Intelligent construction monitoring system for earthwork projects based on data analysis

Abstract

The present invention discloses an intelligent construction monitoring system for earthwork projects based on data analysis, which relates to the field of earthwork projects and comprises a construction module for uploading specification parameters of a building foundation pit and constructing a standard model of the building foundation pit based on the specification parameters of the building foundation pit; a real-scene modeling module for collecting real-time excavation status parameters of the building foundation pit and constructing a real-scene model of the building foundation pit based on the excavation status parameters; the present invention respectively constructs a standard model of the building foundation pit and a real-scene model of the building foundation pit through the design specification parameters of the building foundation pit and the real-scene collection parameters, and compares the similarity of the two models to make a safety judgment on the shape of the building foundation pit in real time in the middle and late stages of the excavation of the building foundation pit, and when the judgment result is unsafe, the defective position of the foundation pit excavation is traced in real time, so that the probability of the shape defect of the building foundation pit is controlled during the excavation and forming process of the building foundation pit, thereby reducing the backfill workload caused by excessive excavation and ensuring the stable progress of the construction foundation pit excavation project.

Description

Technical Field

[0001] The present invention relates to the technical field of earthwork engineering, and in particular to an intelligent construction monitoring system for earthwork engineering based on data analysis. Background Art

[0002] Earthwork involves the excavation, transportation, and placement of earth and rock. This includes site leveling, foundation pit excavation, and roadbed filling. Construction must be planned and implemented based on geological conditions and project requirements, involving techniques such as earthwork volume calculation and slope support. Earthwork is a fundamental component of construction, transportation, and other projects, significantly impacting project progress and safety.

[0003] The invention patent application with application number 202311479617.1 discloses a digital monitoring construction method for airport earthwork projects, including: constructing a monitoring model map of the construction area; obtaining first monitoring data through a first monitoring terminal associated with earthwork rolling equipment; updating the monitoring model map based on the first monitoring data and generating a first prompt message; sending the first prompt message to the first monitoring terminal; obtaining stratigraphic survey data of the construction area; constructing an initial model map based on the stratigraphic survey data; obtaining construction data before the construction step corresponding to the earthwork rolling equipment; correcting the initial model map based on the construction data to obtain the monitoring model map. This application aims to solve the problem that "the quality of airport earthwork projects is mainly reflected in rolling or compaction; the quality control of these two processes is now manual, that is, manual measurement and analysis are carried out and then construction adjustments are made; due to the manual method, there may be subjective analysis errors, and at night due to the influence of factors such as light, measurement will be inconvenient, and the effective implementation of night construction cannot be guaranteed."

[0004] However, during the current excavation process of building foundation pits, although the excavation path is pre-planned, errors are inevitable when using manual excavation equipment. These accumulated errors lead to repeated backfilling or backfilling in the later stages, which not only affects the progress of the construction of the building foundation pit, but also increases the workload of the construction of the building foundation pit.

[0005] Therefore, an intelligent construction monitoring system for earthwork engineering based on data analysis is proposed. Summary of the Invention

[0006] In view of the above-mentioned shortcomings of the prior art, the present invention provides an intelligent construction monitoring system for earthwork engineering based on data analysis, which can effectively solve the problems of the prior art.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0008] The present invention discloses an intelligent construction monitoring system for earthwork engineering based on data analysis, comprising:

[0009] The construction module is used to upload the specification parameters of the building foundation pit and construct the standard model of the building foundation pit based on the specification parameters of the building foundation pit; the real-scene modeling module is used to collect the real-time excavation status parameters of the building foundation pit and construct the real-scene model of the building foundation pit based on the excavation status parameters; the comparison module is used to receive the standard model of the building foundation pit constructed in the construction module and the real-scene model of the building foundation pit constructed in the real-scene modeling module, and compare the similarity of the two models; the judgment module is used to set the building foundation pit morphology safety judgment threshold, and judge whether the current building foundation pit excavation is safe based on the comparison result of the building foundation pit morphology safety judgment threshold and the similarity comparison result in the comparison module; the refresh module is used to refresh the system operation; the traceability module is used to receive the building foundation pit standard model and the building foundation pit real-scene model, and trace the current building foundation pit excavation defect position based on the building foundation pit standard model and the building foundation pit real-scene model

[0010] Furthermore, the real scene modeling module is provided with a drone module, a planning unit and a storage unit at the lower level. The drone module is used to collect real-time excavation status parameters of the construction foundation pit, the planning unit is used to plan the flight path of the drone module, and the storage unit is used to store the real-time excavation status parameters of the construction foundation pit collected by the drone module.

