Internet of things based multi-source sensing data collaborative monitoring method for engineering structure

Abstract

The application discloses an engineering structure multi-source sensing data cooperative monitoring method based on an Internet of Things, relates to the technical field of engineering monitoring, and is used for solving the problem of reduced accuracy of monitoring results in representing actual structure stress and deformation state. The method comprises the following steps: collecting blasting vibration velocity data and extracting vibration velocity peak values within a construction cycle time, dividing the construction cycle time into different disturbance intensity levels, obtaining surrounding rock displacement data and support structure strain data under each disturbance intensity level and performing correlation marking, obtaining crack connection distribution and section convergence of a tunnel section by three-dimensional laser scanning, fusing section geometric features, and judging whether the section geometric state is in a stable state or not. The method further comprises the following steps: screening a historical prediction reference level based on the disturbance intensity level, and updating the historical prediction reference by using the correlation marking detection result of the current construction cycle, so as to realize effective screening of multi-source monitoring data under unified construction conditions and adaptive correction of the reference.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of engineering monitoring, more particularly, the present application relates to a method for collaborative monitoring of multi-source sensing data of engineering structures based on the Internet of Things. BACKGROUND

[0002] With the continuous expansion of the scale of underground projects such as urban rail transit, highway tunnels and underground comprehensive pipe galleries, structural safety monitoring during tunnel construction has become an important technical means to ensure engineering safety and construction quality. Especially under the construction conditions of drill-and-blast method or micro-differential blasting method, the surrounding rock and supporting structure show significant nonlinear response characteristics under the periodic blasting disturbance, and the deformation, strain and crack evolution process has obvious stage and randomness.

[0003] The prior art has the following disadvantages:

[0004] At present, in actual engineering applications, due to the fact that the multi-source sensing data collection process is not combined with the grading management of the blasting disturbance intensity within the construction cycle, there is a lack of mechanism for uniformly associating and screening the monitoring data of surrounding rock displacement and supporting structure strain based on construction disturbance characteristics, and often the monitoring data under different construction stages and different disturbance conditions are mixed and analyzed, resulting in reduced accuracy of the monitoring results in representing the actual structure stress and deformation state, and increased uncertainty in construction safety risk identification and prediction. Therefore, a method for collaborative monitoring of multi-source sensing data of engineering structures based on the Internet of Things is proposed.

[0005] The above information disclosed in the background section is only intended to strengthen the understanding of the background of the present disclosure, and therefore it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0006] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a method for collaborative monitoring of multi-source sensing data of engineering structures based on the Internet of Things, which uses a blasting disturbance intensity grading mechanism within the construction cycle and a multi-source sensing data association updating mechanism under the constraint of cross-sectional geometric stability to solve the problems raised in the above background.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical scheme, a method for collaborative monitoring of multi-source sensing data of engineering structures based on the Internet of Things, comprising the following steps:

[0008] Step S1: Perform construction monitoring on the tunnel cross-section to be measured, retrieve construction state information and extract the current construction stage and construction cycle time, collect blasting vibration velocity data of the tunnel cross-section to be measured within the construction cycle time, and then count the vibration velocity peak value;

[0009] Step S2: Fuse the peak value of vibration velocity with the preliminary disturbance intensity hierarchy of the construction cycle time in the current construction stage, detect the displacement data of the surrounding rock and the strain data of the supporting structure of the tunnel section to be measured, and associate and mark the detection results based on the disturbance intensity hierarchy;

[0010] Step S3: Perform three-dimensional laser scanning processing on the tunnel section to be measured to obtain the crack connectivity distribution and the section convergence of the tunnel section to be measured, evaluate the section geometric characteristics based on the crack connectivity distribution and the section convergence, and determine whether the section geometric state is in a stable state;

[0011] Step S4: When the section geometric state is in a stable state, access the historical database to retrieve the historical prediction reference hierarchy and select the target disturbance intensity hierarchy, and update the historical prediction reference hierarchy with the associated marked detection results of the target disturbance intensity hierarchy.

[0012] In a preferred embodiment, in step S1, the construction state information is retrieved by the tunnel construction management unit, including the current construction stage and the construction cycle time corresponding to the current construction stage;

[0013] When the current construction stage is in the blasting operation stage, the blasting vibration velocity is collected by the blasting vibration monitoring device arranged at the tunnel section to be measured within the construction cycle time, and the blasting vibration velocity is taken as the blasting vibration velocity data.

[0014] In a preferred embodiment, in step S1, if the blasting vibration velocity data is greater than a preset background vibration threshold, the blasting vibration velocity data is marked; otherwise, the blasting vibration velocity data is not marked.

