A landslide stability analysis method, device, equipment and storage medium

Abstract

The present application discloses a landslide stability analysis method, device, equipment and storage medium, which relate to the technical field of electrical digital data processing. The method uses the characteristic of the extreme learning machine that the learning process only needs one forward propagation and no iterative update of weights on the basis of the traditional Swedish strip method to iterate the monitoring data collected by the inclinometer hole, so that the subsequent Swedish strip method calculation results are more accurate. At the same time, the prediction function of the monitoring data is also realized, which effectively reduces the problem of inaccurate calculation of the traditional Swedish strip method.

Description

Technical Field

[0001] The present application relates to the technical field of electronic digital data processing, and in particular to a landslide stability analysis method, device, equipment and storage medium. Background Art

[0002] Landslide prevention and control refers to the prevention and control measures taken when engineering construction is carried out in areas where landslides cannot be avoided or where the slopes are unstable. Landslide stability analysis is the basic work of landslide prevention and control and the prerequisite for subsequent prevention and control work. According to the characteristics of the slope rock and soil, take corresponding measures to prevent the slope from becoming unstable to ensure the safety of production and personnel. Landslide stability analysis is the basic work of landslide prevention and control and the prerequisite for subsequent prevention and control work. The significance of conducting stability research on landslides is that it not only provides a scientific theoretical basis for engineering construction, but also plays an important guiding role in the early warning and forecast of landslide development trends.

[0003] For areas with dense mountains, landslides caused by unstable slopes can cause huge losses to personal safety and economic construction. Therefore, correctly evaluating the stability of the slope and taking corresponding reinforcement measures can greatly reduce such losses.

[0004] The Swedish strip method is mostly used for landslide stability analysis. This method requires continuous trial calculations to determine the potential failure surface where sliding may occur, and does not take into account the differences in friction angles and cohesion between different soil strips. The analyzed landslide safety stability coefficient is somewhat different from the actual one. Summary of the invention

[0005] The main purpose of the present application is to provide a landslide stability analysis method, device, equipment and storage medium to solve the problem that the slope safety stability coefficient in the prior art is somewhat different from the actual one.

[0006] In order to achieve the above objectives, this application provides the following technical solutions:

[0007] A landslide stability analysis method is provided, the analysis method being based on a plurality of inclinometer holes reserved on a monitoring section, the analysis method comprising:

[0008] Step S1, respectively obtaining historical monitoring data of each inclinometer hole based on several historical moments;

[0009] Step S2, obtaining the historical displacement vectors of all historical monitoring data by means of a borehole inclinometer method;

[0010] Step S3, marking the inclinometer holes whose historical displacement vectors exceed a preset displacement threshold as deformation mutation inclinometer holes, and obtaining the historical coordinate points of each deformation mutation inclinometer hole respectively;

[0011] Step S4, learning and training all historical coordinate points of all historical moments by an extreme learning machine to obtain a coordinate point prediction model;

[0012] Step S5, predicting the future coordinate points of each deformation mutation inclinometer hole based on a preset number of steps by using the coordinate point prediction model;

[0013] Step S6, vertically dividing the monitoring section into a plurality of soil strips based on all future coordinate points in the direction from the leading edge point to the trailing edge point on the monitoring section;

[0014] Step S7, obtaining the shear strength index of each soil strip based on the geological drawing of the monitoring section;

[0015] Step S8, calculating the safety factor of each soil strip under the limit equilibrium state by the Swedish strip method based on the shear strength index of each soil strip;

[0016] Step S9, obtaining the minimum value of all safety factors, and marking the soil strip corresponding to the minimum value as the most likely point for landslide to occur.

[0017] As a further improvement of the present application, step S4, learning and training all historical coordinate points of all historical moments by an extreme learning machine to obtain a coordinate point prediction model, includes:

[0018] Step S41, packing all historical coordinate points of the same deformation mutation inclination measuring hole into a coordinate data set;

[0019] Step S42, normalizing all coordinate data sets respectively, and obtaining a normalized data set based on one coordinate data set;

[0020] Step S43, dividing the current normalized data set into a training set and a validation set according to a preset ratio;

[0021] Step S44, defining an extreme learning model according to the data dimension of the training set;

[0022] Step S45, training and learning the extreme learning model through the training set to obtain a coordinate point prediction model of the current deformation mutation inclinometer hole.

[0023] As a further improvement of the present application, step S45, training and learning the extreme learning model through the training set to obtain a coordinate point prediction model of the current deformation mutation inclination hole, includes:

[0024] Step S451, define the training set as ;

[0025] in, , For the The historical coordinate value of a historical moment, for of A collection of dimensional vectors, , For the The future coordinate value of the preset number of steps, for of A collection of dimensional vectors, with superscripts is the transposed matrix;

[0026] Step S452, defining the extreme learning model according to formula (1):

[0027] (1);

[0028] in, , is the first connection weight between the hidden layer and the output layer of the extreme learning model; is the number of nodes in the hidden layer; is the output value of the hidden layer, ; ; is the activation function; is the input layer of the extreme learning model and the first layer of the hidden layer The second connection weight of the nodes; is the hidden layer The threshold of nodes;

[0029] Step S453, randomly generating the second connection weight and the threshold value by a nonlinear mapping function in a feature space based on the extreme learning model;

[0030] Step S454, obtaining the output value of the hidden layer by calculating the randomly generated second connection weight and threshold;

[0031] Step S455: according to formula (2), the optimal solution of the first connection weight is obtained by minimizing the approximate square error:

[0032] (2);

[0033] in, is the output matrix of the hidden layer; is the target matrix of the training sample; is the optimal solution; The output matrix The MP generalized inverse matrix, , The dimension is The set of weight vectors of ;

[0034] Step S456, obtaining the optimal solution and the second connection weight and threshold corresponding to the optimal solution and substituting them into the extreme learning model to obtain the prediction model.

