roads, buildings and pipelines is added to the point cloud information of the surrounding environment. In this way, a point cloud dataset containing comprehensive information of the surrounding environment is obtained. The collected point cloud coordinate information of surrounding roads, buildings and pipelines is added to the point cloud coordinate information, thereby obtaining a dataset containing comprehensive coordinate information of the surrounding environment.

[0054] Further, the surrounding roads are monitored by the three-dimensional laser point cloud collector and the point cloud locator to obtain surrounding road point cloud information, wherein the surrounding road point cloud information has surrounding road point cloud coordinate information, and the execution steps include:

[0055] Construct a target road anchor frame, wherein the target road anchor frame has the same outline as the target road; collect initial point cloud information of surrounding roads based on road distribution orientation through the three-dimensional laser point cloud collector and the point cloud locator; perform density sorting on the initial point cloud information of surrounding roads according to the target road anchor frame to obtain the surrounding road point cloud information, wherein the surrounding road point cloud information has the surrounding road point cloud coordinate information.

[0056] Specifically, first, a target road anchor frame is constructed based on the road profile of the target road. The frame is used to determine the area to be monitored and the scope of data collection. Then, the initial point cloud information of the surrounding roads is collected based on the road distribution orientation through a 3D laser point cloud collector and a point cloud locator. Preferably, the initial point cloud information of the surrounding roads is collected through a vehicle-mounted or airborne 3D laser point cloud collector.

[0057] Optionally, based on the target road anchor frame, the initial point cloud information of the surrounding roads is density sorted to filter out the point cloud information within the anchor frame and exclude other irrelevant information. Through the above process, the point cloud information and point cloud coordinate information of the surrounding roads can be obtained.

[0058] The above steps can help to efficiently collect point cloud information and point cloud coordinate information of surrounding roads, providing an accurate information basis for subsequent analysis and decision-making.

[0059] Further, according to the target road anchor frame, the initial point cloud information of the surrounding roads is density sorted to obtain the point cloud information of the surrounding roads, wherein the point cloud information of the surrounding roads has the point cloud coordinate information of the surrounding roads, and the execution steps include:

[0060] Based on the target road anchor frame, a target road reference axis is constructed, wherein the target road reference axis has an eight-way distance label, and the eight-way distance label is the maximum distance from the eight azimuth boundaries of the target road to the target road reference axis; the target road reference axis is traversed through the initial point cloud information of the surrounding roads, and outlier point clouds are deleted based on the eight-way distance label to obtain a number of road point cloud distribution results; the road point cloud distribution result with the maximum number of point clouds of the several road point cloud distribution results is extracted and set as the surrounding road point cloud information, wherein the surrounding road point cloud information has the surrounding road point cloud coordinate information.

[0061] Specifically, a target road reference axis is constructed based on the target road anchor frame. The target road reference axis may be the central axis of the road, and the target road reference axis has different distance labels in each position and in each direction. Exemplarily, the target road reference axis has eight-way distance labels, indicating the maximum distance from the eight azimuth boundaries of the target road to the target road reference axis.

[0062] Specifically, based on the target road reference axis traversing the initial point cloud information of the surrounding roads, outlier point clouds are removed according to the eight-way distance labels of the target road reference axis to filter out the point cloud information located in the target road anchor frame and exclude other irrelevant information.

[0063] Furthermore, the distribution result with the largest number of point clouds among the several road point cloud distribution results is extracted and set as the surrounding road point cloud information. The above process can more accurately filter and process point cloud data through the reference axis and eight-way distance labels.

[0064] The monitoring settlement extraction module 14 is used to extract the surrounding environment monitoring settlement vector from the surrounding environment settlement modeling result.

[0065] Specifically, based on the settlement vector of each monitoring point in the surrounding environment settlement modeling results, the settlement position and corresponding settlement amount of the surrounding environment are obtained, and the output is the surrounding environment monitoring settlement vector.

[0066] The anomaly analysis module 15 is used to perform settlement anomaly analysis on the settlement vector monitored by the surrounding environment according to the settlement vector predicted by the surrounding environment, so as to obtain the coordinates of the settlement anomaly area.

[0067] Specifically, by comparing the predicted settlement vector of the surrounding environment with the monitored settlement vector of the surrounding environment, the difference between the predicted settlement value and the actual settlement value is obtained, and the points with abnormal settlement are identified and output as the coordinates of the settlement abnormal area. The settlement abnormal area coordinates reflect the location and degree of abnormal settlement, providing data support for subsequent identification and early warning.

[0068] In some embodiments, the settlement anomaly analysis is performed on the settlement anomaly vector of the surrounding environment monitoring according to the settlement anomaly vector of the surrounding environment to obtain the coordinates of the settlement anomaly area, and the execution steps include:

[0069] According to the predicted settlement vector of the surrounding environment, the predicted settlement modulus of the first direction of the first position is extracted; according to the monitored settlement vector of the surrounding environment, the monitored settlement modulus of the first direction of the first position is extracted; when the monitored settlement modulus of the first direction is greater than the predicted settlement modulus of the first direction, the first position is added to the coordinates of the settlement abnormality area.