[0015] The marked blasting vibration velocity data is arranged in time sequence, and adjacent marked blasting vibration velocity data is combined as a blasting event section.

[0016] In each blasting event section, the maximum value of the marked blasting vibration velocity data is taken as the peak value of the blasting event section.

[0017] In a preferred embodiment, in step S2, based on the vibration velocity reference value of the current construction stage retrieved from the historical database, the ratio of the peak value of the vibration velocity to the vibration velocity reference value is taken as the peak value coefficient of the vibration velocity;

[0018] The peak values of the vibration velocity are sorted in time sequence, the peak value change amount of the vibration velocity obtained by subtracting adjacent peak values of the vibration velocity is arranged in descending order of numerical value, and the peak value coefficients of the vibration velocity are selected in turn based on the preset number of hierarchies to obtain the hierarchical division peak value of the vibration velocity.

[0019] The construction cycle time is divided into disturbance intensity hierarchies according to the collection time corresponding to the hierarchical division peak value of the vibration velocity.

[0020] In a preferred embodiment, in step S2, the surrounding rock displacement measurement values of the tunnel section to be measured are obtained by the surrounding rock displacement monitoring device in the time segment corresponding to each disturbance intensity level;

[0021] wherein the mean value of the surrounding rock displacement measurement values measured before the time segment corresponding to the disturbance intensity level is taken as the displacement reference value, and the mean value of the surrounding rock displacement measurement values after the end of the time segment corresponding to the disturbance intensity level is taken as the displacement response value;

[0022] The surrounding rock displacement data is calculated based on the displacement response value and the displacement reference value;

[0023] The support structure strain measurement values of the tunnel section to be measured are obtained by the support structure strain monitoring device;

[0024] The support structure strain data is calculated based on the support structure strain measurement values;

[0025] The surrounding rock displacement data and the support structure strain data corresponding to each disturbance intensity level are taken as the detection results, and the detection results are associated with the disturbance intensity levels for marking.

[0026] In a preferred embodiment, in step S3, after the end of the corresponding construction cycle, the three-dimensional laser scanning device arranged in the tunnel is controlled to scan the tunnel section to be measured according to the preset scanning resolution and scanning angle, and three-dimensional spatial point cloud data of the inner wall surface of the section is obtained;

[0027] The three-dimensional spatial point cloud data is composed of a plurality of discrete points with spatial coordinates, and each spatial point includes its three-dimensional coordinates in the tunnel section coordinate system;

[0028] The three-dimensional spatial point cloud data is locally plane-fitted, the included angle of the normal vectors of adjacent spatial points and the distance between the points are calculated, and when the included angle of the normal vectors between adjacent spatial points is greater than a preset normal vector mutation threshold value and the distance between the points exceeds a preset gap determination threshold value, it is determined that there is a crack feature point in the region;

[0029] The crack feature points are subjected to spatial clustering processing to obtain a plurality of crack clusters, and it is judged whether the crack clusters are interconnected based on the minimum spatial distance between the crack clusters and the consistency of the extension direction, so as to form a crack connected distribution.

[0030] In a preferred embodiment, in step S3, a plurality of characteristic azimuth angles are selected in the tunnel section coordinate system, which are uniformly distributed along the circumference;

[0031] The radial distance of the current scanning section contour point to the geometric center of the section is determined in each characteristic azimuth angle direction, and the radial distance is difference calculated with the radial distance of the same section at the corresponding azimuth angle under the last construction cycle or the initial reference state, to obtain the local convergence amount at each azimuth angle;

[0032] Taking the absolute value of the local convergence amount corresponding to all azimuth angles and averaging them as the section convergence amount;

[0033] The fracture connectivity distribution and the section convergence amount are normalized, and after normalization, the fracture connectivity distribution and the section convergence amount are weighted and summed according to the preset fusion weight to obtain the section geometric feature;

[0034] The change amplitude of the section geometric feature in the adjacent construction cycle is calculated to obtain the section geometric feature change value;

[0035] When the section geometric feature change value is less than the preset geometric stability judgment threshold, it is determined that the geometric state of the tunnel section under test is in a stable state;

[0036] When the section geometric feature change value exceeds the preset geometric stability judgment threshold, it is determined that the section geometric state is in an unstable state.

