1. An anchoring slope safety evaluation method based on a genetic algorithm and a discrete element analysis method is characterized by comprising the following steps: the method comprises the following steps:

s1: collecting basic data of the rock anchoring side slope based on-site investigation and instrument monitoring;

s2: according to the basic data, constructing a rock anchoring slope analysis calculation model based on discrete element analysis software 3 DEC;

s3: determining a creep constitutive model and a creep parameter range based on a rock mass creep test;

s4: assigning the creep constitutive model determined in the step S3 to the rock anchoring slope analysis and calculation model constructed in the step S2 in 3DEC software, taking the creep parameter range determined in the step S2 as an input parameter of a genetic algorithm, combining the responses of the genetic algorithm and the rock anchoring slope analysis and calculation model, and performing combined iterative calculation to determine the optimal parameter of the creep constitutive model in the rock anchoring slope analysis and calculation model;

s5: substituting the optimal parameters of the creep constitutive model obtained in the step S3 and the optimal parameters of the creep constitutive model in the rock anchoring slope analysis and calculation model obtained in the step S4 into the rock anchoring slope analysis and calculation model constructed in the step S2, and extracting characteristic variables of the rock anchoring slope based on the set rock anchoring slope analysis time;

s6: and (4) according to the change condition of the characteristic variable, performing long-term safety evaluation on the rock mass anchoring side slope.

2. The method for evaluating the safety of the anchored slope based on the genetic algorithm and the discrete element analysis method as claimed in claim 1, characterized in that: in step S1, the basic data includes shape data obtained from field survey, deformation data of the anchoring side slope and stress data of the anchoring structure obtained from monitoring by an instrument, and the shape data includes elevation of the anchoring side slope, landslide boundary, shape and distribution characteristics of the structural plane, and deformation characteristics of the side slope.

3. The method for evaluating the safety of the anchored slope based on the genetic algorithm and the discrete element analysis method as claimed in claim 1, characterized in that: in step S3, the process of determining the creep constitutive model and the creep parameter range based on the rock creep test is as follows:

firstly, arranging triaxial creep test data of a rock mass to obtain axial strain-time relation curves under different confining pressures, then determining element types contained in an element model according to the characteristics of the axial strain-time relation curves, finally selecting the element model containing the element types, comparing a typical strain-time curve of the element model with a test result of a triaxial creep test, and determining a creep constitutive model;

and estimating the parameter range of the creep constitutive model based on the triaxial creep test data.

4. The method for evaluating the safety of the anchored slope by genetic algorithm and discrete element analysis method as claimed in claim 1, wherein: in step S4, the process of determining the optimal parameters of the element model of the creep constitutive model in the rock anchored slope analysis calculation model is as follows:

s4-1: combining lithology anchoring slope analysis calculation model response in 3DEC software with an objective function of a genetic algorithm by compiling the genetic algorithm through python;

s4-2: importing the instrument monitoring data collected in the step S1 into python in the step S4-1 as a part of the objective function;

s4-3: by means of an iButton writing environment built in 3DEC software, a genetic algorithm is introduced, and corresponding parameters are set and comprise: the method comprises the steps of performing iterative calculation of a genetic algorithm and analytical calculation of a rock anchoring slope analysis calculation model based on the target function determined in S4-2, combining field monitoring data, continuously obtaining a fitness value, and finally obtaining the optimal constitutive model parameter which best meets actual monitoring data.

5. The method for evaluating the safety of the anchored slope based on the genetic algorithm and the discrete element analysis method as claimed in claim 1, characterized in that: in step S5, the process of extracting the characteristic variables of the rock-anchored slope is as follows:

s5-1: assigning the creep constitutive model determined in the step S3 and the step S4 and the optimal parameters to the rock anchoring slope analysis and calculation model constructed in the step S2;

s5-2: according to the data collected in the step S1, based on the arrangement condition of the on-site monitoring instrument, setting the characteristic position of the rock anchoring slope analysis and calculation model for deformation monitoring, and monitoring the axial force change condition of the anchor cable;

s5-3: and inputting the set rock mass anchoring slope analysis time into 3DEC software, and extracting by taking the monitoring variable of the characteristic position and the axial force value of the anchor cable as characteristic variables.

6. The method for evaluating the safety of the anchored slope based on the genetic algorithm and the discrete element analysis method as claimed in claim 1, characterized in that: in step S6, the process of evaluating the long-term safety of the rock-anchored slope is as follows:

s6-1: constructing a displacement change curve of the change of the displacement along with the time and an axial force change curve of the change of the axial force of the anchor cable along with the time based on the characteristic variables extracted in the step S5;

s6-2: and evaluating the long-term safety of the rock-anchored slope based on the displacement change curve, the axial force change curve of the anchor cable and the maximum change amplitude of the axial force of the anchor cable, wherein when the displacement change curve and the axial force change curve of the anchor cable both show convergence and the maximum change amplitude of the axial force of the anchor cable is less than or equal to 10%, the slope is evaluated as safe, otherwise, the slope is unsafe.

