Slope stability early risk judgment method, device and equipment based on intelligent measurement pile

Abstract

The embodiment of the invention discloses a slope stability early-stage risk judgment method, device and equipment based on an intelligent measurement pile, and relates to the technical field of slope engineering safety prevention and control. The scheme can comprise the following steps: acquiring sensing data of a plurality of intelligent measuring piles in a monitoring area, wherein the sensing data at least comprises displacement data and inclination data; wherein the intelligent measuring pile is arranged on the surface of a slope; based on a preset partitioning algorithm, the monitoring area is divided into a plurality of intelligent measurement pile unit spaces according to the positions of the multiple intelligent measurement piles, and each intelligent measurement pile unit space comprises one intelligent measurement pile; processing the displacement data and the inclination data of each intelligent measuring pile to obtain a deformation parameter vector of each intelligent measuring pile; the deformation parameter vector is compared with a first early warning threshold value, and an over-limit intelligent measurement pile is obtained; a preset clustering algorithm is utilized to cluster the over-limit intelligent testing piles, a plurality of risk partitions in the monitoring area are obtained, and each risk partition comprises at least one intelligent testing pile unit space; according to the deformation parameter vector of each risk subarea intelligent measurement pile, determining a risk subarea comprehensive deformation characterization parameter value, and comparing the risk subarea comprehensive deformation characterization parameter value with a second early warning threshold value to obtain a risk grade of each risk subarea; according to the comprehensive deformation characterization parameter value of the risk partition and the area factor of the comprehensive deformation characterization parameter value, determining the comprehensive deformation characterization parameter value of the overall risk of the slope, and comparing the comprehensive deformation characterization parameter value with a third early warning threshold value to obtain the overall risk level of the slope, so that the early risk state of the slope stability can be comprehensively and accurately judged under the condition of realizing large-range coverage monitoring.

Description

Technical Field

[0001] The present application relates to the technical field of slope engineering monitoring, and in particular to a method, device and equipment for early risk determination of slope stability based on intelligent testing piles. Background Art

[0002] Slope instability is one of the geological disasters that is difficult to avoid and has huge potential hazards during engineering construction and operation. It may not only cause damage to engineering structures, but may even threaten the safety of life and property. Under the condition that slope risk assessment cannot be comprehensive and accurate, monitoring work is very necessary and important, and surface deformation monitoring is one of the main means.

[0003] In recent years, with the rapid development of technologies such as the Internet of Things, big data, and artificial intelligence, automated slope monitoring technology has developed rapidly. However, due to factors such as expensive equipment costs, low integration, insufficient measurement accuracy, and severe weather interference, the existing monitoring technology is difficult to achieve the purpose of universal real-time monitoring covering a large area. Especially in the context of limited construction investment, only a small number of discrete and limited monitoring points can be deployed, resulting in a greatly increased risk of missed detection and misjudgment, which cannot meet the current needs of slope engineering risk safety prevention and control management to effectively reduce missed alarms and false alarms, and to identify and intervene early.

[0004] Therefore, how to achieve universal monitoring goals with large-scale coverage, multi-parameter synchronization, high precision and low cost in the early stage of slope safety risks, and then determine the early risk status of slopes more comprehensively, accurately and effectively, has become a key issue and universal demand that needs to be urgently solved in the industry. Summary of the invention

[0005] The embodiments of this specification provide a method for determining the early risk of slope stability based on intelligent testing piles to solve the problem in the prior art that it is impossible to comprehensively, accurately and effectively determine and analyze the early risk status of the slope.

[0006] To solve the above technical problems, the embodiments of this specification are implemented as follows:

[0007] In a first aspect, an embodiment of this specification provides a method for early risk determination of slope stability based on intelligent pile detection, comprising:

[0008] Acquire sensing data of a plurality of intelligent measuring piles in a monitoring area, wherein the sensing data at least includes displacement data and tilt data; wherein the intelligent measuring piles are arranged on a slope surface;

[0009] Based on a preset partitioning algorithm, according to the positions of the plurality of smart piles, the monitoring area is divided into a plurality of smart pile unit spaces, each of which contains one smart pile;

[0010] Processing the displacement data and the tilt data of each of the smart testing piles to obtain a deformation parameter vector of each of the smart testing piles;

[0011] Compare the deformation parameter vector with the first warning threshold to obtain an over-limit intelligent pile;

[0012] Using a preset clustering algorithm, clustering the over-limit intelligent testing piles to obtain clusters of the over-limit intelligent testing piles;

[0013] According to the clustering clusters, a plurality of risk zones in the monitoring area are obtained, and each of the risk zones includes at least one of the intelligent detection pile unit spaces.

[0014] In a second aspect, an early risk determination device for slope stability based on intelligent pile detection is provided in an embodiment of the present specification, comprising:

[0015] An acquisition module is used to acquire the sensing data of multiple intelligent detection piles in the monitoring area, wherein the sensing data at least includes displacement data and tilt data; wherein the intelligent detection piles are arranged on the surface of the slope;

[0016] A partitioning module, configured to divide the monitoring area into a plurality of intelligent testing pile unit spaces based on a preset partitioning algorithm and according to the positions of the plurality of intelligent testing piles, each intelligent testing pile unit space including one intelligent testing pile;

[0017] A processing module, used for processing the displacement data and the tilt data of each of the smart testing piles to obtain a deformation parameter vector of each of the smart testing piles;

[0018] A comparison module, used for comparing the deformation parameter vector with a first warning threshold value to obtain an over-limit intelligent detection pile;

[0019] A clustering module, used to cluster the over-limit intelligent testing piles using a preset clustering algorithm to obtain clusters of the over-limit intelligent testing piles;

[0020] A determination module is used to obtain a plurality of risk zones within the monitoring area according to the clustering clusters, each of which includes at least one of the intelligent detection pile unit spaces.

[0021] In the third aspect, an embodiment of the present specification provides a device for early risk assessment of slope stability based on intelligent testing piles, comprising a memory, a processor and an embedded computer program thereof, wherein the processor executes the computer program to implement the method for early risk assessment of slope stability based on intelligent testing piles in scheme one.