[0037] In a preferred embodiment, in step S4, when it is determined that the geometric state of the tunnel section under test is in a stable state, the historical database is accessed to retrieve the historical prediction reference hierarchy;

[0038] The historical prediction reference hierarchy refers to a structure response reference formed based on multi-source sensing data in the tunnel engineering construction process and stored according to construction stage and disturbance intensity hierarchy classification;

[0039] Each historical prediction reference hierarchy corresponds to a group of historical detection reference records formed under the same construction stage and the same disturbance intensity hierarchy;

[0040] The historical detection reference record includes the upper and lower boundaries of the reference interval of the surrounding rock displacement feature and the support structure strain feature;

[0041] In the historical detection reference record, further filter out the target disturbance intensity hierarchy consistent with the disturbance intensity hierarchy of the current construction cycle.

[0042] In a preferred embodiment, in step S4, the correlation marker detection result obtained in the current construction cycle is compared with the historical detection reference interval defined by the corresponding historical detection reference record:

[0043] When the correlation marker detection result falls within the historical detection reference interval corresponding to the target disturbance intensity hierarchy, it is determined that the correlation marker detection result is an effective update sample, and the historical prediction reference hierarchy corresponding to the target disturbance intensity hierarchy is updated based on the effective update sample;

[0044] In the updating process, the boundaries of the reference interval in the historical detection reference record are corrected;

[0045] When the correlation mark detection result does not fall into the historical detection reference interval, then the update processing is not performed on the historical prediction reference level.

[0046] Technical effects and advantages of the present application:

[0047] The present application divides the construction cycle time into different disturbance intensity levels by collecting blasting vibration velocity data and extracting the vibration velocity peak value within the construction cycle time, obtains the surrounding rock displacement data and support structure strain data under each disturbance intensity level and performs correlation marking, obtains the crack connectivity distribution and section convergence of the tunnel section by three-dimensional laser scanning, fuses to form the section geometric features, and judges whether the section geometric state is in a stable state, when the section geometric state meets the stable condition, the historical prediction reference level is screened based on the disturbance intensity level, and the historical prediction reference is updated by using the correlation mark detection result of the current construction cycle. BRIEF DESCRIPTION OF DRAWINGS

[0048] Fig. 1 The implementation flowchart of the present application based on the Internet of Things engineering structure multi-source sensing data collaborative monitoring method.

[0049] Fig. 2 The step schematic diagram of the present application based on the Internet of Things engineering structure multi-source sensing data collaborative monitoring method. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0051] The present application divides the construction cycle time into different disturbance intensity levels by collecting blasting vibration velocity data and extracting the vibration velocity peak value within the construction cycle time, obtains the surrounding rock displacement data and support structure strain data under each disturbance intensity level and performs correlation marking, obtains the crack connectivity distribution and section convergence of the tunnel section by three-dimensional laser scanning, fuses to form the section geometric features, and judges whether the section geometric state is in a stable state, when the section geometric state meets the stable condition, the historical prediction reference level is screened based on the disturbance intensity level, and the historical prediction reference is updated by using the correlation mark detection result of the current construction cycle.

[0052] Example 1, as Figs. 1-2The shown Internet of Things-based engineering structure multi-source sensor data collaborative monitoring method comprises the following steps:

[0053] Step S1: Perform construction monitoring on the tunnel section to be measured, call the construction state information and extract the current construction stage and construction cycle time, collect the blasting vibration velocity data of the tunnel section to be measured within the construction cycle time, and then count the peak value of the vibration velocity;

[0054] Step S2: Fuse the peak value of the vibration velocity and the current construction stage to preliminarily divide the disturbance intensity hierarchy of the construction cycle time, detect the surrounding rock displacement data and support structure strain data of the tunnel section to be measured, and associate and mark the detection results based on the disturbance intensity hierarchy;

[0055] Step S3: Perform three-dimensional laser scanning processing on the tunnel section to be measured, obtain the crack connectivity distribution and section convergence of the tunnel section to be measured, and comprehensively evaluate the section geometric characteristics and judge whether the section geometric state is in a stable state;

[0056] Step S4: When the section geometric state is in a stable state, access the historical database to call the historical prediction reference hierarchy and filter the target disturbance intensity hierarchy, and update the historical prediction reference hierarchy using the associated and marked detection results of the target disturbance intensity hierarchy.

[0057] The specific implementation is as follows:

[0058] In step S1, the tunnel section to be measured is monitored, the construction state information is called through the tunnel construction management unit, the construction state information reflects the current construction operation state of the tunnel section, including the current construction stage and the construction cycle time;

[0059] Among them, the current construction stage is the tunnel section operation stage identifier obtained according to the construction process; the construction cycle time is a time interval determined in the current construction stage, taking one complete construction operation cycle as a unit;

[0060] The blasting operation stage is the most concentrated construction operation behavior of applying instantaneous dynamic disturbance to the surrounding rock and support structure in tunnel construction;