[0070] Specifically, first, according to the predicted settlement vector of the surrounding environment, the predicted settlement modulus of the first direction of the first position is extracted; then, according to the monitored settlement vector of the surrounding environment, the monitored settlement modulus of the first direction of the first position is extracted. Then, the predicted settlement modulus of the first direction is compared with the monitored settlement modulus of the first direction, and when the monitored settlement modulus is greater than the predicted settlement modulus, the first position is added to the settlement abnormality area coordinates.

[0071] The rendering mark warning module 16 is used to render and mark the coordinates of the abnormal settlement area in the settlement modeling result of the surrounding environment, and display and warn through a visual interface.

[0072] Specifically, the abnormal settlement area is rendered and marked on the settlement modeling results of the surrounding environment, and displayed through a visual interface. First, the abnormal settlement area and the corresponding abnormal value are marked on the settlement modeling results, and then the model is rendered according to the abnormal settlement area and the abnormal value, so as to clearly display the abnormal area on the visual interface. Exemplary visual effects include adjusting color, brightness, transparency, etc.

[0073] In summary, the construction surrounding environment dynamic monitoring and early warning system based on three-dimensional laser point cloud technology provided by the present invention has the following technical effects:

[0074] Through the state information acquisition module, the foundation pit construction state information of the preset construction area is obtained, including the construction displacement vector; the environmental settlement prediction module processes the construction displacement vector through the stress distribution state reconstruction model of the preset construction area to obtain the surrounding environment predicted settlement vector; the environmental surrounding monitoring module monitors the surrounding environment through the three-dimensional laser point cloud monitoring equipment to obtain the surrounding environment settlement modeling results; the monitoring settlement extraction module extracts the surrounding environment monitoring settlement vector from the surrounding environment settlement modeling results; the abnormal analysis module performs settlement abnormality analysis on the surrounding environment monitoring settlement vector according to the surrounding environment predicted settlement vector to obtain the settlement abnormality area coordinates; the rendering mark warning module renders the settlement abnormality area coordinates in the surrounding environment settlement modeling results and displays the warning through a visual interface. Thus, the technical effect of improving the accuracy and stability of the warning is achieved.

[0075] It should be understood that the embodiments disclosed in the present invention and the above description can enable those skilled in the art to use the present invention to implement the present invention. At the same time, the present invention is not limited to the above-mentioned embodiments. It should be understood that those skilled in the art can still modify the technical solutions recorded in the above-mentioned embodiments, or replace some of the technical features therein by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. A dynamic monitoring and early warning system for the construction surrounding environment based on three-dimensional laser point cloud technology, characterized by: include: A status information acquisition module, the status information acquisition module is used to obtain the construction status information of the foundation pit in the preset construction area, wherein the construction status information includes a construction displacement vector; An environmental settlement prediction module, which is used to reconstruct a model through the stress distribution state of the preset construction area, process the construction displacement vector, and obtain a predicted settlement vector of the surrounding environment; An environmental surrounding monitoring module, which is used to monitor the surrounding environment through a three-dimensional laser point cloud monitoring device to obtain a surrounding environment settlement modeling result; A monitoring settlement extraction module, the monitoring settlement extraction module is used to extract the surrounding environment monitoring settlement vector from the surrounding environment settlement modeling result; An anomaly analysis module, the anomaly analysis module is used to perform settlement anomaly analysis on the settlement vector monitored by the surrounding environment according to the settlement vector predicted by the surrounding environment, and obtain the coordinates of the settlement anomaly area; A rendering mark warning module, which is used to render and mark the coordinates of the abnormal settlement area in the settlement modeling results of the surrounding environment, and display and warn through a visual interface; The stress distribution state of the preset construction area is reconstructed into a model, the construction displacement vector is processed, and the surrounding environment predicted settlement vector is obtained. The execution steps include: Collecting basic information of the preset construction area, wherein the basic information of the preset construction area includes groundwater level, geological type, and annual average soil humidity; According to the groundwater level, the geological type, and the annual average soil moisture, the stress distribution state reconstruction model is configured based on a graph neural network; The construction displacement vector is fitted by the stress distribution state reconstruction model to obtain a stress distribution state prediction result; Performing a surrounding environment settlement simulation based on the stress distribution state prediction result to obtain the surrounding environment predicted settlement vector; According to the groundwater level, the geological type, and the annual average soil moisture, the stress distribution state reconstruction model is configured based on a graph neural network, and the execution steps include: Based on the preset initial three-dimensional model of the construction area and the graph neural network, a topological structure of a soil stress distribution reconstruction model is constructed, wherein the topological structure of the soil stress distribution reconstruction model is the same as the topological structure of the preset initial three-dimensional model of the construction area, and each position has stress distribution monitoring data before construction; Based on the groundwater level, the geological type, the annual average soil moisture and the initial three-dimensional model of the preset construction area, a construction displacement vector record data set and a soil stress distribution state monitoring data set are collected; The topological structure of the soil stress distribution reconstruction model is trained by a stochastic gradient descent algorithm according to the soil stress distribution state monitoring data set and the construction displacement vector recording data set to generate the stress distribution state reconstruction model.

