Deformation monitoring method and apparatus for deep foundation pit slope, and storage medium and electronic device

Abstract

A deformation monitoring method for a deep foundation pit slope, the method comprising: acquiring monitoring data of a deep foundation pit slope (S1); on the basis of the monitoring data and a timestamp, constructing three-dimensional deformation field models of the deep foundation pit slope (S2); segmenting each of the three-dimensional deformation field models, each of which is constructed at a plurality of time points, into a plurality of deformation units, and obtaining time-series deformation information of each deformation unit (S3); inputting the time-series deformation information of the deformation units into a trained prediction network model, so as to obtain a predicted deformation amount of each deformation unit (S4); and on the basis of the predicted deformation amounts of the plurality of deformation units, obtaining the risk status of the deep foundation pit slope (S5). By means of the method, a three-dimensional deformation field model is segmented and refined, and changes of each deformation unit are monitored, thereby enabling early warning before risk statuses occur or at an initial stage of the occurrence of the risk status, and thus improving the timeliness of early warning. Further provided are a deformation monitoring apparatus for a deep foundation pit slope, and a storage medium and an electronic device.

Description

Methods, devices, storage media and electronic equipment for monitoring the deformation of deep foundation pit slopes Technical Field

[0001] This invention relates to the field of deep foundation pit monitoring technology, and more specifically, to a method, device, storage medium, and electronic equipment for monitoring the deformation of deep foundation pit slopes. Background Technology

[0002] With the increasing number of high-rise buildings, the scale of underground foundation pit excavation is also increasing. During the excavation of foundation pits and the construction of underground projects, the stability and safety of the structural facade and the foundation pit slope have become key aspects of project monitoring.

[0003] In order to detect problems in a timely manner and avoid accidents, deformation monitoring of deep foundation pits is required during the construction and operation of buildings. In traditional deformation monitoring, methods such as total stations or levels are often used.

[0004] Single monitoring methods can only monitor the deformation of deep foundation pit slopes in one dimension. Therefore, the potential risks of deep foundation pit slopes analyzed based on data obtained from traditional deformation monitoring methods are also relatively one-sided, resulting in inaccurate deformation predictions and untimely safety warnings. Summary of the Invention

[0005] The problem that this invention aims to solve is that a single monitoring method can only monitor the deformation of deep foundation pit slopes in one dimension, resulting in inaccurate deformation prediction and untimely safety warnings.

[0006] To address the aforementioned problems, in a first aspect, the present invention provides a method for monitoring the deformation of deep foundation pit slopes, comprising:

[0007] Acquire monitoring data of deep foundation pit slopes, wherein the monitoring data includes slope surface data, slope internal horizontal displacement data, slope internal settlement data, and deep foundation pit slope three-dimensional data;

[0008] Based on the monitoring data and timestamps, a three-dimensional deformation field model of the deep foundation pit slope is constructed;

[0009] Each of the three-dimensional deformation field models constructed at multiple time points is divided into multiple deformation units to obtain the temporal deformation information of each deformation unit, wherein the deformation unit is a three-dimensional model obtained by dividing according to a preset size;

[0010] The temporal deformation information of the deformation unit is input into the trained prediction network model to obtain the predicted deformation amount of each deformation unit.

[0011] The risk status of the deep foundation pit slope is obtained based on the predicted deformation amount of the multiple deformation units.

[0012] Optionally, after acquiring the monitoring data of the deep foundation pit slope, the deep foundation pit slope deformation monitoring method further includes:

[0013] The preprocessing of the slope internal settlement data in the monitoring data includes the settlement amount, which is:

[0014] in, Let ε(z) represent the soil settlement measured by the optical fiber between positions z1 and z2, and let f represent the strain of the optical fiber. 0 f(z) represents the initial Brillouin scattering frequency shift of the fiber at position z, and C represents the initial Brillouin scattering frequency shift of the fiber at position z. s C represents the proportionality coefficient between the frequency shift of the backscattered Brillouin light from the optical fiber and the strain of the optical cable. T The coefficient representing the ratio of the frequency shift of the backscattered Brillouin light from the optical fiber to the temperature of the optical cable is ΔT(z), which represents the temperature change of the optical fiber at position z.

[0015] Optionally, constructing a three-dimensional deformation field model of the deep foundation pit slope based on the monitoring data and timestamps includes:

[0016] A timestamp alignment method is used to align the monitoring data uploaded by different devices at different times onto a unified time axis; a coordinate transformation tool is used to align the monitoring data uploaded by different devices into a unified coordinate system.

[0017] Optionally, obtaining the risk status of the deep foundation pit slope based on the predicted deformation of the plurality of deformation units includes:

[0018] Based on the predicted deformation of the multiple deformation units, the average predicted deformation of the three-dimensional deformation field model in each direction is obtained, wherein the average predicted deformation includes the average predicted deformation in the lateral direction, the average predicted deformation in the longitudinal direction, and the average predicted deformation in the vertical direction.

[0019] The average of the predicted deformation values ​​obtained at multiple consecutive time points in each direction is summed to obtain the cumulative value of the predicted deformation.

