Deep foundation pit multi-parameter automatic monitoring method based on digital twin driving

Abstract

The application relates to the technical field of electric digital data processing, in particular to a deep foundation pit multi-parameter automatic monitoring method based on digital twin driving, which comprises the following steps: determining a target region of a current monitoring period by using a target offset vector and an adjacent offset vector of a target point cloud, determining a shape stability degree and a deformation region by using displacement information in the target region; determining a temporary fluctuation possibility and a real deformation region by using the shape stability degree and the deformation region marking times; determining a target period corresponding to an abnormal growth of a vibration frequency time sequence and a stress time sequence in the real deformation region; determining an injury accumulation index of the real deformation region by using a temporary fluctuation possibility time sequence corresponding to each target period; and determining a corresponding early warning priority by using the shape stability degree and the injury accumulation index of the real deformation region. The application can accurately grade and early warn the soil body instability degrees of different regions of a deep foundation pit, and improves the accuracy of deep foundation pit automatic monitoring.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital data processing, in particular to a deep foundation pit multi-parameter automatic monitoring method based on digital twin driving. BACKGROUND

[0002] The digital twin in deep foundation pit engineering is a dynamic mapping body of the physical foundation pit and its virtual model. The existing method usually monitors the deep foundation pit construction process by mapping the physical deep foundation pit to the virtual space through multi-source sensors, Internet of Things, building information model and dynamic simulation technology. For example, in the deep foundation pit scene, the data monitored by the multi-source sensors is transmitted to the twin platform, and the effective stress change of the soil is predicted based on the sensor data, so as to enhance the control of precipitation. The existing method usually monitors the construction process based on the soil conditions at the initial construction.

[0003] In the actual deep foundation pit construction process (especially in the scene of deep and soft soil), during the construction, the vibration generated by the heavy engineering equipment such as excavators and cranes when working may cause a certain change in the stress state of the soil (there may also be a change in the stress of the soil due to precipitation in the construction area), which cannot evaluate the interference of the soil change in the deep foundation pit construction process on the stability of the foundation pit, so as to make the automatic monitoring of the deep foundation pit inaccurate, resulting in misjudgment of the state of the foundation pit, such as being unable to distinguish between normal excavation settlement and abnormal deformation or misjudging the soil deformation, which reduces the accuracy of the engineering construction. SUMMARY

[0004] In order to solve the technical problem that it is difficult to evaluate the interference of the soil change in the deep foundation pit construction process on the stability of the foundation pit, and the accuracy of the automatic monitoring of the deep foundation pit is low, the purpose of the present application is to provide a deep foundation pit multi-parameter automatic monitoring method based on digital twin driving, and the technical solution adopted is as follows:

[0005] The present application provides a deep foundation pit multi-parameter automatic monitoring method based on digital twin driving, which comprises:

[0006] Determine the target offset vector and the adjacent offset vector of the target point cloud and the adjacent point cloud of the deep foundation pit in the current monitoring period relative to the preset construction model respectively;

[0007] Determine the target region of the current monitoring period using the target offset vector and the adjacent offset vector, and determine the morphological stability degree and the deformation region of the target region using the displacement information in the target region;

[0008] Determine the temporary fluctuation possibility and the real deformation region using the morphological stability degree of the deformation region and the deformation region marking number in the historical monitoring period;

[0009] Determine the target period corresponding to the abnormal growth of the vibration frequency time sequence and the stress time sequence in the real deformation area, and determine the damage accumulation index of the real deformation area by using the temporary fluctuation possibility time sequence corresponding to each target period;

[0010] Determine the corresponding early warning priority by using the morphological stability degree and the damage accumulation index of the real deformation area.

[0011] Further, the target offset vector of the target point cloud in the current monitoring period and the adjacent offset vector of the adjacent point cloud relative to the preset construction model of the deep foundation pit, comprising:

[0012] Determine the spatial nearest neighbor point of the target point cloud in the current point cloud data of the deep foundation pit in the current monitoring period and the target point cloud in the preset construction model of the deep foundation pit;

[0013] The vector from the spatial nearest neighbor point of the target point cloud to the target point cloud is used as the target offset vector of the target point cloud, and the adjacent offset vector of the adjacent point cloud of the target point cloud in the current point cloud data is determined.

[0014] Further, the target region of the current monitoring period is determined by using the target offset vector and the adjacent offset vector, comprising:

[0015] Determine the cosine value of the included angle and the offset difference value between the target offset vector and the adjacent offset vector;

[0016] The offset similarity degree between the target point cloud and its adjacent point cloud is calculated by using the cosine value of the included angle and the offset difference value;

[0017] The point cloud with an offset similarity degree greater than a preset similarity threshold is used as the offset similar point of the target point cloud, and the offset similar point and the target point cloud are combined into the same region to obtain the target region of the current monitoring period.

[0018] Further, the morphological stability degree and the deformation area of the target region are determined by using the displacement information in the target region, comprising:

[0019] Determine the morphological stability degree of the target region by using the horizontal displacement and the vertical displacement of each monitoring point in the target region;

[0020] The target region with a morphological stability degree less than a preset stability threshold is used as the deformation area.