[0011] The drone module is equipped with a distance measuring device, and the drone module performs a distance measurement operation through the distance measuring device, and the distance measurement result is recorded as the construction foundation pit excavation state parameter;

[0012] Among them, when the storage unit stores the construction foundation pit excavation status parameters, the drone module position information and collection direction are marked for each construction foundation pit excavation status parameter during collection.

[0013] Furthermore, during the operation phase of the planning unit, a system user manually edits three sets of three-dimensional coordinates, determines a plane based on the three sets of three-dimensional coordinates, sets a point matrix on the determined plane, and performs a distance measurement operation on each point in the point matrix in a direction perpendicular to the plane, so that the distance measurement operation is performed on both sides of the point with the determined plane as a reference;

[0014] The above operation is performed three times in the planning unit. Two of them are performed according to the above logic to complete the distance measurement operation in the four directions of front, back, left, and right of each point. The other one is performed with the excavation surface of the building pit as the plane and the above logic is used to perform the distance measurement operation in the vertical downward direction only.

[0015] Among them, the distance between each adjacent point in the point matrix is ​​equal, and the total number of points in the point matrix obeys the setting logic that the higher the construction accuracy requirement of the building foundation pit real scene model, the more points there are in the point matrix, and vice versa, the fewer points there are in the point matrix. The corresponding geometric representation of the plane determined by the three sets of three-dimensional coordinates in the building foundation pit real scene model is perpendicular to the bottom surface of the building foundation pit real scene model and parallel to the side of the building foundation pit real scene model.

[0016] Furthermore, the three sets of three-dimensional coordinates used to determine the plane do not have a collinear relationship. When the drone module performs a ranging operation through the ranging device, it starts from any corner point in the point matrix, passes through all the midpoints of the point matrix, and performs two ranging operations with each point.

[0017] When storing the excavation state parameters, the storage unit performs differentiated storage based on the position information of the drone module marked with the excavation state parameters, so that each differentiated storage interval stores two sets of excavation state parameters, that is, two ranging results, and the ranging results corresponding to different planes are differentiated and stored again based on the previous differentiated storage operation;

[0018] After the drone and module pass through all the midpoints of the dot matrix and the excavation status parameters collected at each point in the dot matrix are distinguished and stored in the storage unit, the real scene modeling module is triggered to run.

[0019] Furthermore, during the operation phase of the reality modeling module, the excavation state parameters are continuously acquired in the storage unit. Each time the excavation state parameters are acquired, a distinguished storage interval is used as an acquisition target, two ranging results are acquired, and the position information of the drone module marked by the two ranging results is simultaneously picked up in the three-dimensional space. Then, a corresponding line segment is drawn in combination with the ranging results and the acquisition direction in the mark;

[0020] Based on the above operation, the same operation is performed on each of the partitioned storage intervals in the storage unit to draw the same number of line segments as the partitioned storage intervals in the storage unit;

[0021] After all the corresponding line segments that distinguish and store the intervals composed of two distance measurement results are drawn, the endpoints of all the drawn line segments are used as contour points, and the adjacent contour points are connected to each other to complete the construction of the closed building foundation pit real-scene model.

[0022] Furthermore, the similarity comparison logic between the standard model of the building foundation pit and the real-scene model of the building foundation pit in the comparison module is expressed as follows:

[0023] Where: The similarity between the standard model of the building foundation pit to be revised and the real-scene model of the building foundation pit; 、 is the weight; is the centroid position similarity; is the volume similarity; Density distribution similarity; The similarity between the revised standard model of the building foundation pit and the real-scene model of the building foundation pit; is the tolerance factor;

[0024] in, 、 are all positive numbers, , are the centroid coordinates of the two model point clouds, For the overall scale of the scene, take the length of the longest side of the foundation pit.

[0025] Furthermore, the volume similarity Similarity to density distribution The calculation formula is: ;

[0026] Where: It is the minimum value between the volume of the standard model of the building foundation pit and the real-scene model of the building foundation pit; The building foundation is the maximum volume between the standard model of the building foundation pit and the real-scene model of the building foundation pit; is the density vector, by dividing the model point cloud into Macro voxels, calculate the point cloud density within each voxel; is the modulus of the vector; the tolerance factor Values ​​subject to: Tolerance factor belongs to [0.9, 1]. The greater the density of the point matrix used in the operation phase of the real scene modeling module and the smaller the size of the standard model of the building foundation pit, the greater the tolerance factor. The larger the value, the smaller the point matrix density used in the operation phase of the real scene modeling module, and the larger the size of the standard model of the building foundation pit, the greater the tolerance factor. The smaller the value.

