Slope deformation trend space-time hybrid analysis method based on large-scale gridding displacement monitoring information

Abstract

A side slope deformation trend space-time hybrid analysis method based on large-scale gridding displacement monitoring information relates to the technical field of geotechnical engineering deformation monitoring, and comprises the following steps: arranging detection optical fibers; monitoring data captured by the rotation detection unit and the detection optical fiber are collected, and the gridded monitoring data are arranged in a matrix form; the monitoring data are analyzed, and the state of the slope is recognized; vMD is adopted to decompose the time sequence of slope deformation, the tendency, periodicity and volatility of slope deformation are comprehensively considered, and the deformation rule of the slope along with time is mastered; the method comprises the following steps: constructing a data set of slope deformation influence factors, dividing a training set and a prediction set, substituting the training set into an artificial intelligence model, establishing a model through continuous iterative calculation, and analyzing the slope deformation trend by using the model. Meanwhile, time-space mixed analysis can be synchronously carried out while the cost can be effectively reduced, and the effects of accurate judgment of the overall stability of the slope and deformation prediction are achieved.

Description

Technical Field

[0001] The invention relates to the technical field of geotechnical engineering deformation monitoring, and in particular to a spatiotemporal hybrid analysis method for slope deformation trends based on large-scale gridded displacement monitoring information. Background Art

[0002] Due to the uneven spatial distribution of geotechnical media, significant differences in mechanical parameters and limited exploration technology, it is difficult to fully grasp the stability characteristics of geotechnical engineering in the design stage. Therefore, the specification requires the use of a dynamic design method that monitors the displacement response of the geotechnical body and corrects the design by feedback during the construction process. The feedback analysis of the temporal and spatial characteristics of slope deformation during slope construction is very important.

[0003] Traditional slope monitoring instruments that mainly use point-based monitoring (such as total stations, levels, and inclinometers) are easy to install and simple to operate, but are easily affected by multiple factors such as meteorological factors, ground environment, and the technical level of operators. The data collection and analysis process still mainly relies on manual labor, which makes it difficult to meet the high requirements for timeliness and accuracy of slope monitoring and early warning.

[0004] In recent years, new technologies such as oblique photography, three-dimensional laser scanning, and ground-based radar, which are mainly point cloud monitoring methods, are non-contact active measurement methods that can quickly acquire data over a large area to establish a three-dimensional model, but can only monitor the surface displacement of the slope.

[0005] Quasi-distributed and distributed monitoring methods based on optical fiber technology are efficient measurement methods that use light waves as carriers and optical fibers as media to sense changes in physical parameters of the external environment. In slope monitoring, the frequency drift value generated by light waves with changes in external deformation, temperature, vibration and other factors is used to analyze and evaluate slope status information. With its advantages of anti-interference, high precision, distribution and real-time performance, it is increasingly widely used in slope monitoring.

[0006] Rock slopes have typical structural characteristics. The instability process starts with local block sliding, gradually expands to adjacent parts and produces a chain effect to form overall instability. The spatial and temporal characteristics of deformation are equally important for slope stability prediction.

[0007] Supported by the above monitoring technologies, there are currently two methods for slope deformation trend prediction and analysis: time analysis and space analysis. Time trend analysis methods include neural networks, gray systems, regression analysis, time series and other methods, while space analysis methods are represented by point cloud models generated by oblique photography, three-dimensional laser scanning, ground-based radar, etc. The above methods have the following shortcomings: (1) The time trend prediction model is a time series analysis model based on monitoring data. It does not reflect the physical mechanism of deformation, and the accuracy of long-term trend prediction is unpredictable.

[0008] (2) Temporal trend prediction generally only predicts the displacement trend of a single measuring point. The information obtained is the local displacement information of the slope. There is insufficient understanding of the spatial characteristics of slope deformation, making it difficult to deeply understand the slope deformation mechanism.

[0009] (3) In recent years, the emergence of new non-contact monitoring methods has provided new solutions for spatial analysis. However, the accuracy of oblique photography, three-dimensional laser scanning, etc. is difficult to meet the sub-millimeter accuracy requirements. Ground-based radar is expensive. At the same time, the accuracy of non-contact methods is easily affected by the meteorological environment. Summary of the invention

[0010] The embodiment of the present invention provides a spatiotemporal hybrid analysis method of slope deformation trend based on large-scale grid displacement monitoring information, which is used to solve the above problems.