[0022] An embodiment of the present specification can achieve the following beneficial effects: by dividing the slope in the monitoring area into a number of smart testing pile unit spaces, determining the over-limit state of each smart testing pile according to the perception data of the smart testing piles in each smart testing pile unit space, clustering the over-limit smart testing piles, and obtaining a risk partition containing at least one of the smart testing pile unit spaces, the early risk state of the slope can be determined comprehensively, accurately and effectively. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0024] Figure 1 A schematic flow chart of a method for early risk determination of slope stability based on intelligent pile testing provided in an embodiment of this specification;

[0025] Figure 2 A schematic diagram of the structure of the intelligent testing pile provided in the embodiment of this specification;

[0026] Figure 3 A schematic diagram of the network topology of the intelligent pile detection system provided in the embodiments of this specification;

[0027] Figure 4 A schematic diagram of the structure of a device for early risk determination of slope stability based on intelligent testing piles provided in an embodiment of this specification;

[0028] Figure 5 A schematic diagram of the structure of an early risk assessment device for slope stability based on intelligent testing piles provided in an embodiment of this specification. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solutions and advantages of one or more embodiments of this specification clearer, the technical solutions of one or more embodiments of this specification will be clearly and completely described below in combination with the specific embodiments of this specification and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of one or more embodiments of this specification.

[0030] Slope instability is one of the geological disasters that is difficult to avoid and has great potential harm during engineering construction and operation. It may not only cause damage to engineering structures, but may even threaten the safety of life and property. Therefore, how to comprehensively cover, simply, quickly, and accurately judge slope stability risks in a timely manner has become a key issue to ensure engineering safety and disaster prevention and mitigation.

[0031] In recent years, with the rapid development of technologies such as the Internet of Things, big data, and artificial intelligence, new development opportunities have been provided for the realization of automated and intelligent monitoring of slopes. However, the current automated monitoring technology for slope engineering still has some shortcomings and challenges:

[0032] 1. Cost investment of monitoring equipment: The cost of system monitoring equipment is generally high, and most of them are complicated to install and maintain. Currently, they are mainly used in a small number of major slope scenarios such as those with greater safety hazards, complex geological conditions, and important locations, and cannot be promoted for large-scale use.

[0033] 2. Limitations of monitoring technology: Although some automated monitoring technologies have emerged, such as SAR (synthetic aperture radar), GNSS (global navigation satellite system), distributed fiber optic sensing, and visual image measurement, they still face certain challenges in practical applications. For example, SAR and visual image technology are greatly restricted by meteorological conditions, and SAR technology has low measurement accuracy and poor real-time performance; GNSS technology is easily affected by satellite signal coverage and adverse meteorological environment, has low measurement accuracy, and has high installation and maintenance costs; although distributed fiber optic sensing technology has low construction costs and can achieve long-distance continuous measurement, it is mostly based on qualitative measurement, and has high maintenance costs and is prone to false alarms; in addition, some advanced equipment such as deep displacement monitoring array displacement meters are expensive and difficult to apply on a large scale.

[0034] 3. Challenges of automated slope monitoring systems: Against the backdrop of decreasing investment in engineering construction, even though major slopes generally use relatively expensive point-type sensing equipment such as GNSS surface displacement monitoring, visual image surface displacement monitoring, and deep horizontal displacement monitoring, they can only deploy a small number of discrete and limited monitoring points, and are unable to achieve overall coverage of large-scale slopes. This greatly increases the risk of missed detections and misjudgments, and is unable to meet the increasingly stringent safety and control needs of current slope projects.

[0035] Among them, GNSS surface displacement monitoring is one of the main means of slope risk prevention and control monitoring. Due to the high price of the equipment, it is impossible to deploy a large number of coverage monitoring, and the limitation of the working principle leads to low measurement accuracy. Therefore, it is mostly used in the overall obvious development of major slopes. Prevention and control monitoring scenes, but it cannot capture the typical small deformation signs that often appear in the early stage of slope landslide development. That is, once GNSS surface displacement monitoring finds abnormal data, the slope deformation development is generally obvious, especially for more sudden slope landslides, collapses and other disasters. At the same time, considering the complexity and differentiation of the geological environment of the slope, the existing risk assessment system is still imperfect, and the current single-point monitoring methods such as surface displacement GNSS receivers and deep displacement array displacement meters cannot achieve accurate deployment and monitoring, resulting in a high probability of missed detection and misjudgment. In recent years, the industry has also paid a high price and learned profound lessons for this.

[0036] In view of this, for the application scenario of early risk prevention and control of slope engineering, this application combines the research and analysis of typical deformation characteristics of early development of slope landslides, relies on the latest MEMS intelligent sensing technology, and innovatively develops and proposes a portable wireless intelligent detection pile and a universal quantitative monitoring and identification method for early risk screening of slopes based on intelligent detection pile point group technology. It aims to eliminate the technical blind spots of existing application scenarios, effectively solve the pain points of early risk coverage screening and monitoring of large-scale scattered slope landslides, collapses and other disasters, quickly identify the overall and local risk dynamics of the slope in the early stage, identify and calculate the zoning and grading of risk areas, and has significant advantages such as large coverage, high equipment precision and low cost, simple deployment, easy installation and maintenance, and simultaneous measurement of multiple parameters. It provides a reliable basis for further precise special monitoring and implementation, and also gains more time for slope risk prevention and control work.

[0037] The technical solutions provided by the embodiments of this specification are described in detail below in conjunction with the accompanying drawings.

[0038] In order to solve the defects in the prior art, this solution provides the following embodiments:

[0039] A method for early risk determination of slope stability based on intelligent testing piles provided in the embodiment of the specification is specifically described in conjunction with the accompanying drawings.

[0040] Figure 1 A flow chart of a method for early risk assessment of slope stability based on intelligent testing piles provided in an embodiment of this specification.

[0041] like Figure 1 As shown, the process may include the following steps:

[0042] Step S110: Acquire sensing data of a plurality of intelligent measuring piles in a monitoring area, wherein the sensing data at least includes displacement data and tilt data; wherein the intelligent measuring piles are arranged on a slope surface.

[0043] In the embodiments of this specification, the smart detection pile can be a wireless smart detection pile. The wireless smart detection pile is a low-cost and high-precision intelligent sensing device customized for monitoring creep deformation in the early stage of slope landslide development. It has the function of synchronously measuring multi-parameter information such as horizontal displacement, inclination, vibration acceleration, deformation azimuth, temperature, etc. of rock and soil. It is suitable for overall coverage and universal monitoring scenarios for large-area dispersed slope risk prevention and control, and can effectively monitor and screen the early risk status of slopes.

[0044] Multiple wireless intelligent measuring piles can be deployed in a shallow covering manner on the slope to form a slope full-area point group control network. Through each intelligent measuring pile measuring point, perception data can be obtained in real time. In the early stage of slope stability risk, the micro-deformation state and characteristics of the overall surface rock and soil of the slope can be effectively grasped at the first time, and cluster verification analysis, as well as automatic zoning and grading and timely alarm can be performed, thereby buying more time for slope risk prevention and control work.

[0045] Figure 2 This is a schematic diagram of the structure of the intelligent testing pile provided in the embodiments of this specification.