[0021] Further, the morphological stability degree of the target region is determined by using the horizontal displacement and the vertical displacement of each monitoring point in the target region, comprising:

[0022] Determine the range of the horizontal displacement of each monitoring point in the target region and the cumulative value of the horizontal displacement at each time;

[0023] The stable index of the horizontal displacement is calculated by using the range of the horizontal displacement and the cumulative value of the horizontal displacement;

[0024] The shape stability degree of the target region is calculated by using the stable index of the horizontal displacement and the stable index of the vertical displacement.

[0025] Further, the determination of the temporary fluctuation possibility and the real deformation region by using the shape stability degree of the deformation region and the deformation region marking number in the historical monitoring period comprises:

[0026] The deformation region marking number of the deformation region marked as the deformation region in the historical monitoring period is determined, and the average of the deformation region marking number in all historical monitoring periods is determined;

[0027] The temporary fluctuation possibility is determined by using the average of the shape stability degree of the deformation region in all historical monitoring periods and the average of the deformation region marking number;

[0028] The deformation region with the temporary fluctuation possibility less than or equal to the preset fluctuation threshold is taken as the real deformation region.

[0029] Further, the determination of the temporary fluctuation possibility by using the average of the shape stability degree of the deformation region in all historical monitoring periods and the average of the deformation region marking number comprises:

[0030] The monitoring time period corresponding to the deformation region marked as the deformation region for two or more times in the historical monitoring period is combined as a single deformation fluctuation duration, and the average of the duration of all deformation fluctuations is determined;

[0031] The temporary fluctuation possibility is calculated by using the average of the shape stability degree of the deformation region in all historical monitoring periods, the average of the deformation region marking number, and the average of the duration.

[0032] Further, the determination of the target time period corresponding to the abnormal growth of the vibration frequency time sequence and the stress time sequence in the real deformation region comprises:

[0033] The vibration frequency time sequence in the real deformation region and the target growth time period of the vibration frequency time sequence are determined.

[0034] The abnormal growth time period in the target growth time period is determined by using the growth amplitude of the target growth time period, the average of the vibration frequency, and the average of the growth amplitude of all growth time periods.

[0035] The overlapping time period between the abnormal growth time period of the vibration frequency time sequence and the abnormal growth time period of the stress time sequence is taken as the target time period corresponding to the abnormal growth.

[0036] Further, the determination of the damage accumulation index of the real deformation region by using the temporary fluctuation possibility time sequence corresponding to each target time period comprises:

[0037] The temporary fluctuation possibility closest to the starting time in the distance target period is taken as the temporary fluctuation possibility corresponding to the target period;

[0038] A first-order difference sequence of the temporary fluctuation possibility time sequence corresponding to each target period is determined, and a negative value in the first-order difference sequence is taken as an effective damage value;

[0039] The number of effective damage values and the mean of the absolute values thereof and the mean of the time lengths of each effective damage value from the corresponding previous target period are used to calculate a damage accumulation index of the real deformation region.

[0040] Further, the determination of the corresponding early warning priority using the morphological stability degree and the damage accumulation index of the real deformation region comprises:

[0041] The morphological stability degree, the damage accumulation index and the number of voxels of the real deformation region are used to calculate an early warning priority of the real deformation region; the early warning priority is used to output different early warning signals about the real deformation region.

[0042] The present application has the following beneficial effects:

[0043] The present application is based on digital twin technology, analyzes deep foundation pit construction monitoring data obtained by multiple sensors during deep foundation pit construction, analyzes the deformation degree of different regions in the deep foundation pit at the current time based on real-time deformation data, thereby distinguishing temporary deformation regions and real deformation regions formed by heavy construction machinery and load fluctuations, analyzing the damage accumulation degree of each region according to the fluctuation of the deformation degree of each real deformation region at different times in the deep foundation pit, thereby predicting the ability of each region to respond to soil changes, and accurately classifying and warning the soil instability degree of different regions in the deep foundation pit under real-time construction progress based on the deformation development trend of the real deformation region, thereby improving the accuracy of deep foundation pit automatic monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0044] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without creative labor.

[0045] Figure 1 A step flow chart of a deep foundation pit multi-parameter automatic monitoring method based on digital twin driving provided by an embodiment of the present application;

[0046] Figure 2A refinement flowchart of step S2 in a deep foundation multi-parameter automatic monitoring method based on digital twin driving provided by one embodiment of the present application is shown in the figure.

[0047] Figure 3 A refinement flowchart of step S2 in a deep foundation multi-parameter automatic monitoring method based on digital twin driving provided by another embodiment of the present application is shown in the figure.

[0048] Figure 4 A refinement flowchart of step S3 in a deep foundation multi-parameter automatic monitoring method based on digital twin driving provided by one embodiment of the present application is shown in the figure.

[0049] Figure 5 A refinement flowchart of step S4 in a deep foundation multi-parameter automatic monitoring method based on digital twin driving provided by one embodiment of the present application is shown in the figure.

