Landslide mass displacement monitoring method and related product

Abstract

The invention relates to the technical field of intelligent monitoring, in particular to a landslide mass displacement monitoring method and related products, and the method comprises the steps: dividing a landslide mass into a plurality of sub-regions; arranging a reference point in each sub-region; acquiring a displacement change characteristic parameter of the reference point, and judging whether the reference point has significant displacement or not; judging the deformation state of the landslide mass; integrating the displacement data of the plurality of reference points, and performing data fusion to obtain a comprehensive evaluation index of the landslide mass displacement; through zoning processing based on a geological structure, topographic features and a sliding direction, the landslide mass can be scientifically and reasonably zoned, pertinence and uniformity of reference point layout are ensured, and monitoring blind areas and unnecessary resource waste are avoided; and the layout positions and the number of the reference points are determined by adopting an optimization algorithm, so that the layout density and the construction maintenance cost are considered, and the economical efficiency and the practicability of the monitoring scheme are improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring technology, and in particular to a landslide displacement monitoring method and related products. Background Art

[0002] In the field of landslide displacement monitoring, how to scientifically select the appropriate location and number of deformation reference points is a key technical challenge. Factors such as the geological structure, terrain characteristics, and sliding direction of the landslide have an important impact on the layout of reference points. If the reference points are too sparsely laid out, it is difficult to accurately capture and characterize the overall displacement and deformation of the landslide, resulting in incomplete monitoring data and affecting the accuracy of the landslide stability assessment; however, if the layout is too dense, it will greatly increase the construction and maintenance costs and reduce the economy and practicality of the monitoring system.

[0003] In addition, the displacement change amplitude and rate of different reference points may vary greatly, which makes how to effectively weigh the monitoring data of each reference point and integrate and analyze these data to reflect the overall displacement trend of the landslide body an important technical problem that needs to be solved.

[0004] In the process of landslide displacement monitoring, the stability of the reference point itself cannot be ignored. Due to the uneven deformation of the landslide body, some reference points may experience secondary displacements such as tilt and settlement, resulting in deviations or even distortions in the monitoring data. Therefore, how to effectively identify and correct the displacement data of the reference points, eliminate secondary displacement errors, and improve the reliability of the monitoring data is still one of the technical difficulties.

[0005] At the same time, landslide displacement monitoring is often affected by external environmental factors. For example, changes in environmental parameters such as temperature, humidity, rainfall and wind may cause slight changes in the reference point position or equipment monitoring errors, further reducing the accuracy of monitoring data. Therefore, in the data analysis process, how to fully consider the interference of environmental factors and make effective corrections is an important part of improving the accuracy of monitoring data. Summary of the invention

[0006] The technical problem to be solved by the present invention is how to scientifically optimize the layout of reference points for landslide deformation, identify and correct the secondary displacement errors of the reference points, comprehensively analyze the displacement monitoring data and eliminate the interference of environmental factors. The purpose is to provide a landslide displacement monitoring method and related products, which realizes the partitioned layout of reference points based on geological structure, terrain characteristics and sliding direction, combines data fusion analysis with environmental factor correction, and accurately evaluates the overall displacement deformation trend of the landslide.

[0007] The present invention is achieved through the following technical solutions:

[0008] A landslide displacement monitoring method, comprising:

[0009] According to the parameter characteristics of the landslide body, the landslide body is divided into several sub-areas;

[0010] Reference points are laid out in each sub-area to identify the key parts of the landslide mass;

[0011] Obtain real-time displacement monitoring data of the reference point, analyze the displacement change characteristic parameters of the reference point, and determine whether the reference point has significant displacement;

[0012] According to the displacement change characteristic parameters of the reference point, the deformation state of the landslide body is determined;

[0013] The displacement data of multiple reference points are integrated and fused to obtain a comprehensive evaluation index of the landslide displacement.

