A method for early warning of slope safety in geotechnical engineering

Abstract

The present invention discloses a method for slope safety early warning in geotechnical engineering, belonging to the field of geotechnical engineering technology. The method comprises the following steps: obtaining geological conditions and reinforcement measures at the slope geotechnical engineering site, determining a slope model analysis area, and constructing a slope model; selecting target monitoring points of the slope model; defining a slope safety early warning coefficient corresponding to slope instability; obtaining slope characteristic parameters; constructing a back propagation neural network between the slope characteristic parameters and the geotechnical response, performing Bayesian updating, and obtaining the geotechnical response for each excavation step; constructing a relationship mapping model between the geotechnical response and the slope safety early warning coefficient, obtaining the slope safety early warning coefficient for each excavation step, and performing a slope safety early warning. The present invention solves the problem of insufficient slope early warning accuracy in geotechnical engineering.

Description

Technical Field

[0001] The present invention belongs to the technical field of geotechnical engineering, and in particular relates to a method for early warning of slope safety in geotechnical engineering. Background Art

[0002] Against the backdrop of rapid economic development, large-scale land development and resource utilization are inevitable. Resource development and construction projects often involve extensive geotechnical engineering, particularly in steep, mountainous areas with complex terrain, often involving high slope excavation. However, slope instability can lead to geological disasters such as landslides and collapses, posing a serious threat to life and property. To monitor the rate and magnitude of deformation during construction and prevent and mitigate the occurrence of engineering geological hazards, it is necessary to accurately and effectively predict and evaluate deformation caused by excavation, as well as conduct risk assessments of the excavation process's safety.

[0003] Common risk control and safety early warning methods include on-site monitoring, displacement prediction, stability assessment, and reinforcement measures. Numerical simulation methods are also used to predict slope deformation and conduct stability evaluations. Numerical simulation results are closely related to the geomechanical parameters of the rock mass. Although these parameters can be obtained through direct on-site in-situ testing or laboratory analysis, the risk of construction accidents increases exponentially when these parameters are inaccurate due to sampling or errors. Furthermore, while inversion analysis can more reasonably address the lack of measurement parameter accuracy, it typically fails to fully utilize field measurement data for inversion, is insufficient to accurately reflect the key characteristics of slope excavation, and struggles to meet the high accuracy requirements for slope safety early warnings. Summary of the Invention

[0004] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method for early warning of slope safety in geotechnical engineering, which solves the problem of insufficient accuracy of early warning of slopes in geotechnical engineering.

[0005] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0006] The present invention provides a method for early warning of slope safety in geotechnical engineering, comprising the following steps:

[0007] S1. Obtain the geological conditions and reinforcement measures of the slope geotechnical engineering site, determine the slope model analysis area, and construct the slope model;

[0008] S2. Select target monitoring points of the slope model;

[0009] S3. Define the slope safety warning coefficient corresponding to slope instability;

[0010] S4. Analyze the sensitivity between the structural surface parameters at the target monitoring point and the slope safety warning coefficient, and the sensitivity between the rock and soil material parameters at the target monitoring point and the displacement and anchoring force, to obtain the slope characteristic parameters;

[0011] S5. Construct a back propagation neural network between slope characteristic parameters and geotechnical responses, and perform Bayesian update to obtain the geotechnical responses of each excavation step;

[0012] S6. Based on the slope characteristic parameters, a relationship mapping model between geotechnical response and slope safety warning coefficient is constructed to obtain the slope safety warning coefficient of each excavation step and perform slope safety warning.

[0013] The beneficial effects of the present invention are as follows: the present invention provides a method for early warning of slope safety in geotechnical engineering, which determines the slope model analysis area and constructs a slope model by obtaining detailed geological conditions and reinforcement measures at the slope geotechnical engineering site, so that the constructed slope model is closer to the actual situation, providing a reliable basis for subsequent accurate analysis of the safety status of the slope; the present invention analyzes the sensitivity between the structural surface parameters at the target monitoring point and the slope safety early warning coefficient, as well as the sensitivity between the geotechnical material parameters and the displacement and anchoring force, and can accurately identify the characteristic parameters that have the most significant impact on the safety of the slope, thereby achieving dimensionality reduction of associated parameters, reducing the amount of data analysis and processing, and more accurately reflecting the safety status of the slope, greatly improving the accuracy of the early warning; the present invention constructs a back propagation neural network between the slope characteristic parameters and the geotechnical response, and performs Bayesian updating, which can fully utilize the powerful nonlinear mapping ability of the neural network and the adaptability of the Bayesian update, and can continuously optimize the model parameters according to the actual monitoring data, so that the model's prediction of the geotechnical response is more accurate, thereby obtaining a more accurate slope for each excavation step. The safety warning coefficient provides highly reliable data support for slope safety warning; the present invention not only takes into account macro factors such as the geological conditions and reinforcement measures of the slope, but also deeply analyzes the impact of micro factors such as structural surface parameters and rock and soil material parameters on slope safety. By comprehensively considering multi-dimensional parameters, it is possible to fully capture the safety changes of the slope under different conditions, making the warning results more comprehensive and reliable; this solution analyzes the rock and soil response analysis and slope safety warning coefficient mapping for each excavation step, and can track the changes in the safety status of the slope during the excavation process in real time, realize comprehensive safety monitoring of the entire slope excavation process, and provide all-round protection for slope engineering safety; the present invention uses Bayesian updating to enable the back propagation neural network between the slope characteristic parameters and the rock and soil response to adjust the model parameters in real time according to the actual monitoring data, ensuring that the model is always consistent with the actual situation, thereby realizing dynamic warning of the slope safety status; the present invention uses each excavation step as a node to calculate and warn the slope safety warning coefficient, and can dynamically evaluate the safety status of the slope according to the advancement of the excavation progress.