[0046] like Figure 2 As shown, the main components of the wireless intelligent testing pile include an antenna, an antenna fixing clip, a device housing, a measuring module, a communication module, and a power supply module, wherein the measuring module, the communication module, and the power supply module are located in a cavity wrapped by the device housing; the device housing is the main structure of the intelligent testing pile, and is sealed and assembled with stainless steel, aluminum alloy, or high-strength hard plastic material for protection; the measuring module mainly includes a series of acceleration sensors and magnetic declination sensors, which are used to sense, collect, and calculate the main parameter data of the intelligent testing pile such as horizontal displacement, vibration acceleration, inclination, and deformation azimuth, and are the core functional unit of the device; the communication module is mechanically fixed in the device housing, and adopts low-power wireless communication technologies such as ZigBee and LoRa, and is used for data collection and real-time interactive communication functions; the power supply module is mechanically fixed in the device housing, and is composed of a battery pack and a power processing and conversion unit, so as to realize the self-powered, low-power, and long-term stable operation of the intelligent testing pile.

[0047] Figure 3 This is a schematic diagram of the network topology of the intelligent testing pile system provided in the embodiments of this specification.

[0048] By deploying multiple wireless intelligent measuring pile measuring points around the slope and building an intelligent measuring pile point group control network, the dynamic information such as shallow small horizontal displacement and inclination, vibration acceleration and deformation azimuth of the early stage of slope landslide can be accurately sensed in real time over a large range, and sent wirelessly to the Internet via the communication base station, and the background system performs real-time analysis and processing.

[0049] The working principle of the intelligent measuring pile is based on the research and practical development and design of typical deformation characteristics of rock and soil in the early stage of slope landslide development. At this stage, the displacement of slope rock and soil usually presents a typical distribution characteristic of large surface displacement and decreasing as it goes deeper, and the initial displacement is generally small. By using a large number of intelligent measuring pile measuring points arranged in a slope covering manner to accurately sense and measure the early creep deformation of shallow rock and soil, it is possible to effectively and systematically screen and monitor the early risk status of large-area slope landslides.

[0050] It should be noted that although the intelligent measuring pile measures a relative displacement value when measuring horizontal displacement parameters, it can meet the engineering application requirements for accurately capturing small changes in the slope deformation displacement field and its morphological characteristics, as well as the development trend of the typical differentiated horizontal displacement deformation evolution background conditions in the early deformation stage of the slope landslide, that is, focusing on the large-scale capture of the early risk abnormal state of the slope for screening and monitoring purposes. Through the verification, validation and statistical modeling cluster analysis calculation of the measuring point data of the slope intelligent measuring pile system, reliable support can be provided for the early risk zoning and grading of the slope.

[0051] The on-site civil construction of the intelligent pile system includes two parts: wireless intelligent piles and communication base stations. Among them, the intelligent piles are arranged in a point group coverage pattern at intervals on the slope platform or the shallow surface of the slope to form a full-area monitoring network for surface deformation, with the spacing between measuring points ranging from 20 to 40 meters.

[0052] The layout of intelligent testing pile measuring points should be combined with the scale of the slope, geological conditions, historical diseases and existing technical data, and fully combined with economic factors. The intelligent testing pile measuring points can be evenly distributed, or arranged in a sparse and dense manner, to effectively improve the utilization rate and cost-effectiveness of the measuring points.

[0053] Step S120: Based on a preset partitioning algorithm, the monitoring area is divided into a plurality of intelligent detection pile unit spaces according to the positions of the plurality of intelligent detection piles, each intelligent detection pile unit space including one intelligent detection pile.

[0054] In the embodiment of this specification, the preset partitioning algorithm can obtain the position coordinates of each smart detection pile, divide the monitoring area according to the spatial position coordinate relationship of the smart detection pile, and obtain a number of smart detection pile unit spaces covering the entire monitoring area, and each smart detection pile unit space contains a smart detection pile.

[0055] In practical applications, a spatial distribution map of the intelligent measuring pile measuring points is constructed, and the monitoring area is divided into n unit spaces. Each intelligent measuring pile corresponds to a unit space, and the distance to all points in the spatial area is closer to the intelligent measuring pile than to any other intelligent measuring pile.

[0056] The definition of the intelligent pile unit space can be: a set of points P on a given plane area = {p 1 , p 2 , ..., pn}, where n≥2, and p i ≠p i , Point p i The preset partitioning algorithm uses the spatial distribution graph to divide the plane into n unit spaces V (p i ), each unit space V(p i ) contains all distances p on the plane i A point that is closer than any other point in P, that is:

[0057]

[0058] Among them, d(x, p i ) represents the distance from point x to point p in space i The distance can be, but is not limited to, Euclidean distance, Manhattan distance, Chebyshev distance, etc.

[0059] The spatial distribution diagram is a collection of unit spaces, denoted by Vall(P) = {V(p 1 ), V(p 2 ), ..., V(p n )}.

[0060] The steps to construct the spatial distribution map of intelligent piles are as follows:

[0061] (1) Input point set: Assume that the coordinates of the intelligent pile are P = {(x 1 , V 1 ), (x 2 ,y 2 ),...,(x n ,y n )}. Among them, (x n ,y n ) is the coordinate of the nth intelligent measuring pile.

[0062] (2) Calculate the equidistant line: An equidistant line is a straight line obtained from the set of all points in a plane that are equally distant from two fixed points on the plane. Any point on the straight line is equally distant from the two fixed points. i and p j , calculate their equidistant lines B(pi, p j ):

[0063]

[0064] (3) Finding spatial vertices: In the process of constructing a spatial distribution map, it is necessary to calculate the intersection points of equidistant lines. These intersection points are the spatial vertices. Let point set P = {p 1 , p 2 , ..., p n}, where pi =(x i ,y i ), i=1, 2, ..., n.

[0065] For any three different points p i , p j , p k ∈P, the equidistant lines between them are B(p i , p j ), B(p i , p k ), B(p j , p k ). If these three equidistant lines have an intersection, then the intersection is a spatial vertex, denoted by v(p i , p j , p k ).

[0066] Let v(p i , p j , p k )=(x,y), then according to the properties of equidistant lines, we have:

[0067]

[0068] The above system of equations represents the distance from point (x, y) to p i , p j , p k The distances between the three points are equal. Solving the equations, we can get the spatial vertex v(p i , p j , p k )’s coordinates (x, y).

[0069] Let d ij Represents point p i and p j The distance between the space vertices v(p i , p j , p k ) can also be expressed as:

[0070]

[0071] where di x , d jx and d kx similar.