2. The construction surrounding environment dynamic monitoring and early warning system based on three-dimensional laser point cloud technology as claimed in claim 1, characterized in that: The surrounding environment is monitored by 3D laser point cloud monitoring equipment to obtain the surrounding environment settlement modeling results. The execution steps include: The three-dimensional laser point cloud monitoring device includes a three-dimensional laser point cloud collector, a point cloud locator and splicing software; The surrounding environment is monitored by the three-dimensional laser point cloud collector and the point cloud locator to obtain point cloud information of the surrounding environment, wherein the point cloud information of the surrounding environment has point cloud coordinate information; Using the stitching software, stitching the surrounding environment point cloud information based on the point cloud coordinate information to obtain a surrounding environment point cloud modeling result; Based on the surrounding environment point cloud modeling result, the initial modeling result of the surrounding environment is marked with a settlement vector to obtain the surrounding environment settlement modeling result.

3. The construction surrounding environment dynamic monitoring and early warning system based on three-dimensional laser point cloud technology as claimed in claim 2, characterized in that: The surrounding environment is monitored by the three-dimensional laser point cloud collector and the point cloud locator to obtain point cloud information of the surrounding environment, wherein the point cloud information of the surrounding environment has point cloud coordinate information, and the execution steps include: The surrounding roads are monitored by the three-dimensional laser point cloud collector and the point cloud locator to obtain point cloud information of the surrounding roads, wherein the point cloud information of the surrounding roads has point cloud coordinate information of the surrounding roads; The surrounding buildings are monitored by the three-dimensional laser point cloud collector and the point cloud locator to obtain point cloud information of the surrounding buildings, wherein the point cloud information of the surrounding buildings has point cloud coordinate information of the surrounding buildings; The surrounding roads are monitored by the three-dimensional laser point cloud collector and the point cloud locator to obtain the point cloud information of the surrounding pipelines, wherein the point cloud information of the surrounding pipelines has the point cloud coordinate information of the surrounding pipelines; Adding the surrounding road point cloud information, the surrounding building point cloud information and the surrounding pipeline point cloud information into the surrounding environment point cloud information; The surrounding road point cloud coordinate information, the surrounding road point cloud coordinate information and the surrounding pipeline point cloud coordinate information are added into the point cloud coordinate information.

4. The construction surrounding environment dynamic monitoring and early warning system based on three-dimensional laser point cloud technology as claimed in claim 3 is characterized in that: The surrounding roads are monitored by the three-dimensional laser point cloud collector and the point cloud locator to obtain the surrounding road point cloud information, wherein the surrounding road point cloud information has the surrounding road point cloud coordinate information, and the execution steps include: Constructing a target road anchor frame, wherein the target road anchor frame has the same contour as the target road; Collecting initial point cloud information of surrounding roads based on road distribution orientation through the three-dimensional laser point cloud collector and the point cloud locator; According to the target road anchor frame, the initial point cloud information of the surrounding roads is density sorted to obtain the surrounding road point cloud information, wherein the surrounding road point cloud information has the surrounding road point cloud coordinate information.

5. The construction surrounding environment dynamic monitoring and early warning system based on three-dimensional laser point cloud technology as claimed in claim 4, characterized in that: According to the target road anchor frame, density sorting is performed on the initial point cloud information of the surrounding roads to obtain the point cloud information of the surrounding roads, wherein the point cloud information of the surrounding roads has the coordinate information of the point cloud of the surrounding roads, and the execution steps include: Based on the target road anchor frame, construct a target road reference axis, wherein the target road reference axis has an eight-way distance label, and the eight-way distance label is the maximum distance from the eight azimuth boundaries of the target road to the target road reference axis; The target road reference axis is used to traverse the initial point cloud information of the surrounding roads, and outlier point clouds are deleted based on the eight-way distance labels to obtain a number of road point cloud distribution results; A road point cloud distribution result with a maximum point cloud quantity among the plurality of road point cloud distribution results is extracted and set as the surrounding road point cloud information, wherein the surrounding road point cloud information has the surrounding road point cloud coordinate information.

6. The construction surrounding environment dynamic monitoring and early warning system based on three-dimensional laser point cloud technology as claimed in claim 1, characterized in that: Performing settlement anomaly analysis on the settlement vector monitored by the surrounding environment according to the settlement vector predicted by the surrounding environment to obtain the coordinates of the settlement anomaly area, the execution steps include: Extracting a predicted settlement modulus in a first direction at a first position according to the predicted settlement vector of the surrounding environment; Extracting the first direction monitoring sedimentation modulus of the first position according to the surrounding environment monitoring sedimentation vector; When the monitored sedimentation modulus in the first direction is greater than the predicted sedimentation modulus in the first direction, the first position is added to the coordinates of the abnormal sedimentation area.

Images