Slope stability evaluation and control method based on multi-factor deformation analysis

Abstract

The invention provides a slope stability evaluation and control method based on multi-factor deformation analysis, and belongs to the technical field of geotechnical engineering.The slope stability evaluation and control method comprises the following steps that S1, multiple monitoring parameters of a slope are obtained; s2, predicting the deformation trend of the slope by using an adaptive depth hybrid prediction model based on the monitoring parameters to obtain a deformation trend probability prediction result; s3, synchronously with the prediction step, performing real-time evaluation on the slope stability by using a stability dynamic evaluation model based on the monitoring parameters to obtain a slope stability coefficient F; s4, performing risk judgment according to the deformation trend probability prediction result and the slope stability coefficient F, and determining the current risk level of the slope; and S5, triggering a corresponding hierarchical control measure based on the current risk level, and carrying out advanced control on the slope stability. The problem that according to a traditional scheme, after monitoring data reach a threshold value, a control measure is started, the slope which starts to deform is controlled, and the optimal intervention period is missed is solved.

Description

Technical Field

[0001] This invention belongs to the field of geotechnical engineering technology, and specifically relates to a method for evaluating and controlling slope stability based on multi-factor deformation analysis. Background Technology

[0002] Slope stability assessment is a crucial decision-making basis for slope design, slope stability determination, slope reinforcement and treatment, and prevention of geological disasters. Existing slope control methods are mostly based on post-event response, taking measures only when obvious deformation signs appear, lacking a predictive proactive control mechanism.

[0003] Traditional methods rely on monitoring data reaching a preset alarm threshold before initiating control measures. By this time, the slope has often entered an accelerated deformation phase, resulting in limited control effectiveness, high costs, and inaccurate timing, often missing the optimal intervention period. Therefore, a multi-factor deformation analysis method for slope stability evaluation and control is designed. Summary of the Invention

[0004] This invention provides a slope stability evaluation and control method based on multi-factor deformation analysis, which solves the problem of traditional methods that start control measures only after monitoring data reaches a threshold, thus missing the optimal intervention period for controlling slopes that have already begun to deform.

[0005] In view of the above problems, the technical solution proposed by the present invention is as follows:

[0006] This invention provides a method for slope stability evaluation and control based on multi-factor deformation analysis, comprising the following steps:

[0007] S1 acquires various monitoring parameters of the slope, including surface displacement, deep displacement, soil volumetric water content, groundwater level, and rainfall.

[0008] S2, Based on the monitoring parameters, the adaptive deep hybrid prediction model is used to predict the slope deformation trend and obtain the deformation trend probability prediction result;

[0009] S3, in sync with the prediction step, the slope stability is evaluated in real time using a dynamic stability evaluation model based on the monitoring parameters, and the slope stability coefficient F is obtained;

[0010] S4. Based on the probability prediction results of deformation trends and the slope stability coefficient F, risk assessment is conducted to determine the current risk level of the slope.

[0011] S5, based on the current risk level, triggers corresponding graded control measures to proactively control slope stability.

[0012] Preferably, the construction of the adaptive deep fusion prediction model in step S2 includes the following steps:

[0013] A Gaussian filter prediction model was constructed to extract multi-scale features from monitoring data and separate deformation feature components of different frequencies.

[0014] A deep autoregressive prediction network model is constructed, integrating an attention mechanism to receive raw monitoring data and feature components after Gaussian filtering.

[0015] The Gaussian filter prediction model and the deep autoregressive prediction network model are combined in parallel to form the initial adaptive deep hybrid prediction model.

[0016] Preferably, step S2, which uses an initial adaptive depth fusion prediction model to predict the slope deformation trend, includes the following steps:

[0017] The latest time-series monitoring data is simultaneously input into the Gaussian filter prediction model and the deep autoregressive prediction network model to obtain two prediction values ​​respectively.

