CN114692441B - A loess landslide stability prediction method, electronic equipment and storage medium - Google Patents

Abstract

The application provides a loess landslide stability prediction method, electronic equipment and a storage medium, and belongs to the field of stability prediction. The method comprises the following steps: acquiring shear strength parameters of a target area, and constructing a shear strength parameter mean error function; constructing a sliding surface position error function according to sliding surface observation information of first destabilization of a target sliding surface and a critical sliding surface determined by finite difference numerical simulation calculation; calculating a landslide stability coefficient through finite difference strength folding and subtracting, and constructing a landslide stability coefficient error function; combining the three constructed error functions, and carrying out multiple iterations by using a genetic algorithm based on multiple groups of random sampling samples to obtain an optimal parameter group; and constructing a simulated sliding surface according to other parameters of the target landslide and the optimal parameter set, and predicting a stability prediction result of the target landslide under the condition that the consistency characterization is correct. The present application aims to obtain a more reliable inverse analysis result and to predict the stability of landslide more accurately.

Description

A loess landslide stability prediction method, electronic equipment and storage medium

technical field

The embodiments of the present application relate to the technical field of stability prediction, and in particular, relate to a loess landslide stability prediction method, electronic equipment, and a storage medium.

Background technique

With the rapid development of the economy, human activities such as engineering construction, agricultural irrigation and resource development are becoming more and more frequent, but they are also accompanied by the frequent occurrence of landslide geological disasters in the process of development. Landslide geological disasters reflect a large number, wide distribution, Therefore, in order to protect the safety of people's lives and property, the risk assessment and prevention and control of landslides are particularly important.

Obtaining accurate and reliable physical and mechanical calculation parameters is the basic prerequisite for the evaluation of landslide stability. At present, a large number of parameters for the study of landslide stability have been obtained through a large number of experimental studies, such as the residual strength parameters of the sliding zone soil, different confining pressure As well as the shear strength of the saturated remolded loess at the shear rate, these test results provide important data support for the study of the stability of loess landslides, but there are many uncertain factors in the laboratory test research, and it is difficult to obtain the physical and mechanical parameters of the undisturbed loess. At the same time, due to the problem of scale effect, it is difficult for experimental research results to characterize the parameter characteristics of real landslide hazards.

Carrying out parameter back analysis based on landslide disaster observation information provides an important way to obtain reliable loess landslide calculation parameters. Based on the limit equilibrium theory, the display expression of the parameter inversion is deduced through the discrimination of the critical arc sliding surface, and the inversion analysis of the slope stability parameters is carried out; based on the three-dimensional upper limit analysis theory, the reliability inversion of the shear strength parameters of the landslide is studied. Analytical method.

Although the above research results have improved the accuracy of landslide stability calculation parameters to a certain extent, most of them only consider the stability coefficient or deformation characteristics as constraints, which makes the reliability of the back analysis results unsatisfactory. Therefore, how to It is an urgent problem to improve the parametric results of the back analysis and use the parametric results of the back analysis to predict the stability of the landslide.

Contents of the invention

The embodiments of the present application provide a loess landslide stability prediction method, electronic equipment and a storage medium, aiming at obtaining more reliable back analysis results and more accurately predicting the stability of the landslide.

In the first aspect, the embodiment of the present application provides a method for predicting the stability of loess landslides, the method comprising:

Obtain the shear strength parameters of the target area, the shear strength parameters include the effective cohesion of multiple groups of natural loess, the effective internal friction angles of multiple groups of natural loess, the effective cohesion of multiple groups of saturated loess, and the effective cohesion of multiple groups of saturated loess According to the data of the effective internal friction angle, construct the shear strength parameter mean value error function according to the shear strength parameter;

According to the observation information of the sliding surface of the first instability of the target landslide and the critical sliding surface determined by the finite difference numerical simulation calculation, the sliding surface position error function is constructed;

The shear strength parameter is sampled, and according to the cohesion data obtained by sampling and the internal friction angle data, the landslide stability coefficient is calculated by the finite difference strength reduction method, and the error function of the landslide stability coefficient is constructed;

In combination with the error function of the mean value of the shear strength parameter, the error function of the position of the sliding surface and the error function of the coefficient of stability of the landslide, based on multiple groups of random sampling samples, the genetic algorithm is used to perform multiple iterations to obtain the optimal parameter group, wherein, The optimal parameter group includes the effective cohesion of natural loess, the effective internal friction angle of natural loess, the effective cohesion of saturated loess and the effective internal friction angle of saturated loess;

According to other parameters of the target landslide and the optimal parameter group, construct a simulated sliding surface in the Flac model, and compare the consistency of the first unstable sliding surface observation information of the simulated sliding surface and the target landslide;

When the consistency characterization is correct, the stability prediction result of the target landslide is obtained through the Flac model, and the prediction result includes the stability coefficient and sliding surface information of subsequent instability.