[0050] Figure 6 A refinement flowchart of step S4 in a deep foundation multi-parameter automatic monitoring method based on digital twin driving provided by another embodiment of the present application is shown in the figure.

[0051] Figure 7 A structural schematic diagram of a hardware operating environment of a deep foundation multi-parameter automatic monitoring device based on digital twin driving involved in the embodiment scheme of the present application is shown in the figure. DETAILED DESCRIPTION

[0052] In order to further illustrate the technical means and effects taken by the present application to achieve the predetermined purposes, the following describes in detail the specific embodiments, structures, features and effects of a deep foundation multi-parameter automatic monitoring method based on digital twin driving according to the present application, with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0053] 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 the present application belongs.

[0054] The specific scheme of the deep foundation multi-parameter automatic monitoring method based on digital twin driving provided by the present application is described in detail below with reference to the accompanying drawings.

[0055] Embodiment one:

[0056] For the deep foundation multi-parameter automatic monitoring method based on digital twin driving provided by the present application, please refer to Figure 1FIG. 1 shows a step flow chart of a deep foundation multi-parameter automatic monitoring method based on digital twin driving according to an embodiment of the present application.

[0057] The deep foundation multi-parameter automatic monitoring method based on digital twin driving comprises the following steps.

[0058] Step S1, determining a target offset vector and a neighboring offset vector of a target point cloud and a neighboring point cloud of the deep foundation in a current monitoring period relative to a preset construction model, respectively;

[0059] In the embodiment, first, various monitoring data of the deep foundation can be obtained based on the multi-source sensors, and the specific implementation process can be as follows:

[0060] For the monitoring points in the deep foundation construction range, the following steps are performed:

[0061] Taking a deep foundation project in deep soft soil as an example, an initial BIM model (i.e., a preset construction model) of the deep foundation project design is converted into point cloud data (voxel down-sampling processing is performed thereon) by using a BIM (Building Information Modeling) export tool, clustering is performed on the processed point cloud by using DBSCAN (Density-Based Spatial Clustering of Applications with Noise), the neighborhood radius is set to 1.5 m (which can be adjusted according to the actual construction scene), the minimum point number is 20 (which can be adjusted), and the density of each cluster is obtained, denoted as a structure index a = a cluster density - all result cluster average density of a certain (any) cluster; and the initial global monitoring point number is denoted as A (which can be adjusted according to actual engineering requirements).

[0062] Then, the monitoring point number of the cluster is A.

[0063] The monitoring points are uniformly distributed within the range of the cluster.

[0064] After the monitoring points are determined, the multi-source sensors are installed (24 hours are taken as a monitoring period):

[0065] The monitoring points are uniformly set in the deep foundation, and sensors are set at each monitoring point:

[0066] The horizontal displacement at each monitoring point is obtained by using an inclinometer;

[0067] The foundation surface settlement degree at each monitoring point is obtained by using a static level gauge;

[0068] The vibration frequency of each monitoring point is obtained by using a vibration sensor;

[0069] The stress of each monitoring point is obtained by using a strain gauge;

[0070] Acquire a three-dimensional image of the construction range once a day at a fixed time using radar (target auxiliary radar monitoring exists in the foundation pit engineering range);

[0071] Based on the above sensors, real-time measurement is performed on each monitoring point to obtain the above various monitoring data, and the timestamps are aligned.

[0072] The monitoring data obtained by the above sensors is transmitted to the data processing module, stored, and subjected to deformation degree analysis to determine the soil instability risk of different regions.

[0073] For the above step S1, specifically, the step S1 comprises:

[0074] Determine the spatial nearest neighbor point of the target point cloud in the current point cloud data of the current monitoring period and the target point cloud in the preset construction model of the deep foundation pit;

[0075] The vector of the spatial nearest neighbor point of the target point cloud pointing to the target point cloud is taken as the target offset vector of the target point cloud, and the adjacent offset vectors of the adjacent point clouds of the target point cloud in the current point cloud data are determined.

[0076] In this embodiment, taking a single monitoring period, such as the current monitoring period, as an example:

[0077] The current point cloud data obtained by the radar in the current monitoring period is subjected to voxel downsampling at the same density as the preset construction model of the deep foundation pit.

[0078] ICP (Iterative Closest Point) algorithm is used to register the above two point clouds (corresponding to the current point cloud data and the preset construction model) (at this time, the two point clouds are in the same coordinate system). Taking any point (for example, point i, as the target point cloud) in the current monitoring period corresponding point cloud as an example, find the spatial nearest neighbor point (denoted as ) of point i in the preset construction model. The vector from to i is denoted as the target offset vector (denoted as , and the modulus is denoted as ).

[0079] The offset vectors of each point in the current monitoring period corresponding point cloud are obtained according to the above method. Similarly, the adjacent offset vectors of the adjacent point clouds of the target point cloud are also obtained.