[0035] As a further improvement of the present application, step S7, based on the geological drawings of the monitoring section, respectively obtains the shear strength index of each soil strip, including:

[0036] Step S71, calculating the sliding surface length of each soil strip according to the Euclidean distance;

[0037] Step S72, searching for the internal friction angle and cohesion of each soil strip based on the geological drawing;

[0038] Step S73, if the internal friction angle of the current soil strip is equal to the internal friction angle of the next soil strip, the internal friction angle of the current soil strip remains unchanged;

[0039] Step S74, if the internal friction angle of the current soil strip is not equal to the internal friction angle of the next soil strip, then the internal friction angle of the current soil strip is equal to the average value of the internal friction angle of the current soil strip and the internal friction angle of the next soil strip;

[0040] Step S75, if the cohesion of the current soil strip is equal to the cohesion of the next soil strip, the cohesion of the current soil strip remains unchanged;

[0041] Step S76, if the cohesion of the current soil strip is not equal to the cohesion of the next soil strip, the cohesion of the current soil strip is equal to the average of the cohesion of the current soil strip and the cohesion of the next soil strip;

[0042] Step S77, defining the sliding surface length, internal friction angle, and cohesion of the current soil strip as shear strength indicators of the current soil strip.

[0043] As a further improvement of the present application, step S8, based on the shear strength index of each soil strip, the safety factor of each soil strip under the limit equilibrium state is calculated by the Swedish strip method, including:

[0044] Step S81, calculate the safety factor of each soil strip by formula (3):

[0045] (3);

[0046] in, For the The safety factor of each soil strip is is the total number of all soil strips, For the The cohesion of a soil strip, For the The sliding surface length of a soil strip, For the The gravity of a soil bar, For the The sliding surface inclination of the soil strip, For the The internal friction angle of a soil strip.

[0047] As a further improvement of the present application, step S9, obtaining the minimum value of all safety factors, and marking the soil strip corresponding to the minimum value as the most likely point for landslide occurrence, then includes:

[0048] Step S10, outputting each soil strip and the safety factor of each soil strip on the geological drawing, and sending them to an external visualization terminal;

[0049] Step S20: marking the soil strip corresponding to the minimum value in red on an external visualization terminal.

[0050] In order to achieve the above objectives, this application also provides the following technical solutions:

[0051] A device for analyzing landslide stability, the device for analyzing landslide stability is applied to the above-mentioned method for analyzing landslide stability, and the device for analyzing landslide stability comprises:

[0052] The module for acquiring historical monitoring data of the inclinometer hole is used to respectively acquire the historical monitoring data of each inclinometer hole based on several historical moments;

[0053] A historical displacement vector acquisition module is used to obtain historical displacement vectors of all historical monitoring data through a borehole inclinometer method;

[0054] A deformation mutation inclinometer hole marking module is used to mark the inclinometer holes whose historical displacement vectors exceed a preset displacement threshold as deformation mutation inclinometer holes, and obtain the historical coordinate points of each deformation mutation inclinometer hole respectively;

[0055] A coordinate point prediction model acquisition module is used to learn and train all historical coordinate points at all historical moments through an extreme learning machine to obtain a coordinate point prediction model;

[0056] A future coordinate point prediction acquisition module, used to predict the future coordinate points of each deformation mutation inclinometer hole based on a preset number of steps through the coordinate point prediction model;

[0057] A soil strip division module, used to divide the monitoring section vertically into a plurality of soil strips based on all future coordinate points in the direction from the leading edge point to the trailing edge point on the monitoring section;

[0058] A shear strength index acquisition module, used to respectively acquire the shear strength index of each soil strip based on the geological drawings of the monitoring section;

[0059] A safety factor calculation module is used to calculate the safety factor of each soil strip under the limit equilibrium state by Swedish strip method based on the shear strength index of each soil strip;

[0060] The most likely landslide occurrence point marking module is used to obtain the minimum value of all safety factors and mark the soil strip corresponding to the minimum value as the most likely landslide occurrence point.

[0061] In order to achieve the above objectives, this application also provides the following technical solutions:

[0062] An electronic device comprises a processor and a memory coupled to the processor, wherein the memory stores program instructions executable by the processor; when the processor executes the program instructions stored in the memory, the above-mentioned landslide stability analysis method is implemented.

[0063] In order to achieve the above objectives, this application also provides the following technical solutions:

[0064] A storage medium stores program instructions, and when the program instructions are executed by a processor, the above-mentioned landslide stability analysis method can be implemented.

[0065] The present application obtains historical monitoring data of each inclinometer hole based on several historical moments respectively; obtains historical displacement vectors of all historical monitoring data by means of the borehole inclinometer method; marks the inclinometer holes whose historical displacement vectors exceed a preset displacement threshold as deformation mutation inclinometer holes, and obtains historical coordinate points of each deformation mutation inclinometer hole respectively; learns and trains all historical coordinate points of all historical moments by means of an extreme learning machine to obtain a coordinate point prediction model; predicts future coordinate points of each deformation mutation inclinometer hole based on a preset number of steps by means of the coordinate point prediction model; vertically divides the monitoring section into several soil strips based on all future coordinate points in the direction from the leading edge point to the trailing edge point on the monitoring section; obtains the shear strength index of each soil strip based on the geological drawings of the monitoring section; calculates the safety factor of each soil strip under the limit equilibrium state by means of the Swedish strip division method based on the shear strength index of each soil strip; obtains the minimum value among all safety factors, and marks the soil strip corresponding to the minimum value as the most likely point for landslide occurrence. Based on the traditional Swedish strip method, the present application utilizes the characteristic of the extreme learning machine that the learning process only requires one forward propagation and does not require iterative updating of weights to iterate the monitoring data collected from the incline hole, thereby making the subsequent Swedish strip method calculation results more accurate. At the same time, it also realizes the prediction function of the monitoring data, effectively reducing the problem of inaccurate calculations by the traditional Swedish strip method. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1A schematic diagram of the process steps of an embodiment of a method for analyzing landslide stability of the present application;

