BIM-based cofferdam deformation safety monitoring method and system based on photogrammetry

Abstract

The present application relates to the technical field of three-dimensional modeling analysis, in particular to a cofferdam deformation safety monitoring method and system based on BIM oblique photography. The present application registers cofferdam models of different monitoring periods and projects them into elevation grid maps, further determines the ground surface inclination angle between adjacent grids in the elevation grid map, compares it with the preset sliding angle, obtains the earthwork connectivity weight between adjacent grids, then simulates divergent distribution, calculates the earthwork transfer proportion from the earthwork loss grid to the earthwork new grid, combines with the grid distance, constructs the transfer cost model of the grid earthwork and solves the transfer route, further filters the mechanical transfer route to evaluate the cofferdam deformation monitoring result in each monitoring period. The present application introduces the transfer cost model based on the earthwork connectivity weight and transfer distance, so as to evaluate the mechanical handling cost proportion, accurately distinguish natural landslides and artificial mechanical handling, effectively intercept invalid safety alarms caused by daily construction, and improve the reliability of cofferdam deformation monitoring.

Description

Technical Field

[0001] This invention relates to the field of three-dimensional modeling and analysis technology, specifically to a method and system for monitoring the deformation safety of cofferdams based on BIM oblique photography. Background Technology

[0002] The safety of cofferdams in water conservancy projects is crucial to the overall stability of the project and the safety of downstream areas, and their deformation must be closely monitored. Currently, construction units often use drones to obtain current surface point clouds, subtract the elevation values ​​from the previous period's surface point clouds, and thus calculate the change in surface volume of the cofferdam. If the change exceeds a safe value, a structural deformation alarm is triggered to ensure the safety of the cofferdam.

[0003] However, the construction site of the cofferdam is frequently filled by machinery. The elevation subtraction method can only calculate the volume change at the excavation and dumping points, but cannot distinguish whether the difference in elevation is caused by vehicle transportation dumping or natural landslide. In addition, some outsourced vehicles are not connected to the digital platform and there are communication blind spots. It is impossible to obtain complete vehicle GPS tracks and mechanical status to help eliminate construction interference. As a result, there are many misjudgments and low availability of safety monitoring and early warning for cofferdam deformation. Summary of the Invention

[0004] To address the poor performance of existing cofferdam deformation safety monitoring methods, this invention aims to provide a cofferdam deformation safety monitoring method and system based on BIM oblique photography. The specific technical solution adopted is as follows: A method for monitoring the deformation and safety of cofferdams based on BIM oblique photography, the method comprising: The cofferdam model is obtained based on the oblique photogrammetry data of each monitoring period; the cofferdam models of different monitoring periods are registered based on BIM and simultaneously projected into a two-dimensional horizontal elevation grid map; Based on the elevation differences of the same coordinate grids between adjacent monitoring periods, the earthwork loss grids and earthwork addition grids in each elevation grid are obtained; in each monitoring period, the surface tilt angle between adjacent grids is determined according to the elevation and position differences between adjacent grids in the elevation grid, and the earthwork connectivity weight between adjacent grids is obtained based on the deviation of the surface tilt angle from the preset slip angle. Combining earthwork connectivity weights, the earthwork is distributed divergently starting from the earthwork loss grid. The earthwork transfer ratio from each earthwork loss grid to each earthwork addition grid is simulated and calculated within each monitoring period. Combined with the distance between grids, a grid earthwork transfer cost model is constructed. Based on the transfer cost model, the grid earthwork transfer route is solved. Among all the earthwork transfer routes in each monitoring period, mechanical transfer routes are selected based on the earthwork transfer ratio between grids, and the cofferdam deformation monitoring results are evaluated based on the mechanical transfer routes.

[0005] Furthermore, the method for obtaining the elevation grid map includes: The two-dimensional horizontal projection surface of the registered cofferdam model is divided into orthogonal grids to determine the orthogonal grid plane; the grid elevation value is determined based on the z-axis value of all discrete points in each grid and assigned to the corresponding grid to obtain the elevation grid map.

[0006] Furthermore, the method for obtaining the soil loss grid and the soil addition grid includes: For each grid in the elevation grid map of each monitoring period, subtract the grid elevation value of the grid with the same coordinates in the elevation grid map of the adjacent previous monitoring period from its grid elevation value, and combine it with the grid area to calculate the earthwork change volume of the corresponding grid. The grids whose earthwork change volume is greater than the preset increase volume are designated as earthwork increase grids; the grids whose earthwork change volume is less than the preset loss volume are designated as earthwork loss grids.

[0007] Furthermore, the method for obtaining the surface tilt angle includes: Between adjacent grids, the opposite side of the surface tilt angle is determined based on the difference between grid elevation values, and the adjacent side of the surface tilt angle is determined based on the straight-line distance between grid center points. The surface tilt angle is then solved by inversely based on the opposite and adjacent sides.

[0008] Furthermore, between adjacent grids, the tangent of the surface tilt angle is used as the first tangent, and the tangent of the preset slip angle is used as the second tangent; when the first tangent is less than the second tangent, the earthwork connectivity weight is set to 0; when the first tangent is greater than or equal to the second tangent, the earthwork connectivity weight is set to the normalized result of the first tangent.

[0009] Furthermore, the method for obtaining the earthwork transfer ratio includes: In each elevation grid map, the grid is used as a node, and the earthwork connectivity weight between adjacent grids is used as the edge capacity to construct an earthwork transfer topology map; For each earthwork loss grid, it is used as the starting node and assigned a preset unit ratio. Based on the edge capacity, the earthwork transfer topology map is diverged and distributed to determine the termination node of all earthwork distribution topology routes. The distribution cutoff condition is that the earthwork connectivity weight between the latest distributed node and its adjacent nodes is 0 or the cumulative number of distributions reaches a preset interception threshold. Calculate the cumulative product of the edge capacity of all edges between the starting node and the corresponding node of each newly added earthwork grid in each assigned topology route and the preset unit ratio. The sum of the cumulative product of all assigned topology routes between the starting node and the corresponding node of each newly added earthwork grid is used as the earthwork transfer ratio. The earthwork transfer ratio between the starting node and the corresponding nodes of the remaining newly added earthwork grids not included in the assigned topology routes is set to 0.