[0014] Specifically, the parameter characteristics of the landslide body include: geological structure parameters of the landslide body, topographic characteristic parameters of the landslide body and sliding direction of the landslide body;

[0015] The landslide body is partitioned using a clustering algorithm based on geological parameters. The methods include:

[0016] Collect geological structural parameters of the landslide, including soil layer thickness, lithology, water content and joint distribution;

[0017] Collect terrain characteristic parameters, including slope, aspect and terrain relief;

[0018] Determine the sliding direction of the landslide based on the historical sliding data or field observation data of the landslide;

[0019] Normalize the geological structure parameters and terrain characteristic parameters;

[0020] The normalized data were analyzed using a clustering algorithm, and the initial partitioning of each area of ​​the landslide body was generated by calculating the Euclidean distance or cosine similarity between parameters;

[0021] The initial zoning results are iteratively optimized, and the zoning boundaries are readjusted according to the principle of minimizing the objective function. The objective function takes the minimization of the variance of the geological parameters and terrain characteristics within the region as a constraint to generate the final landslide sub-region zoning scheme;

[0022] Output the partition results of landslide sub-areas.

[0023] Specifically, the method for determining the reference point includes:

[0024] Establish an objective function with constraints of maximizing the total coverage of reference point layout, maximizing the importance weight of key parts, controlling the layout density, and minimizing the construction and maintenance costs;

[0025] The objective function is solved by a genetic algorithm to determine the optimal layout position and number of reference points in each sub-area. The method includes:

[0026] The layout positions and quantities of the reference points are represented as chromosomes, and the reference point coordinates and the layout quantities are represented by binary coding or real number coding;

[0027] Randomly generate several populations, each individual represents a reference point layout plan;

[0028] The objective function is used as the fitness function to calculate the fitness of each individual, where the individual with a higher fitness value indicates a better layout plan;

[0029] Generate new populations through selection, crossover and mutation operations;

[0030] Select the individual with the highest fitness value, analyze its chromosome, and determine the optimal layout and number of reference points.

[0031] Specifically, the method for determining whether a reference point has undergone significant displacement includes:

[0032] Obtain displacement data of each reference point;

[0033] Use Kalman filter algorithm to denoise and smooth the displacement data, and synchronize multiple displacement data in time;

[0034] Based on the processed displacement data, the displacement change values ​​are arranged in chronological order to construct the displacement time series of the reference point;

[0035] Wavelet transform is used to perform multi-scale decomposition of displacement time series and extract characteristic parameters of displacement change, including displacement change amplitude, displacement change rate and displacement change direction;

[0036] A preset threshold value is set for each characteristic parameter. If any parameter value of the displacement change amplitude, displacement change rate or displacement change direction exceeds the corresponding preset threshold value, it is determined that a significant displacement has occurred.

[0037] Specifically, the method for determining the deformation state of the landslide body includes:

[0038] Acquire characteristic parameters of the reference point determined to have significant displacement, including displacement change amplitude, displacement change rate, and displacement change direction;

[0039] Screening a number of adjacent reference points that are closest to the reference point, and extracting characteristic parameters of the adjacent reference points;

[0040] The displacement change directions of the determined reference point and its adjacent reference points are analyzed, and the cosine similarity of their direction vectors is calculated; if the cosine similarity is higher than the set threshold, it is determined that the significant displacement originates from the overall deformation of the landslide body; otherwise, the significant displacement is determined to be the secondary displacement of the reference point itself.

[0041] Specifically, the method for obtaining the comprehensive evaluation index of landslide displacement includes:

[0042] After arranging reference points in each sub-area, a weight is assigned to each reference point according to the hazard level assessment results of the landslide sub-area where the reference point is located;

[0043] The displacement characteristic parameters of each reference point are normalized, and the normalization methods include minimum-maximum normalization or Z-score normalization;

[0044] The weighted average method is used to calculate the overall comprehensive evaluation index of the landslide body.

[0045] Furthermore, a warning threshold is set. If the comprehensive evaluation index is greater than the warning threshold, it is determined that the landslide body has a significant deformation trend and a warning signal is triggered.

[0046] Furthermore, the method further comprises:

[0047] Set up environmental monitoring equipment to collect real-time environmental monitoring data of the landslide area and synchronize the environmental monitoring data with the reference point displacement data;

[0048] Using machine learning algorithms, we take environmental factor data as input and reference point displacement data as output to establish a correlation model between the two.