[0020] When any of the cumulative predicted deformation values ​​is greater than the corresponding cumulative warning value, it is determined whether the direction of the average predicted deformation value corresponding to the cumulative predicted deformation value is consistent at multiple time points.

[0021] When the predicted average deformation value is in the same direction at multiple time points, an excessive deformation warning is generated.

[0022] Optionally, obtaining the risk status of the deep foundation pit slope based on the predicted deformation of the plurality of deformation units includes:

[0023] Based on the predicted deformation of two adjacent deformation units in the selected analysis direction, the difference in deformation of the two deformation units in the analysis direction is determined.

[0024] When the deformation difference is greater than the deformation difference threshold corresponding to the analysis direction, it is determined that there is a deformation risk between two adjacent deformation units in the analysis direction, and the contact surface of the two adjacent deformation units is marked as the deformation surface, and the selected analysis direction is marked as the deformation direction;

[0025] Choose any direction as the preset deformation direction, control the preset selection window to move within the current three-dimensional deformation field model, count the number of deformation surfaces within the preset selection window whose deformation direction is the same as the preset deformation direction, and obtain the number of deformation surfaces;

[0026] When the number of deformable surfaces exceeds a preset deformation surface threshold, it is predicted that cracks will be generated in the deep foundation pit slope, and the extension direction of the cracks will be perpendicular to the preset deformation direction.

[0027] Optionally, after predicting that cracks will occur in the deep foundation pit slope when the number of deformable surfaces exceeds a preset deformation surface threshold, and taking the extension direction of the preset range as the extension direction of the cracks, the deep foundation pit slope deformation monitoring method further includes:

[0028] When the preset deformation direction is horizontal or vertical, in the preset deformation direction, it is determined whether the average predicted deformation of the crack on the side closer to the center of the three-dimensional deformation field model is greater than the average predicted deformation of the corresponding other side.

[0029] When the value is greater than 1, it is predicted that the deep foundation pit slope will undergo extrusion and heave deformation.

[0030] When the value is less than 1, it is predicted that the deep foundation pit slope will undergo fracture deformation.

[0031] Optionally, after predicting that cracks will occur in the deep foundation pit slope when the number of deformable surfaces exceeds a preset deformation surface threshold, and taking the extension direction of the preset range as the extension direction of the cracks, the deep foundation pit slope deformation monitoring method further includes:

[0032] When the preset deformation direction is vertical, it is determined whether the average predicted deformation of the upper region of the three-dimensional deformation field model is greater than the average predicted deformation of the lower region of the three-dimensional deformation field model in the preset deformation direction.

[0033] When the value is greater than a certain threshold, it is predicted that the deep foundation pit slope will be at risk of slippage.

[0034] When the value is less than a certain threshold, it is predicted that the deep foundation pit slope will be subject to compression and heave risk.

[0035] Secondly, the present invention also provides a deep foundation pit slope deformation monitoring device, comprising:

[0036] The monitoring data acquisition module is used to acquire monitoring data of the deep foundation pit slope, wherein the monitoring data includes slope surface data, slope internal horizontal displacement data, slope internal settlement data and deep foundation pit slope three-dimensional data.

[0037] The deformation field construction module is used to construct a three-dimensional deformation field model of the deep foundation pit slope based on the monitoring data and timestamps.

[0038] The deformation field segmentation module is used to segment each of the three-dimensional deformation field models constructed at multiple time points into multiple deformation units to obtain the temporal deformation information of each deformation unit, wherein the deformation unit is a three-dimensional model segmented according to a preset size;

[0039] The deformation prediction module is used to input the temporal deformation information of the deformation unit into the trained prediction network model to obtain the predicted deformation of each deformation unit.

[0040] The monitoring and early warning module is used to obtain the risk status of the deep foundation pit slope based on the predicted deformation amount of the multiple deformation units, so as to provide early warning of the risk of the deep foundation pit slope.

[0041] Thirdly, the present invention provides a storage medium storing a computer program that causes a computer to execute the deep foundation pit slope deformation monitoring method as described above.

[0042] Fourthly, the present invention provides an electronic device, comprising:

[0043] One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including methods for performing deep foundation pit slope deformation monitoring as described above.

[0044] This invention provides a method, device, storage medium, and electronic device for monitoring the deformation of deep foundation pit slopes. Compared with the prior art, it has the following advantages:

[0045] By acquiring data from multiple sources, including multiple data points, it is possible to monitor not only the surface deformation of the deep foundation pit slope but also its internal deformation. Combining internal and external monitoring data ensures the accuracy of the constructed three-dimensional deformation field model and subsequent predictions. The three-dimensional deformation field model is further segmented and refined, and a prediction network model is used to predict the deformation of each deformation unit. This allows for monitoring the changes in each deformation unit, anticipating the initial stage of deep foundation pit slope deformation, and providing early warnings at the initial stage before or just after various risks occur, thus improving the timeliness of early warnings. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 is a flowchart illustrating a method for monitoring the deformation of a deep foundation pit slope provided in an embodiment of the present invention;

[0048] Figure 2 is a schematic diagram of the process for early warning of deformation risk of deep foundation pit slope provided in an embodiment of the present invention;

[0049] Figure 3 is a schematic diagram of a deep foundation pit slope deformation monitoring device provided in an embodiment of the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions in the embodiments of this application are described clearly and completely. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0051] This application provides a method, device, storage medium, and electronic device for monitoring the deformation of deep foundation pit slopes. This solves the problem that a single monitoring method can only monitor the deformation of deep foundation pit slopes in one dimension, leading to inaccurate deformation prediction and untimely safety warnings. It enables comprehensive monitoring of deep foundation pit slopes and ensures the timeliness of deformation monitoring and warnings.