[0010] Furthermore, the method for obtaining the transfer cost model includes: Between each earthwork loss grid and each earthwork addition grid, if the earthwork transfer ratio is not 0, the grid center distance is used as the physical transfer cost; otherwise, the physical transfer cost is set to a preset transfer penalty value; preset virtual earthwork supply grids and virtual earthwork receiving grids are used, and the physical transfer cost between the virtual grids and each grid is set to a preset throughput penalty value; a cost matrix is ​​constructed based on the physical transfer cost between grids. The earthwork loss grid and the virtual earthwork supply grid are used as the loss grid, and the earthwork addition grid and the earthwork receiving grid are used as the addition grid; a variable earthwork transfer matrix is ​​preset, and the matrix elements in the variable earthwork transfer matrix are the earthwork transfer volume variables between the loss grid and the addition grid; wherein, the earthwork transfer between the loss grid and the addition grid satisfies the law of conservation of mass. Multiply each element of the cost matrix by the corresponding element in the variable earthwork transfer matrix to obtain the transfer cost model.

[0011] Furthermore, the method for obtaining the grid earthwork transfer route includes: The variable earthwork transfer matrix that minimizes the sum of all matrix elements in the transfer cost model is taken as the optimal earthwork transfer matrix. Each matrix element in the optimal earthwork transfer matrix corresponds to a grid earthwork transfer route.

[0012] Furthermore, the method for obtaining the mechanical transfer route includes: In the optimal earthwork transfer matrix, the physical transfer cost between grids corresponding to each matrix element is determined, and the earthwork transfer route of the grid with a physical transfer cost of a preset transfer penalty value is taken as the mechanical transfer route.

[0013] The system for monitoring the deformation safety of cofferdams based on BIM oblique photography includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the method for monitoring the deformation safety of cofferdams based on BIM oblique photography.

[0014] The present invention has the following beneficial effects: This invention acquires a cofferdam model based on oblique photography data from each monitoring period, providing an analytical basis for subsequent deformation and safety monitoring of the cofferdam. Then, it registers the cofferdam models from different monitoring periods using BIM to eliminate random aerial photography biases and projects them synchronously into a two-dimensional horizontal elevation grid map, providing comparable numerical units for subsequent grid-by-grid comparison and analysis of cofferdam deformation. Furthermore, based on the elevation differences of the same coordinate grids between adjacent monitoring periods, it acquires the soil loss grid and soil replenishment grid in each elevation grid map. In each monitoring period, based on the elevation and positional differences between adjacent grids in the elevation grid map, it determines the surface tilt angle between adjacent grids and, based on the surface tilt angle... For deviations in the preset sliding angle, soil connectivity weights are obtained to characterize whether soil can naturally slide and transfer between adjacent grids and the proportion of soil transfer. Further, starting from the soil loss grid, the distribution diverges outwards and downwards, simulating the control soil's progressive splitting and transmission. The proportion of soil transfer from each soil loss grid to each newly added soil grid is calculated within each monitoring period, and the transfer cost is analyzed in conjunction with the distance between grids, constructing a grid soil transfer cost model. The grid soil transfer routes are then solved based on the transfer cost model. Finally, among all grid soil transfer routes within each monitoring period, mechanical transport routes are selected based on the soil transfer ratio between grids. The cofferdam deformation monitoring results within each monitoring period are then evaluated based on the mechanical transport routes. This invention introduces a transfer cost model based on soil connectivity weights and transport distances to assess the proportion of mechanical transport costs, accurately distinguish between natural landslides and manual mechanical transport, effectively intercept invalid safety alarms caused by daily construction, and improve the reliability and practicality of cofferdam deformation monitoring. Attached Figure Description

[0015] Figure 1 A flowchart illustrating a method for monitoring the deformation safety of a cofferdam based on BIM oblique photography, provided as an embodiment of the present invention; Figure 2 This is a flowchart illustrating a method for obtaining a transfer cost model according to an embodiment of the present invention. Detailed Implementation

[0016] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a cofferdam deformation safety monitoring method and system based on BIM oblique photography proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0018] The following description, in conjunction with the accompanying drawings, details the specific scheme of the cofferdam deformation safety monitoring method and system based on BIM oblique photography provided by this invention.

[0019] Please see Figure 1 The diagram illustrates a flowchart of a method for monitoring the deformation safety of a cofferdam based on BIM oblique photography, according to an embodiment of the present invention, specifically including: Step S1: Obtain the cofferdam model based on the oblique photography data of each monitoring period; register the cofferdam models of different monitoring periods based on BIM and project them synchronously into a two-dimensional horizontal elevation grid map.

[0020] In one embodiment of the present invention, a cofferdam model is first obtained based on oblique photography data collected by UAV during each monitoring period, providing an analytical basis for subsequent deformation and safety monitoring of the cofferdam.

[0021] It should be noted that the cofferdam model is a 3D point cloud model; in this example, a daily monitoring period is used, and oblique photography is performed at 8:00 every day to obtain the cofferdam model.

[0022] Taking any monitoring period as an example, obtaining a cofferdam model based on oblique photogrammetry data is a well-known technical method, which will be briefly described here: The drone is equipped with a multi-lens oblique photography camera to fly along a flight path covering the entire cofferdam construction area, acquiring a multi-view aerial image sequence containing the surface texture and spatial features of the cofferdam. Photogrammetric algorithms are used to extract features from the aerial image sequence and perform aerial triangulation, and dense matching is performed to generate a 3D point cloud model containing X, Y, and Z coordinates.

[0023] Because the terrain at the cofferdam construction site is constantly evolving, and the absolute physical position of the drone's takeoff and flight trajectory may have objective deviations during each drone cruise for tilt photography, these deviations can cause the cofferdam models (3D point cloud models) collected at different monitoring periods to be spatially misaligned. However, the BIM model can provide a fixed 3D Cartesian coordinate system for the cofferdam during the design phase. Based on the fixed geometric benchmark provided by the BIM model, the cofferdam models collected at different monitoring periods can be forcibly registered to the same coordinate system, thereby eliminating the deviations caused by random drift between flights and providing an analytical basis for subsequent changes in the cofferdam surface.