[0080] Step S2, using the target offset vector and the adjacent offset vector to determine the target region of the current monitoring period, using the displacement information in the target region to determine the morphological stability degree and the deformation region of the target region;

[0081] Due to the actual construction process, the specific construction conditions and the preset construction model usually exist certain differences, and under the usual circumstances, the construction process is affected by the soil state and heavy engineering equipment, so that the difference with the preset construction model exists certain fluctuation with time sequence. Taking a single monitoring period as an example, according to the degree of deviation of the actual construction condition of the monitoring period from the preset construction model of the deep foundation pit, the actual construction model is divided into regions, if there is larger displacement and settlement of each monitoring point in the region, the deformation degree of the region in the monitoring period is larger.

[0082] Please refer to Figure 2 In an embodiment, the step S2 comprises:

[0083] Step S21, determine the cosine value of the included angle between the target offset vector and the adjacent offset vector and the offset difference value;

[0084] Step S22, using the cosine value of the included angle and the offset difference value, the offset similarity degree between the target point cloud and its adjacent point cloud is calculated;

[0085] Step S23, the point cloud with the offset similarity degree greater than the preset similarity threshold is taken as the offset similar point of the target point cloud, and the offset similar point and the target point cloud are merged into the same region to obtain the target region of the current monitoring period.

[0086] According to the offset similarity degree, the regions of each point in the point cloud corresponding to the current monitoring period are divided:

[0087] Taking the target point cloud i as an example, the offset similarity degree of i and its adjacent point cloud is analyzed (the points directly adjacent to i and without other points between the two points are the adjacent point cloud of i):

[0088] Let the adjacent point cloud of any target point cloud be , then the offset vector (denoted as , the length of the module is denoted as ) and the cosine value of the included angle of the offset vector of . Then for the offset similarity degree between the target point cloud and its adjacent point cloud ,

[0089]

[0090] In the formula, , the offset similarity degree is denoted as , the offset difference value of point i and , and ​​The points with a normalized result greater than 0.8 (a preset similarity threshold, which can be adjusted) are marked as offset similar points of point i in the current monitoring period, and are merged into the same region as point i. After the merging, all the newly added points are processed according to the above method, and all the qualified points are merged into the region. The above operation is repeated until there is no qualified point cloud.

[0091] Based on the above method, the region division is realized, and the obtained region is marked as a target region in the current monitoring period.

[0092] Please refer to Figure 3 In an embodiment, the step S2 includes:

[0093] The step S201 includes:

[0094] Specifically, the step S201 includes:

[0095] The range of the horizontal displacement of each monitoring point in the target region and the cumulative value of the horizontal displacement at each time point are determined.

[0096] The stable index of the horizontal displacement is calculated by using the range of the horizontal displacement and the cumulative value of the horizontal displacement.

[0097] The stable index of the horizontal displacement and the stable index of the vertical displacement are used to calculate the morphological stability degree of the target region.

[0098] The step S202 includes:

[0099] In the embodiment, taking any single target region j as an example, if the degree of deviation of each point in the point cloud in the target region from the preset construction model is small, and the horizontal displacement of each monitoring point measured at each time point in the current period fluctuates little, the cumulative horizontal displacement is also small, and the surface subsidence degree fluctuates little, and the cumulative subsidence in the vertical direction is also small, then the deformation of the target region in the current monitoring period is small, and the morphological stability degree is large.

[0100] The stable index of the horizontal displacement of each monitoring point in the target region in the current monitoring period is denoted as , wherein is the range of the horizontal displacement of the monitoring point in the target region, and the greater the extreme value, the greater the fluctuation degree of the horizontal displacement, is the average of the cumulative value of the horizontal displacement of all monitoring points in the current monitoring period, and the stable index of the surface subsidence of the region in the current period is obtained in the same way i.e. the stability index of vertical displacement. Further, for the calculation process of the stability degree of the morphological stability is as follows:

[0101]

[0102] denotes the stability index of horizontal displacement, denotes the stability index of vertical displacement, is the mean value of the modulus of the displacement vector of each point in the point cloud corresponding to the target region j, and the smaller the mean value is, the smaller the degree of deviation of the target region from the preset construction model is.

[0103] is normalized to (0, 1), and the target region whose normalized result is less than 0.6 (a preset stability threshold, which can be adjusted specifically) is marked as a deformation region, wherein the monitoring points of the deformation region are deformation monitoring points.

[0104] Step S3, determining the temporary fluctuation possibility and the real deformation region by using the morphological stability degree of the deformation region and the deformation region marking number in the historical monitoring period;

[0105] When the soil is disturbed by vibration, temporary deformation may exist, especially in deep foundation pit engineering under deep soft soil scene. Deep soft soil has strong compressibility and high sensitivity, and is easily affected by vibration caused by heavy equipment operation and short-term load disturbance and other interference factors in the deep foundation pit construction process, so that the soil has temporary deformation phenomenon. After the interference factors disappear, the corresponding region may have partial or full recovery, while the real deformation cannot be recovered. Therefore, the real deformation region and the temporary deformation region are distinguished based on the above differences.