[0014] Furthermore, the S1 includes the following steps:

[0015] S11. Obtain the geological conditions and reinforcement measures of the slope geotechnical engineering site and determine the slope model analysis area;

[0016] S12. setting a number of finite difference grid points, Mohr-Coulomb failure criteria, and excavation steps based on the physical parameters of the rock mass and the physical parameters of the structural surface in the slope model analysis area;

[0017] S13. Construct a slope model based on the set finite difference grid points, the Mohr-Coulomb failure criteria and excavation steps corresponding to different rock masses and structural surfaces.

[0018] The beneficial effects of adopting the above further scheme are: the present invention determines the analysis area based on the geological conditions and reinforcement measures of the slope geotechnical engineering site, and sets the finite difference grid points according to the physical parameters of the rock mass and structural surface, so that the slope model is constructed more reasonably, and can accurately restore the slope state, providing a reliable basis for slope safety early warning.

[0019] Furthermore, the S2 includes the following steps:

[0020] S21. Install appearance monitoring equipment and anchor cable force measurement equipment in the slope model analysis area;

[0021] S22. Acquire monitoring data from the displacement monitoring device and the anchor force measuring device;

[0022] S23. Select the equipment location where the monitoring data is continuous and uninterrupted and is deployed during the slope excavation period as the target observation point.

[0023] The beneficial effect of adopting the above-mentioned further scheme is: the scheme of the present invention selects the equipment positions that are continuous in time and deployed during the excavation as the target observation points, and on the basis of ensuring the integrity and validity of the data, it can more accurately grasp the dynamic changes of the slope during the excavation, and provide a reliable basis for subsequent safety warnings.

[0024] Furthermore, the S3 includes the following steps:

[0025] S31, obtaining the actual cohesion and actual internal friction angle of the failure surface corresponding to each excavation step in the slope model analysis area as actual slope shear strength parameters;

[0026] S32. According to the strength reduction method, actual slope shear strength parameters are reduced, and a slope model excavation simulation is performed simultaneously until the slope model enters a slope instability state, and a proportional coefficient corresponding to the slope instability is used as a slope safety warning coefficient;

[0027] The calculation expression of the slope safety warning coefficient is as follows:

[0028] ,

[0029] in, represents the reduced cohesion, represents the actual cohesion, represents the slope safety warning coefficient, represents the reduced internal friction angle, represents the actual internal friction angle.

[0030] The beneficial effect of adopting the above further scheme is: this scheme reduces the actual slope shear strength parameters by using the strength reduction method and simulates excavation until the slope becomes unstable, thereby realizing the definition of the slope safety warning coefficient. The defined slope safety warning coefficient can intuitively reflect the stability of the slope.

[0031] Furthermore, the S4 includes the following steps:

[0032] S41. Analyze the sensitivity between the deformation modulus, cohesion, internal friction angle, gravity, and Poisson's ratio of the structural surface at the target monitoring point and the slope safety warning coefficient, and select the structural surface parameters that change with the slope safety warning coefficient as the first slope characteristic variable;

[0033] S42, analyzing the sensitivity between the deformation modulus and cohesion of the rock and soil material at the target monitoring point and the displacement and anchoring force, and selecting the structural surface parameter with a relative sensitivity greater than a preset sensitivity threshold as the second slope characteristic variable;

[0034] S43. Use the first slope characteristic variable and the second slope characteristic variable as slope characteristic parameters.

[0035] The beneficial effects of adopting the above further scheme are: this scheme accurately screens out the key structural surface parameters that change with the warning coefficient as the first slope characteristic variable, and analyzes the sensitivity of rock and soil material parameters to displacement and anchoring force, selects high-sensitivity parameters as the second slope characteristic variable, and integrates the slope characteristic parameters to determine the core factors affecting slope safety, which can provide a basis for more accurate slope safety warning and effectively improve the accuracy and reliability of slope safety warning.

[0036] Furthermore, the calculation expression of the relative sensitivity in S42 is as follows:

[0037] ,

[0038] ,

[0039] in, Relative sensitivity, Indicates the The sensitivity ratio of the parameters for sensitivity analysis, represents the number of parameters for sensitivity analysis, Indicates the The displacement response of the slope model corresponding to the upper limit of the parameters for sensitivity analysis is Indicates the The slope model displacement response corresponding to the lower limit of the parameter for sensitivity analysis is represents the displacement response of the slope model for the parameters for which sensitivity analysis is performed, Indicates the The upper limit of the parameters for sensitivity analysis, Indicates the The lower limit of the parameter for sensitivity analysis, represents the parameters for sensitivity analysis, Indicates the The anchoring force response of the slope model corresponding to the upper limit of the parameter for sensitivity analysis is: Indicates the The anchoring force response of the slope model corresponding to the upper limit of the parameter for sensitivity analysis is: Anchor force response of the slope model representing the parameters for which sensitivity analysis is performed.