[0052] The technical solution in this application is to solve the above-mentioned technical problems, and the general idea is as follows:

[0053] To promptly identify problems and prevent accidents, deformation monitoring of the foundation pit is necessary during building construction and operation. This involves monitoring changes in the pit's shape and spatial position under external loads and analyzing structural deformation to assess its stability. Therefore, deformation monitoring of the structural facade is crucial for preventing potential safety hazards, ensuring the structural integrity of the building, and facilitating smooth construction.

[0054] Traditional deformation monitoring often employs methods such as total stations or levels. However, these methods are time-consuming and inefficient, and each method can only monitor the deformation of deep foundation pit slopes in one dimension. Analysis of potential risks based on data from traditional methods is also incomplete, leading to inaccurate deformation predictions and delayed safety warnings. Therefore, this paper attempts to fuse multi-source heterogeneous data collected by various monitoring devices to construct a three-dimensional deformation field. This allows for monitoring of the foundation pit's deformation in every direction, providing comprehensive monitoring and ensuring accuracy.

[0055] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0056] As shown in Figure 1, an embodiment of this application provides a method for monitoring the deformation of a deep foundation pit slope, including:

[0057] S1: Acquire monitoring data of the deep foundation pit slope, wherein the monitoring data includes slope surface data uploaded by a GNSS receiver (Global Navigation Satellite System Receiver), horizontal displacement data inside the slope uploaded by a horizontal optical fiber, settlement data inside the slope uploaded by a settlement optical fiber, and three-dimensional data of the deep foundation pit slope uploaded by a three-dimensional scanner.

[0058] S2: Based on the monitoring data and timestamps, construct a three-dimensional deformation field model of the deep foundation pit slope.

[0059] S3: Divide each of the three-dimensional deformation field models constructed at multiple time points into multiple deformation units to obtain the temporal deformation information of each deformation unit. The deformation unit is a three-dimensional model divided according to a preset size. The temporal deformation information includes the lateral deformation, longitudinal deformation and vertical deformation of the deformation unit at multiple time points.

[0060] S4: Input the temporal deformation information of the deformation unit into the trained prediction network model to obtain the predicted deformation amount of each deformation unit, wherein the predicted deformation amount includes the lateral predicted deformation amount, the longitudinal predicted deformation amount and the vertical predicted deformation amount.

[0061] S5: Based on the predicted deformation of the multiple deformation units, the risk status of the deep foundation pit slope is obtained.

[0062] In this embodiment, the monitoring data includes slope surface data uploaded by a GNSS receiver, internal horizontal displacement data of the slope uploaded by a horizontal optical fiber, internal settlement data of the slope uploaded by a settlement optical fiber, and three-dimensional data of the deep foundation pit slope uploaded by a 3D scanner. It can be seen that multiple devices are used to collect data, obtaining data from multiple aspects of the deep foundation pit slope. Using multiple data sources not only monitors the surface deformation of the deep foundation pit slope but also its internal deformation. Based on the monitoring data and timestamps, a three-dimensional deformation field model of the deep foundation pit slope is constructed. Combining internal and external monitoring data ensures the accuracy of the constructed three-dimensional deformation field model and the accuracy of subsequent predictions. Each of the three-dimensional deformation field models constructed at a given time point is divided into multiple deformation units to obtain the temporal deformation information of each deformation unit. The temporal deformation information of the deformation units is input into the trained prediction network model to obtain the predicted deformation amount of each deformation unit. By refining the three-dimensional deformation field model and using the prediction network model to predict the deformation amount of each deformation unit, it is possible to monitor the changes of each deformation unit and thus anticipate the initial stage of deep foundation pit slope deformation. Based on the predicted deformation amounts of multiple deformation units, the risk status of the deep foundation pit slope is obtained, enabling early warning at the stage when various risk situations have not yet occurred or have just occurred, thereby improving the timeliness of early warning.

[0063] The following is a detailed description of each step.

[0064] Optionally, S1: After acquiring the monitoring data of the deep foundation pit slope, the deep foundation pit slope deformation monitoring method further includes:

[0065] The monitoring data is preprocessed.

[0066] Specifically, based on monitoring needs, multiple GNSS receivers can be deployed, their operating modes configured, and parameters such as sampling rate and data format set. Time synchronization among multiple GNSS receivers should be ensured, and accurate timestamps recorded. The GNSS receivers should be started to begin data acquisition, and their status monitored in real time. Data storage devices should be checked periodically to ensure no data loss or damage. For slope surface data uploaded by the GNSS receivers: the acquired GNSS data is imported into data processing software for preliminary preprocessing and organization. Professional GNSS data processing software (such as Trimble Business Center, Leica Infinity, Topcon Magnet, Bernese GNSS Software, etc.) is used to solve the data, obtaining high-precision coordinate and displacement information. Differential techniques, precise ephemeris analysis, and atmospheric correction methods are applied to correct errors in the GNSS data and improve data accuracy.