Optionally, constructing a shear strength parameter mean error function according to the shear strength parameters, including:

Calculate the effective cohesion of multiple groups of natural loess, the effective internal friction angle of multiple groups of natural loess, the effective cohesion of multiple groups of saturated loess, and the effective internal friction angle of multiple groups of saturated loess respectively. The mean and standard deviation obey the normal distribution;

According to the calculated mean and standard deviation, the error function of the mean value of the shear strength parameters is constructed.

Optionally, the constructed mean error function of the shear strength parameter is:

表示各参数的协方差矩阵的逆矩阵，T表示矩阵转置。In the formula, x is the shear strength parameter vector, μ _x is the mean value of each parameter,

Represents the inverse matrix of the covariance matrix of each parameter, and T represents the matrix transpose.

Optionally, according to the observation information of the sliding surface of the first instability of the target landslide and the critical sliding surface determined by the finite difference numerical simulation calculation, the sliding surface position error function is constructed, including:

According to the observation information of the sliding surface of the first instability of the target landslide, a plurality of characteristic coordinate points on the sliding surface of the first instability are selected;

Determining a plurality of critical sliding surfaces through multiple finite difference numerical simulation calculations, and obtaining a plurality of characteristic coordinate points on each of the critical sliding surfaces;

A sliding surface position error function is constructed according to the multiple characteristic coordinate points on the sliding surface of the first instability and the multiple characteristic coordinate points on each of the critical sliding surfaces.

Optionally, according to the observation information of the sliding surface of the first instability of the target landslide and the critical sliding surface determined by the finite difference numerical simulation calculation, the sliding surface position error function is constructed, including:

According to the observation information of the sliding surface of the first instability of the target landslide, select 5 characteristic coordinate points on the sliding surface of the first instability, denoted as (a _i , b _i ), where i=1, 2, 3 , 4, 5;

Multiple critical sliding surfaces are determined through multiple finite difference numerical simulation calculations, and five characteristic coordinate points on each critical sliding surface are obtained, denoted as (xi _, y _i ), where i=1, 2, 3, 4 , 5;

According to the 5 characteristic coordinate points on the sliding surface of the first instability and the 5 characteristic coordinate points on the critical sliding surface, the position error function of the sliding surface is constructed as follows:

Optionally, the shear strength parameters are sampled, and according to the cohesion data and internal friction angle data obtained by sampling, the landslide stability coefficient is calculated by the finite difference strength reduction method, and the error function of the landslide stability coefficient is constructed, including :

Sampling the shear strength parameters, and calculating the landslide stability coefficient by the finite difference strength reduction method according to the cohesion data and internal friction angle data obtained by sampling;

Assuming that the stability factor when the target landslide is unstable is 1;

The constructed landslide stability coefficient error function is:

f ₃ (x)=(FS(x)-1)

In the formula, FS is the landslide stability coefficient calculated by the finite difference strength reduction method.

In the second aspect, the embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and operable on the processor, and the computer program is implemented when the processor executes the computer program. Embodiment The method described in the first aspect.

In a third aspect, an embodiment of the present application provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the method described in the first aspect of the embodiment is implemented.

Beneficial effect:

By constructing three functions, namely, the error function of the mean value of the shear strength parameters, the error function of the sliding surface position and the error function of the landslide stability coefficient, and combining the genetic algorithm, iteratively calculates multiple groups of random sampling samples to obtain the optimal parameter set. The optimal parameter set includes the optimal natural loess effective cohesion, natural loess effective internal friction angle, saturated loess effective cohesion and saturated loess effective internal friction angle; then using the optimal parameter set data and the target landslide other parameters, construct a simulated sliding surface in the Flac model, and compare the consistency of the simulated sliding surface with the actual observation information of the first unstable sliding surface of the target landslide. The stability prediction result of the target landslide, the prediction result includes the stability coefficient and sliding surface information of the subsequent instability.

This method takes into account the observation information of the sliding surface, the shear strength parameter information and the stability coefficient at the first instability of the landslide, and sets more optimization constraints for the back analysis by setting three functions, so as to obtain a more reliable The back analysis results can then effectively predict the stability state and critical sliding surface of the target landslide.

Description of drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the accompanying drawings that need to be used in the description of the embodiments of the present application. Obviously, the accompanying drawings in the following description are only some embodiments of the present application , for those skilled in the art, other drawings can also be obtained according to these drawings without paying creative labor.