Background

Since the implementation of a series of national hydropower development plans, a large number of water conservancy and hydropower projects are built in the southwest region of China, along with the construction and operation of the projects, a large number of rocky high slope treatment projects are developed, and an anchoring structure is widely applied as one of the most common means in the slope treatment projects. Therefore, a large number of rock anchoring side slopes are formed, and the long-term safety of the side slopes is not only related to the operation of the whole hydropower project, but also seriously related to the life and property safety of people in the reservoir area. In the most common standard for evaluating landslide stability at present, only one force which is equivalent to an evenly distributed anchoring structure is balanced and calculated, the dynamic interaction between the anchoring structure and the side slope is difficult to consider, and meanwhile, the current standard is a static evaluation method for evaluating the side slope stability, and the long-term safety state of the rock anchoring side slope is difficult to accurately judge.

The Discrete Element Method (DEM) is a numerical method for stress analysis of rock mass with structural planes and joints, and was proposed first by Cundall in 1971. The large displacement of the discontinuous rock mass, the phenomena of contact surface slippage, separation and the like can be simulated, and meanwhile, the internal deformation and stress distribution conditions of the structural surface and the jointed rock mass can be reflected more truly. The method takes Newton's second law as a theoretical basis. The method can be applied to the analysis and research of Rigid bodies (Rigid Block) and deformable bodies (Deforma bleb lock). 3DEC is an abbreviation of three-dimensional discrete element program (3 dimensional discrete element Code), and is a mature discrete element numerical simulation software developed and researched by ITASCA corporation, usa. The method can simulate the engineering case with discontinuity, can also simulate the stress and deformation of the rock mass under dynamic or static load, is mainly suitable for analysis and calculation of large deformation and rotation, can automatically judge each contact point between blocks, and can also analyze the medium stress and deformation under dynamic or static load. The method is currently applied to the fields of the influence of structures such as slope evaluation, underground engineering and bedding, joints or faults on rock foundations and the like. However, most of the analysis at present adopts a Mohr model, and the change of the critical strain state and the strength deterioration of the rock mass material caused by the creep characteristic cannot be considered; meanwhile, creep parameters are selected, so that the efficiency of parameter determination is low through trial calculation and experiments, and the response of the rock anchoring slope analysis and calculation model is different from actual monitoring data to a certain extent. The two defects cause the existing research deviation on the evolution mechanism, the long-term safety evaluation and the like of the rock anchoring side slope.

Disclosure of Invention

In order to solve the problems, the invention provides an anchoring slope safety evaluation method based on a genetic algorithm and a discrete element analysis method, which mainly comprises the following steps:

s1: collecting basic data of the rock anchoring side slope based on-site investigation and instrument monitoring;

s2: according to the basic data, constructing a rock anchoring slope analysis calculation model based on discrete element analysis software 3 DEC;

s3: determining a creep constitutive model and a creep parameter range based on a rock mass creep test;

s4: assigning the creep constitutive model determined in the step S3 to the rock anchoring slope analysis and calculation model constructed in the step S2 in 3DEC software, taking the creep parameter range determined in the step S2 as an input parameter of a genetic algorithm, combining the responses of the genetic algorithm and the rock anchoring slope analysis and calculation model, and performing combined iterative calculation to determine the optimal parameter of the creep constitutive model in the rock anchoring slope analysis and calculation model;

s5: substituting the optimal parameters of the creep constitutive model obtained in the step S3 and the optimal parameters of the creep constitutive model in the rock anchoring slope analysis and calculation model obtained in the step S4 into the rock anchoring slope analysis and calculation model constructed in the step S2, and extracting characteristic variables of the rock anchoring slope based on the set rock anchoring slope analysis time;

s6: and (4) according to the change condition of the characteristic variable, performing long-term safety evaluation on the rock mass anchoring side slope.

Further, the basic data comprise form data obtained by site survey, deformation data of the anchoring side slope and stress data of the anchoring structure, wherein the deformation data are obtained by monitoring of instruments, and the form data comprise the elevation of the anchoring side slope, the landslide boundary, the form and distribution characteristics of the structural surface and the deformation characteristics of the side slope.

Further, the process of determining the creep constitutive model and the creep parameter range based on the rock creep test is as follows:

firstly, arranging triaxial creep test data of a rock mass to obtain axial strain-time relation curves under different confining pressures, then determining element types contained in an element model according to the characteristics of the axial strain-time relation curves, finally selecting the element model containing the element types, comparing a typical strain-time curve of the element model with a test result of a triaxial creep test, and determining a creep constitutive model;

and estimating the parameter range of the creep constitutive model based on the triaxial creep test data.