[0067] For the slope internal settlement data acquired and uploaded via fiber optic cable, the micro-strain data is transformed into deformation data, and the settlement (including the internal settlement data of the slope) is calculated as follows:

[0068] in, Let ε(z) represent the soil settlement measured by the optical fiber between positions z1 and z2, and let f represent the strain of the optical fiber. 0 f(z) represents the initial Brillouin scattering frequency shift (MHz) of the fiber at position z, and f(z) represents the initial Brillouin scattering frequency shift (MHz) of the fiber at position z. s The proportionality coefficient (MHz / με) representing the frequency shift of the backscattered Brillouin light from the optical fiber relative to the strain of the optical cable can be provided by the optical cable supplier or determined through optical cable calibration tests. T The proportionality coefficient (MHz / ℃) representing the frequency shift of the backscattered Brillouin light from the optical fiber to the cable temperature can be provided by the optical cable supplier or determined through optical cable calibration tests. ΔT(z) represents the temperature change of the optical fiber at position z.

[0069] In addition, the monitoring data also includes slope internal tilt deformation data uploaded by the inclinometer. The inclinometer monitors the tilt changes of the rock strata inside the slope, which is equivalent to the combined change results of horizontal displacement and vertical settlement inside the slope. Therefore, the slope internal tilt deformation data uploaded by the inclinometer can be used to verify the data uploaded by the horizontal optical fiber and the settlement optical fiber.

[0070] For the 3D data of deep foundation pit slope uploaded by the 3D scanner: import the collected 3D point cloud data into the data processing software, and use professional data processing software (such as Cyclone, RealWorks, etc.) to preprocess the point cloud data, including denoising, alignment, stitching and modeling, to obtain the 3D boundary model of the deep foundation pit slope.

[0071] Optionally, S2: Constructing a three-dimensional deformation field model of the deep foundation pit slope based on the monitoring data and timestamps includes:

[0072] A timestamp alignment method is used to align the monitoring data uploaded by different devices at different times onto a unified time axis, ensuring temporal consistency of the data. A coordinate transformation tool is used to align the monitoring data uploaded by different devices into a unified coordinate system, ensuring spatial consistency of the data.

[0073] Specifically, the same data collection time point or collection cycle is set for multiple devices to ensure that multiple devices collect data at the same time point. When collecting data, the data collection time point is recorded. The data collected at the same time point is processed, and the data uploaded by multiple devices is transformed into the same coordinate system by using the relative positional relationship between each device or the positional relationship between each device and the same reference point, thereby aligning the data in the time and space dimensions.

[0074] For step S3, a three-dimensional deformation field model can be constructed using the data collected at each time point. Initially, data can be collected at multiple time points to obtain multiple three-dimensional deformation field models. Each three-dimensional deformation field model is then segmented, except for the first one. The segmentation point is then fixed as part of the three-dimensional deformation field model, and moves along with the deformation of the model. Therefore, deformable elements can be obtained in the subsequently obtained three-dimensional deformation field models, and the deformation amount of each element can also be obtained. To perform multi-dimensional monitoring of the three-dimensional deformation field model, the deformation of the deformable elements in the lateral, longitudinal, and vertical directions can be monitored. For example, the relative deformation of a deformable element in the lateral direction at multiple time points can be recorded to form the lateral temporal deformation information of that element. Similarly, the longitudinal and vertical temporal deformation information of the deformable element can also be obtained.

[0075] For step S4, the ConvLSTM (Convolutional Long Short-Term Memory) model can be pre-trained using the temporal deformation information of each deformation unit obtained in the previous stage to obtain the prediction network model. Alternatively, a deformation prediction network model can be built using TensorFlow or PyTorch libraries. Furthermore, the actual deformation amounts obtained subsequently can be compared with the predicted deformation amounts to correct the prediction network model, thereby making the prediction network model increasingly accurate.

[0076] Optionally, as shown in Figure 2, S5: Based on the predicted deformation of the multiple deformation units, the risk status of the deep foundation pit slope is obtained, including:

[0077] S511: Based on the predicted deformation of the multiple deformation units, the average predicted deformation of the three-dimensional deformation field model in each direction is obtained, wherein the average predicted deformation includes the average predicted deformation in the lateral direction, the average predicted deformation in the longitudinal direction, and the average predicted deformation in the vertical direction.

[0078] Specifically, the average value of the predicted deformation in the lateral direction is obtained by summing the deformation of all the deformable elements and then dividing by the number of deformable elements. Similarly, the average value of the predicted deformation in the longitudinal direction and the average value of the predicted deformation in the vertical direction can also be calculated.

[0079] S512: The average value of the predicted deformation obtained at multiple consecutive time points in each direction is summed to obtain the cumulative value of the predicted deformation.