[0040] The beneficial effect of adopting the above further scheme is: the present invention provides a relative sensitivity calculation method, which can accurately calculate the sensitivity between rock and soil material parameters and displacement and anchoring force, providing a basis for more accurate slope safety warning, and effectively improving the accuracy and reliability of slope safety warning.

[0041] Furthermore, the S5 includes the following steps:

[0042] S51, obtaining a plurality of slope characteristic parameter combinations, and using them as input parameters of a back propagation neural network model;

[0043] S52, using the Latin hypercube sampling method to select a number of training samples and test samples from the input parameters;

[0044] S53, inputting the training samples and the test samples into the slope model respectively to obtain the geotechnical response of each excavation step, and using the geotechnical response as the output parameter of the corresponding back propagation neural network model, wherein the geotechnical response is displacement or anchoring force;

[0045] S54, constructing a back propagation neural network model for each geotechnical response in each excavation step;

[0046] S55. Assuming that the slope characteristic parameters are independent in the prior distribution, obtain the prior distribution probability density of the slope characteristic parameters;

[0047] The calculation expression of the prior distribution probability density of the slope characteristic parameters is as follows:

[0048] ,

[0049] , ,

[0050] in, represents the prior distribution probability density of slope characteristic parameters, Indicates the Slope characteristic parameters, represents the total number of slope characteristic parameters, Indicates the The standard deviation of the logarithm of the slope characteristic parameters, represents an exponential function with e as the base constant, represents the natural logarithm operation, Indicates the The average value of the logarithms of the slope characteristic parameters, Indicates the The average value of the slope characteristic parameters, Indicates the The standard deviation of the slope characteristic parameters;

[0051] S56. Based on the probability density function of the measurement error of displacement and anchoring force following the normal distribution, a likelihood function for updating the slope characteristic parameters by using displacement and anchoring force in each excavation step is constructed based on the back propagation neural network model;

[0052] S57. Based on the prior distribution probability density of the slope characteristic parameters and the likelihood function of updating the slope characteristic parameters through displacement and anchoring force in each excavation step, Bayesian update is performed to obtain the geotechnical response of each excavation step and the posterior distribution probability density of the slope characteristic parameters.

[0053] The beneficial effect of adopting the above further scheme is that the present invention can continuously, quickly and accurately predict the stability of the slope and the geotechnical response during subsequent excavation by constructing a back propagation neural network between the slope characteristic parameters and the geotechnical response and performing Bayesian updating.

[0054] Furthermore, the calculation expression of the likelihood function for updating the slope characteristic parameters by displacement and anchoring force in each excavation step in S56 is as follows:

[0055] ,

[0056] in, represents the likelihood function for updating the slope characteristic parameters through displacement and anchoring force at each excavation step, Indicates the excavation step when displacement data starts to be monitored. Indicates from arrive The product operation of represents the slope characteristic parameter, represents the on-site measured displacement at the target monitoring point, represents the on-site measured anchoring force at the target monitoring point, represents the total number of excavation steps, represents the standard deviation of the measurement error of the displacement, Indicates the target monitoring point On-site displacement measurement of each excavation step, Indicates the The displacement output by the back propagation neural network model for each excavation step, Indicates the excavation step when the anchor force data starts to be monitored. Indicates from arrive The product operation of represents the standard deviation of the measurement error of the anchoring force, Indicates the target monitoring point On-site measurement of anchoring force for each excavation step, Indicates the The anchoring force output by the back propagation neural network model during each excavation step.

[0057] The beneficial effect of adopting the above-mentioned further scheme is as follows: the present invention provides a calculation method for the likelihood function of updating the slope characteristic parameters through displacement and anchoring force in each excavation step, which can provide a basis for Bayesian updating, thereby providing a basis for accurately obtaining the geotechnical response of the next excavation step and the posterior distribution probability density of the slope characteristic parameters.

[0058] Furthermore, the S6 includes the following steps:

[0059] S61. Based on slope characteristic parameters, a mapping model of the relationship between geotechnical response and slope safety warning coefficient is constructed;

[0060] S62. Mapping the geotechnical response of each excavation step according to a relationship mapping model between geotechnical response and slope safety warning coefficient to obtain a slope safety warning coefficient for each excavation step;

[0061] S63. If the slope safety warning coefficient of a certain excavation step is greater than or equal to the first safety warning coefficient threshold, the slope under the excavation step is in a safe state and no slope safety warning is issued;

[0062] S64: If the slope safety warning coefficient of a certain excavation step is less than the first safety warning coefficient threshold and greater than or equal to the second safety warning coefficient threshold, the slope under the excavation step is in the first warning state, and a slope safety warning in the first warning state is performed;

[0063] S63: If the slope safety warning coefficient of a certain excavation step is less than the second safety warning coefficient threshold and greater than or equal to the third safety warning coefficient threshold, the slope under the excavation step is in the second warning state, and a slope safety warning in the second warning state is performed;

[0064] S64. If the slope safety warning coefficient of a certain excavation step is less than the third safety warning coefficient threshold, the slope under the excavation step is in the third warning state, and a slope safety warning in the third warning state is performed.