[0011] The spatiotemporal hybrid analysis method of slope deformation trend based on large-scale grid displacement monitoring information includes the following steps: A plurality of rotation detection units are arranged on the detection optical fiber in a serial manner to form a multi-measurement point array structure; The reference point is set outside the slope excavation area, and the starting section of the strain optical fiber is buried in the anchor pile. The above detection optical fiber is arranged in an S shape on the slope to form a grid-like monitoring network covering the slope, and the monitoring data is obtained from the slope; Collect the monitoring data captured by the rotation detection unit and the detection optical fiber, and arrange the gridded monitoring data in a matrix form, where the rows represent the pile numbers, the columns represent the elevations, and the matrix elements are the horizontal displacement increments of the corresponding positions on the slope surface; Analyze the above monitoring data to identify the state of the slope; According to the actual project, set different levels of early warning values ​​to monitor the status of the slope; VMD is used to decompose the time series of slope deformation, comprehensively consider the trend, periodicity and volatility of slope deformation, and grasp the deformation law of slope over time; Construct a data set of factors affecting slope deformation, divide it into training set and prediction set, bring the training set into the artificial intelligence model, establish the model through continuous iterative calculation, and use the model to analyze the slope deformation trend.

[0012] Furthermore, the rotation detection unit includes two standard rods, which are connected end to end through a rotating shaft, and an angle sensor for detecting angle changes is provided at the rotating shaft.

[0013] Further, the detection optical fiber includes a strain optical fiber a and a strain optical fiber b, and the strain optical fiber a and the strain optical fiber b are arranged on both sides of the standard rod in the rotation direction.

[0014] Furthermore, the strain optical fiber a and the strain optical fiber b are connected to the standard rod by glue, and the position where the glue is set is defined as a fixed point.

[0015] Furthermore, the position of the fixed point is adjustable, and the distance between the strain optical fiber a and the strain optical fiber b on both sides of the rotating shaft can be adjusted by adjusting the position of the fixed point.

[0016] Furthermore, the process of obtaining the horizontal displacement increment of the position includes: Acquire data from the angle sensor to identify the angle between the standard poles; The horizontal displacement increment is calculated using the acquired angle and the position of the fixed point.

[0017] Furthermore, the steps of identifying the state of the slope using the monitoring data include: Randomly extract at least one submatrix from the elements of the matrix; The local deformation increment mean and localized eigenvalues ​​are defined for the submatrix; The random forest algorithm is introduced to find the submatrix with the largest average value of local deformation increment, and then search for the submatrix with larger localized eigenvalues ​​in turn; The line connecting the centers of these sub-matrices is the potential sliding surface of the slope; If a submatrix has both the largest local deformation increment average and the smallest localized eigenvalue, then the submatrix corresponds to the area with the largest deformation and the smallest difference in the slope area, which is the potential sliding body of the slope.

[0018] The beneficial effects of the above technical solution provided by the embodiment of the present invention include at least: 1. Based on the high-precision, large-capacity sampling and good economy of arrayed optical fibers, by arranging grid-based distributed measuring points on the slope surface, it is possible to realize the recognition of spatial deformation information that cannot be achieved by traditional point-based monitoring methods, and solve the traditional problem of difficulty in ensuring coverage monitoring of local sliding blocks on the slope surface.

[0019] 2. Array fiber Bragg grating technology is used to identify the health status of the slope, which can take into account both the overall and local aspects and increase the information dimension. At the same time, the flexibility of grating fiber technology can easily unify information collection elements such as strain, stress, and dislocation, and use a matrix form, which makes it easier to achieve data integration, management, and analysis.

[0020] 3. The proposed hybrid spatiotemporal analysis method for slope deformation can capture the spatial variability of structured slope displacement information, combine the temporal trend and spatial difference of displacement, improve the sensitivity of slope stability degradation information identification, and realize the early identification of slope instability precursor information.

[0021] 4. Conventional distributed fiber optic monitoring technology outputs a huge amount of data, and simple graphical expressions do not directly reflect the state characteristics. Further analysis is needed to reveal the deformation mechanism and judge safety hazards. The random forest ensemble learning algorithm based on artificial intelligence is introduced to identify the band distribution characteristics of the shear outlet deformation on the slope surface, so as to achieve accurate judgment of the overall stability of the slope and deformation prediction.