[0106] Please refer to Figure 4 In an embodiment, the step S3 comprises:

[0107] Step S31, determining the deformation region marking number of the deformation region marked as a deformation region in the historical monitoring period, and determining the mean value of the deformation region marking number in all historical monitoring periods;

[0108] Step S32, determining the temporary fluctuation possibility by using the mean value of the morphological stability degree and the mean value of the deformation region marking number of the deformation region in all historical monitoring periods;

[0109] Specifically, the step S32 comprises:

[0110] combining the monitoring time periods corresponding to the deformation region marked as a deformation region for two or more times in succession in the historical monitoring period into a single deformation fluctuation duration, and determining the mean value of the durations of all deformation fluctuations;

[0111] ​​​The temporary fluctuation possibility is calculated by using the average of the shape stability degree, the average of the deformation region marking times and the average of the time length in all historical monitoring periods of the deformation region.

[0112] In step S33, the deformation region with the temporary fluctuation possibility less than or equal to the preset fluctuation threshold is regarded as a real deformation region.

[0113] In the embodiment, the deformation region in the period in which the current time is located (the current monitoring period) is taken as an example, and the deformation region determined by the unmanned airborne radar in the current monitoring period is patrolled (at least once every two hours in the current monitoring period).

[0114] The local point cloud and the whole point cloud in the current monitoring period are registered by using the ICP algorithm (and the deformation region screening and the shape stability degree calculation are also performed according to the above implementation process), and the whole point cloud in the current monitoring period is registered with all historical whole point clouds, and the range of all deformation regions in the current monitoring period is found in the historical point clouds.

[0115] Then, according to the above analysis, the more the number of times that a deformation region (for example, deformation region j) is marked as a deformation region in the region involved in the historical monitoring period, the greater the shape stability degree in each marked historical monitoring period, and the faster the deformation region j returns to the normal stable degree after being marked as a deformation region, the greater the possibility of the deformation region j being a temporary fluctuation interference region (temporary fluctuation possibility) , and the smaller the possibility of the deformation region j being a real deformation region.

[0116] Wherein, since the shape of the deformation region determined each time is uncertain, if the number of deformation monitoring points belonging to the deformation region in the corresponding range of the deformation region j in a certain historical monitoring period is greater than 0.6 times the number of monitoring points in the deformation region j, it is considered that the region corresponding to the historical monitoring period is marked as a deformation region in the historical monitoring period, and the average of the number of times that the deformation region j is marked as a deformation region in all historical monitoring periods (the average of the deformation region marking times) is recorded as .

[0117] If a single shape region j is marked as a deformation region for two times or more than two times, the continuous monitoring period of the shape region j is combined and marked as a single (single) deformation fluctuation, and the time length of the single deformation fluctuation is recorded as . For the temporary fluctuation possibility , there is:

[0118]

[0119] In the formula, the average of the shape stability degree of all historical monitoring periods (including local patrol and whole period monitoring) is recorded as , and the average of the deformation region marking times is recorded as The average of the time length of all the deformation fluctuations in the shape area j is represented. The average of the number of labels for the deformation area is marked.

[0120] The Normalized to (0, 1), mark the deformation area whose normalized result is greater than 0.5 (preset fluctuation threshold, which can be adjusted specifically) as a temporary fluctuation interference area, and the rest is the real deformation area.

[0121] Step S4, determine the target time period corresponding to the abnormal growth of the vibration frequency time sequence and the stress time sequence in the real deformation area, and determine the damage accumulation index of the real deformation area by using the temporary fluctuation possibility time sequence corresponding to each target time period;

[0122] The actual deformation area in the deep foundation pit construction, i.e. the real deformation area, is obtained from the above-mentioned various embodiments, wherein the deformation degree of the real deformation area may fluctuate with the interference of factors such as heavy construction equipment vibration and load disturbance. When a certain area has strong ability to respond to soil changes, its deformation degree usually changes less with the fluctuation of the above-mentioned interference factors, and when there is damage accumulation, the ability of the corresponding area to respond to soil changes gradually weakens with the interference of the above-mentioned factors, so that its deformation degree gradually increases.

[0123] Please refer to Figure 5 In an embodiment, the step S4, determining the target time period corresponding to the abnormal growth of the vibration frequency time sequence and the stress time sequence in the real deformation area, comprises:

[0124] Step S41, determine the vibration frequency time sequence in the real deformation area and the target growth time period of the vibration frequency time sequence;

[0125] Step S42, determine the abnormal growth time period in the target growth time period by using the growth amplitude of the target growth time period and the average of the vibration frequency and the average of the growth amplitude of all the growth time periods;

[0126] Step S43, the overlapping time period between the abnormal growth time period of the vibration frequency time sequence and the abnormal growth time period of the stress time sequence is taken as the target time period corresponding to the abnormal growth.

[0127] Arrange the vibration frequency and stress measured by each monitoring point in the real deformation area in time sequence (the values measured by the sensor are denoised by wavelet) to obtain the vibration frequency time sequence and the stress time sequence, and obtain their respective time sequence curves.