[0049] The historical displacement data of the reference point and the corresponding environmental monitoring data are used as training sets to optimize the model parameters and obtain the trained association model;

[0050] The environmental monitoring data collected in real time is input into the trained association model, the displacement data is corrected according to the predicted displacement data, and the corrected displacement data is used as the displacement data for determining whether a reference point has undergone significant displacement.

[0051] A landslide displacement monitoring terminal comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a landslide displacement monitoring method as described in any one of the above items is implemented.

[0052] A computer program product includes a computer program / instruction, and when the computer program / instruction is executed by a processor, the landslide displacement monitoring method described in any one of the above items is implemented.

[0053] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0054] The present invention partitions the landslide body into relatively homogeneous sub-areas through a clustering algorithm based on geological structure, terrain characteristics and sliding direction; uses an optimization algorithm to arrange reference points in each sub-area to maximize coverage of key parts and balance the arrangement density with construction and maintenance costs; obtains real-time displacement data of the reference points, uses Kalman filtering for denoising and wavelet transform for data processing, extracts characteristic parameters of displacement changes and determines whether significant displacement occurs; through analysis of adjacent reference point data, identifies whether the displacement is overall deformation or secondary displacement; and, based on data fusion methods such as weighted average or principal component analysis, comprehensively evaluates the overall displacement deformation trend of the landslide body.

[0055] The present invention can scientifically and reasonably partition the landslide body through zoning processing based on geological structure, terrain characteristics and sliding direction, ensuring that the reference point layout is targeted and uniform, avoiding monitoring blind spots and unnecessary waste of resources; an optimization algorithm is used to determine the layout position and number of reference points, taking into account both the layout density and the construction and maintenance costs, thereby improving the economy and practicality of the monitoring solution.

[0056] By using Kalman filtering and wavelet transform to denoise and smooth the reference point displacement data, the stability and accuracy of the data are significantly improved; by comparing and analyzing the data of adjacent reference points, the secondary displacement of the reference points can be effectively identified and abnormal data can be eliminated, thereby ensuring that the monitoring results truly reflect the actual deformation of the landslide body.

[0057] By using data fusion methods such as weighted average or principal component analysis and integrating data from multiple reference points, the overall displacement and deformation trend of the landslide body can be accurately evaluated, providing a quantitative basis for the stability evaluation of the landslide body. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings illustrate exemplary embodiments of the present invention and, together with the description thereof, are used to explain the principles of the present invention. These drawings are included to provide a further understanding of the present invention, and the accompanying drawings are included in and constitute a part of this specification and do not constitute a limitation of the embodiments of the present invention.

[0059] Figure 1 It is a schematic flow chart of a landslide displacement monitoring method according to the present invention. DETAILED DESCRIPTION

[0060] To make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and implementation methods. It is understood that the specific implementation methods described herein are only used to explain the relevant content, rather than to limit the present invention.

[0061] It should also be noted that, for the convenience of description, only the parts related to the present invention are shown in the drawings.

[0062] In the absence of conflict, the embodiments and features of the embodiments of the present invention may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0063] Embodiment 1

[0064] like Figure 1 As shown, a landslide displacement monitoring method is provided. The method realizes monitoring and evaluation of the overall displacement trend of the landslide body through the steps of partitioning, setting reference points, real-time data acquisition and processing, deformation state analysis and data fusion. The method includes:

[0065] According to the parameter characteristics of the landslide body (such as geological structure, topographic characteristics and sliding direction), the landslide body is divided into several sub-areas to ensure the consistency of geological characteristics within different areas.

[0066] Reference points are arranged in each sub-area to determine the key parts of the landslide body. The key parts are high-risk areas where deformation may occur or areas with significant geological features.

[0067] Acquire real-time displacement monitoring data of reference points, analyze displacement change characteristic parameters of reference points, and determine whether reference points have significant displacement; obtain displacement data of each reference point in real time through sensing equipment (such as GNSS sensors, displacement sensors, etc.). Process the collected reference point displacement data and extract displacement change characteristic parameters, such as displacement amplitude, rate, and direction. Significant displacement is obtained by comparison with the threshold.