[0022] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings.

[0023] The technical solution of the present invention is further described in detail below through the drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 A flow chart of spatiotemporal hybrid analysis disclosed in an embodiment of the present invention; Figure 2 A schematic diagram of a slope deformation monitoring network disclosed in an embodiment of the present invention; Figure 3 It is a schematic diagram of the installation structure of the detection optical fiber and the rotation detection unit disclosed in the embodiment of the present invention; Figure 4 for Figure 3 Schematic diagram of the side view structure in.

[0025] Reference numerals: 1. Detection optical fiber; 11. Strain optical fiber a; 12. Strain optical fiber b; 2. Rotation detection unit; 21. Standard rod; 22. Rotating shaft; 23. Angle sensor; 3. Slope. DETAILED DESCRIPTION

[0026] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0027] like Figure 1As shown in FIG. 1 , a temporal-spatial hybrid analysis method for slope deformation trend based on large-scale grid displacement monitoring information is provided, including a gridded high-coverage monitoring network and a temporal-spatial hybrid analysis method for slope deformation trend. The specific technical scheme is as follows: like Figure 2 As shown, the gridded high-coverage monitoring network includes a detection optical fiber and a plurality of rotation detection units arranged in series on the detection optical fiber, and the two are combined to form a multi-measurement point array structure.

[0028] like Figure 3 and 4 As shown, the detection optical fiber is composed of strain optical fiber a and strain optical fiber b, and the rotation detection unit includes two standard rods, which are connected end to end through a rotating shaft. An angle sensor for detecting angle changes is provided at the rotating shaft, and the strain optical fiber a and strain optical fiber b are distributed in mirror images on both sides of the rotation direction of the standard rod.

[0029] The strain optical fiber a and the strain optical fiber b are connected to the standard rod by glue, and the position where the glue is set is defined as a fixed point. The distance between the strain optical fiber a and the strain optical fiber b on both sides of the shaft can be adjusted by adjusting the position of the fixed point.

[0030] (1) The rotation detection unit converts the grating strain into the relative rotation angle of the standard rods at both ends of the rotating shaft, and uses the rotation angle of the standard rod to convert the relative displacement at both ends of the standard rod (the horizontal displacement increment can be obtained using trigonometric functions); the standard rod connected to the rotating shaft is a rotation detection unit, and multiple measurement units connected by a detection optical fiber form a multi-measurement point array structure.

[0031] (2) By rotating the strained optical fibers a and b arranged on opposite sides of the shaft, the strained optical fibers are converted into a rotation angle by utilizing the characteristic that when one side is pulled, the other side relaxes naturally. The relaxed optical fiber is only affected by temperature. The difference between the two is used to eliminate the influence of temperature, thus achieving temperature self-compensation.

[0032] (3) By adjusting the distance between the optical fiber fixed points at both ends of the shaft, long-distance, large-range quasi-distributed monitoring can be achieved.

[0033] (4) The benchmark point is set outside the slope excavation area, and the starting section of the strain optical fiber is buried in the anchor pile to serve as a comparison of deformation and temperature, while reducing interference with the construction process.

[0034] (5) The detection optical fiber is laid out in an “S” shape on the slope surface to form a high-precision grid-covered monitoring network. The accuracy of each monitoring point can reach sub-millimeter level.

[0035] (6) The gridded monitoring data are arranged in a matrix form, where the rows represent the pile numbers, the columns represent the elevations, and the matrix elements are the horizontal displacement increments at the corresponding positions on the slope surface. As shown in matrix A, the elements in the matrix are Representative Line The horizontal displacement increment of the slope surface at the sensor position is calculated, so the total number of sensors is indivual Matrix A:

[0036] (7) Take at least one submatrix of the elements of matrix A, such as: As the starting point of the k+1 order submatrix, define , Among them, a represents the horizontal displacement increment of the slope surface, which constitutes the elements of the matrix A', and the subscript Indicates the position of the sensor corresponding to the displacement increment, and k indicates The number of elements that expand outward from the starting point; represents the average value of the local deformation increment of the slope within the coverage area of ​​submatrix A', represents the localized eigenvalue of slope deformation within the coverage area of ​​submatrix A'. If the value is larger, it means that the slope deformation in the corresponding area is larger; if It is larger, indicating that the slope deformation in the corresponding area is different.