[0072] This shows that the space vertex v(p i , p j , p k ) is the plane to point p i , p j , p kThe point with the smallest distance to the maximum value.

[0073] Therefore, for any three different points in the point set P, if their equidistant lines have an intersection, then the intersection is a spatial vertex. By calculating all possible combinations of three points, all spatial vertices in the spatial distribution graph can be obtained.

[0074] (4) Constructing spatial edges:

[0075] Connect spatial vertices to form spatial edges.

[0076] 1) According to the definition of unit space, point x belongs to V(p i ) if and only if:

[0077]

[0078] 2) Similarly, point x belongs to V(p j ) if and only if:

[0079]

[0080] 3) If point x also belongs to V(p i ) and V(p j ), then it must satisfy:

[0081]

[0082] This means that x is in p i and p j The equidistant line B(p i , p j ) and to p i and p j The distance to is less than or equal to the distance to any other point.

[0083] 4) Therefore, the spatial edge e(p i , p j ) is the set of all points x that satisfy the above conditions, that is:

[0084]

[0085] Therefore, the spatial edge e(p i , p j ) is the equidistant line B(p i , p j ) can be expressed as:

[0086]

[0087] in:

[0088] p i and pj are two points in the given point set P.

[0089] B(p i , p j ) indicates p i and p j The equidistant line is:

[0090]

[0091] Among them, d(x, p) represents the distance from point x to point p, P\{p i , p j} means that in the point set P except p i and p j For all points other than i , p j ) represents the equidistant line B(p i , p j ) on the set of points from which p i The distance from P is less than or equal to the distance to any other point in P, that is, e(p i , p j ) is B(p i , p j ) belongs to the unit space V(p i ) or V(p j ) part.

[0092] Step S130: Process the displacement data and the tilt data of each of the smart testing piles to obtain a deformation parameter vector of each of the smart testing piles.

[0093] Step S140: Compare the deformation parameter vector with the first warning threshold to obtain an over-limit intelligent detection pile.

[0094] In the embodiments of this specification, the intelligent testing pile can send sensing data at a preset frequency, and the first warning threshold can be determined according to the importance, hazard and seasonality of the slope project, based on experience or with reference to relevant technical specifications, and the first warning threshold can be a number of range values. The inclination data and displacement data of each intelligent testing pile can be processed to obtain a deformation parameter vector of each intelligent testing pile, and the deformation parameter vector can be compared with the first warning threshold to determine whether the intelligent testing pile is in an over-limit state, and obtain an over-limit intelligent testing pile.

[0095] In order to ensure the accuracy of the results, the single perception data, the perception data for three consecutive days, and the cumulative perception data of each intelligent testing pile can be obtained, that is, each intelligent testing pile can obtain multiple deformation parameter vectors. The perception data at the current moment is calculated by difference with the perception data at the previous moment to obtain a single displacement value and a single tilt value; the perception data at the current moment is calculated by difference with the perception data at the same moment three consecutive days ago to obtain a displacement value for three consecutive days and a tilt value for three consecutive days; the perception data at the current moment is calculated by difference with the perception data at the preset initial moment of the intelligent testing pile to obtain a cumulative displacement value and a cumulative tilt value. The single displacement value and the single tilt value, the displacement value and the tilt value for three consecutive days, and the cumulative displacement value and the cumulative tilt value are processed separately to obtain a single two-dimensional vector parameter, a two-dimensional vector parameter for three consecutive days, and a cumulative two-dimensional vector parameter. The above three deformation parameter vectors can be compared with their corresponding first warning thresholds.

[0096] In the embodiment of this specification, if the deformation parameter vector of the smart testing pile is greater than the first warning threshold, it means that the smart testing pile is in an over-limit state; if the deformation parameter vector of the smart testing pile is less than or equal to the first warning threshold, it means that the smart testing pile is in a stable state.

[0097] In practical applications, in order to more accurately determine the early risk status of the slope, multiple deformation parameter vectors of each smart testing pile can be obtained, and each deformation parameter vector can be compared with its corresponding first warning threshold. If any deformation parameter vector of the smart testing pile is greater than the corresponding first warning threshold, it is determined that the smart testing pile is in an over-limit state.

[0098] Specifically, the single horizontal displacement value is Lxy i , the single tilt value is α i ;

[0099] Horizontal displacement value Lxy for 3 consecutive days 3d,i , the tilt value for 3 consecutive days is α 3d，i ;

[0100] The accumulated horizontal displacement value is Lxy sum，i , the cumulative tilt value is α sum，i ;

[0101] These parameters are expressed as two-dimensional vectors:

[0102] v i =(Lxy i , α i )

[0103] V 3d,i =(Lxy 3d,i , α 3d,i )

[0104] V sum，i=(Lxy sum，i , α sum，i )

[0105] Among them, v i is a two-dimensional vector of single deformation parameters, V 3d,i is the two-dimensional vector of deformation parameters for three consecutive days, v sum，i is the two-dimensional vector of cumulative deformation parameters.

[0106] For each deformation parameter two-dimensional vector, set a corresponding first warning threshold vector:

[0107] θ 1 =(θ 11 ,θ 12 ) corresponds to V i ;

[0108] θ 2 =(θ 21 ,θ 22 ) corresponds to V 3d,i ;

[0109] θ 3 =(θ 31 ,θ 32 ) corresponds to V sum，i .

[0110] Among them, θ 1 is the single warning threshold vector, θ 2 is the warning threshold vector for three consecutive days, θ 3 is the cumulative warning threshold vector.

[0111] v i With θ 1 For comparison, v 3d，i With θ 2 For comparison, v sum，i With θ 3 The first component of each warning threshold vector corresponds to the displacement value in the two-dimensional deformation parameter vector, and the second component corresponds to the tilt value in the two-dimensional deformation parameter vector. Any component of the deformation parameter vector can be compared with the corresponding component of the corresponding warning threshold vector, for example, v i Lxy i With θ 1 θ 11 Make a comparison.

[0112] If any component of any deformation parameter vector of the intelligent pile i exceeds the corresponding component of the corresponding warning threshold vector, the intelligent pile is considered to be in an over-limit state, denoted by m i =1; otherwise, it is recorded as m i =0, indicating that the intelligent pile is in a stable state.

[0113] The mathematical expression is:

[0114]

[0115] Wherein, V represents a logical OR operation.

[0116] Step S150: clustering the over-limit intelligent testing piles using a preset clustering algorithm to obtain clusters of the over-limit intelligent testing piles.

[0117] Step S160: According to the clustering clusters, a plurality of risk partitions in the monitoring area are obtained, each of the risk partitions including at least one of the intelligent detection pile unit spaces.