Further, the process of determining the optimal parameters of the element model of the creep constitutive model in the rock anchored slope analysis and calculation model is as follows:

s4-1: combining lithology anchoring slope analysis calculation model response in 3DEC software with an objective function of a genetic algorithm by compiling the genetic algorithm through python;

s4-2: importing the instrument monitoring data collected in the step S1 into python in the step S4-1 as a part of the objective function;

s4-3: by means of an iButton writing environment built in 3DEC software, a genetic algorithm is introduced, and corresponding parameters are set and comprise: the method comprises the steps of performing iterative calculation of a genetic algorithm and analytical calculation of a rock anchoring slope analysis calculation model based on the target function determined in S4-2, combining field monitoring data, continuously obtaining a fitness value, and finally obtaining the optimal constitutive model parameter which best meets actual monitoring data.

Further, the process of extracting the characteristic variables of the rock anchoring side slope is as follows:

s5-1: assigning the creep constitutive model determined in the step S3 and the step S4 and the optimal parameters to the rock anchoring slope analysis and calculation model constructed in the step S2;

s5-2: according to the data collected in the step S1, based on the arrangement condition of the on-site monitoring instrument, setting the characteristic position of the rock anchoring slope analysis and calculation model for deformation monitoring, and monitoring the axial force change condition of the anchor cable;

s5-3: and inputting the set rock mass anchoring slope analysis time into 3DEC software, and extracting by taking the monitoring variable of the characteristic position and the axial force value of the anchor cable as characteristic variables.

Further, the process of evaluating the long-term safety of the rock anchored slope is as follows:

s6-1: constructing a displacement change curve of the change of the displacement along with the time and an axial force change curve of the change of the axial force of the anchor cable along with the time based on the characteristic variables extracted in the step S5;

s6-2: and evaluating the long-term safety of the rock-anchored slope based on the displacement change curve, the axial force change curve of the anchor cable and the maximum change amplitude of the axial force of the anchor cable, wherein when the displacement change curve and the axial force change curve of the anchor cable both show convergence and the maximum change amplitude of the axial force of the anchor cable is less than or equal to 10%, the slope is evaluated as safe, otherwise, the slope is unsafe.

The technical scheme provided by the invention has the beneficial effects that: the method can effectively improve the accuracy and the efficiency of evaluating the safety of the rock anchoring side slope, and can provide favorable conditions for the research on the evolution mechanism of the landslide of the rock anchoring side slope and the evaluation on the dynamic safety.

Drawings

The invention will be further described with reference to the accompanying drawings and examples, in which:

fig. 1 is a flowchart of an anchor slope safety evaluation method based on a genetic algorithm and a discrete element analysis method in an embodiment of the present invention.

FIG. 2 is a flow chart of an implementation of the long-term safety evaluation method for a slope based on a genetic algorithm and a discrete element analysis method in the embodiment of the present invention.

Fig. 3 is a model of a component commonly used in the embodiment of the present invention.

FIG. 4 is a schematic diagram of an intersection operator in an embodiment of the invention.

FIG. 5 is a diagram illustrating mutation operators in accordance with an embodiment of the present invention.

FIG. 6 is a flow chart of a genetic algorithm implementation in an embodiment of the present invention.

FIG. 7 is a flow chart of the calculation and analysis of 3DEC software in the embodiment of the present invention.

FIG. 8 is a 3DEC numerical rock anchoring slope analysis calculation model in the embodiment of the invention.

Figure 9 is a strain-time curve of a rock mass at a confining pressure of 200kpa in an embodiment of the invention.

FIG. 10 is a typical strain-time curve of the Burgers model in an embodiment of the present invention.

FIG. 11 is a Burgers-mohr model in an embodiment of the invention.

FIG. 12 is a graph of objective function values of a genetic algorithm as a function of iteration number in an embodiment of the present invention.

Fig. 13 shows the position of slope surface monitoring and the position of anchor cable in the embodiment of the present invention.

FIG. 14 is a graph showing the time-dependent displacement of the slope table feature point in the X direction in the embodiment of the present invention.

FIG. 15 is a graph of anchor cable axial force versus time in an embodiment of the present invention.

FIG. 16 is a diagram illustrating a displacement field of a rock anchored slope and a stress distribution of anchor lines in an embodiment of the present invention.

Detailed Description

For a more clear understanding of the technical features, objects and effects of the present invention, embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