[0067] Figure 2 A schematic diagram of functional modules of an embodiment of a device for analyzing landslide stability of the present application;

[0068] Figure 3 This is a schematic diagram of the structure of an embodiment of the electronic device of the present application;

[0069] Figure 4 This is a schematic diagram of the structure of an embodiment of the storage medium of the present application. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0071] The terms "first", "second" and "third" in this application are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined as "first", "second" and "third" can explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "multiple" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined. All directional indications (such as up, down, left, right, front, back...) in the embodiments of this application are only used to explain the relative position relationship, movement, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication also changes accordingly. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally also includes steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or devices.

[0072] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0073] like Figure 1As shown, this embodiment provides an embodiment of a method for analyzing landslide stability. In this embodiment, the analysis method is based on a number of inclinometer holes reserved on the monitoring section.

[0074] Preferably, the inclinometer borehole is a method for measuring the deformation of the ground soil layer, determining the deformation of the foundation and the horizontal form of the stratum, which is usually used to monitor building structures, subways, roads, bridges and other infrastructure projects, and can also be used in underground hydrogeological exploration and other fields. The principle of inclinometer borehole is to monitor and record the deformation of the stratum by setting measuring instruments such as inclinometers and deformation meters in the borehole, so as to analyze the deformation characteristics of the soil and further evaluate the foundation stability of the project.

[0075] Preferably, the inclinometer hole is usually achieved by a borehole inclinometer:

[0076] 1) Install the borehole inclinometer.

[0077] ① Select a suitable drilling hole diameter to ensure that the drilling hole diameter can accommodate the inclinometer and its cable.

[0078] ②Installate the inclinometer in the borehole pipe and ensure that the inclinometer can work normally.

[0079] ③Insert the cable and connect the data logger.

[0080] 2) Calibrate the borehole inclinometer.

[0081] ①Install the level and adjust the horizontal position of the borehole inclinometer.

[0082] ②Perform zero point calibration so that the output of the borehole inclinometer is zero when it is in the horizontal position.

[0083] ③ Calibrate the angle and make adjustments based on the indication difference to ensure that the measured angle data is accurate.

[0084] 3) Take measurements.

[0085] ① Turn on the data logger and start recording the angle data of the borehole inclinometer.

[0086] ② Use different depth intervals for recording and ensure that the depth intervals are consistent.

[0087] ③After the measurement is completed, close the data collector, export the data and process it.

[0088] 4) Data processing.

[0089] ①Import the data into professional data processing software.

[0090] ② Perform data analysis and generate measurement reports.

[0091] ③ Data that does not meet the requirements will be initially excluded, and accurate and reliable data will be retained.

[0092] Specifically, the analysis method comprises the following steps:

[0093] Step S1, respectively obtain the historical monitoring data of each inclinometer hole based on several historical moments.

[0094] Step S2, obtaining the historical displacement vectors of all historical monitoring data by means of the borehole inclinometer method.

[0095] Preferably, the borehole inclinometer method can measure the vertical and horizontal displacements of the soil body, and then convert the vertical and horizontal displacement lengths into vectors.

[0096] Step S3, marking the inclinometer holes whose historical displacement vectors exceed a preset displacement threshold as deformation mutation inclinometer holes, and obtaining the historical coordinate points of each deformation mutation inclinometer hole respectively.

[0097] Preferably, the preset displacement threshold may be set to 10 mm, that is, the length of the historical displacement vector does not exceed 10 mm.

[0098] Step S4, learning and training all historical coordinate points of all historical moments through an extreme learning machine to obtain a coordinate point prediction model.

[0099] Preferably, the extreme learning model is a feedforward neural network that can be used to train a single hidden layer. The extreme learning machine is different from the traditional single hidden layer feedforward neural network training algorithm. The extreme learning machine randomly selects the input layer weights and hidden layer biases, and the output layer weights are calculated and analyzed based on the Moore-Penrose (MP) generalized inverse matrix theory by minimizing the loss function composed of the training error term and the regularization term of the output layer weight norm. Even if the extreme learning machine randomly generates hidden layer nodes, the extreme learning machine still maintains the general approximation ability of a single hidden layer feedforward neural network.

[0100] Step S5, predicting the future coordinate points of each deformation mutation inclinometer hole based on a preset number of steps by using the coordinate point prediction model.

[0101] Preferably, the preset number of steps can be set to steps of different durations such as one step per second, one step per minute, etc.

[0102] Step S6: vertically divide the monitoring section into a plurality of soil strips based on all future coordinate points along the direction from the leading edge point to the trailing edge point on the monitoring section.

[0103] Preferably, the thickness of a soil strip is generally 0.5m.

[0104] Step S7, obtaining the shear strength index of each soil strip based on the geological drawing of the monitoring section.

[0105] Step S8, calculating the safety factor of each soil strip under the limit equilibrium state by the Swedish strip method based on the shear strength index of each soil strip.

[0106] Step S9, obtaining the minimum value of all safety factors, and marking the soil strip corresponding to the minimum value as the most likely point for landslide to occur.