[0118] In the embodiment of this specification, a clustering algorithm is used to identify and cluster the over-limit smart test piles in the monitoring area, and the clusters can only contain over-limit and mutually adjacent smart test piles. Among them, each smart test pile points to its corresponding smart test pile unit space, and each cluster can represent a risk zone. In order to better determine the risk zone in the monitoring area, each cluster can contain three smart test piles, that is, each risk zone can be composed of three complete and adjacent smart test pile unit spaces, indicating that the area is in a risk state.

[0119] It should be understood that the order of some steps in the methods described in one or more embodiments of this specification can be interchanged according to actual needs, or some steps can be omitted or deleted.

[0120] In the present application, the slopes in the monitoring area are divided into several smart testing pile unit spaces, the over-limit status of each smart testing pile is determined according to the perception data of the smart testing piles in each smart testing pile unit space, the over-limit smart testing piles are clustered, and a risk partition containing at least one smart testing pile unit space is obtained, which can comprehensively, accurately and effectively determine the early risk status of the slope.

[0121] based on Figure 1 The method in this specification also provides some specific implementation plans of the method, which are described below.

[0122] Optionally, before clustering the over-limit intelligent detection piles using a preset clustering algorithm in the embodiments of this specification, the method may include:

[0123] Identifying a boundary line for dividing the intelligent pile unit space;

[0124] According to the boundary line, determining the adjacent relationship of each of the intelligent testing piles;

[0125] According to the adjacent relationship, a smart testing pile adjacent relationship identification matrix is ​​constructed.

[0126] In the embodiments of the present specification, the boundary line of the intelligent testing pile unit space is identified to determine the adjacent relationship between each intelligent testing pile. Each intelligent testing pile carries unique identification information. The adjacent relationship can be used to determine whether any two intelligent testing piles are adjacent. The proximity of two intelligent testing piles can indicate that there are no other intelligent testing piles between the two intelligent testing piles. For example, the proximity of intelligent testing pile 1 and intelligent testing pile 2 can indicate that there are no intelligent testing piles between intelligent testing pile 1 and intelligent testing pile 2. The non-adjacent between intelligent testing pile 1 and intelligent testing pile 3 can indicate that there are one or more intelligent testing piles between intelligent testing pile 1 and intelligent testing pile 3.

[0127] According to the adjacent relationship of each smart testing pile, an adjacent relationship identification matrix of smart testing piles is constructed. By using the adjacent relationship identification matrix of smart testing piles, mutually adjacent smart testing piles can be quickly found. Then, combined with the over-limit status of each of the smart testing piles, clustering expansion can be performed to cluster the over-limit and mutually adjacent smart testing piles into clusters, and the smart testing pile unit spaces corresponding to the smart testing piles in the clusters are combined to obtain risk partitions in a risky state.

[0128] Specifically, construct an n×n adjacent relationship identification matrix

[0129] Matrix element a ij The value rules are:

[0130] When i≠j, if the intelligent test pile i and the intelligent test pile j are adjacent, then a ij =1; otherwise, a ij =0.

[0131] When i=j, that is, the diagonal elements, let a ii =0.

[0132] In this way, the matrix A becomes a symmetric matrix, and the main diagonal elements are all 0. The matrix element a ij Indicates the adjacent relationship between intelligent testing piles i and j:

[0133]

[0134] By using the adjacent relationship identification matrix A, we can quickly find the intelligent test piles adjacent to the intelligent test pile p. The specific steps are as follows:

[0135] 1) Initialize the cluster set

[0136] 2) Traverse all intelligent test piles i=1, 2, ..., n:

[0137] If the smart test pile i is not assigned to any cluster and m i =1, a new cluster C is created. l =(i), add it to the cluster set Let Q = (i) represent the set of intelligent test piles to be expanded.

[0138] when Repeat the following steps:

[0139] Take a smart pile p from Q and remove it from Q.

[0140] For the p-th row (or p-th column) of matrix A, find all pk =1 and m k = 1, if the smart test pile k is not assigned to any cluster, it will be added to the cluster C l =C l ∪{k}, and add it to the set to be expanded Q=Q∪{k}.

[0141] Repeat step 2) until all the over-limit intelligent pile measurement points are assigned to a cluster. The final cluster set is Among them, m is the number of clusters, C l (l=1, 2, ..., m) is the set of intelligent detection piles of the lth cluster.

[0142] 3) Clustering cluster set Each C l (l=1,2,…,m), determine the number of intelligent test piles n c Does it meet the quantity requirements? If not, this C l Remove from the set and finally get the updated cluster set Among them, M is the number of clusters that meet the number requirements of intelligent testing piles, and the number of clusters in the set C l This is the risk zone within the preset area.

[0143] Optionally, the determining of the adjacent relationship of each of the intelligent detection piles according to the boundary line in the embodiments of this specification may specifically include:

[0144] According to the boundary line, a preset boundary algorithm is used to determine the shared boundary of the intelligent pile unit space;

[0145] According to the shared boundary, the adjacent relationship of each of the intelligent testing piles is determined.

[0146] In an embodiment of the present specification, the preset boundary algorithm may refer to matching the boundary lines of any two smart testing pile unit spaces respectively to determine whether the two smart testing pile unit spaces have a shared boundary, thereby determining whether the two smart testing pile unit spaces are adjacent. By determining which smart testing pile unit spaces are adjacent, it is possible to quickly determine which smart testing piles are adjacent.

[0147] For example, a plurality of intelligent pile unit spaces include at least a first intelligent pile unit space and a second intelligent pile unit space. A preset boundary algorithm is used to determine whether the first intelligent pile unit space and the second intelligent pile unit space contain a common boundary line. If so, the first intelligent pile unit space and the second intelligent pile unit space are determined to be adjacent. If not, the first intelligent pile unit space and the second intelligent pile unit space are determined to be not adjacent.

[0148] If one intelligent pile unit space shares a boundary with another intelligent pile unit space, the intelligent piles corresponding to the two intelligent pile unit spaces are considered to be adjacent.

[0149] Determine the boundary of the unit space: For each space point p i , find its unit space V(p i ) Unit space V(p i ) can be expressed as:

[0150]

[0151] That is, point p i The boundary of the cell space is composed of all i The relevant spatial edge e(p i , p j ) is composed of.

[0152] Determine shared boundary points: For each spatial edge e(p i , p j ),if Then p i and p j are adjacent points. That is, if point p i and p j If the unit spaces of two points have a common boundary, then they are adjacent points.

[0153] Through the above steps, the spatial distribution map can be used to intuitively and accurately judge and identify the surrounding adjacent measuring points of the intelligent pile measuring point, which is suitable for various complex layout situations.