[0061] When the current construction stage is in the blasting operation stage, the blasting operation stage contains multiple blasting events, and within the construction cycle time, the blasting vibration velocity is collected through the blasting vibration monitoring device arranged in the tunnel section to be measured, and the blasting vibration velocity is taken as the blasting vibration velocity data;

[0062] The preset background vibration threshold is compared with the blasting vibration velocity data, the blasting event section is filtered, if the blasting vibration velocity data is greater than the preset background vibration threshold, the blasting vibration velocity data is marked; otherwise, the blasting vibration velocity data is not marked;

[0063] The marked blasting vibration velocity data is arranged in chronological order, and adjacent marked blasting vibration velocity data is combined as a blasting event section;

[0064] In each blasting event section, the maximum value of the marked blasting vibration velocity data is taken as the peak value of the blasting event section.

[0065] It should be noted that the tunnel construction management unit is a construction control platform for recording and managing tunnel construction processes and operation stages; the preset background vibration threshold value can be set according to the statistical distribution of historical non-blasting operation stage vibration data.

[0066] In step S2, the instantaneous dynamic disturbance exerted by different disturbance intensity levels on the surrounding rock and the supporting structure is different in strength, and the peak values of the vibration velocity in the construction cycle time are sorted in chronological order in the current construction stage, and the blasting event section is divided into different disturbance intensity levels based on the distribution of the peak values of the vibration velocity;

[0067] The vibration velocity reference value of the current construction stage is called from the historical database, and the vibration velocity reference value reflects the historical level of the peak value of the blasting event vibration velocity in the current construction stage;

[0068] The ratio of the peak value of the vibration velocity to the vibration velocity reference value is taken as the peak value coefficient of the vibration velocity;

[0069] The peak value change amount obtained by differencing adjacent peak value coefficients of the vibration velocity is arranged in descending order of numerical value, and the peak value coefficients of the vibration velocity are sequentially selected based on the preset number of levels to obtain the level segmentation peak value of the vibration velocity, and the construction cycle time is divided into disturbance intensity levels according to the collection time corresponding to the level segmentation peak value of the vibration velocity;

[0070] In the time section corresponding to each disturbance intensity level, the surrounding rock displacement measurement value of the tunnel section to be measured is obtained by the surrounding rock displacement monitoring device;

[0071] The mean value of the surrounding rock displacement measurement value in the preset baseline time window before the starting time of the time section corresponding to the disturbance intensity level is taken as the displacement reference value; and the mean value of the surrounding rock displacement measurement value in the preset evaluation time window after the end of the time section corresponding to the disturbance intensity level is taken as the displacement response value.

[0072] The difference between the displacement response value and the displacement reference value is taken as the surrounding rock displacement data.

[0073] It should be explained that the preset baseline time window is used to eliminate the influence of monitoring noise and short-time disturbance on the initial state judgment of displacement and strain, and can be set according to the sampling frequency of the monitoring equipment and the historical stable data; the preset evaluation time window can be set according to the length of the disturbance intensity level or the difference between the construction stages.

[0074] The support structure strain measurement value of the tunnel section to be measured is obtained by a support structure strain monitoring device;

[0075] The average of the support structure strain measurement value in a preset baseline time window before the starting moment of the corresponding time section of the disturbance intensity level is taken as a strain reference value, and the average of the support structure strain measurement value in a preset evaluation time window after the end of the corresponding time section of the disturbance intensity level is taken as a strain response value;

[0076] The difference between the strain response value and the strain reference value is taken as the support structure strain data, reflecting the change degree of the stress state of the support structure under the action of the disturbance intensity level;

[0077] The surrounding rock displacement data and the support structure strain data corresponding to each disturbance intensity level are taken as the detection results, and the detection results are associated with the disturbance intensity levels.

[0078] It should be noted that the preset number of levels is used to limit the number of disturbance intensity levels, which can be set according to the demand of construction management on disturbance grading accuracy and the number of historical blasting events; the surrounding rock displacement monitoring device is a structure deformation monitoring device for detecting the displacement change of the surrounding rock of the tunnel section relative to the initial reference position; the support structure strain monitoring device is a structure stress monitoring device for detecting the strain change of the force member of the tunnel support structure under external load.

[0079] In step S3, three-dimensional laser scanning processing is performed on the tunnel section to be measured to obtain geometric parameter data for quantitatively representing the geometric state of the section. Specifically, after the corresponding construction cycle ends, the three-dimensional laser scanning device arranged in the tunnel is controlled to scan the tunnel section to be measured according to a preset scanning resolution and scanning angle to obtain three-dimensional space point cloud data of the inner wall surface of the section.