[0065] The beneficial effects of adopting the above-mentioned further scheme are as follows: the present invention constructs a relationship mapping model between geotechnical response and slope safety warning coefficient based on slope characteristic parameters, and can obtain the corresponding slope safety warning coefficient according to the geotechnical response mapping of each excavation step. At the same time, by setting different levels of safety warning coefficient thresholds, the slope status is subdivided into four states: safe, first warning, second warning and third warning, and corresponding warning measures are taken for different states. Under this hierarchical warning mechanism, warning information can be issued to relevant personnel in a timely and accurate manner according to the severity of the slope safety situation, so that the staff can make precautionary and response preparations in advance, effectively reduce the safety risks brought by slope instability, and ensure the smooth progress of engineering construction and the safety of people's lives and property.

[0066] Furthermore, the calculation expression of the relationship mapping model between the geotechnical response and the slope safety warning coefficient in S61 is as follows:

[0067] ,

[0068] in, represents the slope safety warning coefficient mapped by geotechnical response, represents the geotechnical response of the slope model based on the slope characteristic parameters, represents the inversion error random variable of the slope model, represents the joint probability density prior distribution of slope characteristic parameters, represents the probability density function of the inversion error random variable, represents the differential result of the slope characteristic parameters, represents the differential result of the inversion error random variable, Indicates from 1 to The product operation of represents the total number of geotechnical responses, Indicates the The probability density function of the measurement error corresponding to the geotechnical response is, Indicates the first Geotechnical response, Represents the output of the back-propagation neural network model Geotechnical response.

[0069] The beneficial effect of adopting the above further scheme is: the present invention provides a calculation method for the relationship mapping model between geotechnical response and slope safety warning coefficient, which can accurately map the geotechnical response of each excavation step to the slope safety warning coefficient under the corresponding excavation step, thereby reflecting the slope stability and providing a basis for slope safety warning.

[0070] Other advantages of the present invention will be analyzed in more detail in subsequent embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following is a brief introduction to the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0072] Figure 1 The present invention is a flowchart of a method for early warning of slope safety in geotechnical engineering according to an embodiment of the present invention. DETAILED DESCRIPTION

[0073] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0074] like Figure 1 As shown, in one embodiment of the present invention, the present invention provides a method for slope safety early warning in geotechnical engineering, comprising the following steps:

[0075] S1. Obtain the geological conditions and reinforcement measures of the slope geotechnical engineering site, determine the slope model analysis area, and construct the slope model;

[0076] The S1 comprises the following steps:

[0077] S11. Obtain the geological conditions and reinforcement measures of the slope geotechnical engineering site and determine the slope model analysis area;

[0078] In this embodiment, the geological conditions include slope lithology and slope structure, wherein the slope lithology includes soil slope, rock slope and rock-soil mixed slope, and the slope structure includes block structure slope, layered structure slope and fragmented structure slope;

[0079] The reinforcement measures include retaining measures, technical reinforcement, drainage measures, load reduction and counter-pressure measures, ecological reinforcement measures, and comprehensive reinforcement measures that utilize all of the above reinforcement measures. Retaining measures include retaining walls, anti-slide piles, and prestressed anchor cables. Technical reinforcement includes grouting, anchor reinforcement, and soil nailing. Drainage measures include surface drainage through intercepting ditches and drainage ditches, as well as draining groundwater through blind ditches and upward-sloping drainage holes. Load reduction and counter-pressure measures include removing the upper rock and soil of the slope to reduce the slope and loading soil and rock at the front edge of the landslide to apply counter-pressure. Ecological reinforcement measures include vegetation reinforcement and geotextile reinforcement. Under geological conditions such as high and complex slopes, comprehensive reinforcement measures consisting of prestressed anchor cables, anti-slide piles, and drainage systems are typically used.

[0080] S12. setting a number of finite difference grid points, Mohr-Coulomb failure criteria, and excavation steps based on the physical parameters of the rock mass and the physical parameters of the structural surface in the slope model analysis area;

[0081] In this embodiment, the physical parameters of the rock mass include dry density, deformation modulus, Poisson's ratio, cohesion and stable slope ratio; the physical parameters of the structural surface include structural surface rock cutting type, rock dry density, deformation modulus, Poisson's ratio and cohesion; the Mohr-Coulomb failure criterion can characterize the stress-strain characteristics corresponding to the rock mass and the structural surface. The excavation step is used to reflect the excavation process of slope geotechnical engineering. Each time the excavation command is executed, all the rock masses corresponding to the excavation step are excavated at one time to provide a basis for calculating the displacement, anchoring force and safety warning coefficient corresponding to the excavation step working condition.

[0082] S13. Construct a slope model based on the set finite difference grid points, the Mohr-Coulomb failure criteria and excavation steps corresponding to different rock masses and structural surfaces.

[0083] S2. Select target monitoring points of the slope model;

[0084] The S2 comprises the following steps:

[0085] S21. Install appearance monitoring equipment and anchor cable force measurement equipment in the slope model analysis area;

[0086] S22. Acquire monitoring data from the displacement monitoring device and the anchor force measuring device;

[0087] S23. Select the equipment location where the monitoring data is continuous and uninterrupted and is deployed during the slope excavation period as the target observation point.