[0027] Furthermore, when the determination module determines that the similarity comparison result is greater than or equal to the building foundation pit morphology safety determination threshold, the current building foundation pit excavation is determined to be safe; otherwise, the current building foundation pit excavation is determined to be unsafe;

[0028] The delay determination unit under the determination module is used to record the historical similarity comparison results of the comparison module, and when the similarity comparison results of at least the latest three times are continuously decreasing, it is determined that the current building foundation pit excavation is unsafe;

[0029] Among them, when the judgment results of the judgment module and the delay judgment unit are both safe, the refresh module is triggered to run. When either the judgment results of the judgment module or the delay judgment unit are unsafe, the traceability module is triggered to run, and after the traceability module runs, the refresh module is triggered to run.

[0030] Furthermore, during the operation phase of the traceability module, any four non-coplanar points are picked in the construction pit real scene module, and the four points can form a closed three-dimensional model. Corresponding points of the four non-coplanar points picked in the construction pit standard model are further picked in the construction pit real scene model. Based on the picked points and the corresponding points, the construction pit standard model and the construction pit real scene model are aligned.

[0031] When the standard model of the building foundation pit is aligned with the real-scene model of the building foundation pit, the area where the real-scene model of the building foundation pit protrudes or is concave outside or inside the standard model of the building foundation pit is the location of the current excavation defect of the building foundation pit.

[0032] Furthermore, the construction module is interactively connected with the real-scene modeling module through a wireless network, the real-scene modeling module is interactively connected with the drone module, the planning unit and the storage unit through a wireless network, the real-scene modeling module is interactively connected with the comparison module and the judgment module through a wireless network, the comparison module is interactively connected with the storage unit through a wireless network, the comparison module is interactively connected with the judgment module through a wireless network, the judgment module is interactively connected with the delay judgment unit through a wireless network, and the judgment module and the delay judgment module are interactively connected with the refresh module and the traceability module through a wireless network.

[0033] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0034] The present invention provides an intelligent construction monitoring system for earthwork projects based on data analysis. During operation, the system constructs a standard model of a building foundation pit and a real-scene model of a building foundation pit respectively through design specification parameters of the building foundation pit and real-scene acquisition parameters. By comparing the similarity of the two models, the system makes a real-time safety judgment on the shape of the building foundation pit in the middle and late stages of the building foundation pit excavation. When the judgment result is unsafe, the defective position of the foundation pit excavation is traced in real time, so that the probability of defects in the shape of the building foundation pit is controlled during the excavation and shaping process of the building foundation pit, the backfill workload caused by excessive excavation is reduced, and the stable progress of the building foundation pit excavation project is ensured. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0036] Figure 1 This is a structural diagram of an intelligent construction monitoring system for earthwork projects based on data analysis;

[0037] Figure 2 This is an example schematic diagram of the real scene modeling module drawing line segments based on the distance measurement results in the present invention. DETAILED DESCRIPTION

[0038] To make the purpose, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0039] The present invention will be further described below with reference to the embodiments. Example

[0040] The intelligent construction monitoring system for earthwork engineering based on data analysis in this embodiment is as follows: Figure 1 Shown, including:

[0041] A construction module is used to upload construction foundation pit specification parameters and construct a construction foundation pit standard model based on the construction foundation pit specification parameters;

[0042] The real-scene modeling module is used to collect real-time excavation status parameters of the building foundation pit and build a real-scene model of the building foundation pit based on the excavation status parameters;

[0043] The real scene modeling module is equipped with a drone module, a planning unit and a storage unit. The drone module is used to collect real-time excavation status parameters of the building foundation pit. The planning unit is used to plan the flight path of the drone module. The storage unit is used to store the real-time excavation status parameters of the building foundation pit collected by the drone module.

[0044] The drone module is equipped with a distance measuring device, through which the drone module performs distance measurement operations, and the distance measurement results are recorded as the excavation status parameters of the building foundation pit;

[0045] Among them, when the storage unit stores the construction foundation pit excavation state parameters, the drone module position information and collection direction when each construction foundation pit excavation state parameter is marked;

[0046] During the operation phase of the planning unit, the system user manually edits three sets of 3D coordinates, determines a plane based on the three sets of 3D coordinates, sets a point matrix on the determined plane, and performs distance measurement on each point in the point matrix along the direction perpendicular to the plane, so that the distance measurement is performed on both sides of the point with the determined plane as the reference.