[0068] The deformation state of the landslide body is determined based on the displacement change characteristic parameters of the reference point; the data of multiple reference points are combined to distinguish the overall deformation and local deformation of the landslide body.

[0069] The displacement data of multiple reference points are integrated and fused to extract the overall displacement and deformation trend of the landslide body, thus obtaining a comprehensive evaluation index of the landslide body displacement.

[0070] This method divides the landslide body into relatively homogeneous sub-areas and scientifically arranges reference points in key areas to obtain displacement data in real time and extract displacement characteristic parameters, determine the significant displacement of the reference points, combine the data of multiple reference points to perform deformation state judgment and data fusion analysis, and finally obtain a comprehensive evaluation index of the landslide body displacement.

[0071] Embodiment 2

[0072] The parameter characteristics of the landslide body include: geological structure parameters of the landslide body, topographic characteristic parameters of the landslide body and sliding direction of the landslide body;

[0073] The landslide body is partitioned using a clustering algorithm based on geological parameters. The methods include:

[0074] Collect geological structural parameters of the landslide, including soil layer thickness, lithology, water content and joint distribution;

[0075] Collect terrain characteristic parameters, including slope, aspect and terrain relief;

[0076] Determine the sliding direction of the landslide based on the historical sliding data or field observation data of the landslide;

[0077] The geological structure parameters and terrain characteristic parameters are normalized; the purpose of normalization is to map all parameter data into the same numerical range to ensure the comparability between parameters during cluster analysis. Commonly used methods include minimum-maximum normalization or Z-score normalization.

[0078] Clustering algorithms (such as K-means clustering or hierarchical clustering) are used to analyze the normalized data. The similarity of parameters in each area is determined by calculating the Euclidean distance or cosine similarity between parameters, and the initial partitioning of each area of ​​the landslide body is generated, that is, areas with similar parameter characteristics are classified into the same sub-area.

[0079] The initial zoning results are iteratively optimized, and the zoning boundaries are readjusted according to the principle of minimizing the objective function. The objective function takes the minimization of the variance of the geological parameters and terrain characteristics within the region as a constraint to generate the final landslide sub-region zoning scheme;

[0080] Output the optimized landslide sub-area zoning scheme.

[0081] Embodiment 3

[0082] This embodiment aims at the reference point layout problem in landslide displacement monitoring, uses a genetic algorithm to optimize the position and number of reference points, and determines the optimal layout of reference points in each sub-area by setting an objective function and combining the iterative solution mechanism of the genetic algorithm. The method for determining the reference points includes:

[0083] Establish an objective function with constraints of maximizing the total coverage of reference point layout, maximizing the importance weight of key parts, controlling the layout density, and minimizing the construction and maintenance costs;

[0084] The objective function is solved by a genetic algorithm to determine the optimal layout position and number of reference points in each sub-area. The method includes:

[0085] The layout positions and quantities of reference points are represented as chromosomes, and binary coding or real number coding is used to represent the coordinates of the reference points and the layout quantities; the encoded chromosome represents a reference point layout plan, which includes the reference point positions and quantity information in multiple sub-areas.

[0086] Randomly generate several populations, each individual represents a reference point layout plan;

[0087] The objective function is used as the fitness function to calculate the fitness of each individual, where the individual with a higher fitness value indicates a better layout plan;

[0088] Generate new populations through selection (roulette wheel selection or tournament selection), crossover (single-point crossover or multi-point crossover, etc.) and mutation operations;

[0089] During the multiple iterations of the genetic algorithm, the fitness function is continuously calculated and the population is updated until the convergence conditions are met (such as the fitness value no longer changes significantly or the maximum number of iterations is reached).

[0090] Select the individual with the highest fitness value, analyze its chromosome, and determine the optimal layout and number of reference points.

[0091] Embodiment 4

[0092] This embodiment performs denoising, time series construction and multi-scale feature extraction on the displacement data of the reference point, and combines a preset threshold to determine the significant displacement. The method includes:

[0093] Real-time displacement data is collected by installing monitoring equipment (such as GNSS sensors, displacement sensors, etc.) at each reference point. The displacement data includes the three-dimensional coordinate changes or displacements of the reference points at different time nodes.