[0080] Specifically, the deformation is the change in the current 3D deformation field model relative to the previous time point. Therefore, the cumulative predicted deformation value is the sum of the changes at multiple consecutive time points. Accumulation can begin from the first monitoring moment or from a certain intermediate moment, thus obtaining the global cumulative predicted deformation value or the cumulative predicted deformation value for a specific time period, enabling flexible analysis of the 3D deformation field model. The cumulative predicted deformation value includes the lateral, longitudinal, and vertical cumulative predicted deformation values.

[0081] S513: When any one of the predicted deformation cumulative values ​​is greater than the corresponding cumulative warning value, determine whether the direction of the predicted deformation average value corresponding to the predicted deformation cumulative value is consistent at multiple time points.

[0082] S514: When the predicted average deformation is in the same direction at multiple time points, an excessive deformation warning is generated.

[0083] Specifically, the cumulative predicted deformation value in each direction corresponds to a cumulative warning value. When a cumulative predicted deformation value exceeds the corresponding cumulative warning value, it indicates that the deep foundation pit slope is deforming excessively in the direction corresponding to that cumulative predicted deformation value. Further analysis is needed to determine if the direction of the average predicted deformation value at multiple time points is consistent. If the direction is consistent, it means that the deformation has been continuously moving in the same direction at multiple time points, and the probability of continued deformation in that direction in the future is high. In this case, a warning can be issued if the deformation has already exceeded the limit. If the direction is inconsistent, it means that there are instances of reverse deformation at multiple time points. If such inconsistencies are frequent, it means that even if the current cumulative deformation value exceeds the limit, the deformation may reverse at the next time point, causing the cumulative value to decrease and not exceed the limit. In this case, a warning can be temporarily withheld or a notification can be given without issuing a warning. Determining directional consistency significantly improves the accuracy of warnings and reduces the probability of false alarms.

[0084] Optionally, S5: Based on the predicted deformation of the multiple deformation units, the risk status of the deep foundation pit slope is obtained, including:

[0085] S521: Determine the difference in deformation amount between two adjacent deformation units in the selected analysis direction based on the predicted deformation amount of the two adjacent deformation units in the selected analysis direction.

[0086] Specifically, the analysis direction is any one of the horizontal, vertical, and longitudinal directions. In two adjacent deformation units, the deformation unit closer to the edge of the three-dimensional deformation field model is designated as the first deformation unit, and the other deformation unit is designated as the second deformation unit. The predicted deformation of the first deformation unit is subtracted from the predicted deformation of the second deformation unit to obtain the deformation difference. For example, if the horizontal direction is chosen as the analysis direction, the deformation difference between the two adjacent deformation units in the horizontal direction is calculated based on their predicted deformations. A positive deformation difference indicates that the two deformation units tend to move away from each other in the horizontal direction, while a negative deformation difference indicates that the two deformation units tend to move closer together in the horizontal direction.

[0087] S522: When the deformation difference is greater than the deformation difference threshold corresponding to the analysis direction, it is determined that there is a deformation risk between two adjacent deformation units in the analysis direction, and the contact surface of the two adjacent deformation units is marked as the deformation surface, and the selected analysis direction is marked as the deformation direction.

[0088] Specifically, when the difference in deformation in the lateral direction is greater than the threshold for the difference in deformation in the lateral direction, it is determined that there is a risk of deformation in the lateral direction between the two deformation units, and the contact surface perpendicular to the lateral direction is marked as the deformation surface, and the lateral direction is recorded as the deformation direction.

[0089] S523: Select any direction as the preset deformation direction, control the preset selection window to move within the current three-dimensional deformation field model, count the number of deformation surfaces within the preset selection window whose deformation direction is the same as the preset deformation direction, and obtain the number of deformation surfaces.

[0090] S524: When the number of deformable surfaces is greater than the preset deformation surface threshold, it is predicted that cracks will be generated in the deep foundation pit slope, and the extension direction of the cracks is perpendicular to the preset deformation direction.

[0091] Specifically, if the horizontal direction is selected as the preset deformation direction, for example, the preset selection window is a thin cuboid with a vertical and horizontal dimension of 30cm and a horizontal dimension of 2cm. Select a position and move the preset selection window horizontally. The movement step can be 2cm to ensure no overlap between selected positions, or 1cm to allow overlapping selections of the 3D deformation field model. After moving horizontally from one side of the 3D deformation field model to the opposite side at the selected position, move the preset selection window vertically or horizontally with a movement step of 30cm, 20cm, or 10cm, etc., and then gradually move it horizontally again, repeating the above steps until the preset selection window has traversed the entire current 3D deformation field model. After each selection, the number of deformation surfaces with a lateral deformation direction within the preset selection window is counted. If the number of deformation surfaces is 63 and the preset deformation surface threshold is 30, it means that 63 contact surfaces within this preset selection window exhibit lateral deviation. This indicates that lateral cracks may appear within this narrow preset range, and the cracks extend approximately perpendicular to the preset deformation direction. For example, in the distance analysis above, the preset deformation direction is lateral. If more than the preset deformation surface threshold is found within the preset selection window, it means that significant lateral deformation will occur at many locations within the preset selection window, resulting in approximately lateral cracks in the deformation field model. Based on the identified deformation risk, the risk type of the deep foundation pit slope is further analyzed. It should be noted that cracks may have irregular shapes, and their extension direction will not be a straight line; they may be curved or undulating. Therefore, cracks approximately perpendicular to the preset deformation direction can be considered as cracks perpendicular to the preset deformation direction.