[0047] The above operation is performed three times in the planning unit. Two of them are performed according to the above logic to complete the distance measurement operation in the four directions of front, back, left, and right of each point. The other one is performed with the excavation surface of the building pit as the plane and the above logic is used to perform the distance measurement operation in the vertical downward direction only.

[0048] Among them, the spacing between adjacent points in the point matrix is ​​equal, and the total number of points in the point matrix follows the setting logic that the higher the construction accuracy requirement of the building foundation pit real scene model, the more points there are in the point matrix, and vice versa, the fewer points there are in the point matrix. The corresponding geometric representation of the plane determined by the three sets of three-dimensional coordinates in the building foundation pit real scene model is perpendicular to the bottom surface of the building foundation pit real scene model and parallel to the side surface of the building foundation pit real scene model;

[0049] The three sets of three-dimensional coordinates used to determine the plane do not have a collinear relationship. When the drone module performs ranging operations through the ranging device, it starts from any corner point in the point matrix, passes through all the midpoints of the point matrix, and performs two ranging operations with each point.

[0050] When the storage unit stores the excavation state parameters, it performs differentiated storage based on the position information of the drone module marked with the excavation state parameters, so that each differentiated storage interval stores two sets of excavation state parameters, that is, two ranging results, and the ranging results corresponding to different planes are differentiated and stored again based on the previous differentiated storage operation;

[0051] After the drone and module pass through all the midpoints of the dot matrix and the excavation status parameters collected at each point in the dot matrix are distinguished and stored in the storage unit, the reality modeling module is triggered to run;

[0052] During the operation phase of the reality modeling module, the excavation state parameters are continuously acquired from the storage unit. Each time the excavation state parameters are acquired, a distinguished storage interval is used as the acquisition target. Two ranging results are obtained, and the position information of the drone module marked by the two ranging results is picked up in the three-dimensional space simultaneously. Then, a corresponding line segment is drawn based on the ranging results and the acquisition direction in the mark.

[0053] Based on the above operation, the same operation is performed on each of the partitioned storage intervals in the storage unit to draw the same number of line segments as the partitioned storage intervals in the storage unit;

[0054] After all the corresponding line segments of the storage intervals formed by the two distance measurement results are drawn, the end points of all the drawn line segments are used as contour points, and the adjacent contour points are connected to complete the construction of the closed building foundation pit real scene model;

[0055] A comparison module is used to receive the standard model of the building foundation pit constructed in the construction module and the real-scene model of the building foundation pit constructed in the real-scene modeling module, and compare the similarities between the two models;

[0056] The similarity comparison logic between the standard model of the building foundation pit and the real-scene model of the building foundation pit in the comparison module is expressed as follows: Where: The similarity between the standard model of the building foundation pit to be revised and the real-scene model of the building foundation pit; 、 is the weight; is the centroid position similarity; is the volume similarity; is the density distribution similarity; The similarity between the revised standard model of the building foundation pit and the real-scene model of the building foundation pit; is the tolerance factor; 、 are all positive numbers, , are the centroid coordinates of the two model point clouds, The overall scale of the scene is the length of the longest side of the foundation pit; the volume similarity Similarity to density distribution The calculation formula is: Where: It is the minimum value between the volume of the standard model of the building foundation pit and the real-scene model of the building foundation pit; It is the maximum volume between the standard model of the building foundation pit and the real-scene model of the building foundation pit; is the density vector, by dividing the model point cloud into Macro voxels, calculate the point cloud density within each voxel; is the magnitude of the vector;

[0057] Tolerance Factor The value is subject to:

[0058] Tolerance Factor belongs to [0.9, 1]. The greater the density of the point matrix used in the operation phase of the real scene modeling module and the smaller the size of the standard model of the building foundation pit, the greater the tolerance factor. The larger the value, the smaller the point matrix density used in the operation phase of the real scene modeling module, and the larger the size of the standard model of the building foundation pit, the greater the tolerance factor. The smaller the value;

[0059] The judgment module is used to set the building foundation pit morphology safety judgment threshold, and compare the building foundation pit morphology safety judgment threshold with the similarity comparison result in the comparison module to determine whether the current building foundation pit excavation is safe;

[0060] When the judgment module determines that the similarity comparison result is greater than or equal to the building foundation pit morphology safety judgment threshold, it determines that the current building foundation pit excavation is safe; otherwise, it determines that the current building foundation pit excavation is unsafe;