The embodiment of the invention provides an anchoring slope safety evaluation method based on a genetic algorithm and a discrete element analysis method. The current design of rock-anchored slopes and common evaluation methods such as simple striping, the Bischopper method, the Swedish striping and the like are based on static analysis which only equates the effect of an anchor cable to a uniformly distributed force, and the static stability of the whole slope is calculated, as described in the technical specifications GB50330-2013 of building slope engineering and the technical specifications GB50086-2015 of rock-soil anchor rod and shotcrete support engineering. And the creep effect of the rock mass is not considered. The method has two innovations: firstly, the interaction of the anchor cable and the rock body and the action of the rock slope structural surface (discontinuous surface) can be considered by fully utilizing the discrete element analysis method. Second, the impact of rock creep on the long-term safety of the lithologic anchored slope may be considered. And the safety and time hook is dynamically analyzed, and the change of the slope safety along with the rock creep is fully considered. Thirdly, the experimental data and the field dynamic monitoring data are fully utilized, the genetic algorithm and the discrete element analysis software are combined, and the automatic and accurate confirmation of the rock anchoring slope analysis calculation model parameters is realized. The calculation response of the whole rock anchoring slope analysis and calculation model is more in line with the actual engineering situation, and the evaluation result is more accurate. The implementation of the method is not that the results are combined and optimized after simple respective calculation, but the optimization calculation of the genetic algorithm is carried out by dynamically combining a discrete element analysis method based on python language programming by combining each generation of calculation of the genetic algorithm with the discrete element analysis calculation process, and the combination method is put forward for the first time.

Referring to fig. 1-2, fig. 1 is a flowchart of an anchor slope safety evaluation method based on a genetic algorithm and a discrete element analysis method in an embodiment of the present invention, and fig. 2 is a flowchart of an implementation of a slope long-term safety evaluation method based on a genetic algorithm and a discrete element analysis method in an embodiment of the present invention, which specifically includes the following steps:

s1: and collecting basic data of the rock anchoring side slope based on field investigation and instrument monitoring.

And (3) surveying and collecting landslide form data according to the site of the anchored rock slope, for example: the elevation of the anchoring side slope, the landslide boundary, the form and distribution characteristics of the structural surface, the deformation characteristics of the side slope and the like. And collecting deformation data of the anchoring side slope, stress data of the anchoring structure and the like according to a field monitoring instrument. For example: and collecting slope surface displacement data based on the GPS monitoring pier and collecting the change condition of the anchor cable anchoring force based on the anchor cable dynamometer.

S2: and constructing a rock anchoring slope analysis calculation model based on discrete element analysis software 3 DEC.

And constructing a generalized rock anchoring slope analysis calculation model of the rock anchoring slope in 3DEC discrete element analysis software based on the data collected in the step S1. And secondary influence factors are ignored, and primary influence factors are considered, so that the computing efficiency of the software is improved.

S3: determining a creep constitutive model and a creep parameter range according to a rock creep test, wherein the concrete implementation steps of the step S3 are as follows:

s3-1: a creep constitutive model is determined. At present, creep constitutive models most commonly used for creep analysis of rock and soil are divided into two types, namely empirical models formed based on tests and data fitting and element models obtained through theoretical analysis. However, empirical models lack theoretical basis and are only fit to experimental data. The element model can reflect the mechanical property of the rock-soil body more truly, and is simple and easy to use, so that the element model is more widely used in actual engineering. The most frequently used original model is shown in fig. 3, a is a Maxwell element model, b is a Kelvin element model, c is a Burgers element model, and d is a Poynting-Thomson element model;

the method adopts a method of triaxial creep test and element model curve analogy to determine the basic type of a creep constitutive model. The specific process is that firstly, the triaxial creep test data of the rock mass is collated to obtain the axial strain-time relation curve under different confining pressures. Then, determining the type of elements, such as elastic elements, viscous elements and the like, which the model should contain according to the curve characteristics; finally, a model containing the elements is selected from the models, and a typical strain-time curve is compared with the test result to determine the basic type of the creep structure.

S3-2: and estimating the parameter range of the constitutive model based on the triaxial creep test data.

The Maxwell constitutive model is shown as a in FIG. 3, and the stress-strain relationship is as follows:

where k is the elastic modulus, η is the viscosity coefficient, σ is the stress, and ε is the strain.

The Kelvin constitutive model is shown as b in fig. 3, and the stress-strain relationship is as follows:

where k is the elastic modulus, η is the viscosity coefficient, σ is the stress, and ε is the strain.

Burgers is shown as c in FIG. 3, and the stress-strain relationship is:

wherein k is the modulus of elasticity, wherein k₁Is the elastic modulus, k, of the Kelvin body in the model of the element₂Is the elastic modulus of Maxwell body in the element model, and eta is the viscosity coefficient, wherein eta is₁Is the viscosity coefficient, η, of the Kelvin body in the model of the component₂The viscous coefficient of Maxwell body in the element model is shown, sigma is stress, and epsilon is strain.

The Poynting-Thomson constitutive model is shown as d in FIG. 3, and the relationship between stress and strain is:

in the formula k₁Is the elastic modulus, k, of the Maxwell body in the element model₂The elastic modulus of the spring in the element model, eta is the viscosity coefficient, sigma is the stress, and epsilon is the strain.