[0092] In an optional embodiment of the present invention, after predicting that cracks will occur in the deep foundation pit slope when the number of deformable surfaces exceeds a preset deformation surface threshold, and taking the extension direction of the preset range as the extension direction of the cracks, the deep foundation pit slope deformation monitoring method further includes:

[0093] When the preset deformation direction is horizontal or vertical, in the preset deformation direction, it is determined whether the average predicted deformation on the side of the crack closer to the center of the three-dimensional deformation field model is greater than the average predicted deformation on the other side.

[0094] When the value is greater than a certain threshold, it is predicted that the deep foundation pit slope will undergo compression and heaving deformation.

[0095] When the value is less than 1, it is predicted that the deep foundation pit slope will undergo fracture deformation.

[0096] Specifically, when the preset deformation direction is transverse or longitudinal, it means that when the crack extension direction is perpendicular to the transverse or longitudinal extension, the crack will extend vertically, or the crack will extend relatively along the longitudinal or transverse direction. However, since the deformation direction is transverse or longitudinal, the cracks on the deep foundation pit slope may originate from fracture or internal compression deformation. Therefore, it is necessary to analyze the average value of the predicted deformation on both sides of the crack. If the preset deformation direction is transverse, the average value of the transverse predicted deformation on both sides of the crack is compared; if the preset deformation direction is longitudinal, the average value of the longitudinal predicted deformation on both sides of the crack is compared. When the average predicted deformation on the side closer to the center of the 3D deformation field model is greater than the average predicted deformation on the opposite side, it indicates that the deformation on the side closer to the center is larger. However, due to the obstruction of the smaller soil layer on the periphery, the larger soil layer can only compress the outer layer in the future, and compression heave deformation is highly likely. Conversely, when the average predicted deformation on the side closer to the center of the 3D deformation field model is less than the average predicted deformation on the opposite side, it indicates that the deformation on the side closer to the center is smaller, while the deformation on the side farther from the center is larger. In this case, the side farther from the center is the outer soil layer region, and the deformation of the outer soil layer will not be obstructed, leading to separation from the inner soil layer and fracture deformation. When the average predicted deformation on the side closer to the center of the 3D deformation field model is equal to the average predicted deformation on the opposite side, it indicates that the deformation within the 3D deformation field model is uniform, and cracks will not occur. After analyzing that cracks will occur in the deep foundation pit slope, further locating the source of the cracks and determining a more accurate deformation type is crucial for implementing targeted preventive measures in advance.

[0097] In an optional embodiment of the present invention, after predicting that cracks will occur in the deep foundation pit slope when the number of deformable surfaces exceeds a preset deformation surface threshold, and taking the extension direction of the preset range as the extension direction of the cracks, the deep foundation pit slope deformation monitoring method further includes:

[0098] When the preset deformation direction is vertical, it is determined whether the average predicted deformation of the upper region of the three-dimensional deformation field model is greater than the average predicted deformation of the lower region of the three-dimensional deformation field model in the preset deformation direction. The three-dimensional deformation field model is divided into upper and lower parts at the height center of the three-dimensional deformation field model. The upper part is regarded as the upper region of the three-dimensional deformation field model and the lower part is regarded as the lower region of the three-dimensional deformation field model.

[0099] When the value is greater than a certain threshold, it is predicted that the deep foundation pit slope will be at risk of slippage.

[0100] When the value is less than a certain threshold, it is predicted that the deep foundation pit slope will be subject to compression and heave risk.

[0101] Specifically, similar to the deformation analysis above, when the preset deformation direction is vertical, it indicates that cracks will extend horizontally or vertically. When the average predicted deformation of the upper region is greater than that of the lower region, cracks will form between the upper and lower soil layers. When the cracks are large, the upper soil layer will slide along the lower soil layer, especially in the deep foundation pit slope area, where the slope itself has a certain gradient. In this case, the risk of sliding is greater, so the risk of sliding on the deep foundation pit slope can be predicted in advance. When the average predicted deformation of the upper region is less than that of the lower region, and the preset deformation direction is vertically upward, the lower soil layer will squeeze the upper soil layer, forcing the upper soil layer to bulge. However, when the preset deformation direction is vertically downward, it indicates that both the lower and upper soil layers will settle. In this case, uniform settlement has little impact on the buildings above the foundation pit, but further attention is still needed to prevent large settlement of the foundation pit.

[0102] In summary, compared with existing technologies, it has the following beneficial effects:

[0103] 1. Data is collected using multiple devices to obtain data on deep foundation pit slopes from multiple perspectives. This not only monitors the surface deformation of the deep foundation pit slopes but also their internal deformation. Combining internal and external monitoring data ensures the accuracy of the constructed three-dimensional deformation field model and the accuracy of subsequent predictions. The three-dimensional deformation field model is segmented and refined, and the deformation of each deformation unit is predicted using a prediction network model. This allows for monitoring the changes in each deformation unit, anticipating the initial stage of deep foundation pit slope deformation, and providing early warnings at the initial stage before or just when various risks have occurred, thus improving the timeliness of early warnings.