[0024] Specifically, firstly, the BIM design model of the cofferdam is analyzed, and the three-dimensional rectangular coordinate system of the BIM design model is used as the absolute spatial coordinate system; in the stable hardened area around the cofferdam construction site that is not affected by mechanical excavation and filling operations (such as the surface of the surrounding mountain rock or permanent concrete structure), at least three reference target points with known absolute spatial coordinates are marked. The cofferdam model collected during each monitoring period is registered to the absolute spatial coordinate system of the aforementioned BIM design model to achieve synchronous registration. Taking any monitoring period as an example, the registration operation is a well-known technical method, which will be briefly described here: Identify the reference target points in the cofferdam model and extract their three-dimensional measurement coordinates. Perform target matching between the three-dimensional measurement coordinates of the reference target points and their absolute spatial coordinates. Through this target matching relationship, perform rigid translation and rotation operations on the three-dimensional point set in the cofferdam model to uniformly register and nest the cofferdam model to the absolute spatial coordinate system.

[0025] Since the three-dimensional mapping data corresponding to the cofferdam model is an irregular and discrete set of three-dimensional points, the point density and distribution may be different in different monitoring periods, making it impossible to directly compare point by point to assess the deformation of the cofferdam. Therefore, in this embodiment of the invention, after registering the cofferdam model for different monitoring periods, the cofferdam model is further projected synchronously into an elevation grid map in a two-dimensional horizontal plane. By converting the spatiotemporally heterogeneous point cloud data into regular grid data with the same resolution and topology, comparable numerical units are provided for subsequent grid-by-grid comparative analysis of the cofferdam deformation.

[0026] Preferably, in one embodiment of the present invention, the method for obtaining the elevation grid map includes: The two-dimensional horizontal projection surface of the registered cofferdam model is divided into orthogonal grids to determine the orthogonal grid plane; the grid elevation value is determined based on the z-axis value of all discrete points in each grid and assigned to the corresponding grid to obtain the elevation grid map.

[0027] Specifically, taking any monitoring period as an example, the two-dimensional horizontal projection surface (i.e., the XOY plane of the absolute spatial coordinate system) of the registered cofferdam model is orthogonally meshed to obtain an orthogonal mesh plane; in this example, the mesh size is 1m×1m, but the implementer can adjust it as needed; for each mesh of the orthogonal mesh plane, the z-axis value of the three-dimensional discrete point set that the projection falls into the mesh is averaged, and the average value is used as the mesh elevation value and assigned to the corresponding mesh to obtain an elevation mesh map; when there is no three-dimensional discrete point set in the mesh, bilinear interpolation is used to fill it using the elevation of adjacent meshes.

[0028] Step S2: Based on the elevation differences of the same coordinate grids between adjacent monitoring periods, obtain the earthwork loss grids and earthwork addition grids in each elevation grid map; in each monitoring period, determine the surface tilt angle between adjacent grids according to the elevation and position differences between adjacent grids in the elevation grid map, and obtain the earthwork connectivity weight between adjacent grids based on the deviation of the surface tilt angle from the preset slip angle.

[0029] The construction of a cofferdam is a continuous and dynamic process. If the cofferdam model at the current monitoring period is directly compared with the initial design reference surface from several weeks or even months ago, the result will include the cumulative amount of excavation and filling since construction began. This cumulative change in excavation and filling is meaningless for judging whether the cofferdam has deformed or landslided. Therefore, this embodiment of the invention compares the elevation differences of the same coordinate grids between elevation grid maps of adjacent monitoring periods, thereby analyzing the mechanical excavation damage and soil dumping and accumulation behavior that are newly added in a single monitoring period, assessing the preparation for sudden changes in the cofferdam in a single monitoring period, and obtaining the soil loss grid and soil addition grid in each elevation grid map; the soil loss grid represents the soil that has been excavated or slipped at the grid location, which is the loss end or supply end of soil transfer; the soil addition grid represents the soil that has been dumped at the grid location, which is the addition end or receiving end of soil transfer.

[0030] Preferably, in one embodiment of the present invention, considering that the area of ​​the elevation difference between grids at the same coordinate position in adjacent monitoring periods can be combined to assess the change in earthwork volume, thereby helping to determine whether earthwork is lost or added at the grid; therefore, the method for obtaining the earthwork loss grid and the earthwork addition grid includes: For each grid in the elevation grid map of each monitoring period, subtract the grid elevation value of the grid with the same coordinates in the elevation grid map of the adjacent previous monitoring period from its grid elevation value, and combine it with the grid area to calculate the earthwork change volume of the corresponding grid. Grids with earthwork changes exceeding the preset new volume are designated as new earthwork grids; grids with earthwork changes less than the preset loss volume are designated as earthwork loss grids.

[0031] Specifically, taking any monitoring period k (k≠1) as an example, for the m-th grid in the elevation grid map of monitoring period k, subtract the grid elevation value of the m-th grid in the elevation grid map of monitoring period k-1 from its grid elevation value to obtain the elevation difference between the coordinate grids; multiply the elevation difference by the grid area to obtain the earthwork change volume Vm between the coordinate grids. A preset grid tolerance volume V is defined, which is determined by multiplying the nominal elevation error value (e.g., ±0.05m) of the UAV survey by the grid area; the preset new volume is set to +V, and the preset loss volume is set to -V; where + and - represent the positive and negative signs. When the volume change of earthwork Vm < -V, it is determined that earthwork loss occurred in the m-th grid between adjacent monitoring periods (k and k-1); when the volume change of earthwork Vm > +V, it is determined that earthwork addition occurred in the m-th grid between adjacent monitoring periods (k and k-1); when the volume change of earthwork does not meet the above two conditions, it is determined that there is no change in earthwork.