S4: assigning the creep constitutive model determined in the step S3 to the rock anchoring slope analysis and calculation model constructed in the step S2 in 3DEC software, taking the creep parameter range determined in the step S2 as an input parameter of a genetic algorithm, combining the responses of the genetic algorithm and the rock anchoring slope analysis and calculation model, and performing combined iterative calculation to determine the optimal parameter of the creep constitutive model in the rock anchoring slope analysis and calculation model; the method comprises the following concrete steps:

s4-1: the genetic algorithm is written through python, and model simulation calculation response of the anchoring slope in 3DEC software is combined with an objective function of the genetic algorithm. And simultaneously, determining the optimal parameters of the creep structure by means of the field monitoring data of the anchored slope.

The genetic algorithm implementation process comprises the following steps:

(1) an optimized target parameter is determined. The method selects the parameters of the rock mass creep constitutive as target parameters;

(2) carrying out binary coding on the target parameter;

(3) fitness function selection

The fitness value determines the quality of the individuals in the population and determines whether the excellent individuals in the current generation are selected to be inherited to the next generation. The fitness function value is not negative and a larger value indicates that the individual is more excellent.

(4) Genetic operator

Including selection operators, crossover operators, and mutation operators

The selection operator gives the probability of being selected or eliminated according to the fitness of different individuals to the environment, and the individuals are selected according to a certain rule, and common methods comprise roulette selection, sequencing selection and tournament selection. The evaluation method uses roulette selection.

Suppose that the population comprises n individuals x1, x2, xn, and the fitness of each individual is f (x)_i) Since individuals have a certain probability of being selected, the fitness should satisfy the formula (1):

1≥f(x_i)≥0 (1)

definition of p_iFor the distribution probability, the survival probability of the individual is expressed, and the solution is carried out according to the formula (2)

The distribution law should satisfy the formulas (3) and (4),

p_i≥0 (3)

when the fitness value of an individual is high, the probability of distribution is also high, so that the probability that a good individual is selected is increased. Meanwhile, individuals with poor performance are selected with a certain probability, so that the diversity of the offspring population individuals can be ensured.

The crossover operator refers to the fact that a gene exchange rule is appointed so that parent genes are mutually exchanged to generate a brand new filial generation, and the crossover operator has a certain probability to obtain an individual superior to the parent. Common crossover operators include single-point crossover, double-point crossover, multi-point crossover, partial mapping crossover, sequential crossover, and the like.

The method uses single-point crossing, firstly selects a crossing point, and interchanges the binary codes after the crossing point is crossed by two individuals, as shown in fig. 4. And the mutation operator can keep a good diversity of the population in the genetic algorithm and imitate the genetic variation in the biological evolution. The realization process is as follows: in the first step, some individuals are selected from the population according to the set variation probability, and random one bit in the selected individuals is subjected to 0-1 transformation, wherein the specific process is shown in fig. 5, and the implementation process of the whole genetic algorithm is shown in fig. 6.

S4-2: importing the engineering example monitoring data collected in the step S1 into python as a part of an objective function;

s4-3: and performing iterative calculation of a genetic algorithm and analytical calculation of a rock anchoring slope analytical calculation model, combining on-site monitoring data, continuously acquiring a fitness value, and finally acquiring the optimal constitutive model parameters which best meet actual monitoring data.

Python writes the parameter description required for the genetic algorithm:

the objective function func: an objective function.

Target dimension ndim: the dimensionality of the objective function is typically the number of parameters.

Population size _ pop: the population size represents the number of chromosomes needed in calculation, the more chromosomes, the more information exchanged among the chromosomes, and therefore, the better the optimization performance is, but when the number of chromosomes reaches a certain number, the improvement of the algorithm due to the increase of the number is small, and the efficiency of the genetic algorithm is influenced.

Maximum number of iterations max _ lite: and solving the maximum iteration times of the optimal value process.

Mutation probability prob _ mut: and (3) the probability of mutation when the genetic algorithm performs mutation operator operation.

Parameter ranges lb and ub: refers to the upper and lower limits of the parameter, and generally sets the parameter range of the optimization problem to be the range.

Precision: and solving the precision of the obtained parameters when optimizing the parameters.

S5: extracting characteristic variables of the rock anchoring side slope based on the set rock anchoring side slope analysis time;

the specific implementation steps of step S5 are as follows:

step S5-1: and assigning the creep constitutive model determined in the step S3 and the step S4 and the optimal parameters to the rock mass anchoring slope analysis and calculation model constructed in the step S2.

3DEC software computational analysis rationale:

the discrete unit method can be realized only by satisfying the equilibrium equation, and the deformation body still needs to conform to the constitutive equation.

(1) Physical equation

Two blocks that interact in discrete element analysis need to satisfy the following physical constitutive equations:

ΔF_n＝K_nΔU_n

ΔF_s＝K_sΔU_s

F_n＝F_n1+K_nΔU_n

F_s＝F_s1+K_sΔU_s

in the formula K_n、K_sFor normal and tangential contact stiffness of the joint surfaces, Δ F_n、ΔF_sIncrement of normal and tangential forces between blocks, Δ U_n、ΔU_sIn increments of normal and tangential displacement between blocks, F_n、F_sNormal and tangential forces between blocks, F_n1、F_s1The initial values of the normal force and the tangential force between the blocks or the final value of the last iteration.