Fig. 1 is the step flowchart of the loess landslide stability prediction method that an embodiment of the present application proposes;

Fig. 2 is a schematic diagram of the sliding surface of the destabilized target landslide proposed by an embodiment of the present application;

Fig. 3 is the flowchart of the anti-analysis process that an embodiment of the present application proposes;

Fig. 4 is the landslide vertical profile that an embodiment of the present application proposes;

Fig. 5 is the simulated result when the Dangchuan 2# landslide that an embodiment of the application proposes does not take place landslide;

Fig. 6 is a schematic diagram of the convergence of the genetic algorithm proposed by an embodiment of the present application;

Fig. 7 is a schematic diagram of the Pareto multi-objective optimization result proposed by an embodiment of the present application;

Fig. 8 is a schematic diagram of the comparison between the simulated sliding surface and the actual sliding surface in the first instability proposed by an embodiment of the present application;

Fig. 9 is a schematic diagram of the predicted sliding surface in the second instability proposed by an embodiment of the present application;

Fig. 10 (a) is a schematic diagram of the instability range obtained by the first round of simulation of the second instability proposed by an embodiment of the present application;

Fig. 10(b) is a schematic diagram of the instability range obtained by the second round of simulation of the second instability proposed by an embodiment of the present application;

Fig. 10(c) is a schematic diagram of the instability range obtained from the third round of simulation of the second instability proposed by an embodiment of the present application.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of this application.

In order to protect the safety of people's lives and property, the risk assessment and prevention and control of landslides are particularly important. In order to obtain more reliable back analysis results and effectively predict the stability state and critical sliding surface of landslides, this application proposes a A method for predicting the stability of loess landslides.

With reference to Fig. 1, the step flowchart of a kind of loess landslide stability prediction method in the embodiment of the present invention is shown, and as Fig. 1, described method comprises the following steps:

S101. Obtain the shear strength parameters of the target area, the shear strength parameters include the effective cohesion of multiple groups of natural loess, the effective internal friction angles of multiple groups of natural loess, the effective cohesion of multiple groups of saturated loess, and the effective cohesion of multiple groups of natural loess. The data of the effective internal friction angle of the saturated loess is used to construct the mean error function of the shear strength parameter according to the shear strength parameter.

First, obtain the shear strength parameters of the area where the landslide is located for back analysis and stability prediction, including the effective cohesion of multiple groups of natural loess, the effective internal friction angle of multiple groups of natural loess, and the effective cohesion of multiple groups of saturated loess. force and the effective internal friction angle data of multiple groups of saturated loess, the acquisition method is not limited in this application, for example, the effective cohesion of 27 groups of natural loess, the effective internal friction of 27 groups of natural loess can be obtained by consulting the literature angle, the effective cohesion of 18 groups of saturated loess and the effective internal friction angle of 18 groups of saturated loess, and the obtained shear strength parameters constitute a data set.

In one implementation, this embodiment also provides a method for constructing a mean error function of the shear strength parameter, including:

Calculate the effective cohesion of multiple groups of natural loess, the effective internal friction angle of multiple groups of natural loess, the effective cohesion of multiple groups of saturated loess, and the effective internal friction angle of multiple groups of saturated loess respectively.

According to the calculated mean and standard deviation, construct the mean error function of the shear strength parameters;

Exemplarily, the constructed mean error function of the shear strength parameter is:

表示各参数的协方差矩阵的逆矩阵，T表示矩阵转置。In the formula, x is the shear strength parameter vector, μ _x is the mean value of each parameter,

Represents the inverse matrix of the covariance matrix of each parameter, and T represents the matrix transpose.

Since the shear strength parameters include the above four parameters, and the influence of the mean value of each parameter is different, the unbiased model can also be used to reduce the influence of the mean value of each parameter.

S102. Construct a sliding surface position error function according to the sliding surface observation information of the first instability of the target landslide and the critical sliding surface determined through the finite difference numerical simulation calculation.

In a feasible implementation manner, this step may include the following steps:

S1: According to the observation information of the sliding surface of the first instability of the target landslide, select multiple characteristic coordinate points on the sliding surface of the first instability;

S2: Determining a plurality of critical sliding surfaces through multiple finite difference numerical simulation calculations, and obtaining a plurality of characteristic coordinate points on each of the critical sliding surfaces;

In this step, according to multiple sets of parameter data in each iteration of the genetic algorithm, multiple finite difference numerical simulations are performed to obtain multiple critical sliding surfaces, and then multiple characteristic coordinate points are obtained on each critical sliding surface. The number of characteristic coordinate points obtained on the critical sliding surface is the same as the number of characteristic coordinate points on the first unstable sliding surface.

S3: Construct a sliding surface position error function according to the multiple characteristic coordinate points on the first unstable sliding surface and the multiple characteristic coordinate points on the critical sliding surface.