[0154] In order to determine the risk level of the risk zone, optionally, after obtaining the plurality of risk zones in the monitoring area in the embodiment of this specification, the method may further include:

[0155] Identify the over-limit intelligent testing piles corresponding to each of the risk zones;

[0156] Using a parameter synthesis algorithm, the parameter data of each of the over-limit intelligent testing piles is converted into a comprehensive deformation representation value of the risk zone;

[0157] The risk level of the risk zone is determined according to the comprehensive deformation characterization value and the second warning threshold.

[0158] In the embodiments of this specification, by systematically identifying and evaluating the over-limit state of the intelligent piles in each risk zone, the parameter synthesis algorithm may refer to a method of converting displacement data into a comprehensive deformation characterization value. Specifically, the displacement value of each over-limit intelligent pile in the risk zone is decomposed in a preset potential sliding direction using a vector decomposition algorithm to obtain the sub-item displacement value of the risk zone in the preset potential sliding direction. The comprehensive deformation characterization value of the risk zone is calculated based on the sub-item displacement value of the over-limit intelligent pile in each risk zone using a weighted average algorithm.

[0159] The vector decomposition algorithm can refer to the preset potential sliding direction of the slope, and the displacement value of the smart pile is vector-decomposed in the preset potential sliding direction. The displacement value of each over-limit smart pile in the preset potential sliding direction can be obtained by the vector decomposition algorithm, and the average value of the displacement value of the preset potential sliding direction of the over-limit smart pile in the risk zone can be calculated to obtain the comprehensive deformation characterization value of the risk zone.

[0160] For each risk zone, a pre-set over-limit judgment method is used to determine the risk level of the risk zone.

[0161] The comprehensive deformation characterization value may include the comprehensive displacement characterization value. Taking the comprehensive displacement characterization value as an example, suppose the lth risk partition C l Contains m l There are intelligent test piles, numbered respectively The preset potential sliding direction angle of the slope is θ 0 .

[0162] For each intelligent test pile i in this risk partition k , using the current accumulated water displacement value At the preset sliding direction angle θ 0 Perform vector decomposition on the vector to obtain the cumulative horizontal displacement sub-item values ​​and save them:

[0163]

[0164] in, For intelligent pile testing k The deformation azimuth at this moment is It is the sub-item value of the cumulative horizontal displacement.

[0165] Then, the intelligent pile i k The accumulated horizontal displacement sub-item value at the current moment minus the accumulated horizontal displacement sub-item value at the corresponding moment three days ago is the horizontal displacement sub-item value for three consecutive days. Smart pile test kThe cumulative horizontal displacement item value at the current moment minus the cumulative horizontal displacement item value at the previous moment, that is, the continuous single horizontal displacement item value is obtained.

[0166] Then, calculate the risk partition C l All intelligent piles in the preset sliding direction angle θ 0 The comprehensive displacement representation value on is calculated as follows:

[0167]

[0168] in, is the single comprehensive displacement representation value, is the comprehensive displacement representation value for 3 consecutive days, is the cumulative comprehensive displacement representation value.

[0169] Through the three comprehensive displacement characterization values ​​of the above risk partition, the risk partition at the preset sliding direction angle θ can be comprehensively characterized. 0 The overall displacement degree.

[0170] In practical applications, the risk zones may be divided into different risk levels according to the second warning threshold, for example, the risk levels may be divided into mild, moderate and severe.

[0171] Assume that the three second warning thresholds of the jth comprehensive deformation representation value are θ j1 ,θ j2 and θ j3 ,satisfy:

[0172] θ j1 <θ j2 <θ j3 ,j=1,2,3

[0173] The risk level of the risk zone can be determined by comparing the comprehensive deformation representation value of the risk zone with the second warning threshold. For example, if the comprehensive deformation representation value of the risk zone is less than θ j1 It can be said that the risk level of the risk zone is mild, and the comprehensive deformation characterization value of the risk zone is located at θ j1 and θ j2 It can be said that the risk level of the risk zone is moderate, and the comprehensive deformation characterization value of the risk zone is between θ j2 and θ j3 In between, it can be indicated that the risk level of the risk zone is severe.

[0174] Optionally, in the embodiments of this specification, determining the risk level of the risk zone according to the comprehensive deformation characterization value and the second warning threshold may specifically include:

[0175] According to the comprehensive deformation characterization value and the second warning threshold, the risk level of the risk zone is determined using a risk grading rule system; the risk grading rule system includes one or more of a threshold judgment method, a weighted average method, a fuzzy logic system, a decision tree, a random forest, and a neural network.

[0176] In the embodiments of this specification, the second warning threshold can be the same as the first warning threshold, or it can be set based on experience. Based on the above-mentioned comprehensive deformation characterization value and the second warning threshold, the risk zone level is assessed, and a risk grading rule system can be introduced. The risk grading rule system can include a threshold judgment method, a weighted average method, a fuzzy logic system, a decision tree, a random forest, a neural network, etc., which are not specifically limited here. The system can receive the comprehensive deformation characterization value of the risk zone and the second preset threshold as input, and output the risk level of the risk zone.

[0177] For example, let the system be a function f, then:

[0178]

[0179] Among them, S l Risk Zone C l level of risk.

[0180] The risk grading rule system may include threshold judgment method, weighted average method, fuzzy logic system, decision tree, random forest, neural network, etc. The method for implementing risk grading can be selected according to the requirements of the engineering scenario and practical experience, and no specific limitation is made here.

[0181] Specifically, according to the risk classification rule system output S l , determine the risk zone C l For example, S l Mapped to the following levels:

[0182] If S l ≤1, it is judged as no risk level;

[0183] If 1 l ≤2, it is judged as a mild risk level;

[0184] If 2 l ≤3, it is judged as a moderate risk level;

[0185] If S l >3, it is judged as a severe risk level.

[0186] In order to obtain the total area factor of each risk zone, optionally, the method described in the embodiment of this specification may also include:

[0187] ​​Based on a preset space area algorithm, calculate the unit area of ​​the intelligent testing pile unit space corresponding to the over-limit intelligent testing pile;

[0188] According to the unit area, the total area factor of the risk zone is obtained.

[0189] In the embodiments of the present specification, the preset spatial area algorithm is a method for calculating the area of ​​a polygon. According to the results of cluster analysis, all over-limit intelligent detection piles belonging to the same risk zone can be determined. By adding up the unit areas corresponding to each over-limit intelligent detection pile, the total area factor of the risk zone can be obtained.

[0190] Specifically, the risk zone is divided into units based on the unit space of each intelligent testing pile. Then, the i-th intelligent testing pile unit space is V(p i ). Unit space V(p i ) can be obtained by calculating the area of ​​its polygon using a preset spatial area algorithm.