(2) Equation of motion

The fundamental newton's second law of the discrete element method, the force acting on a discrete mass causes the mass to move, and the centroid of the mass should satisfy the following equation:

a＝F_{combination of Chinese herbs}/m

In the formula, F_xi、F_yiSubjecting the block to the action of another block, F_x、F_yThe resultant horizontal and vertical forces to which the block is subjected. M is the resultant moment, x, to which the block is subjected_i、y_iIs the coordinate of the action point, x and y are the coordinates of the centroid of the block body, a is the acceleration, m is the mass of the rock mass,angular acceleration, I moment of inertia.

(3) And (4) carrying out numerical integration by adopting a forward difference format to obtain the speed and displacement of the block along the resultant force direction and the rotation quantity of the block.

V(t₁)＝V(t₀)+aΔt

U(t₁)＝U(t₀)+V(t₁)Δt

In the formula, t₀For the initial time, Δ t is the calculation time step, U (t)₁)、V(t₁) Respectively is a block body at t₁Displacement and velocity in time, θ (t)₁) Is t₁Displacement and velocity in time, θ (t)₁) Is t₁Moment of inertia of time.

The 3DEC software calculation analysis procedure is shown in fig. 7.

Step S5-2: and according to the data collected in the step S1, setting the characteristic position of the rock anchoring side slope analysis and calculation model of the rock anchoring side slope for deformation monitoring based on the arrangement condition of the on-site monitoring instrument, and monitoring the axial force change condition of the anchor cable.

Step S5-3: and based on the set rock mass anchoring side slope analysis time, extracting the monitoring variable of the characteristic position and the axial force value of the anchor cable as characteristic variables.

S6: and (4) according to the change condition of the characteristic variable, performing long-term safety evaluation on the rock mass anchoring side slope.

The specific implementation steps of S6 are as follows:

s6-1: constructing a change curve of displacement along with time, a change curve of the anchor cable axial force along with time and the maximum change amplitude of the anchor cable axial force based on the characteristic variables extracted in the step S5;

s6-2: and evaluating the long-term safety of the rock-anchored slope based on the displacement change curve, the axial force change curve of the anchor cable and the maximum change amplitude of the axial force of the anchor cable, wherein when the displacement change curve and the axial force change curve of the anchor cable both show convergence and the maximum change amplitude of the axial force of the anchor cable is less than or equal to 10%, the slope is evaluated as safe, otherwise, the slope is unsafe.

The method is based on a genetic algorithm and a discrete element analysis method, a rock anchoring side slope creep rock anchoring side slope analysis calculation model is constructed in 3DEC software, the genetic algorithm is combined with field monitoring data to determine the optimal parameters of a creep constitutive model, the two analysis methods are combined through a python writing program, the accuracy and the efficiency of evaluation are improved, meanwhile, the long-term deformation characteristics and the evolution mechanism of the rock anchoring side slope under the creep condition are disclosed, the dynamic evaluation on the long-term safety of the rock anchoring side slope is realized, and a foundation is provided for prediction, early warning and prevention of the same type of side slope.

The selected example of the method is a rock anchoring side slope of a certain hydropower project in the southwest region. The side slope is positioned on the right bank of a river with the upstream of a hydropower station dam site being about 11.5km, the length of the landslide is about 700m, the width of the landslide is about 320-400 m, and the area of the landslide is about 0.25km²The total amount of the formula is about 1300 km³。

In this embodiment, the step S1 is shown in table 1 below based on-site survey, construction data, design data, test data, and instrument monitoring and collecting basic data of the rock-anchored slope:

TABLE 1 landslide rock mass related parameter values

Wherein rho is the density of the rock mass, G is the shear modulus of the rock mass, K is the bulk modulus of the rock mass, C is the cohesion of the rock mass, dip is the inclination angle of the rock stratum, dip-direction is the inclination of the rock stratum, spacing is the rock stratum gap, num is the number of the rock stratum, stiffness-normal is the normal stiffness of the rock stratum, and stiffness-shear is the tangential stiffness of the rock stratum.

The relevant parameters of the cable bolt are shown in table 2 below:

TABLE 2 Anchorage Cable related parameter values

The area is the cross-sectional area of the anchor cable, e is the elastic modulus of the anchor cable, grout _ stiff _ is the rigidity of the grouting body, cable _ yield is the axial yield strength of the anchor cable, cable _ strain _ limit is the axial strain limit of the anchor cable, dowel _ stiff _ is the shear deformation rigidity of the anchor cable, dowel _ yield _ is the shear yield strength of the anchor cable, and dowel _ strain _ limit is the shear strain limit of the anchor cable.