[0104] 2. When the cumulative predicted deformation value exceeds the corresponding cumulative warning value, it indicates that the deep foundation pit slope is deforming excessively in the direction corresponding to that cumulative predicted deformation value. Further analysis is needed to determine whether the direction of the average predicted deformation value at multiple time points is consistent. If the direction is consistent, it means that the deformation has been continuously moving in the same direction at multiple time points, and there is a high probability that it will continue to deform in that direction in the future. In cases where the deformation has exceeded the standard, an early warning can be issued. Determining directional consistency can significantly improve the accuracy of early warnings and reduce the probability of false alarms.

[0105] 3. After analyzing that cracks will appear in the slope of the deep foundation pit, further locate the source of the cracks and determine the more accurate deformation type so as to carry out targeted preventive measures in advance.

[0106] As shown in Figure 3, an embodiment of this application provides a deep foundation pit slope deformation monitoring device, comprising:

[0107] The monitoring data acquisition module 100 is used to acquire monitoring data of the deep foundation pit slope, wherein the monitoring data includes slope surface data, slope internal horizontal displacement data, slope internal settlement data and deep foundation pit slope three-dimensional data.

[0108] The deformation field construction module 200 is used to construct a three-dimensional deformation field model of the deep foundation pit slope based on the monitoring data and timestamps.

[0109] The deformation field segmentation module 300 is used to segment each of the three-dimensional deformation field models constructed at multiple time points into multiple deformation units to obtain the temporal deformation information of each deformation unit. The deformation unit is a three-dimensional model segmented according to a preset size. The temporal deformation information includes the lateral deformation, longitudinal deformation and vertical deformation of the deformation unit at multiple time points.

[0110] The deformation prediction module 400 is used to input the temporal deformation information of the deformation unit into the trained prediction network model to obtain the predicted deformation of each deformation unit, wherein the predicted deformation includes lateral predicted deformation, longitudinal predicted deformation and vertical predicted deformation.

[0111] The monitoring and early warning module 500 is used to obtain the risk status of the deep foundation pit slope based on the predicted deformation amount of the multiple deformation units, so as to provide early warning of the risk of the deep foundation pit slope.

[0112] An optional embodiment of this application provides a storage medium storing a computer program that causes a computer to execute the deep foundation pit slope deformation monitoring method as described above.

[0113] An optional embodiment of this application provides an electronic device, comprising:

[0114] One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including methods for performing deep foundation pit slope deformation monitoring as described above.

[0115] The beneficial effects of the deep foundation pit slope deformation monitoring device, storage medium, and electronic equipment in this example are the same as those of the deep foundation pit slope deformation monitoring method described above, and will not be repeated here.

[0116] The present invention describes electronic devices that can serve as servers or clients of this application, which are examples of hardware devices that can be applied to various aspects of this application. Electronic devices are intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic devices can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the application described and / or claimed herein.

[0117] Electronic devices include a computing unit that can perform various appropriate actions and processes based on a computer program stored in read-only memory (ROM) or loaded from a storage unit into random access memory (RAM). The RAM can also store various programs and data required for device operation. The computing unit, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0118] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc. In this application, the units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this application according to actual needs. Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units can be implemented in hardware or as software functional units.

[0119] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0120] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for monitoring the deformation of deep foundation pit slopes, characterized in that, include: Acquire monitoring data of deep foundation pit slopes, wherein the monitoring data includes slope surface data, slope internal horizontal displacement data, slope internal settlement data, and deep foundation pit slope three-dimensional data; Based on the monitoring data and timestamps, a three-dimensional deformation field model of the deep foundation pit slope is constructed; Each of the three-dimensional deformation field models constructed at multiple time points is divided into multiple deformation units to obtain the temporal deformation information of each deformation unit, wherein the deformation unit is a three-dimensional model obtained by dividing according to a preset size; The temporal deformation information of the deformation unit is input into the trained prediction network model to obtain the predicted deformation amount of each deformation unit. The risk status of the deep foundation pit slope is obtained based on the predicted deformation amount of the multiple deformation units.

2. The method for monitoring the deformation of deep foundation pit slopes as described in claim 1, characterized in that, After acquiring the monitoring data of the deep foundation pit slope, the process also includes: The preprocessing of the slope internal settlement data in the monitoring data includes the settlement amount, which is: Where D(z1-z2) represents the soil settlement measured by the optical fiber between positions z1 and z2, ε(z) represents the strain of the optical fiber, and f 0 f(z) represents the initial Brillouin scattering frequency shift of the fiber at position z, and C represents the initial Brillouin scattering frequency shift of the fiber at position z. s C represents the proportionality coefficient between the frequency shift of the backscattered Brillouin light from the optical fiber and the strain of the optical cable. T The coefficient representing the ratio of the frequency shift of the backscattered Brillouin light from the optical fiber to the temperature of the optical cable is ΔT(z), which represents the temperature change of the optical fiber at position z.