[0061] The delay judgment unit under the judgment module is used to record the historical similarity comparison results of the comparison module. When the similarity comparison results of at least the latest three times are continuously decreasing, it is determined that the current building foundation pit excavation is unsafe;

[0062] Among them, when the judgment results of the judgment module and the delay judgment unit are both safe, the refresh module is triggered to run; when either the judgment results of the judgment module or the delay judgment unit are unsafe, the traceability module is triggered to run, and after the traceability module runs, the refresh module is triggered to run;

[0063] Refresh module, used to refresh system operation;

[0064] The traceability module receives the standard model of the building foundation pit and the real-scene model of the building foundation pit, and traces the location of the current building foundation pit excavation defects based on the standard model of the building foundation pit and the real-scene model of the building foundation pit;

[0065] During the traceability module operation phase, four non-coplanar points are picked in the building foundation pit real scene module, and the four points can form a closed three-dimensional model. Then, corresponding points of the four non-coplanar points picked in the building foundation pit standard model are picked in the building foundation pit real scene model. Based on the picked points and the corresponding points, the building foundation pit standard model and the building foundation pit real scene model are aligned.

[0066] When the standard building foundation pit model and the real-life building foundation pit model are aligned, the area where the real-life building foundation pit model protrudes or is recessed outside or inside the standard building foundation pit model is the location of the current building foundation pit excavation defect;

[0067] The construction module is interactively connected to the real-scene modeling module through a wireless network, the real-scene modeling module is interactively connected to the drone module, the planning unit and the storage unit through a wireless network, the real-scene modeling module is interactively connected to the comparison module and the judgment module through a wireless network, the comparison module is interactively connected to the storage unit through a wireless network, the comparison module is interactively connected to the judgment module through a wireless network, the judgment module is interactively connected to the delay judgment unit through a wireless network, and the judgment module and the delay judgment module are interactively connected to the refresh module and the traceability module through a wireless network.

[0068] In this embodiment, the construction module runs to upload the specification parameters of the building foundation pit, and builds the standard model of the building foundation pit based on the specification parameters of the building foundation pit. The real-scene modeling module runs to collect the real-time excavation status parameters of the building foundation pit, and builds the real-scene model of the building foundation pit based on the excavation status parameters. The drone module synchronously collects the real-time excavation status parameters of the building foundation pit, and the planning unit plans the flight path of the drone module in real time. The storage unit then stores the real-time excavation status parameters of the building foundation pit collected by the drone module. The comparison module then receives the standard model of the building foundation pit constructed in the construction module and the real-scene model of the building foundation pit constructed in the real-scene modeling module, and compares the two. The similarity of the models is determined, and the judgment module further sets the building foundation pit morphology safety judgment threshold. Based on the comparison between the building foundation pit morphology safety judgment threshold and the similarity comparison result in the comparison module, it is determined whether the current building foundation pit excavation is safe. The delayed judgment unit records the historical similarity comparison results of the comparison module in real time. When the similarity comparison results of no less than the latest three times are continuously decreasing, it is determined that the current building foundation pit excavation is unsafe. The refresh module refreshes the system operation in real time. The traceability module synchronously receives the building foundation pit standard model and the building foundation pit real-scene model, and traces the current building foundation pit excavation defect position based on the building foundation pit standard model and the building foundation pit real-scene model.

[0069] Through the system in the above embodiment, a standard model is constructed by uploading foundation pit specification parameters, and a real-life model is generated by combining the real-time excavation status parameters collected by drones. A unique similarity comparison algorithm is used to calculate model similarity based on dimensions such as center of mass position, volume, and density distribution. At the same time, a morphological safety judgment threshold and a delay judgment mechanism are set to accurately identify the excavation safety status. When an anomaly is found, the defect location is quickly traced through model alignment technology. This method of deeply integrating three-dimensional modeling, drone ranging, and data analysis breaks through the limitations of traditional manual monitoring and realizes dynamic visual monitoring and defect location of foundation pit excavation. It not only improves the accuracy and efficiency of construction monitoring, but also can provide early warning of safety hazards. It provides an innovative technical solution for the intelligent construction of earthwork projects and effectively ensures construction safety and project quality.

[0070] The following is an example of the application of the system in the above embodiment:

[0071] During the late stages of excavation work for a construction foundation pit in a certain city's commercial center, the pit, designed to be 20 meters deep, 100 meters long, and 80 meters wide, would house a three-story underground parking garage and supporting commercial facilities. Located in a busy area with densely populated surrounding buildings and complex underground pipelines, the excavation required extremely high safety and precision. To ensure construction safety and improve quality and efficiency, an intelligent construction monitoring system for earthwork projects based on data analysis was implemented.