Referring to Fig. 2, it shows a schematic diagram of the sliding surface of the target landslide that has become unstable. As an example, in the actual implementation process, the sliding surface is equally divided in the x direction according to the observation information of the sliding surface of the first instability of the target landslide For four segments, 5 characteristic coordinate points can be obtained as (a _i , b _i ), where i=1, 2, 3, 4, 5, and the position information of these 5 coordinate points is used as the position of the actual sliding surface Information; among them, the leftmost characteristic coordinate point is the position of the shear outlet, and the rightmost characteristic coordinate point is the position of the trailing edge tension crack.

Multiple critical sliding surfaces are determined through multiple finite difference numerical simulation calculations, and five characteristic coordinate points are selected for each critical sliding surface, denoted as (xi _, y _i ), where i=1, 2, 3, 4 , 5.

Then, according to the 5 characteristic coordinate points on the sliding surface of the first instability and the 5 characteristic coordinate points on the critical sliding surface, the position error function of the sliding surface is constructed as follows:

In other implementation manners, other numerical values may also be set for i.

This embodiment uses Flac to extract the critical sliding surface. Specifically, in order to obtain the critical sliding surface of the loess landslide under given parameter conditions, the unit node displacement information is extracted through Flac built-in Fish language programming, and imported into MATLAB to use the K-means clustering algorithm Identify the sliding unit body, so as to realize the extraction of the critical sliding surface.

The goal of the K-means algorithm is to gather n objects into k specified clusters according to the similarity between objects, and each object belongs to and only belongs to one cluster with the smallest distance to the center of the cluster. The Euclidean distance from each object to each cluster center is as follows:

In the formula, A _i represents the i-th object, C _j represents the j-th clustering center; A _i,t represents the t-th attribute of the i-th object, and C _j,t represents the t-th property of the j-th clustering center attributes, t=1, 2,..., m.

Compare the distance from each object to each cluster center in turn, assign the object to the cluster with the closest distance to the cluster center, and obtain k clusters {S1, S2, ..., Sk}. This application only considers the node displacement attribute, so t=1; its attributes can be divided into node displacement and no node displacement, ie k=2.

S103. Sampling the shear strength parameters, calculating the landslide stability coefficient through the finite difference strength reduction method according to the cohesion data and internal friction angle data obtained by sampling, and constructing an error function of the landslide stability coefficient.

The shear strength parameters obtained in step S101 include effective cohesion of multiple groups of natural loess, effective internal friction angles of multiple groups of natural loess, effective cohesion of multiple groups of saturated loess, and effective internal friction of multiple groups of saturated loess The corner data is randomly sampled to obtain a random sample of the genetic algorithm.

In one embodiment, according to the cohesion data and internal friction angle data obtained by random sampling, the slope stability analysis is performed using the strength reduction method that comes with the FLAC finite difference software, and the shear strength parameters are reduced by the following formula , until the target landslide reaches the limit equilibrium state, the reduction factor at this time is the stability factor FS of the target landslide, and the specific formula is as follows:

c _r =c/FS

为内摩擦角数据；c_r折减后的土体黏聚力；和

为折减后的内摩擦角。In the formula, c is cohesion data;

is the internal friction angle data; the soil cohesion after reduction of c _r ; and

is the reduced internal friction angle.

Assuming that the stability coefficient of the target landslide is 1, the constructed landslide stability coefficient error function is:

f ₃ (x)=(FS(x)-1)

In the formula, FS is the landslide stability coefficient calculated by the finite difference strength reduction method.

In this method, the three error functions are written into the genetic algorithm in advance, and the corresponding parameters are brought into the error function during the execution of the genetic algorithm.

S104. Combining the error function of the mean value of the shear strength parameter, the error function of the position of the sliding surface, and the error function of the coefficient of stability of the landslide, multiple iterations are performed using a genetic algorithm based on multiple groups of random sampling samples to obtain an optimal parameter set, Among them, the optimal parameter group includes the effective cohesion of natural loess, the effective internal friction angle of natural loess, the effective cohesion of saturated loess and the effective internal friction angle of saturated loess.

The genetic algorithm used in this embodiment is the NSGA-II genetic algorithm, which is a fast non-dominated multi-objective optimization genetic algorithm based on the Pareto optimal solution and with an elite retention strategy. It is mainly divided into the following three calculation parts:

1. Pareto solution set fast non-dominated sorting;

2. Calculate the congestion degree and the congestion degree comparison operator;

3. Cross-iteration, survival of the fittest and selection of optimal parameter points by elite retention strategy.

By repeating the above three steps, the target number of iterative calculations or convergence conditions are finally reached, and the optimal Pareto solution set is obtained and output.