[0018] Calculate the prediction errors of the two prediction models, and dynamically allocate the contribution weights of the two prediction models using an inverse error ratio algorithm;

[0019] The two predicted values ​​are weighted and fused based on the assigned weights to generate the final probability prediction result.

[0020] Preferably, the construction of the stability dynamic evaluation model in step S3 includes the following steps:

[0021] Based on real-time monitored soil volumetric water content data, the matrix suction of unsaturated soil is calculated using a soil-water characteristic curve model, and the suction stress is calculated based on the matrix suction.

[0022] A dynamic stability evaluation model considering changes in absorption stress was established based on the actual geological model of the slope, and the slope stability coefficient F was solved using an iterative method.

[0023] Preferably, the stress absorption calculation in step S32 includes the following steps:

[0024] The soil volumetric water content at different depths is obtained from the time domain reflectometer embedded in the slope soil. The effective saturation is calculated based on the volumetric water content to reflect the saturation state of the soil.

[0025] The matric suction of unsaturated soil is obtained by multiplying the matric suction of unsaturated soil with the effective saturation obtained based on the volumetric water content using the soil-water characteristic curve model.

[0026] Preferably, the solution for the slope stability coefficient F includes the following steps:

[0027] By spatially superimposing and comparing the weak zones identified through geological exploration with the abrupt displacement zones identified through deep displacement monitoring, the shape and location of the potential sliding surface of the slope can be determined.

[0028] The soil on the sliding surface is considered as a whole, and the forces acting on it are analyzed, including sliding force, anti-sliding force, cohesion and friction.

[0029] The stability coefficient F of the slope at the current moment is obtained through iterative calculation.

[0030] Preferably, the risk determination in step S4 includes the following steps:

[0031] A four-level risk classification system is established, including stable status, low risk, medium risk, and high risk.

[0032] Risk level assessment is performed based on both the slope stability coefficient F and the predicted deformation trend probability.

[0033] Preferably, the risk level determination under the dual conditions includes:

[0034] When the stability coefficient F ≥ the safety standard threshold A and the deformation trend is stable, it is determined to be a stable state;

[0035] When the warning activation threshold B ≤ stability coefficient F < safety standard threshold A or deformation prediction shows slight acceleration, it is judged as a slight risk;

[0036] When the critical state threshold C ≤ stability coefficient F < early warning activation threshold B or deformation prediction shows a significant acceleration, it is judged as a moderate risk.

[0037] When the stability coefficient F is less than the critical state threshold C or the deformation prediction shows a sharp acceleration, it is judged as high risk.

[0038] Preferably, the graded control measures in step S5 include the following steps:

[0039] For stable conditions, routine monitoring will be implemented, with a monitoring frequency of once per day.

[0040] For mild risks, enhanced monitoring will be implemented, with the monitoring frequency increased to twice a day, and on-site inspections will be strengthened.

[0041] For moderate risks, engineering intervention measures will be implemented, the monitoring frequency will be increased to 4 times / day, and surface drainage and local load reduction measures will be initiated.

[0042] For high-risk situations, implement emergency response measures, including immediate traffic control, personnel evacuation, and slope counter-pressure measures.

[0043] Preferably, the engineering intervention measures also include implementing graded load reduction at the rear edge of the slope, controlling the load reduction rate and volume, repairing and improving the surface drainage system to ensure smooth drainage, and implementing limited counterpressure in the slope toe area to improve anti-sliding capacity.

[0044] Compared with the prior art, the beneficial effects of the present invention are:

[0045] This invention achieves advanced early warning and precise control of slope stability by constructing a synchronous analysis mechanism that integrates an adaptive deep hybrid prediction model and a dynamic stability evaluation model.