[0080] It should be noted that the three-dimensional laser scanning device is a non-contact space measurement device arranged inside the tunnel, which is used to obtain the geometric shape of the inner wall of the tunnel section during construction. Based on the time of flight of laser, the propagation time of laser beam from emission to reception is calculated, and the spatial distance from the laser scanning point to the scanning device is obtained accordingly. Combined with the laser emission direction information, the spatial distance is converted into three-dimensional spatial coordinates in the tunnel coordinate system, thereby forming three-dimensional space point cloud data composed of a plurality of spatial coordinate points.

[0081] The three-dimensional space point cloud data is composed of a plurality of discrete points with spatial coordinates, and each spatial point contains its three-dimensional coordinates in the tunnel section coordinate system, which is used as the original data for section geometric state analysis.

[0082] Based on the obtained three-dimensional spatial point cloud data, the crack connectivity distribution is extracted. Specifically, local plane fitting is performed on the three-dimensional spatial point cloud data, the surface normal vectors corresponding to each spatial point are calculated, and a geometric mutation judgment condition is constructed based on the angle between the normal vectors of adjacent spatial points and the distance between the points. When the angle between the normal vectors of adjacent spatial points is greater than a preset normal vector mutation threshold, and the distance between the points exceeds a preset gap judgment threshold, it is determined that there is a crack feature point in the region.

[0083] It should be noted that the normal vector mutation threshold is an angle threshold for determining whether the surface orientation of adjacent spatial points changes significantly, which is defined as the upper limit of the angle between the normal vectors of adjacent spatial points. By statistically analyzing the angle distribution of the normal vectors of adjacent spatial points in the point cloud data of the crack-free region, the upper quantile value of the statistical distribution is selected as the normal vector mutation threshold; the gap judgment threshold is a distance threshold for determining whether there is an actual geometric discontinuity between adjacent spatial points, which is defined as the minimum abnormal judgment value of the distance between adjacent spatial points. By statistically analyzing the distance distribution between points on a normal continuous surface, a distance value higher than the fluctuation range of the normal distance between points is selected as the gap judgment threshold.

[0084] The spatial clustering processing is performed on the crack feature points to obtain a plurality of crack clusters, and whether the crack clusters are connected to each other is judged based on the minimum spatial distance between the crack clusters and the consistency of the extension direction, so as to form the crack connectivity distribution. The crack connectivity distribution is used to quantitatively describe whether the cracks in the tunnel section to be measured form a continuous through structure, and the numerical result reflects the local geometric discontinuity degree of the section.

[0085] At the same time of extracting the crack connectivity distribution, the section convergence is calculated based on the three-dimensional spatial point cloud data. Specifically, a plurality of characteristic azimuth angles are selected in the tunnel section coordinate system, the radial distance from the current scanning section contour point to the geometric center of the section is determined in each characteristic azimuth angle direction, and the radial distance is difference calculated with the radial distance of the same section at the corresponding azimuth angle in the last construction cycle or the initial reference state, to obtain the local convergence at each azimuth angle.

[0086] Further, the absolute values of the local convergences corresponding to all azimuth angles are taken and the average value is taken as the section convergence, which is used to quantitatively represent the shrinkage level of the overall geometric size of the tunnel section to be measured.

[0087] The crack connectivity distribution and the section convergence are normalized to map to a unified numerical interval, so as to eliminate the influence of dimension difference on the fusion calculation; after normalization, the crack connectivity distribution and the section convergence are weighted and summed according to the preset fusion weight to obtain the section geometric feature.

[0088] It should be noted that the fusion weight is used to reflect the relative importance of the crack connectivity distribution and the cross-section convergence in the evaluation of the cross-section geometric stability, and its setting is based on the historical statistical results of the influence degree of the two on the cross-section geometric stability. Specifically, historical samples that have completed construction and whose cross-section geometric state has been clearly determined to be stable or unstable are selected in the historical database, and the value variation range and discrimination ability of the crack connectivity distribution and the cross-section convergence under different geometric states are respectively counted. The normalized sensitivity degree is taken as the fusion weight by calculating the sensitivity degree of each parameter to the cross-section geometric state determination result.

[0089] The cross-section geometric feature represents the local crack connectivity evolution and the overall cross-section convergence evolution of the cross-section, and the variation value of the cross-section geometric feature in the adjacent construction cycle is obtained by calculating the variation range of the cross-section geometric feature in the adjacent construction cycle.

[0090] When the cross-section geometric feature variation value is less than the preset geometric stability determination threshold, it is determined that the geometric state of the tunnel cross-section under test is in a stable state.

[0091] When the cross-section geometric feature variation value exceeds the preset geometric stability determination threshold, it is determined that the cross-section geometric state is in an unstable state.