[0107] Furthermore, in step S4, all historical coordinate points of all historical moments are learned and trained by an extreme learning machine to obtain a coordinate point prediction model, including:

[0108] Step S41, packing all historical coordinate points of the same deformation mutation inclinometer hole into a coordinate data set.

[0109] Step S42, normalizing all coordinate data sets respectively, and obtaining a normalized data set based on one coordinate data set.

[0110] Preferably, this embodiment prefers the normalization method of zero-mean normalization (Z-score normalization), which gives the mean and standard deviation of the original data to standardize the data, and the processed data conforms to the standard normal distribution, that is, the mean is 0 and the standard deviation is 1. For the normalization method, this embodiment can also use batch normalization. Compared with simple normalization, in the previous neural network training, only the input layer data is normalized, but not normalized in the middle layer. Although the data set of the input node is normalized, the data distribution of the input data after matrix multiplication is likely to change greatly, and as the number of hidden layer network layers continues to deepen, the change in data distribution will become greater and greater. Therefore, the normalization processing performed by batch normalization in the middle layer of the neural network makes the training effect better.

[0111] Step S43, dividing the current normalized data set into a training set and a validation set according to a preset ratio.

[0112] Preferably, the preset ratio is usually 80%:20% to divide the image data into a training set and a validation set, that is, 80% of the data is the training set and 20% of the data is the validation set.

[0113] Step S44, defining an extreme learning model according to the data dimension of the training set.

[0114] Step S45, training and learning the extreme learning model through the training set to obtain the coordinate point prediction model of the current deformation mutation inclination hole.

[0115] Further, in step S45, the extreme learning model is trained and learned through the training set to obtain the coordinate point prediction model of the current deformation mutation inclination hole, including:

[0116] Step S451, define the training set as .

[0117] in, , For the The historical coordinate value of a historical moment, for of A collection of dimensional vectors, , For the The future coordinate value of the preset number of steps, for of A collection of dimensional vectors, with superscripts is the transposed matrix.

[0118] Step S452, define the extreme learning model according to formula (1):

[0119] (1).

[0120] in, , is the first connection weight between the hidden layer and the output layer of the extreme learning model; is the number of nodes in the hidden layer; is the output value of the hidden layer, ; ; is the activation function; is the input layer and the hidden layer of the extreme learning model The second connection weight of the nodes; The hidden layer The threshold of nodes.

[0121] Preferably, It can be set to activation functions such as Sigmoid function and Gaussian function.

[0122] Preferably, if the Sigmoid function is selected, then It can be expressed as:

[0123] , here .

[0124] Step S453: randomly generate the second connection weight and threshold value through a nonlinear mapping function in the feature space based on the extreme learning model.

[0125] Step S454, obtaining the output value of the hidden layer through the randomly generated second connection weight and threshold calculation.

[0126] Step S455, according to formula (2), the optimal solution of the first connection weight is obtained by minimizing the approximate square error:

[0127] (2).

[0128] in, is the output matrix of the hidden layer; is the target matrix of the training samples; is the optimal solution; The output matrix The MP generalized inverse matrix, , The dimension is The set of weight vectors .

[0129] Step S456, obtaining the optimal solution and the second connection weight and threshold corresponding to the optimal solution and substituting them into the extreme learning model to obtain a prediction model.

[0130] Preferably, the Moore-Penrose generalized inverse matrix can be calculated by singular value decomposition .

[0131] It can be understood that the training of a single hidden layer feedforward neural network by an extreme learning machine is divided into two main stages:

[0132] ① Random feature mapping:

[0133] The hidden layer parameters are randomly initialized, and then some nonlinear mapping is used as the activation function to map the input data to a new feature space (called the extreme learning machine feature space). In simple terms, the weights and biases on the hidden layer nodes of the extreme learning machine are randomly generated. The random feature mapping stage is different from many existing learning algorithms, such as support vector machines that use kernel functions for feature mapping, restricted Boltzmann machines in deep neural networks, and autoencoders / autodecoders for feature learning. The nonlinear mapping function in the extreme learning machine can be any nonlinear piecewise continuous function. In the extreme learning machine, the hidden layer node parameters are randomly generated according to any continuous probability distribution and are independent of the training data.

[0134] (2) Linear parameter solution.

[0135] After the first stage, the second connection weight and threshold are randomly generated. In the second stage, only the first connection weight needs to be solved. In order to obtain the first connection weight with good effect on the training sample set, it is necessary to ensure that its training error is minimized. The target matrix of the training samples Minimize the square error as the objective function for evaluating the training error, and then find the solution with the smallest objective function is the optimal solution.

[0136] Preferably, the eigenspace is a vector space that represents a special property of a matrix. The eigenspace is a vector space consisting of all eigenvectors of the matrix. For example, for an n×n matrix A, if there exists a scalar λ and a non-zero vector x such that Ax=λx, then x is the eigenvector of A, and the vector space where the vector x is located is the eigenspace of A.

[0137] It should be noted that the meaning of the symbols in the above additional content is not interoperable with other parts of the embodiment.

[0138] Further, step S7, based on the geological drawings of the monitoring section, the shear strength index of each soil strip is obtained respectively, including:

[0139] Step S71, calculating the sliding surface length of each soil strip according to the Euclidean distance.

[0140] Step S72, searching for the internal friction angle and cohesion of each soil strip based on the geological drawing.

[0141] Preferably, the cohesion and internal friction angle are generally predetermined in advance through direct shear uniaxial compression or triaxial compression tests and reflected on geological drawings. The Engineering Rock Mass Classification Standard GB50218-94 also directly gives the above physical parameters.