[0191] Let V(p i ) are v 1 , v 2 , ..., v m , where m is the number of vertices, the calculation formula of the preset space area algorithm can be:

[0192] Among them, S i V(p i ) area, (x j ,y j ) is the vertex v j The coordinates of (x m+1 ,y m+1 )=(x 1 ,y 1 ).

[0193] By calculating the unit space area of ​​each intelligent pile in the risk zone, we can get an area array S = [S 1 , S 2 , ..., S n ], where S i Represents the unit area of ​​the i-th intelligent testing pile in the risk zone.

[0194] Suppose there are k risk areas, namely C 1 , C 2 , ..., C k For the first risk area C l , let the number of the intelligent piles it contains be Where m l Represents C l The number of internal intelligence test piles.

[0195] Since the polygonal area of ​​each intelligent testing pile is obtained through the spatial distribution map, and there is no overlap between these polygonal areas, C l Area A l It can be expressed as the sum of the unit space areas of all the intelligent test piles inside it:

[0196]

[0197] in, Indicates number i i The unit space area of ​​the intelligent measuring pile.

[0198] In practical applications, the outer envelope of the risk partition can also be obtained. For the lth risk partition C l , the set of intelligent test piles it contains is The corresponding unit space set is

[0199] C l The outer envelope of can be obtained by following the steps below:

[0200] 1) Initialize C l The outer envelope of

[0201] 2) For Each unit space in Traverse all its edges

[0202] If the edge no The common edge of any two unit spaces in is:

[0203]

[0204] Then Join E l middle.

[0205] 3) The final E l That is C l The outer envelope of C l range.

[0206] Assume E l Contains l The line segments are For each line segment e u (u=1,2,...,q l ), let the coordinates of its endpoints be (x u，1 ,y u，1 ) and (x u，2 ,yu，2 ), then e u It can be expressed as a parametric equation:

[0207]

[0208] Therefore, C l The outer envelope of can be expressed as:

[0209]

[0210] In order to determine the risk level of the preset area, optionally, in the embodiment of this specification, after obtaining the total area factor of the risk zone, the method may further include:

[0211] Obtaining a comprehensive deformation characterization value of the overall risk of the slope according to the total area factor and the comprehensive deformation characterization value;

[0212] According to the comprehensive deformation characterization value of the overall risk of the slope and the third warning threshold, the risk level of the monitoring area is determined using a risk grading rule system; the risk grading rule system includes one or more of a threshold judgment method, a weighted average method, a fuzzy logic system, a decision tree, a random forest, and a neural network.

[0213] In the embodiments of this specification, the third warning threshold may be the same as the parameter value corresponding to the first warning threshold, or may be set based on experience.

[0214] The overall risk comprehensive deformation characterization value of the slope can include the overall risk comprehensive displacement characterization value of the slope. Taking the overall risk comprehensive displacement characterization value of the slope as an example, assuming that there are m risk zones in the monitoring area, and the area of ​​the lth risk zone is A l , and its corresponding comprehensive displacement characterization values ​​include: (single comprehensive displacement representation value), (Comprehensive displacement representation value for 3 consecutive days), (Cumulative comprehensive displacement characterization value).

[0215] The calculation formula of the comprehensive displacement characterization value of the overall risk of the slope is:

[0216]

[0217] in, is the single comprehensive displacement representation value of the overall risk, It is the comprehensive displacement representation value of the overall risk for three consecutive days. The cumulative comprehensive displacement representation value is the overall risk.

[0218] The risk grading rule system can include threshold judgment method, weighted average method, fuzzy logic system, decision tree, random forest, neural network, etc., which are not specifically limited here. The overall risk comprehensive deformation characterization value of the slope and the third threshold are used as input to output the risk level of the monitoring area. This can comprehensively and accurately evaluate the early risk status of the slope, provide a reliable basis for further precise special monitoring implementation, and also gain more time for slope risk prevention and control work.

[0219] In the embodiments of this specification, an innovative method and complete set of equipment for monitoring the early risk coverage of landslides and slopes based on intelligent piles, as well as a method for automatic identification and determination of risk zoning and grading are proposed. The technical theory is reliable, the method is reasonable and effective, and the implementation is fast and economical. It can realize the early risk coverage screening and risk zoning and grading determination of disasters such as large-scale landslides and collapses, providing a strong basis for the next step of precise monitoring and implementation, and also gaining more valuable time for the deployment of slope risk prevention and control, filling the gap in the online monitoring of the early risk of landslides and slopes and their zoning and grading determination technology. At the same time, this set of technical products can also be used in the early risk screening and prevention and control scenarios of geological disasters such as debris flows and collapses, and has important market promotion and application value and social benefits.

[0220] Acquire the perception data of multiple smart testing piles in the monitoring area, the perception data at least including displacement data and inclination data; wherein the smart testing piles are set on the surface of the slope; based on the preset partitioning algorithm, divide the monitoring area into several smart testing pile unit spaces according to the positions of the multiple smart testing piles, and each smart testing pile unit space contains one smart testing pile; process the displacement data and inclination data of each smart testing pile to obtain the deformation parameter vector of each smart testing pile; compare the deformation parameter vector with the first warning threshold to obtain the over-limit smart testing pile; cluster the over-limit smart testing piles using the preset clustering algorithm to obtain several risk zones in the monitoring area, each risk zone contains at least one smart testing pile unit space; determine the comprehensive deformation characterization parameter value of the risk zone according to the deformation parameter vector of the smart testing pile in each risk zone, and compare it with the second warning threshold to obtain the risk classification of each risk zone; determine the comprehensive deformation characterization parameter value of the overall risk of the slope according to the comprehensive deformation characterization parameter value of the risk zone and its area factor, and compare it with the third warning threshold to obtain the overall risk level of the slope, which can comprehensively and accurately determine the early risk state of slope stability while achieving large-scale coverage monitoring.

[0221] Figure 4 This is a schematic diagram of the structure of a device for determining early risk of slope stability based on intelligent testing piles proposed in an embodiment of this specification.

[0222] The device for determining early risk of slope stability based on intelligent piles described in the embodiments of this specification may include:

[0223] The acquisition module 402 is used to acquire the sensing data of multiple intelligent measuring piles in the monitoring area, wherein the sensing data at least includes displacement data and tilt data; wherein the intelligent measuring piles are arranged on the surface of the slope;

[0224] A partitioning module 404 is used to divide the monitoring area into a plurality of intelligent testing pile unit spaces based on a preset partitioning algorithm and according to the positions of the plurality of intelligent testing piles, each intelligent testing pile unit space including one intelligent testing pile;

[0225] A processing module 406 is used to process the displacement data and the tilt data of each of the smart piles to obtain a deformation parameter vector of each of the smart piles;

[0226] A comparison module 408 is used to compare the deformation parameter vector with a first warning threshold to obtain an over-limit intelligent pile;

[0227] A clustering module 410 is used to cluster the over-limit intelligent testing piles using a preset clustering algorithm to obtain clusters of the over-limit intelligent testing piles;

[0228] The determination module 412 is used to obtain a plurality of risk zones in the monitoring area according to the clustering clusters, each of which includes at least one smart detection pile unit space.