In this embodiment, the specific implementation steps of the step S2 of constructing the rock anchoring slope analysis calculation model based on the discrete element analysis software 3DEC are as follows:

a numerical rock mass anchoring slope analysis calculation model is constructed based on the data collected in step S1, the length of the model is 700m, the height of the model is 400m, and the inclination angle is about 55 °. Because two-dimensional computational analysis of the profile of the anchored rock slope needs to be carried out, the width, i.e. the Y direction, is only 5 m. The joint model is inclined to the inclination angle of 70 degrees and inclined to the inclination angle of 270 degrees, the length of anchor cables is 50m and the interval is 20m, and the installation angle is about 35 degrees. The other calculation parameters required for calculation in 3DEC are given to the rock mass anchored slope analysis calculation model according to the values shown in tables 3 and 4, and the established numerical rock mass anchored slope analysis calculation model is shown in fig. 8.

In this embodiment, the concrete implementation steps of determining the creep constitutive model and the creep parameter range according to the rock creep test in step S3 are as follows:

s2-1: according to the creep test of the rock sample, the test data is obtained, and figure 9 is a test curve obtained when the confining pressure is 200 kPa. The curve exhibits 2 more obvious characteristics: 1. the specimen undergoes significant instantaneous elastic deformation in a short time after the application of the load, so the model should contain elastic elements; 2. as the test progresses, the specimen exhibits a pronounced creep-slowing phenomenon, which over time remains unchanged in creep rate, exhibiting viscoelastic characteristics, so that the model should comprise viscous elements. Through comparative analysis with a common creep constitutive model, a typical strain-time curve (fig. 10) of the Burgers model is found to be most similar to the curve. Therefore, the constitutive model of the rock mass is preliminarily determined to be a Burgers model. As the damage condition of the rock slope needs to be considered in the long-term safety evaluation, an M-C element is connected in series, so that the combined molar coulomb damage rule is combined. Further, the model becomes a Burgers-mohr model (FIG. 11).

S2-2: the Burgers-mohr model comprises 6 parameters in total, c,E_M、η_M、E_K、η_KRespectively, cohesion, internal friction angle, elastic modulus, maxwell viscosity coefficient, viscoelastic modulus, and kelvin viscosity coefficient. C andthe parameters have been determined experimentally for routine strength. Four parameters are therefore also determined. The stress-time relationship of the Burgers-mohr model can be simplified as:

wherein σ is stress; ε is the strain.

From the test curve (FIG. 10), the initial strain ε 0 and the slope m at steady creep of the model can be found. When t is 0, epsilon 0 is σ/EM based on formula (5); when t takes a large value, the value of,is close to 0, so M is 1/η M; the parameters EM and eta M can be determined based on a least square method. The fitting parameters of the final constitutive model of rock are shown in table 3.

TABLE 3 rock creep parameters

The value ranges determined according to table 3 are shown in table 4 below:

TABLE 4 range of constitutive model parameters of rock mass

From fig. 12, it can be known that the slope surface displacement response of the rock anchoring slope analysis and calculation model is closer to the field monitoring data along with the genetic algorithm, the target function of about 55 generations basically tends to be stable, the convergence is fast, and the parameter value approaches the optimal value fast.

The calculation result shows that_M＝19958.7661MPa、η_M＝1.35E+07MPa·h、E_K＝41200.39718MPa、η_KAnd when the rock slope analysis and calculation model is 254000.3466 MPa.h, the response of the characteristic position characteristic variable of the rock slope analysis and calculation model is close to the monitoring value, and the group of parameters are selected as the parameters of the rock slope constitutive model.

In this example, the specific implementation steps of extracting the characteristic variables of the rock mass anchored slope based on the set rock mass anchored slope analysis time in the step S5 are as follows:

and substituting the creep constitutive model determined in the step S3 and the optimal parameters determined in the step S4 into the rock mass anchoring slope analysis and calculation model constructed in the step S2. The set creep analysis time is input, 120 days are set as the creep analysis time in the embodiment, and the horizontal displacement of 4 characteristic positions of the slope table and the change data of anchor cable load along with time are extracted. The characteristic positions of the slope surface and the anchor line positions are shown in fig. 13, and the extracted characteristic variables and the data of the anchor lines are shown in fig. 14 and 15.

In this embodiment, the specific implementation steps of the step S6 for evaluating the long-term safety of the rock mass anchored slope according to the change condition of the characteristic variable are as follows:

from fig. 14 and 15, it can be seen that the displacement in the X direction of the slope characteristic point exhibits a tendency to fluctuate initially and then to stabilize with time. The anchor rope axial force shows rapid increase along with slope surface deformation at the beginning because the anchor rope is just installed, and later because the interact of anchor rope and rock mass, the anchor rope axial force shows the phenomenon that the acceleration slows down later to descend along with time, and final anchor rope axial force shows stable trend. And after the anchor cable is stable, the change rate of the axial force of the anchor cable is less than 10%, and the anchor cable is judged to be safe based on a judgment threshold value, namely the change amplitude of the axial force of the anchor cable is less than 10%, which is described in the text of the long-term performance and safety evaluation of geotechnical anchoring engineering. And (3) analyzing the change trend of the characteristic variable and the axial force of the anchor cable, and judging the long-term safety of the anchoring side slope as follows: and (4) safety.