[0142] Step S73: If the internal friction angle of the current soil strip is equal to the internal friction angle of the next soil strip, the internal friction angle of the current soil strip remains unchanged.

[0143] Step S74: if the internal friction angle of the current soil strip is not equal to the internal friction angle of the next soil strip, the internal friction angle of the current soil strip is equal to the average value of the internal friction angle of the current soil strip and the internal friction angle of the next soil strip.

[0144] Preferably, the following soil strip number is used To assist in the explanation, The internal friction angle of the soil strip With The internal friction angle of the soil strip If they are not equal, then The internal friction angle of the soil strip .

[0145] Step S75: If the cohesion of the current soil strip is equal to the cohesion of the next soil strip, the cohesion of the current soil strip remains unchanged.

[0146] Step S76: if the cohesion of the current soil strip is not equal to the cohesion of the next soil strip, the cohesion of the current soil strip is equal to the average value of the cohesion of the current soil strip and the cohesion of the next soil strip.

[0147] Preferably, the following soil strip number is used To assist in the explanation, Cohesion of soil strips With Cohesion of soil strips If they are not equal, then Cohesion of soil strips .

[0148] Step S77, defining the sliding surface length, internal friction angle, and cohesion of the current soil strip as shear strength indicators of the current soil strip.

[0149] Further, step S8, based on the shear strength index of each soil strip, the safety factor of each soil strip under the limit equilibrium state is calculated by the Swedish strip method, including:

[0150] Step S81, calculate the safety factor of each soil strip by formula (3):

[0151] (3).

[0152] in, For the The safety factor of each soil strip is is the total number of all soil strips, For the The cohesion of a soil strip, For the The sliding surface length of a soil strip, For the The gravity of a soil bar, For the The sliding surface inclination of the soil strip, For the The internal friction angle of a soil strip.

[0153] Further, in step S9, the minimum value of all safety factors is obtained, and the soil strip corresponding to the minimum value is marked as the most likely point of landslide occurrence, and then, the following steps are included:

[0154] Step S10: Output each soil strip and the safety factor of each soil strip on the geological drawing and send them to an external visualization terminal.

[0155] Step S20: Mark the soil strip corresponding to the minimum value in red on the external visualization terminal.

[0156] This embodiment obtains historical monitoring data of each inclinometer hole based on several historical moments respectively; obtains historical displacement vectors of all historical monitoring data by drilling inclinometer method; marks the inclinometer hole whose historical displacement vector exceeds a preset displacement threshold as a deformation mutation inclinometer hole, and obtains historical coordinate points of each deformation mutation inclinometer hole respectively; learns and trains all historical coordinate points of all historical moments by an extreme learning machine to obtain a coordinate point prediction model; predicts the future coordinate points of each deformation mutation inclinometer hole based on a preset number of steps by the coordinate point prediction model; vertically divides the monitoring section into several soil strips based on all future coordinate points along the direction from the leading edge point to the trailing edge point on the monitoring section; obtains the shear strength index of each soil strip based on the geological drawings of the monitoring section; calculates the safety factor of each soil strip under the limit equilibrium state by the Swedish strip method based on the shear strength index of each soil strip; obtains the minimum value of all safety factors, and marks the soil strip corresponding to the minimum value as the most likely point for landslide occurrence. Based on the traditional Swedish strip method, this embodiment utilizes the characteristic of the extreme learning machine that the learning process only requires one forward propagation and does not require iterative updating of weights to iterate the monitoring data collected from the incline hole, so that the subsequent Swedish strip method calculation results are more accurate. At the same time, the prediction function of the monitoring data is also realized, effectively reducing the problem of inaccurate calculation of the traditional Swedish strip method.

[0157] like Figure 2 As shown, this embodiment provides an embodiment of a device for analyzing landslide stability. In this embodiment, the analysis device is applied to the analysis method as in the above embodiment.

[0158] Specifically, the analysis device includes an inclinometer hole historical monitoring data acquisition module 1, a historical displacement vector acquisition module 2, a deformation mutation inclinometer hole marking module 3, a coordinate point prediction model acquisition module 4, a future coordinate point prediction acquisition module 5, a soil strip division module 6, a shear strength index acquisition module 7, a safety factor calculation module 8, and a landslide most likely occurrence point marking module 9, which are electrically connected in sequence.

[0159] Among them, the inclinometer hole historical monitoring data acquisition module 1 is used to respectively obtain the historical monitoring data of each inclinometer hole based on several historical moments; the historical displacement vector acquisition module 2 is used to obtain the historical displacement vectors of all historical monitoring data through the drilling inclinometer method; the deformation mutation inclinometer hole marking module 3 is used to mark the inclinometer hole whose historical displacement vector exceeds the preset displacement threshold as the deformation mutation inclinometer hole, and respectively obtain the historical coordinate points of each deformation mutation inclinometer hole; the coordinate point prediction model acquisition module 4 is used to learn and train all historical coordinate points of all historical moments through the extreme learning machine to obtain the coordinate point prediction model; the future coordinate point prediction acquisition module 5 is used to obtain the future coordinate point prediction model through the coordinate point prediction model The future coordinate points of each deformation mutation inclinometer hole based on a preset number of steps are predicted respectively; the soil strip division module 6 is used to divide the monitoring section vertically into a number of soil strips based on all future coordinate points along the direction from the leading edge point to the trailing edge point on the monitoring section; the shear strength index acquisition module 7 is used to obtain the shear strength index of each soil strip based on the geological drawings of the monitoring section; the safety factor calculation module 8 is used to calculate the safety factor of each soil strip under the limit equilibrium state based on the shear strength index of each soil strip by the Swedish strip division method; the landslide most likely occurrence point marking module 9 is used to obtain the minimum value of all safety factors, and mark the soil strip corresponding to the minimum value as the most likely landslide occurrence point.