[0092] It should be noted that the geometric stability determination threshold is a numerical limit for determining whether the cross-section geometric state is stable, which is defined as the maximum allowable variation range of the cross-section geometric feature between consecutive construction cycles. The specific setting method is based on the statistical analysis of the variation range of the cross-section geometric feature in the historical construction cycle data under the condition that the cross-section geometric state has been determined to be stable. The sum of the mean and standard deviation is taken as the geometric stability determination threshold by calculating the distribution range, mean and standard deviation of the cross-section geometric feature variation value in multiple construction cycles.

[0093] Through the quantitative fusion calculation of the crack connectivity distribution and the cross-section convergence, the objective expression of the cross-section geometric feature is realized, and the one-sidedness caused by the single geometric parameter determination is avoided, which provides a unified and quantifiable geometric criterion for subsequent monitoring data screening and prediction benchmark updating.

[0094] In step S4, when it is determined that the geometric state of the tunnel cross-section under test is in a stable state, the historical prediction benchmark updating process is started. Specifically, when the cross-section geometric feature value meets the corresponding geometric stability determination condition, it is considered that the structural response detection result collected in the current construction cycle is representative of the project and can be used as an effective sample for prediction benchmark updating. At this time, the historical database is accessed to retrieve the historical prediction benchmark hierarchy.

[0095] It should be noted that the historical database is a structured data storage unit for storing and managing multi-source monitoring historical data formed in the tunnel engineering construction process and its corresponding prediction benchmark information. The tunnel section is taken as the basic index object in the historical database, and the historical prediction benchmark hierarchy is stored in stages according to the construction stage and the disturbance intensity hierarchy, which is used to distinguish the structural response characteristics under different construction conditions.

[0096] The historical prediction benchmark hierarchy refers to the structural response benchmark classified and stored according to the construction stage and the disturbance intensity hierarchy based on multi-source sensing data formed in the tunnel engineering construction process.

[0097] The historical prediction benchmark hierarchy is classified and stored in the historical database according to the construction stage and the disturbance intensity hierarchy in advance, and each historical prediction benchmark hierarchy corresponds to a group of historical detection benchmark records formed under the same construction stage and the same disturbance intensity hierarchy.

[0098] The historical detection benchmark record is derived from the construction cycle in which the section geometry state has been determined to be in a stable state, and is used to describe the normal variation interval of the surrounding rock displacement data and the supporting structure strain data under the corresponding construction condition.

[0099] The historical detection benchmark record includes the upper and lower boundaries of the benchmark interval of the surrounding rock displacement characteristics and the supporting structure strain characteristics, which serves as the basis for subsequent detection result screening and updating.

[0100] Subsequently, based on the disturbance intensity hierarchy, the target disturbance intensity hierarchy consistent with the disturbance intensity hierarchy of the current construction cycle is further screened out from the historical detection benchmark record.

[0101] The target disturbance intensity hierarchy is used to limit the subsequent detection result screening and prediction benchmark updating to be performed only under the same construction condition, thereby avoiding the mixing of structural responses between different disturbance conditions.

[0102] After determining the target disturbance intensity hierarchy, the historical detection benchmark interval corresponding to the target disturbance intensity hierarchy is retrieved, and the associated labeled detection result obtained in the current construction cycle is compared with the historical detection benchmark interval.

[0103] The associated labeled detection result is the surrounding rock displacement data and the supporting structure strain data bound with the disturbance intensity hierarchy label formed in step S2, which is used to determine whether the associated labeled detection result conforms to the normal structural response characteristics under the same construction stage and the same disturbance intensity hierarchy by judging whether the associated labeled detection result falls within the historical detection benchmark interval defined by the corresponding historical detection benchmark record.

[0104] When the correlation mark detection result falls within the historical detection reference interval corresponding to the target disturbance intensity level, it is determined that the correlation mark detection result is an effective update sample, and the historical prediction reference level corresponding to the target disturbance intensity level is updated based on the effective update sample.

[0105] The updating process is achieved by modifying the reference interval boundaries in the historical detection reference record. Specifically, the surrounding rock displacement data and support structure strain data obtained in the current construction cycle are introduced into the historical detection reference record corresponding to the disturbance intensity level as new statistical samples. By updating the statistical distribution of the structural response characteristics under the disturbance intensity level, the upper and lower boundaries of the original reference interval are recalculated numerically, so that the modified reference interval covers the representative range of structural response values in the updated sample set, and the historical prediction reference level can gradually reflect the range of structural response under real engineering conditions as the construction process advances and data accumulates.

[0106] When the correlation mark detection result does not fall within the historical detection reference interval, no update processing is performed on the historical prediction reference level to prevent abnormal data or non-representative data from interfering with the prediction reference.