[0229] Based on the same idea, the embodiments of this specification also provide a device corresponding to the above method.

[0230] Figure 5 A schematic diagram of the structure of an early risk determination device for slope stability based on intelligent testing piles provided in an embodiment of this specification.

[0231] like Figure 5 As shown, an embodiment of this specification provides a device for determining early risk of slope stability based on intelligent testing piles, including a memory 530, a processor 510 and an embedded computer program 520 thereof, and the processor 510 executes the computer program 520 to implement the method for determining early risk of slope stability based on intelligent testing piles described in any of the above embodiments.

[0232] The slope risk zoning and the overall area factor of the slope in the embodiments of this specification refer to the use of the slope risk assessment model to calculate and analyze the early risk zoning and grading of slope stability. It is necessary to consider the influence of slope characteristics (slope gradient, slope shape, etc.) on the calculation weights, especially in the case of extremely irregular slopes or large slopes, which have a certain impact on the model analysis results. The risk grading method can calculate the partition and overall spatial area in three-dimensional space, obtain the slope risk zoning and overall area factor parameters, and effectively improve the accuracy and effectiveness of model analysis.

[0233] The above describes specific embodiments of this specification, and other embodiments are within the scope of the attached claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily have to be performed in the specific order or continuous order shown to achieve the desired results. The various embodiments in this specification are described in a progressive manner, and the same and similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0234] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0235] The above is only an embodiment of this specification and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for early risk determination of slope stability based on intelligent testing piles, characterized in that: include: Acquire sensing data of a plurality of intelligent measuring piles in a monitoring area, wherein the sensing data at least includes displacement data and tilt data; wherein the intelligent measuring piles are arranged on a slope surface; Based on a preset partitioning algorithm, according to the positions of the plurality of smart piles, the monitoring area is divided into a plurality of smart pile unit spaces, each of which contains one smart pile; Processing the displacement data and the tilt data of each of the smart testing piles to obtain a deformation parameter vector of each of the smart testing piles; Compare the deformation parameter vector with the first warning threshold to obtain an over-limit intelligent pile; Using a preset clustering algorithm, clustering the over-limit intelligent testing piles to obtain clusters of the over-limit intelligent testing piles; According to the clustering clusters, a plurality of risk zones in the monitoring area are obtained, and each of the risk zones includes at least one of the intelligent detection pile unit spaces.

2. The method according to claim 1, characterized in that Before clustering the over-limit intelligent testing piles using a preset clustering algorithm, the method includes: Identifying a boundary line for dividing the intelligent pile unit space; According to the boundary line, determining the adjacent relationship of each of the intelligent testing piles; According to the adjacent relationship, a smart testing pile adjacent relationship identification matrix is ​​constructed.

3. The method according to claim 2, characterized in that Determining the adjacent relationship of each of the intelligent detection piles according to the boundary line specifically includes: According to the boundary line, a preset boundary algorithm is used to determine the shared boundary of the intelligent pile unit space; According to the shared boundary, the adjacent relationship of each of the intelligent testing piles is determined.

4. The method according to claim 1, characterized in that: After obtaining the plurality of risk zones within the monitoring area, the method further includes: Identify the over-limit intelligent testing piles corresponding to each of the risk zones; Using a parameter synthesis algorithm, the parameter data of each of the intelligent testing piles is converted into a comprehensive deformation representation value of the risk partition; The risk level of the risk zone is determined according to the comprehensive deformation characterization value and the second warning threshold.

5. The method according to claim 4, characterized in that The step of determining the risk level of the risk zone according to the comprehensive deformation characterization value and the second warning threshold specifically includes: According to the comprehensive deformation characterization value and the second warning threshold, the risk level of the risk zone is determined using a risk grading rule system; the risk grading rule system includes one or more of a threshold judgment method, a weighted average method, a fuzzy logic system, a decision tree, a random forest, and a neural network.

6. The method according to claim 4, characterized in that The method further comprises: Based on a preset space area algorithm, calculate the unit area of ​​the intelligent testing pile unit space corresponding to the over-limit intelligent testing pile; According to the unit area, the total area factor of the risk zone is obtained.

7. The method according to claim 6, characterized in that After obtaining the total area factor of the risk zone, the method further includes: Obtaining a comprehensive deformation characterization value of the overall risk of the slope according to the total area factor and the comprehensive deformation characterization value; According to the comprehensive deformation characterization value of the overall risk of the slope and the third warning threshold, the risk level of the monitoring area is determined using a risk grading rule system; the risk grading rule system includes one or more of a threshold judgment method, a weighted average method, a fuzzy logic system, a decision tree, a random forest, and a neural network.

8. The method according to any one of claims 1 to 7, characterized in that: The intelligent testing pile includes an antenna, an antenna fixing clip, a device housing, a measuring module, a communication module, and a power supply module.

9. An early risk determination device for slope stability based on intelligent testing piles, characterized in that: include: An acquisition module is used to acquire the sensing data of multiple intelligent detection piles in the monitoring area, wherein the sensing data at least includes displacement data and tilt data; wherein the intelligent detection piles are arranged on the surface of the slope; A partitioning module, configured to divide the monitoring area into a plurality of intelligent testing pile unit spaces based on a preset partitioning algorithm and according to the positions of the plurality of intelligent testing piles, each intelligent testing pile unit space including one intelligent testing pile; A processing module, used for processing the displacement data and the tilt data of each of the smart testing piles to obtain a deformation parameter vector of each of the smart testing piles; A comparison module, used for comparing the deformation parameter vector with a first warning threshold value to obtain an over-limit intelligent detection pile; A clustering module, used to cluster the over-limit intelligent testing piles using a preset clustering algorithm to obtain clusters of the over-limit intelligent testing piles; A determination module is used to obtain a plurality of risk zones within the monitoring area according to the clustering clusters, each of which includes at least one of the intelligent detection pile unit spaces.

10. An early risk determination device for slope stability based on intelligent testing piles, comprising a memory, a processor and an embedded computer program thereof, characterized in that: The processor executes the embedded computer program to implement the steps of the method according to any one of claims 1 to 7.

Images