[0160] Furthermore, the coordinate point prediction model acquisition module 4 specifically includes a first coordinate point prediction model acquisition submodule, a second coordinate point prediction model acquisition submodule, a third coordinate point prediction model acquisition submodule, a fourth coordinate point prediction model acquisition submodule, and a fifth coordinate point prediction model acquisition submodule, which are electrically connected in sequence; the first coordinate point prediction model acquisition submodule is electrically connected to the deformation mutation inclination hole marking module 3, and the fifth coordinate point prediction model acquisition submodule is electrically connected to the future coordinate point prediction acquisition module 5.

[0161] Among them, the first coordinate point prediction model acquisition submodule is used to package all historical coordinate points of the same deformation mutation inclinometer hole into a coordinate data set; the second coordinate point prediction model acquisition submodule is used to normalize all coordinate data sets respectively, and obtain a normalized data set based on one coordinate data set; the third coordinate point prediction model acquisition submodule is used to divide the current normalized data set into a training set and a verification set according to a preset ratio; the fourth coordinate point prediction model acquisition submodule is used to define an extreme learning model according to the data dimension of the training set; the fifth coordinate point prediction model acquisition submodule is used to train and learn the extreme learning model through the training set to obtain the coordinate point prediction model of the current deformation mutation inclinometer hole.

[0162] Furthermore, the fifth coordinate point prediction model acquisition submodule specifically includes a first punctuation prediction model acquisition unit, a second punctuation prediction model acquisition unit, a third punctuation prediction model acquisition unit, a fourth punctuation prediction model acquisition unit, a fifth punctuation prediction model acquisition unit, and a sixth punctuation prediction model acquisition unit, which are electrically connected in sequence; the first punctuation prediction model acquisition unit is electrically connected to the fourth coordinate point prediction model acquisition submodule, and the sixth punctuation prediction model acquisition unit is electrically connected to the future coordinate point prediction acquisition module 5.

[0163] The first punctuation prediction model acquisition unit is used to define the training set as .

[0164] in, , For the The historical coordinate value of a historical moment, for of A collection of dimensional vectors, , For the The future coordinate value of the preset number of steps, for of A collection of dimensional vectors, with superscripts is the transposed matrix.

[0165] The second punctuation prediction model acquisition unit is used to define the extreme learning model according to formula (1):

[0166] (1).

[0167] in, , is the first connection weight between the hidden layer and the output layer of the extreme learning model; is the number of nodes in the hidden layer; is the output value of the hidden layer, ; ; is the activation function; is the input layer and the hidden layer of the extreme learning model The second connection weight of the nodes; The hidden layer The threshold of nodes.

[0168] The third punctuation prediction model acquisition unit is used to randomly generate the second connection weight and threshold through a nonlinear mapping function in a feature space based on the extreme learning model.

[0169] The fourth punctuation prediction model acquisition unit is used to obtain the output value of the hidden layer through the randomly generated second connection weight and threshold calculation.

[0170] The fifth punctuation prediction model acquisition unit is used to solve the optimal solution of the first connection weight by minimizing the approximate square error according to formula (2):

[0171] (2).

[0172] in, is the output matrix of the hidden layer; is the target matrix of the training samples; is the optimal solution; The output matrix The MP generalized inverse matrix, , The dimension is The set of weight vectors .

[0173] The sixth punctuation prediction model acquisition unit is used to obtain the optimal solution and the second connection weight and threshold corresponding to the optimal solution and substitute them into the extreme learning model to obtain the prediction model.

[0174] Furthermore, the shear strength index acquisition module 7 specifically includes a first shear strength index acquisition submodule, a second shear strength index acquisition submodule, a third shear strength index acquisition submodule, a fourth shear strength index acquisition submodule, a fifth shear strength index acquisition submodule, a sixth shear strength index acquisition submodule, and a seventh shear strength index acquisition submodule, which are electrically connected in sequence; the first shear strength index acquisition submodule is electrically connected to the soil strip division module 6, and the seventh shear strength index acquisition submodule is electrically connected to the safety factor calculation module 8.

[0175] Among them, the first shear strength index acquisition submodule is used to calculate the sliding surface length of each soil strip according to the Euclidean distance; the second shear strength index acquisition submodule is used to find the internal friction angle and cohesion of each soil strip based on the geological map; the third shear strength index acquisition submodule is used to if the internal friction angle of the current soil strip is equal to the internal friction angle of the next soil strip, then the internal friction angle of the current soil strip remains unchanged; the fourth shear strength index acquisition submodule is used to if the internal friction angle of the current soil strip is not equal to the internal friction angle of the next soil strip, then the internal friction angle of the current soil strip is equal to the internal friction angle of the current soil strip The fifth shear strength index acquisition submodule is used for if the cohesion of the current soil strip is equal to the cohesion of the next soil strip, the cohesion of the current soil strip remains unchanged; the sixth shear strength index acquisition submodule is used for if the cohesion of the current soil strip is not equal to the cohesion of the next soil strip, the cohesion of the current soil strip is equal to the average value of the cohesion of the current soil strip and the cohesion of the next soil strip; the seventh shear strength index acquisition submodule is used to define the sliding surface length, internal friction angle and cohesion of the current soil strip as the shear strength index of the current soil strip.

[0176] Furthermore, the safety factor calculation module 8 is specifically used to calculate the safety factor of each soil strip by formula (3):

[0177] (3).

[0178] in, For the The safety factor of each soil strip is is the total number of all soil strips, For the The cohesion of a soil strip, For the The sliding surface length of a soil strip, For the The gravity of a soil bar, For the The sliding surface inclination of the soil strip, For the The internal friction angle of a soil strip.