[0088] S3. Define the slope safety warning coefficient corresponding to slope instability;

[0089] The S3 comprises the following steps:

[0090] S31, obtaining the actual cohesion and actual internal friction angle of the failure surface corresponding to each excavation step in the slope model analysis area as actual slope shear strength parameters;

[0091] S32. According to the strength reduction method, actual slope shear strength parameters are reduced, and a slope model excavation simulation is performed simultaneously until the slope model enters a slope instability state, and a proportional coefficient corresponding to the slope instability is used as a slope safety warning coefficient;

[0092] The calculation expression of the slope safety warning coefficient is as follows:

[0093] ,

[0094] in, represents the reduced cohesion, represents the actual cohesion, represents the slope safety warning coefficient, represents the reduced internal friction angle, represents the actual internal friction angle.

[0095] S4. Analyze the sensitivity between the structural surface parameters at the target monitoring point and the slope safety warning coefficient, and the sensitivity between the rock and soil material parameters at the target monitoring point and the displacement and anchoring force, to obtain the slope characteristic parameters;

[0096] In this embodiment, the structural surface parameters include the deformation modulus, cohesion, internal friction angle, density and Poisson's ratio of the structural surface, and the rock and soil material parameters include the deformation modulus and cohesion of the rock and soil material;

[0097] The S4 comprises the following steps:

[0098] S41. Analyze the sensitivity between the deformation modulus, cohesion, internal friction angle, gravity, and Poisson's ratio of the structural surface at the target monitoring point and the slope safety warning coefficient, and select the structural surface parameters that change with the slope safety warning coefficient as the first slope characteristic variable;

[0099] S42, analyzing the sensitivity between the deformation modulus and cohesion of the rock and soil material at the target monitoring point and the displacement and anchoring force, and selecting the structural surface parameter with a relative sensitivity greater than a preset sensitivity threshold as the second slope characteristic variable;

[0100] The calculation expression of the relative sensitivity in S42 is as follows:

[0101] ,

[0102] ,

[0103] in, Relative sensitivity, Indicates the The sensitivity ratio of the parameters for sensitivity analysis, represents the number of parameters for sensitivity analysis, Indicates the The displacement response of the slope model corresponding to the upper limit of the parameters for sensitivity analysis is Indicates the The slope model displacement response corresponding to the lower limit of the parameter for sensitivity analysis is represents the displacement response of the slope model for the parameters for which sensitivity analysis is performed, Indicates the The upper limit of the parameters for sensitivity analysis, Indicates the The lower limit of the parameter for sensitivity analysis, represents the parameters for sensitivity analysis, Indicates the The anchoring force response of the slope model corresponding to the upper limit of the parameter for sensitivity analysis is: Indicates the The anchoring force response of the slope model corresponding to the upper limit of the parameter for sensitivity analysis is: Anchor force response of the slope model representing the parameters for which sensitivity analysis is performed.

[0104] In this embodiment, sensitivity analysis is performed on the deformation modulus and cohesion of the rock and soil material at the target monitoring point in each excavation step, and the displacement and anchoring force at the target monitoring point. The parameter with a relative sensitivity greater than 5% is used as the second slope characteristic variable, that is, the sensitivity threshold is 5%.

[0105] S43. Use the first slope characteristic variable and the second slope characteristic variable as slope characteristic parameters.

[0106] S5. Construct a back propagation neural network between slope characteristic parameters and geotechnical responses, and perform Bayesian update to obtain the geotechnical responses of each excavation step;

[0107] The S5 comprises the following steps:

[0108] S51, obtaining a plurality of slope characteristic parameter combinations, and using them as input parameters of a back propagation neural network model;

[0109] S52, using the Latin hypercube sampling method to select a number of training samples and test samples from the input parameters;

[0110] In this embodiment, both the training samples and the test samples are combinations of slope characteristic parameters;

[0111] S53, inputting the training samples and the test samples into the slope model respectively to obtain the geotechnical response of each excavation step, and using the geotechnical response as the output parameter of the corresponding back propagation neural network model, wherein the geotechnical response is displacement or anchoring force;

[0112] In this embodiment, the slope characteristic parameters are used as the input layer neurons of the back-propagation neural network, and the displacement or anchoring force is used as the output layer neurons of the back-propagation neural network; the displacement monitoring data is evaluated from the second excavation step, and the anchoring force monitoring data is evaluated from the third excavation step, so the Bayesian update is performed from the second excavation step. In addition, the slope safety warning coefficient is also evaluated from the second excavation step.

[0113] S54, constructing a back propagation neural network model for each geotechnical response in each excavation step;

[0114] In this embodiment, each back propagation neural network represents a geotechnical response of a target monitoring point after one excavation.

[0115] S55. Assuming that the slope characteristic parameters are independent in the prior distribution, obtain the prior distribution probability density of the slope characteristic parameters;

[0116] The calculation expression of the prior distribution probability density of the slope characteristic parameters is as follows:

[0117] ,

[0118] , ,

[0119] in, represents the prior distribution probability density of slope characteristic parameters, Indicates the Slope characteristic parameters, represents the total number of slope characteristic parameters, Indicates the The standard deviation of the logarithm of the slope characteristic parameters, represents an exponential function with e as the base constant, represents the natural logarithm operation, Indicates the The average value of the logarithms of the slope characteristic parameters, Indicates the The average value of the slope characteristic parameters, Indicates the The standard deviation of the slope characteristic parameters;

[0120] S56. Based on the probability density function of the measurement error of displacement and anchoring force following the normal distribution, a likelihood function for updating the slope characteristic parameters by using displacement and anchoring force in each excavation step is constructed based on the back propagation neural network model;