[0094] The Kalman filter algorithm is used to denoise and smooth the displacement data, filter out data noise introduced by equipment errors, environmental interference, etc., and smooth the data to obtain stable displacement data. Multiple displacement data are time-synchronized to ensure that the data of each reference point are compared and analyzed under the same time reference to avoid misjudgment caused by time differences.

[0095] Based on the processed displacement data, the displacement change values ​​are arranged in chronological order to construct a displacement time series of the reference point; that is, displacement data points arranged in increasing order of time are obtained.

[0096] Wavelet transform is used to perform multi-scale decomposition of displacement time series and extract characteristic parameters of displacement change, including displacement change amplitude, displacement change rate and displacement change direction;

[0097] Set preset thresholds for each characteristic parameter (based on historical monitoring data of the landslide or the experience of geological experts, namely, displacement change amplitude threshold, rate threshold and direction threshold). If any parameter value of the displacement change amplitude, displacement change rate or displacement change direction exceeds the corresponding preset threshold, it is determined that significant displacement has occurred.

[0098] Embodiment 5

[0099] In this embodiment, by analyzing the displacement characteristic parameters of the reference point with significant displacement and its adjacent reference points, the cosine similarity of the direction vector is used to determine the source of the displacement, and the overall deformation of the landslide body is distinguished from the secondary displacement of the reference point itself. The method for determining the deformation state of the landslide body includes:

[0100] Acquire characteristic parameters of the reference point determined to have significant displacement, including displacement change amplitude, displacement change rate, and displacement change direction;

[0101] Screening a number of adjacent reference points that are closest to the reference point, and extracting characteristic parameters of the adjacent reference points;

[0102] The displacement change directions of the determined reference point and its adjacent reference points are analyzed, the cosine similarity of their direction vectors is calculated, and the calculated cosine similarity is compared with a preset similarity setting threshold.

[0103] If the cosine similarity is higher than the set threshold, it means that the direction of the significant displacement is consistent with the displacement direction of the adjacent reference point, and it is determined that the significant displacement originates from the overall deformation of the landslide body.

[0104] If the cosine similarity is lower than the set threshold, it means that the direction of the significant displacement is inconsistent with the surrounding reference points, and the significant displacement is determined to be the secondary displacement of the reference point itself, such as displacement caused by local settlement, tilt or equipment abnormality.

[0105] Embodiment 6

[0106] This embodiment calculates the comprehensive evaluation index of the landslide body by steps such as weight allocation, standardization processing and weighted average to quantitatively reflect the overall displacement deformation trend of the landslide body. The method for obtaining the comprehensive evaluation index of the landslide body displacement includes:

[0107] After setting up reference points in each sub-area, a weight is assigned to each reference point according to the hazard level assessment result of the landslide sub-area where the reference point is located; the hazard level assessment can be determined based on comprehensive factors such as the geological characteristics of the landslide body, terrain conditions, historical sliding records, etc.

[0108] The displacement characteristic parameters of each reference point are normalized, and the normalization methods include minimum-maximum normalization or Z-score normalization;

[0109] The standardized displacement characteristic parameters are combined with the weight of the reference point, and the weighted average method is used to calculate the overall comprehensive evaluation index of the landslide body. Through weighted averaging, the reference point data in different areas are integrated into an overall evaluation index.

[0110] A warning threshold is set (based on the historical monitoring data, geological characteristics and expert experience of the landslide body). If the comprehensive evaluation index is greater than the warning threshold, it is judged that the landslide body has a significant deformation trend and a warning signal is triggered; if the comprehensive evaluation index is not greater than the warning threshold, it is judged that the landslide body as a whole is in a relatively stable state.

[0111] Embodiment 7

[0112] This embodiment sets up environmental monitoring equipment to collect environmental data in real time, combines machine learning algorithms to establish a correlation model between environmental factors and reference point displacement data, and corrects the real-time displacement data to eliminate the interference of environmental factors on the monitoring data, thereby ensuring the accuracy and reliability of the data. The method includes:

[0113] Set up environmental monitoring equipment (such as temperature sensors, humidity sensors, rain gauges, etc.) to collect real-time environmental monitoring data in the landslide area, and synchronize the environmental monitoring data with the reference point displacement data;

[0114] The historical displacement data of the reference point and the environmental monitoring data of the corresponding time period are used as training sets to form a paired data set of input and output.