[0072] 2. System Application Process

[0073] (1) Constructing a Standard Model of the Building Foundation Pit: Construction personnel uploaded the specifications of the building foundation pit through the system's construction module, including detailed information such as length, width, depth, and slope. Based on these parameters, the system automatically constructed a standard model of the building foundation pit, which accurately reflected the shape and dimensions of the foundation pit as required by the design.

[0074] (2) Planning UAV flight paths and data collection

[0075] Planning Unit Operation: The system user manually edited three sets of non-collinear 3D coordinates and defined a plane based on these three sets of coordinates. This plane was perpendicular to the base of the real-life model of the building pit and parallel to its sides. A point matrix was set on this plane, with the spacing between adjacent points set to 2 meters based on the accuracy requirements of the real-life model. Due to the high accuracy requirements of this project, the number of points in the point matrix was relatively large.

[0076] Flight Path Planning: The planning unit plots the drone module's flight path according to specific logic. Starting from any corner point in the point matrix, the drone passes through all points in the matrix. At each point, ranging is performed along the perpendicular plane, with a measurement performed on both sides of the plane to complete ranging in all four directions: front, back, left, and right. Furthermore, using the excavation surface of the building pit as the plane, ranging is performed only in the vertical downward direction according to the aforementioned logic.

[0077] Data Collection: The drone module, equipped with ranging equipment, flies along a planned path and performs ranging operations. At each location, the drone performs two ranging operations, acquiring two sets of excavation status parameters. Simultaneously, a storage unit records the drone module's location and direction at the time of collection for each excavation status parameter.

[0078] (3) Constructing a real-life model of the building foundation pit

[0079] Once the drone module completes data collection at all points and distinguishes and stores the excavation status parameters in the storage unit, the reality modeling module is triggered. The module continuously acquires excavation status parameters from the storage unit, each time targeting a distinct storage interval, obtaining two ranging results. It then simultaneously picks up the drone module's position information, marked by these two ranging results, in three-dimensional space and draws a corresponding line segment based on the ranging results and the acquisition direction. The same operation is repeated for each distinct storage interval in the storage unit, drawing a large number of line segments. Finally, the endpoints of all drawn line segments are used as contour points, and adjacent contour points are connected to complete the construction of a closed reality model of the construction pit.

[0080] (IV) Model similarity comparison

[0081] The comparison module receives the standard model of the building foundation pit and the real-life model, and processes them according to the similarity comparison logic. Although no formula is involved, the principle is to comprehensively consider the similarity of the center of mass position, volume similarity, and density distribution similarity, and calculate the similarity of the two models through the set weights. Among them, the volume similarity takes into account the relationship between the minimum and maximum values ​​of the volumes of the two models, and the density distribution similarity is determined by dividing the model point cloud into macroscopic voxels and calculating the modulus of the point cloud density vector within each voxel. The tolerance factor is determined according to the point matrix density applied in the operation phase of the real-life modeling module and the size of the standard model of the building foundation pit. Since the point matrix density of this project is large and the standard model size is moderate, the tolerance factor is large.

[0082] (V) Safety Assessment

[0083] The judgment module sets a threshold for judging the safety of the building foundation pit morphology. When the similarity comparison result is greater than or equal to the threshold, the current building foundation pit excavation is judged to be safe; otherwise, it is unsafe. At the same time, the subordinate delayed judgment unit of the judgment module records the historical similarity comparison results of the comparison module. During the monitoring process, if no less than the latest three similarity comparison results are continuously declining, the current building foundation pit excavation is also judged to be unsafe. When the judgment results of the judgment module and the delayed judgment unit are both safe, the refresh module is triggered to run, and the system performs a new round of data collection and analysis; if any judgment result is unsafe, the traceability module is triggered to run.

[0084] (6) Defect location tracing

[0085] During a particular monitoring session, the judgment module and the delayed judgment unit produced unsafe results, triggering the traceability module. The module picked four arbitrary non-coplanar points in the real-life model of the building pit, which together form a closed three-dimensional model. It then found corresponding points for these four points in the standard model of the building pit, and aligned the two models based on these picked and corresponding points. During alignment, it was discovered that the southwest corner of the real-life model of the building pit protruded from the standard model. This area represented the defective location of the current building pit excavation.

[0086] (7) System refresh

[0087] After the traceability module is completed, the refresh module is triggered to run, and the system restarts a new round of work flow to continuously monitor the excavation status of the construction foundation pit to ensure construction safety and project quality.