[0037] Submatrix A':

[0038] As for The 2nd-order submatrix starting from ,k=1,j=1,s=1,

[0039] For The 3rd-order submatrix starting from ,k=1,j=2,s=2

[0040]

[0041] (8) Introduce the random forest algorithm to analyze and process the data and find out The largest matrix, search around the matrix in turn The line connecting the centers of these matrices is the potential sliding surface of the slope. Maximum and The smallest value indicates that the matrix includes the area with the largest deformation and the smallest difference in the corresponding slope area, which is the potential sliding body of the slope. According to the actual project, different levels of early warning values ​​are set to judge the stability of the slope in advance and take corresponding disposal measures.

[0042] Find out the above or The maximum or minimum or larger or smaller matrix is ​​obtained by setting a threshold, and the maximum or minimum or larger or smaller matrix is ​​determined when the set threshold is reached.

[0043] (9) VMD (variational mode decomposition) is used to decompose the time series of slope deformation, comprehensively consider the trend, periodicity and volatility of slope deformation, and understand the deformation law of slope over time.

[0044] (10) Construct a data set of factors affecting slope deformation and divide it into a training set and a prediction set. Bring the training set into the artificial intelligence model, establish the model through continuous iterative calculation, verify the reliability of the model, and perform prediction analysis on the prediction set. MSE (mean square error), RMSE (root mean square error), MAE (mean absolute error) and determination coefficient R2 are used to evaluate the prediction accuracy of the model.

[0045] The proposed method of spatiotemporal hybrid analysis of slope deformation trends based on large-scale grid displacement monitoring information is based on the high-precision, large-capacity sampling and good economy of array optical fibers. By arranging grid-based distributed measuring points on the slope surface, it can realize the recognition of spatial deformation information that cannot be achieved by traditional point-based monitoring methods, and solve the traditional problem that it is difficult to ensure coverage monitoring of local sliding blocks on the slope surface. Array fiber grating technology is used for slope health status identification, which can achieve both overall and local considerations and increase the information dimension; at the same time, the flexibility of grating optical fiber technology can easily realize the unification of information collection elements such as strain, stress, and dislocation, and the use of a matrix form makes it easier to achieve data integration, management and analysis. The proposed spatiotemporal hybrid analysis method of slope deformation can capture the spatial variability of structured slope displacement information, combine the temporal trend and spatial difference of displacement, improve the sensitivity of slope stability degradation information identification, and realize early identification of slope instability precursor information. Conventional distributed fiber optic monitoring technology outputs a huge amount of data, and simple graphical expressions do not directly reflect the state characteristics. Further analysis is needed to reveal the deformation mechanism and identify safety hazards. This paper introduces a random forest ensemble learning algorithm based on artificial intelligence to identify the band distribution characteristics of the shear outlet deformation on the slope surface, and realize accurate judgment of the overall stability of the slope and deformation prediction.

[0046] It should be understood that the specific order or hierarchy of steps in the disclosed process is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process can be rearranged without departing from the scope of protection of the present disclosure. The attached method claims present the elements of the various steps in an exemplary order and are not intended to be limited to the specific order or hierarchy described.

[0047] In the above detailed description, various features are grouped together in a single embodiment to simplify the disclosure. This method of disclosure should not be interpreted as reflecting an intention that the embodiments of the claimed subject matter require more features than are clearly stated in each claim. On the contrary, as reflected in the appended claims, the invention is in a state of having less than all the features of the disclosed individual embodiments. Therefore, the appended claims are hereby expressly incorporated into the detailed description, with each claim standing on its own as a separate preferred embodiment of the invention.

[0048] Those skilled in the art will also appreciate that the various illustrative logic blocks, modules, circuits, and algorithmic steps described in conjunction with the embodiments herein can all be implemented as electronic hardware, computer software, or a combination thereof. In order to clearly illustrate the interchangeability between hardware and software, various illustrative components, blocks, modules, circuits, and steps are generally described above around their functions. Whether such functions are implemented as hardware or software depends on specific applications and the design constraints imposed on the entire system. A skilled person can implement the described functions in an alternative manner for each specific application, but such implementation decisions should not be interpreted as departing from the scope of protection of the present disclosure.