[0032] Conventional cofferdam deformation monitoring can only identify changes in earth volume and cannot distinguish whether the changes in earth volume are caused by natural landslides or the transportation of engineering vehicles. However, the natural sliding of soil under gravity requires the surface slope to be greater than or equal to the internal friction angle of the soil (i.e., the critical angle of natural sliding). In each monitoring period, by analyzing the elevation differences and positional differences between adjacent grids in the elevation grid map, it is possible to help assess the angle between the slope and the horizontal plane, and thus help assess the surface tilt angle between adjacent grids. The surface tilt angle is the surface slope, which prepares for subsequent measurement of the deviation from the preset sliding angle.

[0033] Considering that in trigonometric geometry, the elevation difference between adjacent grids can represent the opposite side of the surface slope angle, and the line connecting the centers of adjacent grids can represent the adjacent side (not the hypotenuse) of the surface slope angle, the surface tilt angle can be solved by using trigonometric functions. Based on this, in a preferred embodiment of the present invention, the method for obtaining the surface tilt angle includes: Between adjacent grids, the opposite side of the surface tilt angle is determined based on the difference between grid elevation values, and the adjacent side of the surface tilt angle is determined based on the straight-line distance between grid center points. The surface tilt angle is then solved by inversely based on the opposite and adjacent sides.

[0034] Specifically, taking any monitoring period as an example, any grid in the elevation grid map is taken as the target grid, and its 8 neighboring grids (including diagonally adjacent grids) are taken as reference grids. Between the target grid and each reference grid, the grid elevation value of the target grid is subtracted from the grid elevation value of the reference grid, and the difference in grid elevation is taken as the opposite side length of the surface tilt angle. The straight-line distance between the center of the target grid and the center of the reference grid is taken as the adjacent side length of the surface tilt angle. The opposite side length is taken as the numerator, and the adjacent side length is taken as the denominator (which cannot be 0). The ratio is then subjected to arctangent calculation to solve for the surface tilt angle θ.

[0035] Specifically, when the surface tilt angle θ > 0, it means that the target grid is higher than the reference grid, and the target grid points towards the reference grid as a downhill slope; when the surface tilt angle θ = 0, it means that the target grid and the reference grid are at the same elevation and horizontal; when the surface tilt angle θ < 0, it means that the target grid is lower than the reference grid, and the target grid points towards the reference grid as an uphill slope.

[0036] After determining the surface tilt angle between the target grid and the reference grid, the earthwork connectivity weight between the target grid and each reference grid is obtained based on the deviation of the surface tilt angle from the preset slip angle. The earthwork connectivity weight is a directional weight, which represents the proportion of earthwork in the target grid that naturally slips and transfers to the reference grid, in preparation for subsequent analysis of the transfer route.

[0037] Preferably, in one embodiment of the present invention, the method for obtaining the earthwork connectivity weight includes: Between adjacent grids, the tangent of the surface tilt angle is used as the first tangent, and the tangent of the preset slip angle is used as the second tangent. When the first tangent is less than the second tangent, the earthwork connectivity weight is set to 0. When the first tangent is greater than or equal to the second tangent, the earthwork connectivity weight is set to the normalized result of the first tangent.

[0038] Specifically, a preset slip angle is set. The preset slip angle is the critical angle for the natural slip of the construction filling soil (representing the maximum slope limit at which the soil can remain stable and not slide down by its own static friction without the aid of external mechanical devices or external forces). In this example, taking general clay as an example, the preset slip angle is set to 30°. Between the target grid and each reference grid, the tangent of the surface tilt angle is used as the first tangent (or the tangent can be determined directly without solving the surface tilt angle in the above process); the tangent of the preset slip angle is used as the second tangent. When the first tangent value is less than the second tangent value, it indicates that the surface condition of the target grid pointing to the reference grid is uphill or a relatively gentle downhill slope, which is insufficient to cause soil collapse and landslide under natural gravity. Therefore, the soil connectivity weight is set to 0. When the first tangent value is greater than or equal to the second tangent value, it indicates that the surface condition of the target grid pointing to the reference grid is a steep downhill slope, which has the slope conditions to cause soil to slide downhill. The larger the first tangent value, the greater the degree of landslide. Therefore, the soil connectivity weight is set to the normalized result of the first tangent value.

[0039] The specific method of normalization is as follows: calculate the sum of earthwork connectivity weights between the target grid and all reference grids. When the sum is 0, it indicates that the target grid is neither lost nor added, and is in a stable state. When the sum is not 0, divide the earthwork connectivity weight between the target grid and each reference grid by the sum value to perform normalization, so that the sum of all earthwork connectivity weights radiating outward from the target grid is constant at 1.

[0040] By changing the target grid and redetermining the reference grid, the surface tilt angle between all adjacent grids in the elevation grid map for each monitoring period can be obtained. Furthermore, the earthwork connectivity weight between all adjacent grids can be determined.

[0041] Step S3: Combining earthwork connectivity weights, the earthwork is distributed divergently starting from the earthwork loss grid. The earthwork transfer ratio from each earthwork loss grid to each earthwork addition grid is simulated and calculated within each monitoring period. Combined with the distance between grids, a grid earthwork transfer cost model is constructed. The grid earthwork transfer route is solved based on the transfer cost model.

[0042] Since real-world landslides do not move along a single fixed straight line, but rather exhibit a fan-shaped collapse distribution from a high-altitude source to the surrounding lower-altitude areas, in order to simulate soil loss, this embodiment of the invention will use the grid where soil loss occurs as the divergence source, and start the divergence distribution from the soil loss grid. Combined with the soil connectivity weight, which characterizes the proportion of soil transfer between adjacent grids, the invention simulates the outward splitting and downward transmission of soil, and calculates the proportion of soil transfer from each soil loss grid to each newly added soil grid in each monitoring period. The earthwork transfer ratio simulation quantifies the proportion of high-altitude soil within the earthwork loss grid that naturally rolls down steep slopes and settles in various low-lying newly added earthwork grids, rather than precisely calculating the physical transfer quality of the soil.