[0088] In summary, in the above embodiment, during operation, the system constructs a standard model of the building pit and a real-scene model of the building pit respectively through the design specification parameters of the building pit and the real-scene acquisition parameters. By comparing the similarity of the two models, the safety of the building pit morphology is judged in real time in the middle and late stages of the building pit excavation. When the judgment result is unsafe, the defective position of the foundation pit excavation is traced in real time, so that the probability of defects in the foundation pit morphology is controlled during the excavation and forming process of the building pit, the backfill workload caused by excessive excavation is reduced, and the stable progress of the building pit excavation project is ensured.

[0089] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. The intelligent construction monitoring system for earthwork engineering based on data analysis is characterized by: include: This system is used in the middle and late stages of construction foundation pit excavation projects; A construction module is used to upload construction foundation pit specification parameters and construct a construction foundation pit standard model based on the construction foundation pit specification parameters; The real-scene modeling module is used to collect real-time excavation status parameters of the building foundation pit and build a real-scene model of the building foundation pit based on the excavation status parameters; The real scene modeling module is provided with a drone module, a planning unit and a storage unit at the lower level. The drone module is used to collect real-time excavation status parameters of the building foundation pit, the planning unit is used to plan the flight path of the drone module, and the storage unit is used to store the real-time excavation status parameters of the building foundation pit collected by the drone module. The drone module is equipped with a distance measuring device, and the drone module performs a distance measurement operation through the distance measuring device, and the distance measurement result is recorded as the construction foundation pit excavation state parameter; Among them, when the storage unit stores the construction foundation pit excavation state parameters, the drone module position information and collection direction when each construction foundation pit excavation state parameter is marked; A comparison module is used to receive the standard model of the building foundation pit constructed in the construction module and the real-scene model of the building foundation pit constructed in the real-scene modeling module, and compare the similarities between the two models; The judgment module is used to set the building foundation pit morphology safety judgment threshold, and compare the building foundation pit morphology safety judgment threshold with the similarity comparison result in the comparison module to determine whether the current building foundation pit excavation is safe; Refresh module, used to refresh system operation; The traceability module receives the standard model of the building foundation pit and the real-scene model of the building foundation pit, and traces the location of the current building foundation pit excavation defects based on the standard model of the building foundation pit and the real-scene model of the building foundation pit.

2. The intelligent construction monitoring system for earthwork engineering based on data analysis according to claim 1 is characterized in that: During the operation phase of the planning unit, a system user manually edits three sets of three-dimensional coordinates, determines a plane based on the three sets of three-dimensional coordinates, sets a point matrix on the determined plane, and performs a distance measurement operation on each point in the point matrix in a direction perpendicular to the plane, so that the distance measurement operation is performed on both sides of the point with the determined plane as the reference; The above operation is performed three times in the planning unit, two of which are performed according to the above operation to complete the distance operation in the four directions of front, back, left, and right of each point. The other time, the excavation surface of the building foundation pit is used as the plane, and the above operation is performed only in the vertical downward direction. Among them, the distance between each adjacent point in the point matrix is ​​equal, and the total number of points in the point matrix obeys the setting logic that the higher the construction accuracy requirement of the building foundation pit real scene model, the more points there are in the point matrix, and vice versa, the fewer points there are in the point matrix. The corresponding geometric representation of the plane determined by the three sets of three-dimensional coordinates in the building foundation pit real scene model is perpendicular to the bottom surface of the building foundation pit real scene model and parallel to the side of the building foundation pit real scene model.

3. The intelligent construction monitoring system for earthwork engineering based on data analysis according to claim 2 is characterized in that: The three sets of three-dimensional coordinates used to determine the plane do not have a collinear relationship. When the drone module performs ranging operations through the ranging device, it starts from any corner point in the point matrix, passes through all the midpoints of the point matrix, and performs two ranging operations with each point. When storing the excavation state parameters, the storage unit performs differentiated storage based on the position information of the drone module marked with the excavation state parameters, so that each differentiated storage interval stores two sets of excavation state parameters, that is, two ranging results, and the ranging results corresponding to different planes are differentiated and stored again based on the previous differentiated storage operation; After the drone and module pass through all the midpoints of the dot matrix and the excavation status parameters collected at each point in the dot matrix are distinguished and stored in the storage unit, the real scene modeling module is triggered to run.