[0107] Finally, it should be noted that in this document, relational terms such as first and second and the like can only be used to distinguish one entity or action from another entity or action, and do not necessarily require or imply that these entities or actions have any such actual relationship or order.

[0108] Moreover, the terms "include", "have" or any other variant thereof are intended to cover non-exclusive inclusion, so that processes, methods, articles or devices including a series of elements not only include those elements, but also include other elements not explicitly listed or inherent to such processes, methods, articles or devices. Without more limitations, the element defined by the statement "includes a" does not exclude the presence of additional identical elements in the process, method, article or device that includes the element.

[0109] In this document, the singular forms "a", "an" and "the" can also include plural forms unless the context clearly indicates otherwise. It should also be understood that the terms "include", "have" or "comprise" or the like specify the presence of stated features, integers, steps, operations, components, parts or combinations thereof, but do not exclude the presence or addition of one or more other features, integers, steps, operations, components, parts or combinations thereof. The possibility, while the term "and / or" used in this specification includes any and all combinations of the related listed items.

[0110] The various embodiments described in this specification are intended to be combinable unless otherwise indicated herein. Each embodiment described in this specification is intended to be focused on the differences from other embodiments, and each embodiment can be combined with any of the other embodiments as appropriate. Where the same or similar reference numerals are used in different figures, those reference numerals are intended to refer to the same or similar parts throughout this specification.

[0111] Numerous modifications to the embodiments described herein will be apparent to those skilled in the art, and the principles defined herein can be applied to other embodiments without departing from the spirit or scope of the application. Therefore, the application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A collaborative monitoring method for multi-source sensor data of engineering structures based on the Internet of Things, characterized in that: Includes the following steps: Step S1: Conduct construction monitoring on the tunnel section to be tested, retrieve construction status information and extract the current construction stage and construction cycle time, collect blasting vibration velocity data of the tunnel section to be tested within the construction cycle time, and then count the peak vibration velocity. Step S2: Integrate the peak vibration velocity with the current construction stage to initially divide the construction cycle time into disturbance intensity levels, detect the surrounding rock displacement data and support structure strain data of the tunnel section to be tested, and associate and mark the detection results based on the disturbance intensity levels; Step S3: Perform three-dimensional laser scanning on the cross-section of the tunnel to be tested to obtain the crack connectivity distribution and cross-sectional convergence of the cross-section. Evaluate the geometric characteristics of the cross-section by combining the crack connectivity distribution and cross-sectional convergence and determine whether the geometric state of the cross-section is in a stable state. In step S3, after the corresponding construction cycle is completed, the three-dimensional laser scanning device deployed in the tunnel is controlled to scan the tunnel section to be tested according to the preset scanning resolution and scanning angle to obtain the three-dimensional spatial point cloud data of the inner wall surface of the section. Three-dimensional spatial point cloud data consists of several discrete points with spatial coordinates, and each spatial point contains its three-dimensional coordinates in the tunnel cross-section coordinate system. Local plane fitting is performed on the three-dimensional spatial point cloud data to calculate the angle between the normal vectors of adjacent spatial points and the distance between points. When the angle between the normal vectors of adjacent spatial points is greater than the preset normal vector mutation threshold and the distance between points exceeds the preset gap judgment threshold, the crack feature points are judged to exist in the region. Spatial clustering is performed on the crack feature points to obtain several crack clusters. The interconnection of crack clusters is determined based on the minimum spatial distance and the consistency of the extension direction between crack clusters, thus forming a crack interconnection distribution. In step S3, multiple characteristic azimuth angles uniformly distributed along the circumference are selected in the tunnel cross-section coordinate system; In each characteristic azimuth direction, the radial distance from the current scan section profile point to the geometric center of the section is determined, and the difference between this distance and the radial distance of the corresponding azimuth angle of the same section in the previous construction cycle or initial reference state is calculated to obtain the local convergence amount at each azimuth angle. Take the absolute value of the local convergence for all azimuth angles and calculate its average value as the cross-sectional convergence. The crack connectivity distribution and cross-sectional convergence are normalized. After normalization, the crack connectivity distribution and cross-sectional convergence are weighted and summed according to the preset fusion weights to obtain the cross-sectional geometric features. The change value of cross-sectional geometric features is obtained by calculating the magnitude of change in cross-sectional geometric features in adjacent construction cycles; When the change value of the cross-sectional geometric characteristics is less than the preset geometric stability judgment threshold, the geometric state of the tunnel cross-section under test is determined to be in a stable state. When the change value of the cross-sectional geometric characteristics exceeds the preset geometric stability judgment threshold, the cross-sectional geometric state is determined to be unstable. Step S4: When the cross-sectional geometry is in a stable state, access the historical database to retrieve the historical prediction baseline level and filter the target disturbance intensity level. Update the historical prediction baseline level using the correlation marker detection results of the target disturbance intensity level.