[0121] The calculation expression of the likelihood function for updating the slope characteristic parameters by displacement and anchoring force in each excavation step in S56 is as follows:

[0122] ,

[0123] in, represents the likelihood function for updating the slope characteristic parameters through displacement and anchoring force at each excavation step, Indicates the excavation step when displacement data starts to be monitored. Indicates from arrive The product operation of represents the slope characteristic parameter, represents the on-site measured displacement at the target monitoring point, represents the on-site measured anchoring force at the target monitoring point, represents the total number of excavation steps, represents the standard deviation of the measurement error of the displacement, Indicates the target monitoring point On-site displacement measurement of each excavation step, Indicates the The displacement output by the back propagation neural network model for each excavation step, Indicates the excavation step when the anchor force data starts to be monitored. Indicates from arrive The product operation of represents the standard deviation of the measurement error of the anchoring force, Indicates the target monitoring point On-site measurement of anchoring force for each excavation step, Indicates the The anchoring force output by the back propagation neural network model during each excavation step.

[0124] S57. Based on the prior distribution probability density of the slope characteristic parameters and the likelihood function of updating the slope characteristic parameters through displacement and anchoring force in each excavation step, Bayesian update is performed to obtain the geotechnical response of each excavation step and the posterior distribution probability density of the slope characteristic parameters.

[0125] In this scheme, the posterior distribution probability density of the slope characteristic parameters can be used to determine the probability density function of the prior distribution at the next excavation step, thereby obtaining the geotechnical response of each excavation step through Bayesian updating.

[0126] S6. Based on the slope characteristic parameters, a relationship mapping model between geotechnical response and slope safety warning coefficient is constructed to obtain the slope safety warning coefficient of each excavation step and perform slope safety warning.

[0127] The S6 comprises the following steps:

[0128] S61. Based on slope characteristic parameters, a mapping model of the relationship between geotechnical response and slope safety warning coefficient is constructed;

[0129] The calculation expression of the relationship mapping model between geotechnical response and slope safety warning coefficient in S61 is as follows:

[0130] ,

[0131] in, represents the slope safety warning coefficient mapped by geotechnical response, represents the geotechnical response of the slope model based on the slope characteristic parameters, represents the inversion error random variable of the slope model, represents the joint probability density prior distribution of slope characteristic parameters, represents the probability density function of the inversion error random variable, represents the differential result of the slope characteristic parameters, represents the differential result of the inversion error random variable, Indicates from 1 to The product operation of represents the total number of geotechnical responses, Indicates the The probability density function of the measurement error corresponding to the geotechnical response is, Indicates the first Geotechnical response, Represents the output of the back-propagation neural network model Geotechnical response.

[0132] S62. Mapping the geotechnical response of each excavation step according to a relationship mapping model between geotechnical response and slope safety warning coefficient to obtain a slope safety warning coefficient for each excavation step;

[0133] S63. If the slope safety warning coefficient of a certain excavation step is greater than or equal to the first safety warning coefficient threshold, the slope under the excavation step is in a safe state and no slope safety warning is issued;

[0134] The safe state usually corresponds to a stable slope structure with no obvious deformation or cracks, monitoring data are all within the normal range, there are no signs of potential instability, and no early warning is issued, but routine monitoring is continued to ensure that data records are complete for subsequent analysis and comparison.

[0135] S64: If the slope safety warning coefficient of a certain excavation step is less than the first safety warning coefficient threshold and greater than or equal to the second safety warning coefficient threshold, the slope under the excavation step is in the first warning state, and a slope safety warning in the first warning state is performed;

[0136] The first warning state typically corresponds to minor deformation or localized cracks on the slope, with monitoring data showing slight anomalies but not reaching dangerous levels. These anomalies may be due to seasonal fluctuations, rainfall, or minor human activity. The slope safety warning that must be implemented during the first warning state is a blue alert. Relevant managers and monitoring teams are notified via text message, email, or internal systems. Inspections of monitoring points are increased, data collection intervals are shortened, and trends are closely monitored. Professional personnel are organized to conduct on-site slope surveys, assess potential risks, clean up drainage systems, and reinforce local areas to prevent the situation from deteriorating.

[0137] S63: If the slope safety warning coefficient of a certain excavation step is less than the second safety warning coefficient threshold and greater than or equal to the third safety warning coefficient threshold, the slope under the excavation step is in the second warning state, and a slope safety warning in the second warning state is performed;

[0138] The second warning state typically corresponds to accelerated slope deformation, enlarged cracks, and significant deviations from normal monitoring data. This indicates the potential risk of localized sliding or collapse, posing a threat to surrounding facilities or personnel. The slope safety warning required for the second warning state is to issue a yellow alert, widely communicate through broadcasts, social media, and emergency contact systems to ensure awareness among all relevant personnel. Automated monitoring equipment transmits real-time data to ensure timely detection of anomalies. Cordons are set up to prohibit unauthorized entry into the danger zone and minimize human interference. Evacuation routes are prepared, emergency drills are organized, and rapid response is ensured in emergencies. Geological and engineering experts are invited to conduct on-site assessments and develop detailed remediation plans.

[0139] S64. If the slope safety warning coefficient of a certain excavation step is less than the third safety warning coefficient threshold, the slope under the excavation step is in the third warning state, and a slope safety warning in the third warning state is performed.