4. The intelligent construction monitoring system for earthwork engineering based on data analysis according to claim 3 is characterized in that: During the operation phase of the real scene modeling module, the excavation state parameters are continuously acquired from the storage unit. Each time the excavation state parameters are acquired, a distinguished storage interval is used as an acquisition target, two ranging results are acquired, and the position information of the drone module marked by the two ranging results is simultaneously picked up in three-dimensional space. Then, a corresponding line segment is drawn based on the ranging results and the acquisition direction in the mark; Based on the above operation, the same operation is performed on each of the partitioned storage intervals in the storage unit to draw the same number of line segments as the partitioned storage intervals in the storage unit; After all the corresponding line segments that distinguish and store the intervals composed of two distance measurement results are drawn, the endpoints of all the drawn line segments are used as contour points, and the adjacent contour points are connected to each other to complete the construction of the closed building foundation pit real-scene model.

5. The intelligent construction monitoring system for earthwork engineering based on data analysis according to claim 1 is characterized in that: The similarity comparison logic between the standard model of the building foundation pit and the real-scene model of the building foundation pit in the comparison module is expressed as follows: ； Where: The similarity between the standard model of the building foundation pit to be revised and the real-scene model of the building foundation pit; 、 is the weight; is the centroid position similarity; is the volume similarity; is the density distribution similarity; The similarity between the revised standard model of the building foundation pit and the real-scene model of the building foundation pit; is the tolerance factor; in, 、 are all positive numbers, , are the centroid coordinates of the two model point clouds, For the overall scale of the scene, take the length of the longest side of the foundation pit.

6. The intelligent construction monitoring system for earthwork engineering based on data analysis according to claim 5 is characterized in that: The volume similarity Similarity to density distribution The calculation formula is: ; Where: It is the minimum value between the volume of the standard model of the building foundation pit and the real-scene model of the building foundation pit; It is the maximum volume between the standard model of the building foundation pit and the real-scene model of the building foundation pit; is the density vector, by dividing the model point cloud into Macro voxels, calculate the point cloud density within each voxel; is the magnitude of the vector; The tolerance factor The value is subject to: Tolerance Factor belongs to [0.9, 1]. The greater the density of the point matrix used in the operation phase of the real scene modeling module and the smaller the size of the standard model of the building foundation pit, the greater the tolerance factor. The larger the value, the smaller the point matrix density used in the operation phase of the real scene modeling module, and the larger the size of the standard model of the building foundation pit, the greater the tolerance factor. The smaller the value.

7. The intelligent construction monitoring system for earthwork engineering based on data analysis according to claim 1 is characterized in that: When the determination module determines that the similarity comparison result is greater than or equal to the building foundation pit morphology safety determination threshold, the current building foundation pit excavation is determined to be safe; otherwise, the current building foundation pit excavation is determined to be unsafe; The delay determination unit under the determination module is used to record the historical similarity comparison results of the comparison module, and when the similarity comparison results of at least the latest three times are continuously decreasing, it is determined that the current building foundation pit excavation is unsafe; Among them, when the judgment results of the judgment module and the delay judgment unit are both safe, the refresh module is triggered to run. When either the judgment results of the judgment module or the delay judgment unit are unsafe, the traceability module is triggered to run, and after the traceability module runs, the refresh module is triggered to run.

8. The intelligent construction monitoring system for earthwork engineering based on data analysis according to claim 1 is characterized in that: During the operation phase of the traceability module, four non-coplanar points are picked in the construction pit real scene module, and the four points can form a closed three-dimensional model. Corresponding points of the four non-coplanar points picked in the construction pit standard model are further picked in the construction pit real scene model. Based on the picked points and the corresponding points, the construction pit standard model and the construction pit real scene model are aligned. When the standard model of the building foundation pit is aligned with the real-scene model of the building foundation pit, the area where the real-scene model of the building foundation pit protrudes or is concave outside or inside the standard model of the building foundation pit is the location of the current excavation defect of the building foundation pit.

9. The intelligent construction monitoring system for earthwork engineering based on data analysis according to claim 1 is characterized in that: The construction module is interactively connected to the real-scene modeling module through a wireless network, the real-scene modeling module is interactively connected to the drone module, the planning unit and the storage unit through a wireless network, the real-scene modeling module is interactively connected to the comparison module and the judgment module through a wireless network, the comparison module is interactively connected to the storage unit through a wireless network, the comparison module is interactively connected to the judgment module through a wireless network, the judgment module is interactively connected to the delay judgment unit at the lower level through a wireless network, and the judgment module and the delay judgment module are interactively connected to the refresh module and the traceability module through a wireless network.

Images