[0043] It should be noted that, although landslides from different sources can occur simultaneously, they are physically independent events. Therefore, the proportion of earthwork transfer from each earthwork loss grid to each earthwork addition grid is simulated and tested separately, i.e., asynchronously, in order to accurately distinguish the divergent distribution contribution of each source (earthwork loss grid) and avoid introducing interference from other sources.

[0044] Preferably, in one embodiment of the present invention, considering that the evaluation of the earthwork transfer ratio can be regarded as the topological connectivity of the grid, an earthwork transfer topology graph (network) is first constructed using the earthwork connectivity weight of the earthwork transfer ratio between adjacent grids. Then, starting with the earthwork loss grid as the starting node (divergent source), a step-by-step outward connectivity transfer (divergent allocation) is performed. The mathematical expectation (i.e., the sliding ratio) of soil sliding down the steep slope topological channel (with a larger earthwork connectivity weight) to each grid under pure gravity drive is iteratively calculated, thereby determining the final earthwork transfer ratio. The method for obtaining the earthwork transfer ratio includes: In each elevation grid map, the grid is used as a node, and the earthwork connectivity weight between adjacent grids is used as the edge capacity to construct an earthwork transfer topology map. For each earthwork loss grid, it is used as the starting node and assigned a preset unit ratio. Based on the edge capacity, the earthwork transfer topology map is diverged and distributed to determine the termination node of all earthwork distribution topology routes. The distribution cutoff condition is that the earthwork connectivity weight between the latest distributed node and its adjacent nodes is 0 or the cumulative number of distributions reaches a preset interception threshold. Calculate the cumulative product of the edge capacity of all edges between the starting node and the corresponding node of each newly added earthwork grid in each assigned topology route and the preset unit ratio. The sum of the cumulative product of all assigned topology routes between the starting node and the corresponding node of each newly added earthwork grid is used as the earthwork transfer ratio. The earthwork transfer ratio between the starting node and the corresponding nodes of the remaining newly added earthwork grids not included in the assigned topology routes is set to 0.

[0045] Specifically, taking any monitoring period as an example, in the elevation grid map, all grids are treated as nodes, directed edges are constructed between corresponding nodes of adjacent grids, and the earthwork connectivity weight between adjacent grids is assigned as the edge capacity to the corresponding directed edge. Then, the earthwork transfer topology map is constructed by combining all nodes and all edge capacities. Existing technical methods will not be elaborated here. In the earthwork transfer topology map, take any earthwork loss grid corresponding node as the starting node, set the initial ratio of the starting node to the preset unit ratio 1 (representing 100%), and set the initial ratio of the remaining nodes to 0. Starting from the initial node, the distribution is distributed outwards level by level along the topology: For each directed edge connected to the initial node i, when the edge capacity (earthwork connectivity weight) of the directed edge is > 0, the initial proportion 1 of the initial node is multiplied by the edge capacity, and the product is assigned to the other node q connected by the directed edge, thus obtaining the proportion share received by node q from the initial node i; further, the distribution is passed down level by level from node q (divergent distribution) until the distribution ends. The allocation cutoff condition is that the earthwork connectivity weight between the latest allocated node and its adjacent nodes is 0 (i.e., it cannot continue to connect and pass down); or the cumulative number of allocations reaches the preset interception threshold. The preset interception threshold represents the longest transfer path and is set to a minimum of 1000. In this example, it is set to 1000 to prevent the divergent allocation from failing to meet the theoretical convergence condition and looping infinitely due to numerical errors or extreme terrain. When any of the above allocation cutoff conditions are met, allocation stops; the node corresponding to the cutoff point is taken as the termination node of the earthwork allocation topology route that starts from the initial node i. In each assigned topology route, all newly added earthwork grid nodes along the path are selected. Between the starting node and each corresponding node of the newly added earthwork grid, the cumulative product of the edge capacity of all edges along the path and the preset unit ratio of 1 is calculated. Since there may be multiple assigned topology routes between the starting node and each corresponding node of the newly added earthwork grid, the sum of the cumulative product of all assigned topology routes between the starting node and each corresponding node of the newly added earthwork grid is used as the earthwork transfer ratio between the starting node and each corresponding node of the newly added earthwork grid. In addition, the earthwork transfer ratio between the starting node and the other corresponding nodes of the newly added earthwork grid not included in the assigned topology routes is set to 0.

[0046] When the earthwork transfer ratio Pij > 0, it means that there is one or more continuous paths consisting entirely of steep slopes between the earthwork loss grid i and the earthwork addition grid j, and the soil can roll down from i to j by gravity. When the earthwork transfer ratio Pij = 0, it means that the earthwork loss grid i and the earthwork addition grid j are blocked by at least a section of gentle or uphill terrain, and natural sliding is not possible. Earthwork transfer needs to rely on mechanical handling or other external forces.

[0047] Since the earthwork transfer ratio can indirectly characterize whether there is a continuous steep slope path between the earthwork loss grid and the earthwork addition grid, it can help assess whether it can slide naturally. However, from an engineering physics perspective, soil slides naturally tend to accumulate within a short distance. Therefore, it is also necessary to combine the distance between the earthwork loss grid and the earthwork addition grid to measure the transfer cost, in order to prepare for the subsequent solution of the grid earthwork transfer route. Based on this, after obtaining the earthwork transfer ratio from the earthwork loss grid to the earthwork addition grid, this embodiment of the invention further combines the distance between grids to construct a grid earthwork transfer cost model. The grid earthwork transfer cost model, combining the earthwork transfer ratio and transfer distance cost, transforms the real earthwork transfer problem into a computable operations research optimization problem, so as to solve the global optimal matching of multiple sources and sinks (multiple earthwork loss grids and multiple earthwork addition grids) and prepare for simulating and determining the earthwork transfer routes of all grids.

[0048] Preferably, in one embodiment of the present invention, please refer to Figure 2 The flowchart illustrates a method for obtaining a transfer cost model according to an embodiment of the present invention, specifically including: Step S301: Between each earthwork loss grid and each earthwork addition grid, if the earthwork transfer ratio is not 0, the grid center distance is used as the physical transfer cost; otherwise, the physical transfer cost is set to a preset transfer penalty value; preset virtual earthwork supply grids and virtual earthwork receiving grids, and set the physical transfer cost between the virtual grids and each grid to a preset throughput penalty value; construct a cost matrix based on the physical transfer cost between grids.