[0179] Furthermore, the analysis device also includes a soil strip and safety factor sending module and a minimum soil strip red marking module which are electrically connected in sequence; the soil strip and safety factor sending module is electrically connected to the landslide most likely occurrence point marking module 9.

[0180] Among them, the soil strip and safety factor sending module is used to output each soil strip and the safety factor of each soil strip on the geological drawing, and send them to the external visualization terminal; the minimum value soil strip red marking module is used to mark the soil strip corresponding to the minimum value in red on the external visualization terminal.

[0181] It should be noted that this embodiment is a functional module embodiment based on the above method embodiment. The optimization, expansion, limitation, example, and principle description of this embodiment can be referred to the above embodiment, and this embodiment will not be repeated.

[0182] This embodiment obtains historical monitoring data of each inclinometer hole based on several historical moments respectively; obtains historical displacement vectors of all historical monitoring data by drilling inclinometer method; marks the inclinometer hole whose historical displacement vector exceeds a preset displacement threshold as a deformation mutation inclinometer hole, and obtains historical coordinate points of each deformation mutation inclinometer hole respectively; learns and trains all historical coordinate points of all historical moments by an extreme learning machine to obtain a coordinate point prediction model; predicts the future coordinate points of each deformation mutation inclinometer hole based on a preset number of steps by the coordinate point prediction model; vertically divides the monitoring section into several soil strips based on all future coordinate points along the direction from the leading edge point to the trailing edge point on the monitoring section; obtains the shear strength index of each soil strip based on the geological drawings of the monitoring section; calculates the safety factor of each soil strip under the limit equilibrium state by the Swedish strip method based on the shear strength index of each soil strip; obtains the minimum value of all safety factors, and marks the soil strip corresponding to the minimum value as the most likely point for landslide occurrence. Based on the traditional Swedish strip method, this embodiment utilizes the characteristic of the extreme learning machine that the learning process only requires one forward propagation and does not require iterative updating of weights to iterate the monitoring data collected from the incline hole, so that the subsequent Swedish strip method calculation results are more accurate. At the same time, the prediction function of the monitoring data is also realized, effectively reducing the problem of inaccurate calculation of the traditional Swedish strip method.

[0183] Figure 3 Schematic diagram of the structure of an electronic device according to an embodiment of the present application. Figure 3 As shown, the electronic device 10 includes a processor 101 and a memory 102 coupled to the processor 101 .

[0184] The memory 102 stores program instructions for implementing the landslide stability analysis method of any of the above embodiments.

[0185] The processor 101 is used to execute program instructions stored in the memory 102 to analyze the stability of the landslide.

[0186] The processor 101 may also be referred to as a CPU (Central Processing Unit). The processor 101 may be an integrated circuit chip having a signal processing capability. The processor 101 may also be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0187] Further, Figure 4 This is a schematic diagram of the structure of a storage medium according to an embodiment of the present application. Figure 4The storage medium 11 of the embodiment of the present application stores program instructions 111 that can implement all the above methods, wherein the program instructions 111 can be stored in the above storage medium in the form of a software product, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes, or terminal devices such as computers, servers, mobile phones, tablets, etc.

[0188] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0189] In addition, each functional unit in each embodiment of the present application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. The above is only an implementation method of the present application, and does not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the specification and drawings of this application, or directly or indirectly used in other related technical fields, is also included in the patent protection scope of the present application.

[0190] The specific implementation methods of the present application are described in detail above, but they are only examples, and the present application is not limited to the specific implementation methods described above. For those skilled in the art, any equivalent modification or substitution of the present application is also within the scope of the present application, and therefore, the equalization, modification, and improvement made without departing from the spirit and principle of the present application should be included in the scope of the present application.

Claims

1. A landslide stability analysis method, the analysis method is based on a number of inclinometer holes reserved on the monitoring section, characterized in that: The analysis method comprises: Step S1, respectively obtaining historical monitoring data of each inclinometer hole based on several historical moments; Step S2, obtaining the historical displacement vectors of all historical monitoring data by means of a borehole inclinometer method; Step S3, marking the inclinometer holes whose historical displacement vectors exceed a preset displacement threshold as deformation mutation inclinometer holes, and obtaining the historical coordinate points of each deformation mutation inclinometer hole respectively; Step S4, learning and training all historical coordinate points of all historical moments by an extreme learning machine to obtain a coordinate point prediction model; Step S5, predicting the future coordinate points of each deformation mutation inclinometer hole based on a preset number of steps by using the coordinate point prediction model; Step S6, vertically dividing the monitoring section into a plurality of soil strips based on all future coordinate points in the direction from the leading edge point to the trailing edge point on the monitoring section; Step S7, obtaining the shear strength index of each soil strip based on the geological drawing of the monitoring section; Step S8, calculating the safety factor of each soil strip under the limit equilibrium state by the Swedish strip method based on the shear strength index of each soil strip; Step S9, obtaining the minimum value of all safety factors, and marking the soil strip corresponding to the minimum value as the most likely point for landslide to occur.

2. The analysis method according to claim 1, characterized in that Step S4, learning and training all historical coordinate points of all historical moments by an extreme learning machine to obtain a coordinate point prediction model, including: Step S41, packing all historical coordinate points of the same deformation mutation inclination measuring hole into a coordinate data set; Step S42, normalizing all coordinate data sets respectively to obtain a normalized data set based on one coordinate data set; Step S43, dividing the current normalized data set into a training set and a validation set according to a preset ratio; Step S44, defining an extreme learning model according to the data dimension of the training set; Step S45, training and learning the extreme learning model through the training set to obtain a coordinate point prediction model of the current deformation mutation inclination hole.