[0128] Taking a single monitoring point as an example, filter the time period in which the vibration frequency and stress have abnormal growth:

[0129] Taking the time series curve of vibration frequency as an example, we extract the growth periods (i.e., the curve is segmented at extreme points, with the left endpoint being the minimum and the right endpoint being the maximum). If the growth amplitude of a certain growth period (e.g., period y, as the target growth period) is larger than that of the other growth periods, and the average vibration frequency in that period is also larger, then the target growth period y is likely to be an abnormal growth period. for:

[0130]

[0131] In the formula, The growth rate of y during the target growth period. This represents the average growth rate across all growth periods. Let y be the average vibration frequency during the target growth period.

[0132] Will Normalize to (0, 1), and mark the period when the normalization result is greater than 0.5 (preset growth threshold, which can be adjusted) as the period of abnormal growth in vibration frequency.

[0133] Similarly, the above implementation process is used to screen for periods of abnormal stress growth.

[0134] The period during which the abnormal increase in vibration frequency and the abnormal increase in stress overlap is marked as the target period of abnormal increase.

[0135] Please refer to Figure 6 In another embodiment, step S4, which uses the temporal sequence of temporary fluctuations corresponding to each target time period to determine the damage accumulation index of the actual deformation area, includes:

[0136] Step S401: The probability of the temporary fluctuation closest to the start time in the target time period is taken as the probability of the temporary fluctuation corresponding to the target time period.

[0137] Step S402: Determine the first-order difference sequence of the temporary fluctuation probability time series corresponding to each target time period, and take the negative value in the first-order difference sequence as the effective damage value.

[0138] Step S403: Using the number of effective damage values, their average absolute value, and the average time between each effective damage value and its corresponding previous target time period, the damage accumulation index of the actual deformation area is calculated.

[0139] In this embodiment, if the temporary fluctuation probability corresponds to each target time period If there is a clear decreasing trend, the soil at the current monitoring point is less able to cope with disturbances such as vibration and load fluctuations, and the damage accumulation is more severe.

[0140] The single target period of the monitoring point corresponds to the value closest to the start time of the period.

[0141] The value corresponding to each target period is arranged in time sequence to obtain a temporary fluctuation possibility time sequence and obtain a first-order difference sequence.

[0142] If the abnormal growth of vibration and stress can cause the temporary fluctuation possibility to decrease, the damage degree of the soil body caused by the abnormal growth is greater, and the value of the first-order difference in the first-order difference sequence is marked as an effective damage value.

[0143] If the number of effective damage values corresponding to a monitoring point is greater and the damage amplitude is greater, and the time length of each effective damage value from the previous target period is shorter, the damage accumulation degree of the soil body at the monitoring point is greater:

[0144]

[0145] In the formula, is the number of effective damage values corresponding to the monitoring point, is the mean of the absolute values of the effective damage values, and the greater the mean of the absolute values, the greater the damage amplitude, is the mean of the time lengths of each effective damage value from the previous target period corresponding to itself, and the smaller the mean of the time lengths, the faster the damage accumulation.

[0146] According to the above implementation process, the damage accumulation degrees of all monitoring points are obtained, and the mean of all monitoring points in a single real deformation region is the damage accumulation index of the real deformation region .

[0147] Step S5, using the morphological stability degree and the damage accumulation index of the real deformation region to determine the corresponding early warning priority.

[0148] Specifically, the step S5 comprises:

[0149] The early warning priority of the real deformation region is calculated using the morphological stability degree, the damage accumulation index, and the number of voxels of the real deformation region; and the early warning priority is used to output different early warning signals about the real deformation region.

[0150] In the embodiment, the damage accumulation index of each real deformation region is analyzed to analyze the soil instability degree, the greater the range of a real deformation region, the greater the damage accumulation degree, the more unstable the morphology of the real deformation region at the current time, and the higher the risk of soil instability, and the greater the early warning priority.​

[0151] The early warning priority (index) of a single real deformation region is:

[0152]

[0153] In the formula, is the number of voxels in the real deformation region j in the actual construction point cloud data (current point cloud data) corresponding to the current time, and the larger the number of voxels, the larger the range of the current real deformation region; is the damage accumulation index of the real deformation region j; is the morphological stability degree of the real deformation region; is a normalization function, and the value range is (0, 1).

[0154] In combination with the early warning priority, further, the above-mentioned various embodiments mainly include:

[0155] Based on digital twin technology, the physical entity of deep foundation construction is digitized (digital perception): combining multi-source sensors to monitor the deep foundation construction process under deep soft soil, and combining the difference between the actual construction model and the preset construction model to partition the actual construction range.

[0156] Based on the digitalization result, the temporary deformation region disturbed by mechanical vibration or load fluctuation is screened out.

[0157] And based on the deformation development degree of each real deformation region, the soil recovery ability of the corresponding position is simulated and predicted to obtain the soil instability degree (corresponding early warning priority) of each region.