[0049] Considering that if the earthwork transfer ratio is not 0, it means that the earthwork loss grid and the earthwork new grid can be transferred by natural sliding, so the distance between grids can be directly used as the physical transfer cost of earthwork; while if the earthwork transfer ratio is 0, it means that the earthwork transfer between the earthwork loss grid and the earthwork new grid needs to rely on mechanical handling or other external forces, so a preset transfer penalty value needs to be set to characterize the physical transfer cost of earthwork. Furthermore, considering that there may be situations where soil is purchased from outside the site and transported in, or waste soil is transported out of the site, such external intervention will cause the total amount of soil loss during the monitoring period to be completely different from the total amount of soil added. If a forced match is made directly under the unbalanced total supply and demand, the calculation process will directly report an error and cannot be solved. Therefore, it is necessary to introduce a virtual throughput grid with a preset throughput penalty value to balance the difference between soil inside and outside the site. Among them, the preset throughput penalty value is used to characterize the physical transfer cost of external transfer other than natural sliding and mechanical handling, and it has the highest penalty cost. Furthermore, a cost matrix can be constructed based on the physical transfer cost, where each matrix element represents the physical transfer cost of earthwork between the earthwork loss grid and the earthwork addition grid.

[0050] Specifically, taking any monitoring period as an example, firstly, between each earthwork loss grid and each earthwork addition grid, if the earthwork transfer ratio is not 0, the grid center distance (Euclidean distance in the two-dimensional horizontal projection plane) is used as the physical transfer cost C1; otherwise, the physical transfer cost is set as the preset transfer penalty value; the preset transfer penalty value C2 is determined by extracting the diagonal length of the two-dimensional horizontal projection plane of the cofferdam's BIM design model, multiplying the diagonal length by a preset multiple such as 1000, and obtaining the preset transfer penalty value to characterize the maximum penalty for mechanical transfer; Then, a virtual earthwork supply grid and a virtual earthwork receiving grid are preset; the virtual earthwork supply grid is used to simulate the transportation of waste soil out of the site, and the virtual earthwork receiving grid is used to simulate the transportation of soil purchased from outside the site; the physical transfer cost between the virtual grid and each grid is set as a preset throughput penalty value; the preset throughput penalty value C3 is determined by multiplying the preset transfer penalty value by a preset penalty multiplier, such as 10, to obtain the preset throughput penalty value, which is used to characterize the highest penalty cost for external transfers other than natural sliding and mechanical handling; The total loss volume is calculated by summing the accumulated earthwork changes of all earthwork loss grids and the total new volume is calculated by summing the accumulated earthwork changes of all earthwork new grids. The maximum value of the two total volumes is taken as the off-site earthwork compensation capacity (i.e., earthwork change volume during the monitoring period) to accommodate any imbalance difference, and is assigned to the virtual earthwork supply grid and the virtual earthwork receiving grid respectively. The earthwork loss grid and the virtual earthwork supply grid are taken as loss grids and set as the row dimension of the matrix. The earthwork addition grid and the virtual earthwork receiving grid are taken as addition grids and set as the column dimension of the matrix. Then, a cost matrix is ​​constructed. The matrix element in the p-th row and q-th column of the cost matrix represents the physical transfer cost of earthwork in the p-th loss grid to the q-th addition grid.

[0051] It should be noted that implementers can also adjust the above penalty values ​​themselves, but the physical transfer cost (grid distance) C1 must be less than the preset transfer penalty value C2 and the preset throughput penalty value C3.

[0052] Step S302: The earthwork loss grid and the virtual earthwork supply grid are used as loss grids, and the earthwork addition grid and the earthwork receiving grid are used as addition grids; a variable earthwork transfer matrix is ​​preset, and the matrix elements in the variable earthwork transfer matrix are the earthwork transfer volume variables between the loss grid and the addition grid; wherein, the earthwork transfer between the loss grid and the addition grid satisfies the law of conservation of mass.

[0053] A variable earthwork transfer matrix is ​​further preset, that is, the matrix elements are variable, and its size and row and column sorting are consistent with the cost matrix; the matrix element in the p-th row and q-th column of the variable earthwork transfer matrix represents the earthwork transfer volume variable Xpq transferred from the p-th lost grid to the q-th newly added grid; Among them, the earthwork transfer of lost grids and newly added grids satisfies the law of conservation of mass; that is, for each lost grid, the cumulative earthwork transfer volume variable distributed outward must be strictly equal to the absolute value of its earthwork change volume; for each newly added grid, the cumulative earthwork transfer volume variable received in total must be strictly equal to the absolute value of its earthwork change volume.

[0054] It should be noted that a (variable) earthwork transfer matrix represents an earthwork transfer scheme, and each matrix element represents the specific earthwork transfer volume transferred from the lost grid to the new grid under the earthwork transfer scheme.

[0055] Step S303: Multiply each element of the cost matrix with the corresponding element in the variable earthwork transfer matrix to obtain the transfer cost model.

[0056] It should be noted that the transfer cost model is in matrix form and changes with the variable earthwork transfer matrix. Each matrix element in the transfer cost model represents the transfer cost from the lost grid to the new grid under this earthwork transfer scheme, which prepares for the subsequent solution of the optimal earthwork transfer scheme to obtain the earthwork transfer routes of all grids.

[0057] Further, the grid earthwork transfer route is solved based on the transfer cost model.

[0058] Preferably, in one embodiment of the present invention, the method for obtaining the grid earthwork transfer route includes: The variable earthwork transfer matrix that minimizes the sum of all matrix elements in the transfer cost model is taken as the optimal earthwork transfer matrix. Each matrix element in the optimal earthwork transfer matrix corresponds to a grid earthwork transfer route.