[0046] By combining probabilistic prediction of deformation trends with dynamic evaluation of stability coefficients, this invention can identify early signs of stability deterioration, solving the problem of delayed control timing caused by traditional methods relying on threshold alarms. Through dual-condition risk assessment based on stability coefficients and deformation trend prediction results, a more reliable risk identification mechanism is established, overcoming the limitations of single-indicator assessment. By precisely mapping risk levels to tiered control measures, a step-by-step response from routine monitoring to emergency handling is achieved, optimizing the efficiency of control resource allocation.

[0047] This invention, through its innovative design of simultaneous prediction and evaluation, dual risk assessment, and tiered control response, effectively solves the technical problems of existing slope control methods, such as reactive responses, inaccurate control timing, and limited effectiveness. This solution transforms slope stability management from passive emergency response to proactive safeguarding. In summary, this invention addresses the technical problems of delayed slope control timing and poor control effectiveness caused by the lack of prediction and proactive mechanisms in existing technologies.

[0048] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and in order to make the above and other objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0049] Figure 1 This is a flowchart illustrating a slope stability evaluation and control method based on multi-factor deformation analysis disclosed in this invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0052] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0053] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0054] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0055] Example 1

[0056] See attached document Figure 1 As shown, the present invention provides a technical solution: a method for slope stability evaluation and control based on multi-factor deformation analysis, comprising the following steps:

[0057] S1 acquires various monitoring parameters of the slope, including surface displacement, deep displacement, soil volumetric water content, groundwater level, and rainfall.

[0058] S2, Based on the monitoring parameters, the adaptive deep hybrid prediction model is used to predict the slope deformation trend and obtain the deformation trend probability prediction result;

[0059] S3, in sync with the prediction step, the slope stability is evaluated in real time using a dynamic stability evaluation model based on the monitoring parameters, and the slope stability coefficient F is obtained;

[0060] S4. Based on the probability prediction results of deformation trends and the slope stability coefficient F, risk assessment is conducted to determine the current risk level of the slope.

[0061] S5, based on the current risk level, triggers corresponding graded control measures to proactively control slope stability.

[0062] In addition, the execution of steps S1-S5 is carried out through a computer-readable storage medium storing program instructions. The computer-readable storage medium is located in any of a computer, server, or microcontroller, and these devices are equipped with a microprocessor that executes the program instructions stored in the computer-readable storage medium.

[0063] The embodiments of the present invention are also implemented through the following technical solutions.

[0064] In an embodiment of the present invention, the monitoring of multiple parameters in step S1 specifically includes: surface displacement monitoring by deploying GNSS (Global Navigation Satellite System) monitoring stations or total station prisms in key areas of the slope surface (such as slope shoulder, slope toe, and both sides of cracks) to obtain its displacement amount, displacement rate and direction;

[0065] Deep displacement monitoring involves drilling holes in the slope body, installing fixed inclinometers, calculating the angle changes at different depths, obtaining the horizontal displacement and deformation trend at various depths inside the slope, and accurately determining the location and depth of potential sliding surfaces.

[0066] Soil volumetric water content monitoring directly measures real-time volumetric water content data by burying time domain reflectometer probes or soil moisture sensors at different depths (e.g., 0.5m, 1.0m, 2.0m) on the slope.

[0067] Groundwater level monitoring involves drilling groundwater monitoring boreholes (water level boreholes) within the slope, installing automatic water level gauges inside the boreholes, and continuously measuring the elevation of the water surface inside the boreholes (usually pressure type or float type) to obtain data on changes in groundwater level over time.

[0068] Rainfall monitoring is achieved by installing automatic rain gauges near the slope area. The rain gauges record the start and end times of rainfall, as well as the rainfall intensity (mm / hour) and cumulative rainfall in real time.

[0069] All sensor data is transmitted in real time or near real time to a computer-readable storage medium via wireless communication technologies (such as 4G / 5G or LoRa) to form a complete, spatiotemporally synchronized multi-source monitoring database.

[0070] In an embodiment of the present invention, the construction of the adaptive deep fusion prediction model in step S2 includes the following steps:

[0071] A Gaussian filter prediction model was constructed to extract multi-scale features from monitoring data and separate deformation feature components of different frequencies.