2. The method for collaborative monitoring of multi-source sensor data of engineering structures based on the Internet of Things according to claim 1, characterized in that: In step S1, the construction status information, including the current construction stage and the corresponding construction cycle time, is retrieved through the tunnel construction management unit. When the current construction phase is in the blasting operation phase, the blasting vibration velocity is collected by the blasting vibration monitoring device deployed on the cross section of the tunnel to be tested during the construction cycle time, and the blasting vibration velocity is used as the blasting vibration speed data.

3. The method for collaborative monitoring of multi-source sensor data of engineering structures based on the Internet of Things according to claim 2, characterized in that: In step S1, if the blasting vibration velocity data is greater than the preset background vibration threshold, the blasting vibration velocity data is marked; otherwise, the blasting vibration velocity data is not marked. Arrange the marked blasting vibration velocity data in chronological order, and combine adjacent marked blasting vibration velocity data into blasting event segments; Within each blasting event segment, the maximum value of the marked blasting vibration velocity data is taken as the peak vibration velocity of the blasting event segment.

4. The method for collaborative monitoring of multi-source sensor data of engineering structures based on the Internet of Things according to claim 1, characterized in that: In step S2, the vibration velocity reference value of the current construction stage is retrieved from the historical database, and the ratio of the peak vibration velocity to the reference value is used as the peak vibration velocity coefficient. The peak vibration velocity is sorted in chronological order. The change in peak vibration velocity is obtained by subtracting the coefficients of adjacent peak vibration velocity. The change in peak vibration velocity is then sorted in descending order of numerical value. Based on the preset number of layers, the peak vibration velocity coefficients are selected sequentially to obtain the layered segmented peak vibration velocity. The construction cycle time is divided into disturbance intensity levels based on the acquisition time corresponding to the peak vibration velocity of each level.

5. The method for collaborative monitoring of multi-source sensor data of engineering structures based on the Internet of Things according to claim 4, characterized in that: In step S2, the surrounding rock displacement measurement value of the tunnel section to be tested is obtained through the surrounding rock displacement monitoring device in the time interval corresponding to each level of disturbance intensity. The mean value of the surrounding rock displacement measured before the time interval corresponding to the disturbance intensity level is used as the displacement reference value; the mean value of the surrounding rock displacement measured after the time interval corresponding to the disturbance intensity level is used as the displacement response value. The surrounding rock displacement data is calculated based on the displacement response value and the displacement reference value; The strain measurement value of the support structure of the tunnel section under test is obtained by using a support structure strain monitoring device. The strain data of the support structure are calculated based on the strain measurement values ​​of the support structure. The surrounding rock displacement data and support structure strain data corresponding to each disturbance intensity level are used as the detection results, and the detection results are associated with the disturbance intensity level.

6. The method for collaborative monitoring of multi-source sensor data of engineering structures based on the Internet of Things according to claim 1, characterized in that: In step S4, when it is determined that the geometric state of the tunnel section to be tested is in a stable state, the historical database is accessed to retrieve the historical prediction baseline level. Historical prediction benchmark hierarchy refers to the structural response benchmark formed by multi-source sensor data during tunnel construction and stored according to construction stage and disturbance intensity hierarchy. Each historical prediction benchmark level corresponds to a set of historical detection benchmark records formed under the same construction stage and the same disturbance intensity level. Historical monitoring benchmark records include the upper and lower boundaries of the benchmark interval for surrounding rock displacement characteristics and support structure strain characteristics; Further screening was conducted in historical testing benchmark records to identify target disturbance intensity levels that are consistent with the current construction cycle disturbance intensity level.

7. The method for collaborative monitoring of multi-source sensor data of engineering structures based on the Internet of Things according to claim 6, characterized in that: In step S4, the associated marker detection results obtained in the current construction cycle are compared with the historical detection benchmark interval defined by the corresponding historical detection benchmark record: When the detection result of the associated marker falls within the historical detection benchmark interval corresponding to the target perturbation intensity level, the detection result of the associated marker is determined to be a valid update sample, and the historical prediction benchmark level corresponding to the target perturbation intensity level is updated based on the valid update sample. The boundary of the benchmark interval in the historical test benchmark record is corrected during the update process; If the associated marker detection result does not fall within the historical detection benchmark range, no update processing will be performed on the historical prediction benchmark level.

Images