[0158] The soil instability degree of the real deformation region is fed back (i.e. the real physical construction scene is fed back through the virtual prediction of digital twin). The grading early warning information of the soil instability degree of the real deformation region can be visualized based on the digital twin platform:

[0159] The early warning priority is mapped to the construction model corresponding to the current time, and each real deformation region is marked, and the soil instability priority index of the real deformation region is output in descending order according to , wherein, is a high-risk instability early warning (corresponding to a red flashing region), is a medium-risk instability early warning (corresponding to an orange flashing region), and the remaining real deformation regions are instability early warnings (corresponding to yellow flashing regions).

[0160] ​The application is based on digital twin technology, analyzes deep foundation pit construction monitoring data obtained by multiple source sensors in the deep foundation pit construction process, analyzes the deformation degree of different regions in the deep foundation pit at the current time based on real-time deformation data, thereby distinguishing temporary deformation regions and real deformation regions formed by heavy construction machinery and load fluctuations, correlating the deformation degree of each real deformation region at different times in the deep foundation pit to analyze the damage accumulation degree of each region, thereby predicting the ability of each region to respond to soil changes, and based on the deformation development trend of the real deformation region, accurately classifying and warning the soil instability degree of different regions in the deep foundation pit under real-time construction progress, thereby improving the accuracy of deep foundation pit automatic monitoring.

[0161] Embodiment two:

[0162] The embodiment of the application also provides a deep foundation pit multi-parameter automatic monitoring device based on digital twin driving.

[0163] As shown in Figure 7 , Figure 7 is a structural schematic diagram of a hardware running environment of the deep foundation pit multi-parameter automatic monitoring device based on digital twin driving involved in the embodiment of the application.

[0164] As shown in Figure 7 , the deep foundation pit multi-parameter automatic monitoring device based on digital twin driving can include a processor 1001 such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. The communication bus 1002 is used to realize the connection and communication between the components. The user interface 1003 can include a display (Display), an input unit such as a control panel, and the optional user interface 1003 can also include a standard wired interface, a wireless interface. The network interface 1004 can optionally include a standard wired interface, a wireless interface (such as a WIFI interface). The memory 1005 can be a high-speed RAM memory or a stable memory (non-volatile memory) such as a magnetic disk memory. The memory 1005 can optionally be a storage device independent of the aforementioned processor 1001. The memory 1005, as a computer storage medium, can include a deep foundation pit multi-parameter automatic monitoring program.

[0165] Those skilled in the art can understand that Figure 7 the hardware structure shown in the foregoing embodiments does not constitute a limitation on the device, and can include more or fewer components than those shown, or combine certain components, or different component arrangements.

[0166] Continuing to refer to Figure 7 , Figure 7The memory 1005 as a computer readable storage medium can include an operation device, a user interface module, a network communication module, and a deep foundation pit multi-parameter automatic monitoring program.

[0167] In ​ In the above, the network communication module is mainly used for connecting a server and can communicate data with the server; and the processor 1001 can call the deep foundation pit multi-parameter automatic monitoring program stored in the memory 1005 and execute the steps in the above various embodiments.

[0168] Based on the hardware structure of the deep foundation pit multi-parameter automatic monitoring device based on digital twin driving, various embodiments of the deep foundation pit multi-parameter automatic monitoring method based on digital twin driving of the present application are implemented.

[0169] In addition, the present application also provides a computer readable storage medium. The computer readable storage medium of the present application stores a deep foundation pit multi-parameter automatic monitoring program, wherein the deep foundation pit multi-parameter automatic monitoring program, when executed by a processor, implements the steps of the deep foundation pit multi-parameter automatic monitoring method based on digital twin driving as described above.

[0170] The method implemented by the deep foundation pit multi-parameter automatic monitoring program when executed can refer to various embodiments of the deep foundation pit multi-parameter automatic monitoring method based on digital twin driving of the present application, which will not be described here.

[0171] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or can be advantageous.

[0172] Each embodiment in the specification is described in a progressive manner, and the same and similar parts between each embodiment can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0173] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, device, or computer program product. Therefore, the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can be in the form of a computer program product implemented on one or more computer usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer usable program code.

[0174] The above merely describes preferred embodiments of the present application, and is not intended to limit the protection scope of the present application, and any equivalent structure / method transformation made under the inventive concept of the present application, or direct / indirect application in other related technical fields, is included in the protection scope of the present application.