[0072] A deep autoregressive prediction network model is constructed, integrating an attention mechanism to receive raw monitoring data and feature components after Gaussian filtering.

[0073] The Gaussian filter prediction model and the deep autoregressive prediction network model are run in parallel. Through weighted fusion, the prediction outputs of the two models are given equal weights (e.g., 0.5 each) in the initial state. The two prediction results are then weighted and averaged to output a comprehensive initial prediction value combination, forming the initial adaptive deep fusion prediction model.

[0074] Next, the slope deformation trend is predicted based on the initial adaptive depth fusion prediction model, including the following steps:

[0075] The latest time-series monitoring data is simultaneously input into the Gaussian filter prediction model and the deep autoregressive prediction network model to obtain two prediction values ​​respectively.

[0076] The prediction errors of the two prediction models are calculated, and the contribution weights of the two prediction models are dynamically allocated using an inverse error ratio algorithm. The contribution weights are adaptively adjusted according to the changing trend of the prediction errors to optimize the performance of the prediction models.

[0077] The two predicted values ​​are weighted and fused based on the assigned weights to generate the final probability prediction result.

[0078] The specific process of allocating the contribution weights of the prediction model using the inverse error ratio algorithm is as follows:

[0079] The specific process of dynamically allocating the contribution weights of the two prediction models using the inverse error ratio algorithm is as follows:

[0080] First, record the prediction errors of the Gaussian filter prediction model and the deep autoregressive prediction network model at each step in the same period of time. Then, calculate the average level of the absolute value of the error of each model during this period. A lower average level indicates that the prediction is more accurate. The more accurate model has more weight. Divide the average error of the other model by the sum of the average errors of the two models to get the weight of the Gaussian filter prediction model. The remaining weight is for the deep autoregressive prediction network model.

[0081] Adjusting Gaussian filter parameters according to error changes: If the error fluctuation of the Gaussian filter prediction model increases during this period, it is assumed that high-frequency noise is increasing. Therefore, the smoothing window is widened to make the curve smoother. Conversely, if the fluctuation decreases, the window is narrowed to retain more detailed features.

[0082] In an embodiment of the present invention, the construction of the stability dynamic evaluation model in step S3 includes the following steps:

[0083] S31, based on real-time monitored soil volumetric water content data, calculates the matrix suction of unsaturated soil through a soil-water characteristic curve model, and calculates the suction stress based on the matrix suction.

[0084] Among them, the soil-water characteristic curve model is a bridge connecting soil moisture content and matrix suction. By fitting the experimental data points to a mature soil-water characteristic curve mathematical model (such as the van Genuchten model), the key characteristic parameters in the model (soil air intake value, soil pore size distribution, etc.) are determined through parameter fitting, which represent the water-holding characteristics of this type of soil.

[0085] The stress absorption calculation includes the following steps:

[0086] The soil volumetric water content at different depths is obtained from the time domain reflectometer embedded in the slope soil. The effective saturation is calculated based on the volumetric water content. The effective saturation is the current measured volumetric water content minus the residual volumetric water content of the soil, and then divided by the difference between the saturated volumetric water content and the residual volumetric water content of the soil, reflecting the saturation state of the soil.

[0087] The matric suction of unsaturated soil is obtained by multiplying the matric suction of unsaturated soil with the effective saturation obtained based on the volumetric water content, and the matric suction stress of unsaturated soil is obtained, thus quantifying the contribution of matric suction to stability.

[0088] S32. Based on the actual geological model of the slope, a dynamic evaluation model for stability considering changes in absorption stress is established. The slope stability coefficient F is solved by iterative calculation. The solution of the slope stability coefficient F includes the following steps:

[0089] By spatially superimposing and comparing the weak zones identified through geological exploration with the abrupt displacement zones identified through deep displacement monitoring, the shape and location of the potential sliding surface of the slope can be determined.