[0140] The second warning state usually corresponds to impending or already existing slope instability, such as large-scale sliding or collapse, with drastic changes in monitoring data, posing a serious threat to human life and property. The corresponding slope safety warning for the third warning state is to issue a red alert, immediately activating the highest level of warning mechanism. This alert is widely disseminated through all available channels, including emergency broadcasts and mobile phone emergency notifications. All personnel in the danger zone are evacuated quickly and orderly according to the pre-established evacuation plan. Traffic in the affected area is closed to prevent entry of unauthorized vehicles and personnel. Professional rescue teams are mobilized, equipped with necessary equipment, and prepared for search and rescue operations. While ensuring safety, slope dynamics are continuously monitored to provide data support for subsequent management. The public is promptly informed of the latest situation to avoid panic and seek social support and understanding.

[0141] In this scheme, the first safety warning coefficient threshold, the second safety warning coefficient threshold and the third safety warning coefficient threshold can be set according to actual conditions. The recommended value range of the first safety warning coefficient threshold is 1.25~1.30, the recommended value range of the second safety warning coefficient threshold is 1.05~1.15, and the recommended value of the third safety warning coefficient threshold is 1.

[0142] This scheme can achieve accurate early warning of slope instability risks in geotechnical engineering, and provide a basis for timely and accurate implementation of slope safety early warning measures under corresponding early warning states.

[0143] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered by the scope of protection of the present invention.

Claims

1. A method for early warning of slope safety in geotechnical engineering, characterized in that: The steps include: S1. Obtain the geological conditions and reinforcement measures of the slope geotechnical engineering site, determine the slope model analysis area, and construct the slope model; S2. Select target monitoring points of the slope model; S3. Define the slope safety warning coefficient corresponding to slope instability; S4. Analyze the sensitivity between the structural surface parameters at the target monitoring point and the slope safety warning coefficient, and the sensitivity between the rock and soil material parameters at the target monitoring point and the displacement and anchoring force, to obtain the slope characteristic parameters; S5. Construct a back propagation neural network between slope characteristic parameters and geotechnical responses, and perform Bayesian update to obtain the geotechnical responses of each excavation step; S6. Based on the slope characteristic parameters, a relationship mapping model between geotechnical response and slope safety warning coefficient is constructed to obtain the slope safety warning coefficient for each excavation step and perform slope safety warning; The calculation expression of the relationship mapping model between geotechnical response and slope safety warning coefficient in S6 is as follows: ， in, represents the slope safety warning coefficient mapped by geotechnical response, represents the geotechnical response of the slope model based on the slope characteristic parameters, represents the inversion error random variable of the slope model, represents the joint probability density prior distribution of slope characteristic parameters, represents the probability density function of the inversion error random variable, represents the differential result of the slope characteristic parameters, represents the differential result of the inversion error random variable, Indicates from 1 to The product operation of represents the total number of geotechnical responses, Indicates the The probability density function of the measurement error corresponding to the geotechnical response is, Indicates the first Geotechnical response, Represents the output of the back-propagation neural network model Geotechnical response.

2. The method for early warning of slope safety in geotechnical engineering according to claim 1, characterized in that: The S1 comprises the following steps: S11. Obtain the geological conditions and reinforcement measures of the slope geotechnical engineering site and determine the slope model analysis area; S12. setting a number of finite difference grid points, Mohr-Coulomb failure criteria, and excavation steps based on the physical parameters of the rock mass and the physical parameters of the structural surface in the slope model analysis area; S13. Construct a slope model based on the set finite difference grid points, the Mohr-Coulomb failure criteria and excavation steps corresponding to different rock masses and structural surfaces.

3. The method for early warning of slope safety in geotechnical engineering according to claim 2, characterized in that: The S2 comprises the following steps: S21. Install appearance monitoring equipment and anchor cable force measurement equipment in the slope model analysis area; S22. Acquire monitoring data from the displacement monitoring device and the anchor force measuring device; S23. Select the equipment location where the monitoring data is continuous and uninterrupted and is deployed during the slope excavation period as the target observation point.

4. The method for early warning of slope safety in geotechnical engineering according to claim 3, characterized in that: The S3 comprises the following steps: S31, obtaining the actual cohesion and actual internal friction angle of the failure surface corresponding to each excavation step in the slope model analysis area as actual slope shear strength parameters; S32. According to the strength reduction method, actual slope shear strength parameters are reduced, and a slope model excavation simulation is performed simultaneously until the slope model enters a slope instability state, and a proportional coefficient corresponding to the slope instability is used as a slope safety warning coefficient; The calculation expression of the slope safety warning coefficient is as follows: ， in, represents the reduced cohesion, represents the actual cohesion, represents the slope safety warning coefficient, represents the reduced internal friction angle, represents the actual internal friction angle.

5. The method for early warning of slope safety in geotechnical engineering according to claim 4, characterized in that: The S4 comprises the following steps: S41. Analyze the sensitivity between the deformation modulus, cohesion, internal friction angle, gravity, and Poisson's ratio of the structural surface at the target monitoring point and the slope safety warning coefficient, and select the structural surface parameters that change with the slope safety warning coefficient as the first slope characteristic variable; S42, analyzing the sensitivity between the deformation modulus and cohesion of the rock and soil material at the target monitoring point and the displacement and anchoring force, and selecting the structural surface parameter with a relative sensitivity greater than a preset sensitivity threshold as the second slope characteristic variable; S43. Use the first slope characteristic variable and the second slope characteristic variable as slope characteristic parameters.