Referring to Fig. 3, it shows the flow chart of the reverse analysis process that the embodiment of the present application provides, first constructs 100 groups of random sampling samples, wherein includes four sample parameters in each group of random sampling samples, including the effective cohesion of natural loess, The effective internal friction angle of natural loess, the effective cohesion of saturated loess, and the effective internal friction angle of saturated loess, the sampling range of each parameter in 100 groups of random sampling samples in genetic algorithm can be, the data set composed of shear strength parameters Within the range from the minimum value to the maximum value of the corresponding parameter, the 100 sets of values of each parameter in each group of random sampling samples obey the normal distribution of the mean and standard deviation calculated in step S101.

When the initial number of iterations n=0, according to 100 groups of random samples, numerical simulations were carried out in the Flac model, and the results of each simulation were input into the mean value error function f ₁ (x) of the shear strength parameters to obtain the first function value, Put 100 groups of random sampling samples into the sliding surface calculation data obtained by numerical simulation calculation into the sliding surface position error function f ₂ (x) to obtain the second function value, and bring the stability coefficient calculation data obtained from numerical simulation into the stable The coefficient error function f ₃ (x) obtains the value of the third function, passes the constraints of the three functions, and then inputs the obtained calculation result into the step of cross-variation in the genetic algorithm. When n=1 iterations are performed, the three functions are used Each function value is used to optimize the calculation of the sample, and in this way, the number of iterations is guided to n=n _max , where n _max is the target number of iterations set by the user. At this time, the Pareto solution set is output, and the Pareto solution set includes the values of three functions.

As an example, based on the initial 100 groups of random sampling samples, the genetic algorithm retains the remaining 10 relatively optimal samples through the calculation results of the three optimization functions, and then randomly samples 90 groups of samples, combining the 10 relatively optimal samples and 90 groups of newly drawn samples form 100 groups of samples together, and continue to be constrained by three optimization functions, and so on until the iteration is completed, and the optimal Pareto solution set is obtained, and the best point can be selected from the optimal Pareto solution set; among them, the three An error function is written into the genetic algorithm in advance, and at each iteration, the corresponding parameters can be brought in according to multiple sets of samples.

Three functions are constructed in this method, so the Pareto solution set of this application should be a three-dimensional coordinate system. If the level of a set of non-dominated Pareto solutions is defined as 1, it is removed from the solution set, the Pareto solution level is defined as 2 in the remaining solution set, and the above process is repeated until all the Pareto solution set levels are divided, then it can be Get the ranks of all Pareto solutions in the solution set, and when the target number of iteration calculations or convergence conditions are reached, the Pareto solution closest to the coordinate origin is the optimal Pareto solution set.

After obtaining the optimal Pareto solution set, it is necessary to select the optimal point as the final solution of the multi-objective optimization back analysis, because the error dimensions of the shear strength parameter error, sliding surface error and stability coefficient are different, so this embodiment first To perform dimensionless, specifically the Euclidean dimensionless method is chosen:

表示无量纲化后的Pareto解集，F_i表示在多目标优化后的Pareto解曲线上的点，n代表Pareto解集上的点个数。In the formula,

Represents the dimensionless Pareto solution set, F _i represents the point on the Pareto solution curve after multi-objective optimization, and n represents the number of points on the Pareto solution set.

This embodiment selects the LINMAP method to determine the optimal point of the Pareto solution set, first calculates the distance from the point in the Pareto solution set to the optimal solution point:

表示在q个目标优化下的理想点。In the formula, q is the target quantity,

Denotes the ideal point under optimization of q objectives.

Finally, the selected optimal point i _final is:

i _final =i∈min(d _i+ )

In the formula, d _i+ is the distance from the point in the Pareto solution set to the optimal solution point.

After obtaining the optimal point of Pareto solution set, determine the optimal parameter group corresponding to the optimal point, the optimal parameter group includes the optimal effective cohesion of natural loess, the effective internal friction angle of natural loess, and the effective cohesion of saturated loess and the effective internal friction angle of saturated loess.

S105. According to other parameters of the target landslide and the optimal parameter group, construct a simulated sliding surface in the Flac model, and compare the consistency between the simulated sliding surface and the first unstable sliding surface observation information of the target landslide sex.

Other parameters of the target landslide include size parameters, structural parameters and seepage parameters, etc., which are combined with the optimal parameter group to construct a simulated sliding surface in the Flac model, and then combine the simulated sliding surface with the first instability of the target landslide. The observation information of the sliding surface is compared, that is, the consistency between the simulated sliding surface and the actual sliding surface is judged.

S106. If the consistency characterization is correct, obtain a stability prediction result of the target landslide through the Flac model, and the prediction result includes a stability coefficient and slip surface information of subsequent instability.