[0049] The steps of the method or algorithm described in conjunction with the embodiments herein may be directly embodied as hardware, a software module executed by a processor, or a combination thereof. The software module may be located in a RAM memory, a flash memory, a ROM memory, an EPROM memory, an EEPROM memory, a register, a hard disk, a mobile disk, a CD-ROM, or any other form of storage medium well known in the art. An exemplary storage medium is connected to the processor so that the processor can read information from the storage medium and can write information to the storage medium. Of course, the storage medium may also be an integral part of the processor. The processor and the storage medium may be located in an ASIC. The ASIC may be located in a user terminal. Of course, the processor and the storage medium may also be present in a user terminal as discrete components.

[0050] For software implementation, the techniques described in this application can be implemented with modules (e.g., procedures, functions, etc.) that perform the functions described in this application. These software codes can be stored in a memory unit and executed by a processor. The memory unit can be implemented within the processor or outside the processor. In the latter case, it is coupled to the processor in a communication manner via various means, which are well known in the art.

[0051] The above description includes examples of one or more embodiments. Of course, it is impossible to describe all possible combinations of components or methods for the purpose of describing the above embodiments, but it should be recognized by those skilled in the art that the various embodiments may be further combined and arranged. Therefore, the embodiments described herein are intended to cover all such changes, modifications and variations that fall within the scope of protection of the appended claims. In addition, with respect to the term "comprising" used in the specification or claims, the word is covered in a manner similar to the term "including", just as "including," is explained as a transitional word in the claims. In addition, any term "or" used in the specification of the claims is intended to mean "non-exclusive or".

Claims

1. A spatiotemporal hybrid analysis method for slope deformation trends based on large-scale grid displacement monitoring information, characterized in that: The steps include: A plurality of rotation detection units are arranged on the detection optical fiber in a serial manner to form a multi-measurement point array structure; The reference point is set outside the slope excavation area, and the starting section of the strain optical fiber is buried in the anchor pile. The above detection optical fiber is arranged in an S shape on the slope to form a grid-like monitoring network covering the slope, and the monitoring data is obtained from the slope; Collect the monitoring data captured by the rotation detection unit and the detection optical fiber, and arrange the gridded monitoring data in a matrix form, where the rows represent the pile numbers, the columns represent the elevations, and the matrix elements are the horizontal displacement increments of the corresponding positions on the slope surface; Analyze the above monitoring data to identify the state of the slope; According to the actual project, set different levels of early warning values ​​to monitor the status of the slope; VMD is used to decompose the time series of slope deformation, comprehensively consider the trend, periodicity and volatility of slope deformation, and grasp the deformation law of slope over time; Construct a data set of factors affecting slope deformation, divide it into training set and prediction set, bring the training set into the artificial intelligence model, establish the model through continuous iterative calculation, and use the model to analyze the slope deformation trend.

2. The method according to claim 1, characterized in that The rotation detection unit comprises two standard rods, which are connected end to end through a rotating shaft, and an angle sensor for detecting angle changes is arranged at the rotating shaft.

3. The method according to claim 2, characterized in that The detection optical fiber includes a strain optical fiber a and a strain optical fiber b, and the strain optical fiber a and the strain optical fiber b are arranged on both sides of the standard rod in the rotation direction.

4. The method according to claim 3, characterized in that The strain optical fiber a and the strain optical fiber b are connected to the standard rod through glue, and the position where the glue is set is defined as a fixed point.

5. The method according to claim 4, characterized in that The position of the fixed point is adjustable, and the distance between the strain optical fiber a and the strain optical fiber b on both sides of the rotating shaft can be adjusted by adjusting the position of the fixed point.

6. The method according to claim 5, characterized in that The process of obtaining the horizontal displacement increment of the position includes: Collect data from the angle sensor to obtain the angle between the standard poles; The horizontal displacement increment is calculated using the acquired angle and the position of the fixed point.

7. The method according to claim 1, characterized in that The steps to identify the slope status using monitoring data include: Randomly extract at least one submatrix from the elements of the matrix; The local deformation increment mean and localized eigenvalues ​​are defined for the submatrix; The random forest algorithm is introduced to find the submatrix with the largest average value of local deformation increment, and then search for the submatrix with larger localized eigenvalues ​​in turn; The line connecting the centers of these sub-matrices is the potential sliding surface of the slope; If a submatrix has both the largest local deformation increment average and the smallest localized eigenvalue, then the submatrix corresponds to the area with the largest deformation and the smallest difference in the slope area, which is the potential sliding body of the slope.

Images