6. The method for early warning of slope safety in geotechnical engineering according to claim 5, characterized in that: The calculation expression of the relative sensitivity in S42 is as follows: ， ， in, Relative sensitivity, Indicates the The sensitivity ratio of the parameters for sensitivity analysis, represents the number of parameters for sensitivity analysis, Indicates the The displacement response of the slope model corresponding to the upper limit of the parameters for sensitivity analysis is Indicates the The slope model displacement response corresponding to the lower limit of the parameter for sensitivity analysis is represents the displacement response of the slope model for the parameters for which sensitivity analysis is performed, Indicates the The upper limit of the parameters for sensitivity analysis, Indicates the The lower limit of the parameter for sensitivity analysis, represents the parameters for sensitivity analysis, Indicates the The anchoring force response of the slope model corresponding to the upper limit of the parameter for sensitivity analysis is: Indicates the The anchoring force response of the slope model corresponding to the upper limit of the parameter for sensitivity analysis is: Anchor force response of the slope model representing the parameters for which sensitivity analysis is performed.

7. The method for early warning of slope safety in geotechnical engineering according to claim 6, characterized in that: The S5 comprises the following steps: S51, obtaining a plurality of slope characteristic parameter combinations, and using them as input parameters of a back propagation neural network model; S52, using the Latin hypercube sampling method to select a number of training samples and test samples from the input parameters; S53, inputting the training samples and the test samples into the slope model respectively to obtain the geotechnical response of each excavation step, and using the geotechnical response as the output parameter of the corresponding back propagation neural network model, wherein the geotechnical response is displacement or anchoring force; S54, constructing a back propagation neural network model for each geotechnical response in each excavation step; S55. Assuming that the slope characteristic parameters are independent in the prior distribution, obtain the prior distribution probability density of the slope characteristic parameters; The calculation expression of the prior distribution probability density of the slope characteristic parameters is as follows: ， ， ， in, represents the prior distribution probability density of slope characteristic parameters, Indicates the Slope characteristic parameters, represents the total number of slope characteristic parameters, Indicates the The standard deviation of the logarithm of the slope characteristic parameters, represents an exponential function with e as the base constant, represents the natural logarithm operation, Indicates the The average value of the logarithms of the slope characteristic parameters, Indicates the The average value of the slope characteristic parameters, Indicates the The standard deviation of the slope characteristic parameters; S56. Based on the probability density function of the measurement error of displacement and anchoring force following the normal distribution, a likelihood function for updating the slope characteristic parameters by using displacement and anchoring force in each excavation step is constructed based on the back propagation neural network model; S57. Based on the prior distribution probability density of the slope characteristic parameters and the likelihood function of updating the slope characteristic parameters through displacement and anchoring force in each excavation step, Bayesian update is performed to obtain the geotechnical response of each excavation step and the posterior distribution probability density of the slope characteristic parameters.

8. The method for early warning of slope safety in geotechnical engineering according to claim 7, characterized in that: The calculation expression of the likelihood function for updating the slope characteristic parameters by displacement and anchoring force in each excavation step in S56 is as follows: ， in, represents the likelihood function for updating the slope characteristic parameters through displacement and anchoring force at each excavation step, Indicates the excavation step when displacement data starts to be monitored. Indicates from arrive The product operation of represents the slope characteristic parameter, represents the on-site measured displacement at the target monitoring point, represents the on-site measured anchoring force at the target monitoring point, represents the total number of excavation steps, represents the standard deviation of the measurement error of the displacement, Indicates the target monitoring point On-site displacement measurement of each excavation step, Indicates the The displacement output by the back propagation neural network model for each excavation step, Indicates the excavation step when the anchor force data starts to be monitored. Indicates from arrive The product operation of represents the standard deviation of the measurement error of the anchoring force, Indicates the target monitoring point On-site measurement of anchoring force for each excavation step, Indicates the The anchoring force output by the back propagation neural network model during each excavation step.

9. The method for early warning of slope safety in geotechnical engineering according to claim 7, characterized in that: The S6 comprises the following steps: S61. Based on slope characteristic parameters, a mapping model of the relationship between geotechnical response and slope safety warning coefficient is constructed; S62. Mapping the geotechnical response of each excavation step according to a relationship mapping model between geotechnical response and slope safety warning coefficient to obtain a slope safety warning coefficient for each excavation step; S63. If the slope safety warning coefficient of a certain excavation step is greater than or equal to the first safety warning coefficient threshold, the slope under the excavation step is in a safe state and no slope safety warning is issued; S64: If the slope safety warning coefficient of a certain excavation step is less than the first safety warning coefficient threshold and greater than or equal to the second safety warning coefficient threshold, the slope under the excavation step is in the first warning state, and a slope safety warning in the first warning state is performed; S65. If the slope safety warning coefficient of a certain excavation step is less than the second safety warning coefficient threshold and greater than or equal to the third safety warning coefficient threshold, the slope under the excavation step is in the second warning state, and a slope safety warning in the second warning state is performed; S66. If the slope safety warning coefficient of a certain excavation step is less than the third safety warning coefficient threshold, the slope under the excavation step is in the third warning state, and a slope safety warning in the third warning state is performed.

Images