By constructing three functions, namely, the error function of the mean value of the shear strength parameters, the error function of the sliding surface position and the error function of the landslide stability coefficient, and combining the genetic algorithm, iteratively calculates multiple groups of random sampling samples to obtain the optimal parameter set. The optimal parameter set includes the optimal effective cohesion of natural loess, effective internal friction angle of natural loess, effective cohesion of saturated loess and effective internal friction angle of saturated loess; then use the optimal parameter set data and the target landslide other parameters, construct a simulated sliding surface in the Flac model, and compare the consistency of the simulated sliding surface with the actual observation information of the first unstable sliding surface of the target landslide. The stability prediction result of the target landslide, the prediction result includes the stability coefficient and sliding surface information of the subsequent instability. In one embodiment, in addition to predicting the stability of the target landslide, it is also possible to predict similar landslides.

This method takes into account the observation information of the sliding surface, the shear strength parameter information and the stability coefficient at the first instability of the landslide, and sets more optimization constraints for the back analysis by setting three functions, so as to obtain a more reliable The back analysis results can then effectively predict the stability state and critical sliding surface of the target landslide.

As an example, taking Heifangtai Dangchuan 2# landslide as an example, this method is used to back-analyze the first instability and predict the second instability.

Heifangtai is located on the north bank of the Yellow River in Yongjing County, Gansu Province. The area of the Heifangtai platform is about 11.5km ² . The longest Hulanggou in the gully divides Heifangtai into two parts: the smaller area in the west is Fangtai, About 1.5km ² ; the larger area in the east is Heitai, about 9km ² . The Dangchuan 2# landslide has suffered two static liquefaction instability failures, among which the length of the first instability slide is about 20m, the average width is about 115m, and the area is about 8396m ² ; The area is about 27422m ² .

A1: Construct the Flac model based on the Dangchuan 2# landslide.

Referring to Figure 4, it shows a longitudinal section of the landslide. The lithology of the landslide formation from top to bottom is loess, about 20-50m thick; silty clay, about 4-20m thick; sandy pebble layer, about 1-8m thick; Mudstone, sandy mudstone, occurrence 135°∠11°.

Referring to Figure 5, it shows the simulation results of the Dangchuan 2# landslide when no landslide occurred. The instability and damage of the landslide mainly occurred in the loess layer, and part of the silty clay layer was scraped during the movement. Therefore, the loess layer and some The Flac model is established in the silty clay layer, the model is 50m high, 260m wide at the top, and 300m wide at the bottom, using a constant head boundary.

The seepage pattern in the Heifangtai area is relatively complex, which is a combination of matrix seepage and dominant seepage. Due to the poor permeability of silty clay, groundwater is enriched at the bottom of the loess and seeps out along the edge of the plateau. Seepage plays a leading role, so only dominant seepage is considered here. In the GWflow mode in FLAC, the fast saturated flow method is used to make the bottom loess quickly reach a saturated state, simulating high irrigation water flow through the gaps and fissures of the loess layer to quickly reach the bottom; Specifically, the seepage simulation parameters are shown in Table 1.

Table 1 seepage parameters of each soil layer

The seepage simulation shows that the seepage water flows out from the junction of the loess layer and the silty clay layer, which is consistent with the actual situation. It is determined that the saturated loess layer is below the water level line, and finally the F l ac model of the Dangchuan 2# landslide is obtained.

A2: Obtain the shear strength parameters of the loess in the Heifangtai area, and calculate the mean and standard deviation of each parameter.

By collecting a large number of test results of sliding zone soil of Heifangtai landslide, the shear strength parameters are obtained, that is, the effective cohesion and effective internal friction angle of natural loess and saturated loess are respectively obtained, and statistical analysis is performed on them, and the average value is calculated. and the standard deviation, the results are shown in Table 2. The cohesion of the silty clay layer is 50kPa, and the internal friction angle is 30°.

Table 2 Parameter mean and standard deviation

A3: Construct three optimization objective functions, namely, the error function of the mean value of shear strength parameters, the error function of the position of the sliding surface, and the error function of the landslide stability coefficient.

A4: Determine the optimal parameter set for inverse analysis and perform inversion.

Referring to Figure 6, it shows a schematic diagram of the convergence of the genetic algorithm after 25 iterations, and a schematic diagram of the Pareto multi-objective optimization result is shown with reference to Figure 7; 100 groups of random sampling samples of the genetic algorithm are constructed, and 100 groups of them are combined with three optimization objective functions The random sampling sample was iterated 25 times to obtain the Pareto solution set of multi-objective optimization, and the LINMAP method was used to determine the optimal point of the Pareto solution set. The asterisk point in Figure 7 is the optimal point of the Pareto solution set.

According to the optimal point of Pareto solution set to determine the optimal parameter group for back analysis, the results obtained are: the effective cohesion of natural loess is 20.09kPa, the effective internal friction angle is 22.7°; the effective cohesion of saturated loess is 12.21kPa, and the effective internal friction angle is 22.7°. The friction is 29.6°; the corresponding stability coefficient is 0.97.