[0115] Using machine learning algorithms (such as support vector machines (SVM), random forests (RF) or long short-term memory networks (LSTM)), taking environmental factor data as input and reference point displacement data as output, a nonlinear association model between the two is established;

[0116] The historical displacement data of the reference point and the corresponding environmental monitoring data are used as training sets to optimize the model parameters. The model parameters are optimized through the training process and the model performance is evaluated. Common indicators include mean square error (MSE) and determination coefficient (R 2 ).

[0117] Get the trained association model.

[0118] The real-time collected environmental monitoring data is input into the trained association model, and the displacement data is corrected according to the predicted displacement data. The corrected displacement data = the original collected displacement data - the predicted displacement value caused by environmental factors.

[0119] The corrected displacement data is used as displacement data for determining whether a significant displacement has occurred at the reference point.

[0120] Embodiment 8

[0121] A landslide displacement monitoring terminal comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, a landslide displacement monitoring method as described in any one of the above items is implemented.

[0122] The memory can be used to store software programs and modules. The processor executes various functional applications and data processing of the terminal by running the software programs and modules stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an execution program required for at least one function, etc.

[0123] The data storage area can store data created according to the use of the terminal, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0124] A computer program product includes a computer program / instruction, and when the computer program / instruction is executed by a processor, the landslide displacement monitoring method described in any one of the above items is implemented.

[0125] A computer program product includes a computer program or set of instructions for performing specific tasks or implementing specific functions. These programs or instructions are designed to be executed by a processor to implement a series of predefined steps or operations. The program product may be stored in various forms of computer storage media, such as memory, hard disk, solid-state drive, optical disk or other forms of digital storage devices. It may exist in the form of compiled binary code or in the form of scripts or bytecodes that can be executed by an interpreter. The program product uses carefully designed algorithms and logical instructions to enable the processor to process data in a specific order and manner to complete various functions such as data analysis, user interaction, device control, etc.

[0126] In the description of this specification, the description with reference to the terms "one embodiment / method", "some embodiments / methods", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment / method or example are included in at least one embodiment / method or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment / method or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments / methods or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments / methods or examples described in this specification and the features of the different embodiments / methods or examples, unless they are contradictory.

[0127] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0128] It should be understood by those skilled in the art that the above embodiments are only for the purpose of clearly illustrating the present invention, and are not intended to limit the scope of the present invention. For those skilled in the art, other changes or modifications may be made based on the above invention, and these changes or modifications are still within the scope of the present invention.

Claims

1. A landslide displacement monitoring method, characterized in that: include: According to the parameter characteristics of the landslide body, the landslide body is divided into several sub-areas; Reference points are laid out in each sub-area to identify the key parts of the landslide mass; Obtain real-time displacement monitoring data of the reference point, analyze the displacement change characteristic parameters of the reference point, and determine whether the reference point has significant displacement; According to the displacement change characteristic parameters of the reference point, the deformation state of the landslide body is determined; The displacement data of multiple reference points are integrated and fused to obtain a comprehensive evaluation index of the landslide displacement.

2. A landslide displacement monitoring method according to claim 1, characterized in that: The parameter characteristics of the landslide body include: geological structure parameters of the landslide body, topographic characteristic parameters of the landslide body and sliding direction of the landslide body; The landslide body is partitioned using a clustering algorithm based on geological parameters. The methods include: Collect geological structural parameters of the landslide, including soil layer thickness, lithology, water content and joint distribution; Collect terrain characteristic parameters, including slope, aspect and terrain relief; Determine the sliding direction of the landslide based on the historical sliding data or field observation data of the landslide; Normalize the geological structure parameters and terrain characteristic parameters; The normalized data were analyzed using a clustering algorithm, and the initial partitioning of each area of ​​the landslide body was generated by calculating the Euclidean distance or cosine similarity between parameters; The initial zoning results are iteratively optimized, and the zoning boundaries are readjusted according to the principle of minimizing the objective function. The objective function takes the minimization of the variance of the geological parameters and terrain characteristics within the region as a constraint to generate the final landslide sub-region zoning scheme; Output the partition results of landslide sub-areas.