It can be seen from fig. 12 that the accuracy of the response of the simulation computation model finally tends to be stable as the iteration of the genetic algorithm is continuously increased, which indicates that the genetic algorithm can rapidly and accurately determine the parameters of the rock mass creep constitutive according to the actual monitoring condition. The change of the characteristic variable along with the time is shown in fig. 14, which shows that the long-term safety of the rock-anchored slope is a dynamic change process, and the conventional evaluation methods are static and inaccurate at present. The method provided by the method considers the interaction between the anchoring structure and the rock mass and the influence of the characteristics (creep, structural plane and the like) of the rock slope rock-soil mass, hooks the safety of the rock slope and the time, realizes the dynamic evaluation on the safety of the rock slope, and can obtain the displacement distribution condition of the rock slope after the designated analysis time and the stress distribution condition of the anchor cable as shown in figure 16.

The genetic algorithm is characterized in that:

(1) self-organizing, adaptive, and intelligent. It is not necessary to describe all features of the problem in advance in the design of the genetic algorithm and to specify the measures to be taken by the algorithm for each different feature of the problem. Therefore, the robustness of the genetic algorithm is strong, and the complex unstructured problem can be solved.

(2) Evaluation information based on the objective function value is used in the searching process, and the requirements of continuity and conductibility constraint of the optimization function do not need to be met.

(3) Easy parallelization and reduced expense due to the use of very powerful computer hardware.

(4) The basic idea of the algorithm is very simple, and the algorithm has extremely standard operation modes and implementation steps and can be directly used for specific use.

(5) The genetic algorithm uses search information of a plurality of search points simultaneously. Genetic algorithms begin the search process for the best solution from an initial population of many individuals, rather than from a single individual. The group is subjected to operations such as selection, intersection, variation and the like, and a new generation group is generated. This information avoids searching for unnecessary search points and therefore effectively amounts to searching for more points, an implicit parallelism held by genetic algorithms.

(6) Genetic algorithms use probabilistic searching techniques. The genetic algorithm belongs to a self-adaptive search technology, and the operations of selection, intersection, mutation and the like are all performed in a probabilistic mode, so that the flexibility of the search process is improved.

At present, the parameter confirmation mode of the constitutive model in the discrete element analysis is mainly to endow a determined value through indoor tests. The inversion analysis is carried out by combining with on-site monitoring data, whether the response of the rock anchoring slope analysis and calculation model meets on-site monitoring, deformation signs and the like is mainly checked through manual trial, and the efficiency and the accuracy are low. The method combines a genetic algorithm and a discrete element analysis method to realize automatic optimization of constitutive parameters of the rock anchoring slope analysis and calculation model, improves the working efficiency and increases the accuracy. The following advantages of the genetic algorithm are fully utilized:

(1) self-organizing, adaptive, and intelligent. It is not necessary to describe all features of the problem in advance in the design of the genetic algorithm and to specify the measures to be taken by the algorithm for each different feature of the problem. Therefore, the robustness of the genetic algorithm is strong, and the complex unstructured problem can be solved.

(2) Evaluation information based on the objective function value is used in the searching process, and the requirements of continuity and conductibility constraint of the optimization function do not need to be met.

(3) Easy parallelization and reduced expense due to the use of very powerful computer hardware.

(4) The genetic algorithm uses search information of a plurality of search points simultaneously. Genetic algorithms begin the search process for the best solution from an initial population of many individuals, rather than from a single individual. The group is subjected to operations such as selection, intersection, variation and the like, and a new generation group is generated. This information avoids searching for unnecessary search points and therefore effectively amounts to searching for more points, an implicit parallelism held by genetic algorithms.

(5) Genetic algorithms use probabilistic searching techniques. The genetic algorithm belongs to a self-adaptive search technology, and the operations of selection, intersection, mutation and the like are all performed in a probabilistic mode, so that the flexibility of the search process is improved.

The method combines genetic algorithm and discrete element analysis method through an IPython platform carried by 3DEC software by using python language programming. The evaluation method proposed by the method is not simple to combine results after independently calculating the two evaluation methods. Instead, the objective function calculated in each generation dynamically extracts the analysis result from the discrete element analysis, and the new chromosome generation dynamically assigns new parameter values to the discrete elements for dynamic calculation as shown in fig. 2. And finally, determining constitutive model parameters which best meet the field monitoring data after iterative computation of a genetic algorithm and a discrete element analysis method. The method is currently applied for the first time.

The invention has the beneficial effects that: the method can effectively improve the accuracy and the efficiency of evaluating the safety of the rock anchoring side slope, and can provide favorable conditions for the research on the evolution mechanism of the landslide of the rock anchoring side slope and the evaluation on the dynamic safety.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.