Referring to FIG. 8 , it shows a schematic diagram of the comparison between the simulated sliding surface and the actual sliding surface in the first instability. As shown in Figure 8, the optimal parameter group is substituted into the Flac model for inversion, and then the corresponding simulated sliding surface is obtained, and compared with the actual sliding surface in the first instability sliding, the comparison shows that although there is a slight However, the critical sliding surface calculated by back analysis is in good agreement with the actual observed sliding surface.

A5: Use the Flac model to predict the second destabilization slide of the landslide, and generate the stability prediction results. The prediction results include the sliding surface information and stability coefficient of the subsequent destabilization.

Referring to Fig. 9, a schematic diagram of the predicted sliding surface in the second instability is shown. Since the interval between the second destabilizing slide and the first destabilizing sliding is only about 3 hours, the physical and mechanical parameters of the loess and the water level do not change much in a short period of time, so they can be ignored. The shear strength parameters of loess obtained from information back analysis are used for the second instability sliding prediction.

The scope of the second unsteady sliding is larger, divided into 3 rounds of sliding, the interval between the 3 rounds of sliding is too short, only the sliding surface information of the third round of sliding is recorded, and only the first and second rounds of sliding are recorded The location information of the trailing edge point.

Fig. 10(a) is a schematic diagram of the instability range obtained in the first round of simulation of the second instability, and the predicted stability coefficient is 1.02; Fig. 10(b) is the result obtained in the second round of simulation of the second instability Schematic diagram of the instability range, the predicted stability coefficient is 1.01; 10(c) is a schematic diagram of the instability range obtained from the third round of simulation of the second instability, and the predicted calculated stability coefficient is 1.12.

Comparing the prediction results of the second instability sliding of Dangchuan 2# landslide with the actual instability observation results, the coincidence degree is high, and the three rounds of sliding behavior in the second instability are successfully predicted by using this method. The predicted critical sliding surface is basically the same as the actually observed sliding surface, only the shape of the third wheel sliding surface is quite different. Secondly, the stability coefficients of the three rounds of instability prediction are all close to 1. The stability coefficients of the first and second round predictions are 1.02 and 1.01 respectively, indicating that the landslide is basically in an under-stable state and close to the limit equilibrium state; the stability coefficient of the third round prediction is 1.12, indicating that the landslide is in a basically stable state.

Through the above examples, it is shown that the back analysis results and prediction results of this method are excellent. By combining three optimization objective functions, namely, the error function of the mean value of shear strength parameters, the error function of the position of the sliding surface and the error function of the landslide stability coefficient, a more reliable The results of the back analysis can predict the stability of the subsequent landslide more accurately.

An embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and operable on the processor, and the processor implements the computer program described in the embodiment when executing the computer program. described method.

An embodiment of the present application is a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the method described in the embodiment is implemented.

Each embodiment in this specification is described in a progressive manner, each embodiment focuses on the difference from other embodiments, and the same and similar parts of each embodiment can be referred to each other.

Those skilled in the art should understand that the embodiments of the embodiments of the present application may be provided as methods, devices, or computer program products. Therefore, the embodiment of the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

Embodiments of the present application are described with reference to flowcharts and/or block diagrams of methods, terminal devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each procedure and/or block in the flowchart and/or block diagram, and a combination of procedures and/or blocks in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, embedded processor or processor of other programmable data processing terminal equipment to produce a machine such that instructions executed by the computer or processor of other programmable data processing terminal equipment Produce means for realizing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the The instruction means implements the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions can also be loaded into a computer or other programmable data processing terminal equipment, so that a series of operational steps are performed on the computer or other programmable terminal equipment to produce computer-implemented processing, thereby The instructions executed above provide steps for implementing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

While the preferred embodiments of the embodiments of the present application have been described, additional changes and modifications can be made to these embodiments by those skilled in the art once the basic inventive concept is understood. Therefore, the appended claims are intended to be interpreted to cover the preferred embodiment and all changes and modifications that fall within the scope of the embodiments of the application.

Finally, it should also be noted that in this text, relational terms such as first and second etc. are only used to distinguish one entity or operation from another, and do not necessarily require or imply that these entities or operations, any such actual relationship or order exists. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article, or terminal equipment comprising a set of elements includes not only those elements, but also includes elements not expressly listed. other elements identified, or also include elements inherent in such a process, method, article, or end-equipment. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article or terminal device comprising said element.

In this paper, specific examples are used to illustrate the principles and implementation methods of the application. The descriptions of the above embodiments are only used to help understand the method and core idea of the application; meanwhile, for those of ordinary skill in the art, according to the application There will be changes in the specific implementation and scope of application. In summary, the content of this specification should not be construed as limiting the application.