3. The analysis method according to claim 2, characterized in that Step S45, training and learning the extreme learning model through the training set to obtain a coordinate point prediction model of the current deformation mutation inclination hole, including: Step S451, define the training set as ; in, , For the The historical coordinate value of a historical moment, for of A collection of dimensional vectors, , For the The future coordinate value of the preset number of steps, for of A collection of dimensional vectors, with superscript is the transposed matrix; Step S452, defining the extreme learning model according to formula (1): （1）； in, , is the first connection weight between the hidden layer and the output layer of the extreme learning model; is the number of nodes in the hidden layer; is the output value of the hidden layer, ; ; is the activation function; is the input layer of the extreme learning model and the hidden layer The second connection weight of the nodes; is the hidden layer The threshold of nodes; Step S453, randomly generating the second connection weight and the threshold value by a nonlinear mapping function in a feature space based on the extreme learning model; Step S454, obtaining the output value of the hidden layer by calculating the randomly generated second connection weight and threshold; Step S455: according to formula (2), the optimal solution of the first connection weight is obtained by minimizing the approximate square error: （2）； in, is the output matrix of the hidden layer; is the target matrix of the training set; is the optimal solution; The output matrix The MP generalized inverse matrix, , The dimension is The set of weight vectors of ; Step S456, obtaining the optimal solution and the second connection weight and threshold corresponding to the optimal solution and substituting them into the extreme learning model to obtain the prediction model.

4. The analysis method according to claim 1, characterized in that Step S7, obtaining the shear strength index of each soil strip based on the geological drawing of the monitoring section, including: Step S71, calculating the sliding surface length of each soil strip according to the Euclidean distance; Step S72, searching for the internal friction angle and cohesion of each soil strip based on the geological drawing; Step S73, if the internal friction angle of the current soil strip is equal to the internal friction angle of the next soil strip, the internal friction angle of the current soil strip remains unchanged; Step S74, if the internal friction angle of the current soil strip is not equal to the internal friction angle of the next soil strip, then the internal friction angle of the current soil strip is equal to the average value of the internal friction angle of the current soil strip and the internal friction angle of the next soil strip; Step S75, if the cohesion of the current soil strip is equal to the cohesion of the next soil strip, the cohesion of the current soil strip remains unchanged; Step S76, if the cohesion of the current soil strip is not equal to the cohesion of the next soil strip, the cohesion of the current soil strip is equal to the average of the cohesion of the current soil strip and the cohesion of the next soil strip; Step S77, defining the sliding surface length, internal friction angle, and cohesion of the current soil strip as shear strength indicators of the current soil strip.

5. The analysis method according to claim 4, characterized in that Step S8, based on the shear strength index of each soil strip, the safety factor of each soil strip under the limit equilibrium state is calculated by the Swedish strip method, including: Step S81, calculate the safety factor of each soil strip by formula (3): （3）； in, For the The safety factor of each soil strip is is the total number of all soil strips, For the The cohesion of a soil strip, For the The sliding surface length of a soil strip, For the The gravity of a bar of soil, For the The sliding surface inclination of the soil strip, For the The internal friction angle of a soil strip.

6. The analysis method according to claim 4, characterized in that Step S9, obtaining the minimum value of all safety factors, and marking the soil strip corresponding to the minimum value as the most likely point for landslide to occur, and then comprising: Step S10, outputting each soil strip and the safety factor of each soil strip on the geological drawing, and sending them to an external visualization terminal; Step S20: marking the soil strip corresponding to the minimum value in red on an external visualization terminal.

7. A device for analyzing landslide stability, the device for analyzing landslide stability being applied to the method for analyzing landslide stability as claimed in any one of claims 1 to 6, characterized in that: The landslide stability analysis device comprises: The module for acquiring historical monitoring data of the inclinometer hole is used to respectively acquire the historical monitoring data of each inclinometer hole based on several historical moments; A historical displacement vector acquisition module is used to obtain historical displacement vectors of all historical monitoring data through a borehole inclinometer method; A deformation mutation inclinometer hole marking module is used to mark the inclinometer holes whose historical displacement vectors exceed a preset displacement threshold as deformation mutation inclinometer holes, and obtain the historical coordinate points of each deformation mutation inclinometer hole respectively; A coordinate point prediction model acquisition module is used to learn and train all historical coordinate points at all historical moments through an extreme learning machine to obtain a coordinate point prediction model; A future coordinate point prediction acquisition module, used to predict the future coordinate points of each deformation mutation inclinometer hole based on a preset number of steps through the coordinate point prediction model; A soil strip division module, used to divide the monitoring section vertically into a plurality of soil strips based on all future coordinate points in the direction from the leading edge point to the trailing edge point on the monitoring section; A shear strength index acquisition module, used to respectively acquire the shear strength index of each soil strip based on the geological drawings of the monitoring section; A safety factor calculation module is used to calculate the safety factor of each soil strip under the limit equilibrium state by Swedish strip method based on the shear strength index of each soil strip; The most likely landslide occurrence point marking module is used to obtain the minimum value of all safety factors and mark the soil strip corresponding to the minimum value as the most likely landslide occurrence point.

8. An electronic device, characterized in that: It comprises a processor and a memory coupled to the processor, wherein the memory stores program instructions executable by the processor; when the processor executes the program instructions stored in the memory, the landslide stability analysis method as described in any one of claims 1 to 6 is implemented.

9. A storage medium, characterized in that: The storage medium stores program instructions, and when the program instructions are executed by the processor, the landslide stability analysis method according to any one of claims 1 to 6 can be implemented.

Images