Claims

1. A deep foundation multi-parameter automatic monitoring method based on digital twin driving, characterized in that, The method comprises: determining target offset vectors and adjacent offset vectors of a target point cloud of a deep foundation pit in a current monitoring period and its adjacent point clouds relative to a preset construction model respectively; determining a target region of the current monitoring period by using the target offset vectors and the adjacent offset vectors, and determining a morphological stability degree and a deformation region of the target region by using displacement information in the target region; determining a temporary fluctuation possibility and a real deformation region by using the morphological stability degree of the deformation region and a deformation region marking number in a historical monitoring period; determining target time periods in which vibration frequency time series and stress time series of the real deformation region correspond to abnormal growth together, and determining an injury accumulation index of the real deformation region by using a temporary fluctuation possibility time series corresponding to each target time period; determining a corresponding early warning priority by using the morphological stability degree and the injury accumulation index of the real deformation region; The determination of the temporary fluctuation possibility and the real deformation region comprises: determining a deformation region marking number of the deformation region marked as a deformation region in the historical monitoring period, and determining a deformation region marking number average in all historical monitoring periods; determining the temporary fluctuation possibility by using a morphological stability degree average and the deformation region marking number average in all historical monitoring periods of the deformation region; regarding the deformation region with a temporary fluctuation possibility less than or equal to a preset fluctuation threshold as a real deformation region; The determination of the temporary fluctuation possibility by using the morphological stability degree average and the deformation region marking number average in all historical monitoring periods of the deformation region comprises: merging monitoring time periods in which the deformation region is marked as a deformation region for two or more times in succession into a single deformation fluctuation time length, and determining a time length average of all deformation fluctuations; calculating the temporary fluctuation possibility by using the morphological stability degree average, the deformation region marking number average and the time length average in all historical monitoring periods of the deformation region.

2. The deep foundation multi-parameter automatic monitoring method based on digital twin driving according to claim 1, characterized in that, The determination of target offset vectors and adjacent offset vectors of a target point cloud of a deep foundation pit in a current monitoring period and its adjacent point clouds relative to a preset construction model respectively comprises: determining the target point cloud in current point cloud data of the deep foundation pit in the current monitoring period and a spatial nearest neighbor point of the target point cloud in the preset construction model of the deep foundation pit; regarding a vector from the spatial nearest neighbor point of the target point cloud to the target point cloud as a target offset vector of the target point cloud, and determining an adjacent offset vector of an adjacent point cloud of the target point cloud in the current point cloud data.

3. The deep foundation multi-parameter automatic monitoring method based on digital twin driving according to claim 1, characterized in that, The determination of a target region of the current monitoring period by using the target offset vectors and the adjacent offset vectors comprises: determining a cosine value of an included angle and an offset difference value between the target offset vectors and the adjacent offset vectors; calculating an offset similarity degree between the target point cloud and its adjacent point cloud by using the cosine value of the included angle and the offset difference value; regarding a point cloud with an offset similarity degree greater than a preset similarity threshold as an offset similar point of the target point cloud, and merging the offset similar point and the target point cloud into a same region to obtain the target region of the current monitoring period.

4. The deep foundation multi-parameter automatic monitoring method based on digital twin driving according to claim 1, characterized in that, The determination of a morphological stability degree and a deformation region of the target region by using displacement information in the target region comprises: determining the morphological stability degree of the target region by using horizontal displacement and vertical displacement of each monitoring point in the target region; The target region with a morphological stability degree less than a preset stability threshold is taken as a deformation region.

5. The deep foundation multi-parameter automatic monitoring method based on digital twin driving according to claim 4, characterized in that, The morphological stability degree of the target region is determined by using horizontal displacement and vertical displacement of each monitoring point in the target region, and the method comprises the following steps: determining the range of horizontal displacement of each monitoring point in the target region and the cumulative value of horizontal displacement at each time point; calculating the stability index of horizontal displacement by using the range of horizontal displacement and the cumulative value of horizontal displacement; calculating the morphological stability degree of the target region by using the stability index of horizontal displacement and the stability index of vertical displacement.

6. The deep foundation multi-parameter automatic monitoring method based on digital twin driving according to claim 1, characterized in that, The target time period corresponding to abnormal growth of the vibration frequency time sequence and the stress time sequence in the real deformation region is determined, and the method comprises the following steps: determining the vibration frequency time sequence in the real deformation region and the target growth time period of the vibration frequency time sequence; determining the abnormal growth time period in the target growth time period by using the growth amplitude of the target growth time period, the average value of the vibration frequency and the average value of the growth amplitude of all growth time periods; taking the overlapping time period between the abnormal growth time period of the vibration frequency time sequence and the abnormal growth time period of the stress time sequence as the target time period corresponding to abnormal growth.

7. The deep foundation multi-parameter automatic monitoring method based on digital twin driving according to claim 1, characterized in that, The damage accumulation index of the real deformation region is determined by using the temporary fluctuation possibility time sequence corresponding to each target time period, and the method comprises the following steps: taking the temporary fluctuation possibility closest to the starting time point in the target time period as the temporary fluctuation possibility corresponding to the target time period; determining the first-order difference sequence of the temporary fluctuation possibility time sequence corresponding to each target time period, and taking the negative value in the first-order difference sequence as an effective damage value; calculating the damage accumulation index of the real deformation region by using the number of effective damage values, the average value of the absolute values and the average time length of each effective damage value from the corresponding previous target time period.

8. The deep foundation multi-parameter automatic monitoring method based on digital twin driving according to claim 1, characterized in that, The corresponding early warning priority is determined by using the morphological stability degree and the damage accumulation index of the real deformation region, and the method comprises the following steps: calculating the early warning priority of the real deformation region by using the morphological stability degree, the damage accumulation index and the number of voxels of the real deformation region; the early warning priority is used for outputting different early warning signals about the real deformation region.

Images