Claims (8)

1. A loess landslide stability prediction method, characterized in that the method comprises:

acquiring shear strength parameters of a target area, wherein the shear strength parameters comprise data of effective cohesive force of a plurality of groups of natural loess, effective internal friction angles of a plurality of groups of natural loess, effective cohesive force of a plurality of groups of saturated loess and effective internal friction angles of a plurality of groups of saturated loess, and constructing a mean error function of the shear strength parameters according to the shear strength parameters;

constructing a sliding surface position error function according to sliding surface observation information of first destabilization of a target sliding surface and a critical sliding surface determined by finite difference numerical simulation calculation;

sampling the shear strength parameters, obtaining landslide stability coefficients through finite difference strength folding and subtracting calculation according to cohesive force data and internal friction angle data obtained by sampling, and constructing a landslide stability coefficient error function;

performing multiple iterations by using a genetic algorithm based on multiple groups of random sampling samples by combining the shear strength parameter mean error function, the sliding surface position error function and the landslide stability coefficient error function to obtain an optimal parameter set, wherein the optimal parameter set comprises optimal effective cohesive force of natural loess, effective internal friction angle of natural loess, effective cohesive force of saturated loess and effective internal friction angle of saturated loess;

according to other parameters of the target landslide and the optimal parameter set, a simulated sliding surface is constructed in a Flac model, and consistency of sliding surface observation information of the simulated sliding surface and first instability of the target landslide is compared, wherein the other parameters comprise a size parameter, a structural parameter and a seepage parameter;

and under the condition that the consistency characterization is correct, obtaining a stability prediction result of the target landslide through a Flac model, wherein the prediction result comprises a stability coefficient and sliding surface information of subsequent instability.

2. The method of claim 1, wherein constructing a shear strength parameter mean error function from the shear strength parameters comprises:

respectively calculating the average value and the standard deviation of each of the effective cohesive force of a plurality of groups of natural loess, the effective internal friction angle of a plurality of groups of natural loess, the effective cohesive force of a plurality of groups of saturated loess and the effective internal friction angle of a plurality of groups of saturated loess, and assuming that the average value and the standard deviation are compliant with normal distribution;

and constructing a shear strength parameter mean error function according to the calculated mean value and standard deviation.

3. The method according to claim 2, wherein the shear strength parameter mean error function is constructed as:

wherein x is a shear strength parameter vector, mu _x As the average value of each parameter,

an inverse matrix of the covariance matrix of each parameter is represented, and T represents a matrix transposition.

4. The method of claim 1, wherein constructing a slip position error function from slip observations of a first destabilization of a target slip and a critical slip determined by finite difference numerical simulation calculations comprises:

selecting a plurality of characteristic coordinate points on the first destabilizing sliding surface according to the sliding surface observation information of the first destabilizing sliding surface of the target landslide;

determining a plurality of critical sliding surfaces through a plurality of finite difference numerical simulation calculations, and respectively acquiring a plurality of characteristic coordinate points on each critical sliding surface;

and constructing a sliding surface position error function according to the plurality of characteristic coordinate points on the sliding surface which is unstable for the first time and the plurality of characteristic coordinate points on each critical sliding surface.

5. The method of claim 1, wherein constructing a slip position error function from slip observations of a first destabilization of a target slip and a critical slip determined by finite difference numerical simulation calculations comprises:

according to the slip surface observation information of the first destabilization of the target landslide, 5 characteristic coordinate points on the slip surface of the first destabilization are selected and marked as (a) _i ，b _i ) Wherein i=1, 2,3,4,5;

determination of critical slip by multiple finite difference numerical simulation calculationsA moving surface, 5 characteristic coordinate points on each critical sliding surface are obtained and marked as (x) _i ，y _i ) Wherein i=1, 2,3,4,5;

according to 5 characteristic coordinate points on the sliding surface of the first destabilization and 5 characteristic coordinate points on the critical sliding surface, the constructed sliding surface position error function is as follows:

6. the method of claim 1, wherein sampling the shear strength parameter, calculating a landslide stability coefficient by finite difference strength folding and subtracting from the cohesive force data and the internal friction angle data obtained by sampling, and constructing a landslide stability coefficient error function, comprising:

sampling the shear strength parameter, and obtaining a landslide stability coefficient through finite difference strength folding and subtracting calculation according to cohesive force data and internal friction angle data obtained by sampling;

assuming that the stability coefficient of the target landslide is 1;

the constructed landslide stability coefficient error function is as follows:

f ₃ (x)＝(FS(x)-1)

wherein FS is a landslide stability coefficient obtained by finite difference strength folding and subtracting method.

7. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any one of claims 1 to 6 when executing the computer program.

8. A computer readable storage medium, characterized in that the computer readable storage medium has stored thereon a computer program which, when executed by a processor, implements the method according to any of claims 1 to 6.

Images