[0059] Specifically, as the variable earthwork transfer matrix changes, the sum of all matrix elements in the transfer cost model under each variable earthwork transfer matrix is ​​calculated. The sum represents the total global earthwork transfer cost for multiple sources and sinks (multiple lost grids and multiple newly added grids). The variable earthwork transfer matrix corresponding to the minimum cumulative sum is selected as the optimal earthwork transfer matrix. This optimal earthwork transfer matrix approximates and restores the most labor-saving physical earthwork transfer scheme in real engineering on a numerical level. Furthermore, the route from the lost grid to the new grid corresponding to each matrix element in the optimal earthwork transfer matrix is ​​taken as a grid earthwork transfer route.

[0060] Step S4: Among all the earthwork transfer routes in each monitoring period, mechanical transfer routes are selected based on the earthwork transfer ratio between grids, and the cofferdam deformation monitoring results are evaluated based on the mechanical transfer routes.

[0061] Considering that when the earthwork transfer ratio is equal to 0, it means that the earthwork loss grid and the earthwork addition grid are blocked by at least a section of gentle or uphill terrain, and natural sliding is not possible, earthwork transfer needs to rely on mechanical handling or other external forces; therefore, in this embodiment of the invention, among all the earthwork transfer routes in each monitoring period, the mechanical transportation route for earthwork transfer is selected based on the earthwork transfer ratio, in order to prepare for the subsequent differentiation of cofferdam deformation into mechanical earthwork transportation or natural sliding.

[0062] Preferably, in one embodiment of the present invention, considering that when the earthwork transfer ratio is equal to 0, the physical transfer cost between grids is set to a preset transfer penalty value, the transfer routes that require mechanical handling or other external forces can be screened from all earthwork transfer routes in the optimal earthwork transfer matrix; therefore, the method for obtaining mechanical transfer routes includes: In the optimal earthwork transfer matrix, the physical transfer cost between grids corresponding to each matrix element is determined, and the earthwork transfer route of the grid with a physical transfer cost of a preset transfer penalty value is taken as the mechanical transfer route.

[0063] Finally, the monitoring results of cofferdam deformation during each monitoring period are evaluated based on the mechanical transport route.

[0064] In one embodiment of the present invention, the matrix elements corresponding to each mechanical transport route are extracted from the optimal earthwork transfer matrix, the matrix elements under the same row and column index in the cost matrix are extracted, and the product of the two matrix elements is taken as the mechanical transport cost of the mechanical transport route; the sum of the mechanical transport costs of all mechanical transport routes is taken as the cumulative mechanical cost, and the cost ratio of the cumulative mechanical cost in the total global earthwork transfer cost is calculated. It should be noted that the total cost of global earthwork transfer refers to the sum of all matrix elements in the optimal earthwork transfer matrix. In addition, matrix elements containing virtual grids should be removed from the sum to avoid the penalty value being too large, which would seriously dilute the proportion of mechanical transfer and cause the alarm to fail. Then, a cost percentage threshold is set to characterize the cost percentage of earthwork transfer during conventional mechanical excavation, filling, and transportation, serving as a basis for distinguishing between conventional excavation, filling, and transportation and natural landslides. The cost percentage threshold is determined in stages: In the early stages of monitoring (e.g., the first 3 days), the cost percentage threshold is directly set to the operational experience threshold (e.g., 50%, indicating that half of the cost is generated by conventional mechanical excavation, filling and transportation). In the later stages of monitoring (after 3 days), a dynamic calculation method is used to calculate the average cost percentage in each historical monitoring period in which no landslides or other deformation hazards have occurred. The average is then added to a preset fluctuation tolerance (e.g., 10%) to obtain the cost percentage threshold. Comparing the cost ratio with the cost ratio threshold during the current monitoring period, when the cost ratio is greater than or equal to the cost ratio threshold, it indicates that a large amount of earthwork transfer during the monitoring period may be caused by daily manual excavation and filling operations by excavators or dump trucks, and the possibility of natural landslide deformation of the cofferdam is relatively low; when the cost ratio is less than the cost ratio threshold, it indicates that it may be caused by a real natural landslide, triggering structural deformation or landslide warnings, which will not be elaborated further.

[0065] Based on the same inventive concept, this invention also proposes a cofferdam deformation safety monitoring system based on BIM oblique photography. The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the cofferdam deformation safety monitoring method based on BIM oblique photography described in steps S1-S4.

[0066] In summary, this invention acquires a cofferdam model based on oblique photography data for each monitoring period; registers the cofferdam models for different monitoring periods using BIM and projects them synchronously into a two-dimensional horizontal elevation grid map; based on the elevation differences of the same coordinate grids between adjacent monitoring periods, it acquires the soil loss grid and soil addition grid in each elevation grid map; in each monitoring period, based on the elevation and position differences between adjacent grids in the elevation grid map, it determines the surface tilt angle between adjacent grids, and based on the deviation of the surface tilt angle from the preset slip angle, it acquires the soil connectivity weight between adjacent grids; combined with the soil connectivity weight, it diverges and distributes soil starting from the soil loss grid, simulating and calculating the soil transfer ratio from each soil loss grid to each soil addition grid in each monitoring period, and constructs a grid soil transfer cost model based on the distance between grids; it solves the grid soil transfer route based on the transfer cost model; among all grid soil transfer routes in each monitoring period, it selects mechanical transfer routes based on the soil transfer ratio between grids, and evaluates the cofferdam deformation monitoring results in each monitoring period based on the mechanical transfer routes. This invention introduces a transfer cost model based on earthwork connectivity weight and transport distance to assess the proportion of mechanical handling costs, accurately distinguish between natural landslides and manual mechanical handling, effectively intercept invalid safety alarms caused by daily construction, and improve the reliability and practicality of cofferdam deformation monitoring.