3. A landslide displacement monitoring method according to claim 1, characterized in that: Methods for determining reference points include: Establish an objective function with constraints of maximizing the total coverage of reference point layout, maximizing the importance weight of key parts, controlling the layout density, and minimizing the construction and maintenance costs; The objective function is solved by genetic algorithm to determine the optimal layout position and number of reference points in each sub-area. The method includes: The layout positions and quantities of the reference points are represented as chromosomes, and the reference point coordinates and the layout quantities are represented by binary coding or real number coding; Randomly generate several populations, each individual represents a reference point layout plan; The objective function is used as the fitness function to calculate the fitness of each individual, where the individual with a higher fitness value indicates a better layout plan; Generate new populations through selection, crossover and mutation operations; Select the individual with the highest fitness value, analyze its chromosome, and determine the optimal layout and number of reference points.

4. A landslide displacement monitoring method according to claim 1, characterized in that: Methods for determining whether a reference point has undergone significant displacement include: Obtain displacement data of each reference point; Use Kalman filter algorithm to denoise and smooth the displacement data, and synchronize multiple displacement data in time; Based on the processed displacement data, the displacement change values ​​are arranged in chronological order to construct the displacement time series of the reference point; Wavelet transform is used to perform multi-scale decomposition of displacement time series and extract characteristic parameters of displacement change, including displacement change amplitude, displacement change rate and displacement change direction; A preset threshold value is set for each characteristic parameter. If any parameter value of the displacement change amplitude, displacement change rate or displacement change direction exceeds the corresponding preset threshold value, it is determined that a significant displacement has occurred.

5. A landslide displacement monitoring method according to claim 1, characterized in that: Methods for determining the deformation state of a landslide include: Acquire characteristic parameters of the reference point determined to have significant displacement, including displacement change amplitude, displacement change rate, and displacement change direction; Screening a number of adjacent reference points that are closest to the reference point, and extracting characteristic parameters of the adjacent reference points; The displacement change directions of the determined reference point and its adjacent reference points are analyzed, and the cosine similarity of their direction vectors is calculated; if the cosine similarity is higher than the set threshold, it is determined that the significant displacement originates from the overall deformation of the landslide body; otherwise, the significant displacement is determined to be the secondary displacement of the reference point itself.

6. A landslide displacement monitoring method according to claim 1, characterized in that: Methods for obtaining comprehensive evaluation indicators of landslide displacement include: After arranging reference points in each sub-area, a weight is assigned to each reference point according to the hazard level assessment results of the landslide sub-area where the reference point is located; The displacement characteristic parameters of each reference point are normalized, and the normalization methods include minimum-maximum normalization or Z-score normalization; The weighted average method is used to calculate the overall comprehensive evaluation index of the landslide body.

7. A landslide displacement monitoring method according to claim 6, characterized in that: A warning threshold is set. If the comprehensive evaluation index is greater than the warning threshold, it is determined that the landslide body has a significant deformation trend and a warning signal is triggered.

8. A landslide displacement monitoring method according to claim 1, characterized in that: Also includes: Set up environmental monitoring equipment to collect real-time environmental monitoring data of the landslide area and synchronize the environmental monitoring data with the reference point displacement data; Using machine learning algorithms, we take environmental factor data as input and reference point displacement data as output to establish a correlation model between the two. The historical displacement data of the reference point and the corresponding environmental monitoring data are used as training sets to optimize the model parameters and obtain the trained association model; The environmental monitoring data collected in real time is input into the trained association model, the displacement data is corrected according to the predicted displacement data, and the corrected displacement data is used as the displacement data for determining whether a reference point has undergone significant displacement.

9. A landslide displacement monitoring terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, a landslide displacement monitoring method according to any one of claims 1 to 8 is implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, a landslide displacement monitoring method as described in any one of claims 1 to 8 is implemented.

Images