3. The method for monitoring the deformation of deep foundation pit slopes as described in claim 1, characterized in that, The step of constructing a three-dimensional deformation field model of the deep foundation pit slope based on the monitoring data and timestamps includes: A timestamp alignment method is used to align the monitoring data uploaded by different devices at different times onto a unified time axis; a coordinate transformation tool is used to align the monitoring data uploaded by different devices into a unified coordinate system.

4. The method for monitoring the deformation of deep foundation pit slopes as described in any one of claims 1 to 3, characterized in that, The step of determining the risk status of the deep foundation pit slope based on the predicted deformation of the multiple deformation units includes: Based on the predicted deformation of the multiple deformation units, the average predicted deformation of the three-dimensional deformation field model in each direction is obtained, wherein the average predicted deformation includes the average predicted deformation in the lateral direction, the average predicted deformation in the longitudinal direction, and the average predicted deformation in the vertical direction. The average of the predicted deformation values ​​obtained at multiple consecutive time points in each direction is summed to obtain the cumulative value of the predicted deformation. When any of the cumulative predicted deformation values ​​is greater than the corresponding cumulative warning value, it is determined whether the direction of the average predicted deformation value corresponding to the cumulative predicted deformation value is consistent at multiple time points. When the predicted average deformation value is in the same direction at multiple time points, an excessive deformation warning is generated.

5. The method for monitoring the deformation of deep foundation pit slopes as described in any one of claims 1 to 3, characterized in that, The step of determining the risk status of the deep foundation pit slope based on the predicted deformation of the multiple deformation units includes: Based on the predicted deformation of two adjacent deformation units in the selected analysis direction, the difference in deformation of the two deformation units in the analysis direction is determined. When the deformation difference is greater than the deformation difference threshold corresponding to the analysis direction, it is determined that there is a deformation risk between two adjacent deformation units in the analysis direction, and the contact surface of the two adjacent deformation units is marked as the deformation surface, and the selected analysis direction is marked as the deformation direction; Choose any direction as the preset deformation direction, control the preset selection window to move within the current three-dimensional deformation field model, count the number of deformation surfaces within the preset selection window whose deformation direction is the same as the preset deformation direction, and obtain the number of deformation surfaces; When the number of deformable surfaces exceeds a preset deformation surface threshold, it is predicted that cracks will be generated in the deep foundation pit slope, and the extension direction of the cracks will be perpendicular to the preset deformation direction.

6. The method for monitoring the deformation of deep foundation pit slopes as described in claim 5, characterized in that, After predicting that cracks will occur in the deep foundation pit slope when the number of deformable surfaces exceeds a preset deformable surface threshold, and taking the extension direction of the preset range as the extension direction of the cracks, the method further includes: When the preset deformation direction is horizontal or vertical, in the preset deformation direction, it is determined whether the average predicted deformation of the crack on the side closer to the center of the three-dimensional deformation field model is greater than the average predicted deformation of the corresponding other side. When the value is greater than 1, it is predicted that the deep foundation pit slope will undergo extrusion and heave deformation. When the value is less than 1, it is predicted that the deep foundation pit slope will undergo fracture deformation.

7. The method for monitoring the deformation of deep foundation pit slopes as described in claim 5, characterized in that, After predicting that cracks will occur in the deep foundation pit slope when the number of deformable surfaces exceeds a preset deformable surface threshold, and taking the extension direction of the preset range as the extension direction of the cracks, the method further includes: When the preset deformation direction is vertical, it is determined whether the average predicted deformation of the upper region of the three-dimensional deformation field model is greater than the average predicted deformation of the lower region of the three-dimensional deformation field model in the preset deformation direction. When the value is greater than a certain threshold, it is predicted that the deep foundation pit slope will be at risk of slippage. When the value is less than a certain threshold, it is predicted that the deep foundation pit slope will be subject to compression and heave risk.

8. A deep foundation pit slope deformation monitoring device, characterized in that, include: The monitoring data acquisition module is used to acquire monitoring data of the deep foundation pit slope, wherein the monitoring data includes slope surface data, slope internal horizontal displacement data, slope internal settlement data and deep foundation pit slope three-dimensional data. The deformation field construction module is used to construct a three-dimensional deformation field model of the deep foundation pit slope based on the monitoring data and timestamps. The deformation field segmentation module is used to segment each of the three-dimensional deformation field models constructed at multiple time points into multiple deformation units to obtain the temporal deformation information of each deformation unit, wherein the deformation unit is a three-dimensional model segmented according to a preset size; The deformation prediction module is used to input the temporal deformation information of the deformation unit into the trained prediction network model to obtain the predicted deformation of each deformation unit. The monitoring and early warning module is used to obtain the risk status of the deep foundation pit slope based on the predicted deformation amount of the multiple deformation units, so as to provide early warning of the risk of the deep foundation pit slope.

9. A storage medium, characterized in that, It stores a computer program that causes a computer to perform the deep foundation pit slope deformation monitoring method as described in any one of claims 1 to 7.

10. An electronic device, characterized in that, include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the programs including methods for performing the deep foundation pit slope deformation monitoring method as described in any one of claims 1 to 7.

Images