Claims

1. A method for monitoring the deformation and safety of cofferdams based on BIM oblique photography, characterized in that, The method includes: The cofferdam model is obtained based on the oblique photogrammetry data of each monitoring period; the cofferdam models of different monitoring periods are registered based on BIM and simultaneously projected into a two-dimensional horizontal elevation grid map; Based on the elevation differences of the same coordinate grids between adjacent monitoring periods, the earthwork loss grids and earthwork addition grids in each elevation grid are obtained; in each monitoring period, the surface tilt angle between adjacent grids is determined according to the elevation and position differences between adjacent grids in the elevation grid, and the earthwork connectivity weight between adjacent grids is obtained based on the deviation of the surface tilt angle from the preset slip angle. Combining earthwork connectivity weights, the earthwork is distributed divergently starting from the earthwork loss grid. The earthwork transfer ratio from each earthwork loss grid to each earthwork addition grid is simulated and calculated within each monitoring period. Combined with the distance between grids, a grid earthwork transfer cost model is constructed. Based on the transfer cost model, the grid earthwork transfer route is solved. Among all the earthwork transfer routes in each monitoring period, mechanical transfer routes are selected based on the earthwork transfer ratio between grids, and the cofferdam deformation monitoring results are evaluated based on the mechanical transfer routes.

2. The method for monitoring the deformation and safety of cofferdams based on BIM oblique photography according to claim 1, characterized in that, The method for obtaining the elevation grid map includes: The two-dimensional horizontal projection surface of the registered cofferdam model is divided into orthogonal grids to determine the orthogonal grid plane; the grid elevation value is determined based on the z-axis value of all discrete points in each grid and assigned to the corresponding grid to obtain the elevation grid map.

3. The method for monitoring the deformation and safety of cofferdams based on BIM oblique photography according to claim 1, characterized in that, The methods for obtaining the soil loss grid and the soil addition grid include: For each grid in the elevation grid map of each monitoring period, subtract the grid elevation value of the grid with the same coordinates in the elevation grid map of the adjacent previous monitoring period from its grid elevation value, and combine it with the grid area to calculate the earthwork change volume of the corresponding grid. The grids whose earthwork change volume is greater than the preset increase volume are designated as earthwork increase grids; the grids whose earthwork change volume is less than the preset loss volume are designated as earthwork loss grids.

4. The method for monitoring the deformation and safety of cofferdams based on BIM oblique photography according to claim 1, characterized in that, The method for obtaining the surface tilt angle includes: Between adjacent grids, the opposite side of the surface tilt angle is determined based on the difference between grid elevation values, and the adjacent side of the surface tilt angle is determined based on the straight-line distance between grid center points. The surface tilt angle is then solved by inversely based on the opposite and adjacent sides.

5. The method for monitoring the deformation and safety of cofferdams based on BIM oblique photography according to claim 1, characterized in that, The method for obtaining the earthwork connectivity weight includes: Between adjacent grids, the tangent of the surface tilt angle is used as the first tangent, and the tangent of the preset slip angle is used as the second tangent. When the first tangent is less than the second tangent, the earthwork connectivity weight is set to 0. When the first tangent is greater than or equal to the second tangent, the earthwork connectivity weight is set to the normalized result of the first tangent.

6. The method for monitoring the deformation and safety of cofferdams based on BIM oblique photography according to claim 1, characterized in that, The method for obtaining the earthwork transfer ratio includes: In each elevation grid map, the grid is used as a node, and the earthwork connectivity weight between adjacent grids is used as the edge capacity to construct an earthwork transfer topology map; For each earthwork loss grid, it is used as the starting node and assigned a preset unit ratio. Based on the edge capacity, the earthwork transfer topology map is diverged and distributed to determine the termination node of all earthwork distribution topology routes. The distribution cutoff condition is that the earthwork connectivity weight between the latest distributed node and its adjacent nodes is 0 or the cumulative number of distributions reaches a preset interception threshold. Calculate the cumulative product of the edge capacity of all edges between the starting node and the corresponding node of each newly added earthwork grid in each assigned topology route and the preset unit ratio. The sum of the cumulative product of all assigned topology routes between the starting node and the corresponding node of each newly added earthwork grid is used as the earthwork transfer ratio. The earthwork transfer ratio between the starting node and the corresponding nodes of the remaining newly added earthwork grids not included in the assigned topology routes is set to 0.

7. The method for monitoring the deformation and safety of cofferdams based on BIM oblique photography according to claim 1, characterized in that, The method for obtaining the transfer cost model includes: Between each earthwork loss grid and each earthwork addition grid, if the earthwork transfer ratio is not 0, the grid center distance is used as the physical transfer cost; otherwise, the physical transfer cost is set to a preset transfer penalty value; preset virtual earthwork supply grids and virtual earthwork receiving grids are used, and the physical transfer cost between the virtual grids and each grid is set to a preset throughput penalty value; a cost matrix is ​​constructed based on the physical transfer cost between grids. The earthwork loss grid and the virtual earthwork supply grid are used as the loss grid, and the earthwork addition grid and the earthwork receiving grid are used as the addition grid; a variable earthwork transfer matrix is ​​preset, and the matrix elements in the variable earthwork transfer matrix are the earthwork transfer volume variables between the loss grid and the addition grid; wherein, the earthwork transfer between the loss grid and the addition grid satisfies the law of conservation of mass. Multiply each element of the cost matrix by the corresponding element in the variable earthwork transfer matrix to obtain the transfer cost model.

8. The method for monitoring the deformation and safety of cofferdams based on BIM oblique photography according to claim 7, characterized in that, The method for obtaining the grid earthwork transfer route includes: The variable earthwork transfer matrix that minimizes the sum of all matrix elements in the transfer cost model is taken as the optimal earthwork transfer matrix. Each matrix element in the optimal earthwork transfer matrix corresponds to a grid earthwork transfer route.

9. The method for monitoring the deformation and safety of cofferdams based on BIM oblique photography according to claim 8, characterized in that, The method for obtaining the mechanical transfer route includes: In the optimal earthwork transfer matrix, the physical transfer cost between grids corresponding to each matrix element is determined, and the earthwork transfer route of the grid with a physical transfer cost of a preset transfer penalty value is taken as the mechanical transfer route.

10. A cofferdam deformation safety monitoring system based on BIM oblique photography, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the cofferdam deformation safety monitoring method based on BIM oblique photography as described in any one of claims 1 to 9.

Images