[0090] Specifically, firstly, based on engineering geological mapping and drilling data, identify the stratification, lithological characteristics, geological structures (such as the development direction and density of faults, joints, and fissures) of the slope rock and soil, as well as the location and occurrence of weak interlayers;

[0091] Deep displacement monitoring data identification: By analyzing the deep displacement-depth curves obtained by the inclinometers arranged inside the slope, the locations where the displacement changes significantly, i.e. the depth points where the displacement suddenly increases from a small displacement to a significant displacement, are connected to form a potential sliding surface. If multiple monitoring holes show displacement changes near the same depth, the spatial shape and location of the sliding surface can be determined more confidently.

[0092] Treating the soil on the sliding surface as a whole, we use the slice method to analyze the forces acting on it, including the force that causes the soil to slide downward, which is the sliding force, consisting of the component of the soil's own weight along the sliding surface, and the force that prevents the soil from sliding, which is the antisliding force, consisting of the soil's own cohesion and the frictional force on the sliding surface.

[0093] The dynamic evaluation model for stability considering the variation of suction stress essentially introduces the real-time calculated suction stress of unsaturated soil as an additional strength component into the shear strength criterion of the sliding surface. Based on the Mohr-Coulomb strength theory of unsaturated soil, the contribution of the original matrix suction to the strength is equivalently replaced by suction stress, simplifying the shear strength expression to the sum of effective cohesion, frictional strength generated by total normal stress, and suction stress. Subsequently, in the limit equilibrium slice method calculation, the enhanced shear strength is used to determine the sliding resistance of the bottom surface of each soil slice. The total sliding resistance is dynamically updated with the change of water content, so that the slope stability coefficient can reflect the influence of water field fluctuations on soil strength in real time.

[0094] Based on this, the stability coefficient F of the slope at the current moment is calculated by the slice method as the ratio of the enhanced total anti-sliding force to the total sliding force. The stability coefficient F of the slope at the current moment is obtained by iterative calculation. The stability coefficient F is the ratio of the total available anti-sliding capacity to the total sliding threat on the sliding surface (the total anti-sliding capacity and the total sliding threat are obtained by the slice method). The specific solution process is as follows: First, a reasonable initial estimate of the stability coefficient F is set. The estimated value is substituted into the established mechanical equilibrium equation for calculation. The solution result of the equation is compared with the initial estimate. According to the difference in comparison, the value of the stability coefficient is corrected according to the predetermined algorithm. The calculation results of the estimated value are compared again. When the difference between the stability coefficient F calculated by two consecutive calculations is less than the preset tolerance, it is determined that the calculation has converged. The final converged stability coefficient F is used as the quantitative evaluation index of the current slope stability.

[0095] In an embodiment of the present invention, the risk determination in step S4 includes the following steps:

[0096] A four-level risk classification system was established, including stable state, low risk, medium risk, and high risk. The risk level was determined based on the slope stability coefficient F and the probability prediction results of deformation trends under dual conditions.

[0097] In embodiments of the present invention, the risk level determination under dual conditions includes:

[0098] When the stability coefficient F ≥ the safety standard threshold A and the deformation trend is stable, it is determined to be a stable state;

[0099] When the warning activation threshold B ≤ stability coefficient F < safety standard threshold A or deformation prediction shows slight acceleration, it is judged as a slight risk;

[0100] When the critical state threshold C ≤ stability coefficient F < early warning activation threshold B or deformation prediction shows a significant acceleration, it is judged as a moderate risk.

[0101] When the stability coefficient F is less than the critical state threshold C or the deformation prediction shows a sharp acceleration, it is judged as high risk.

[0102] Among them, the safety standard threshold A adopts the safety factor value required by the engineering design specifications as the criterion for judging the stable state. The early warning activation threshold B is set as a certain proportion of the safety standard threshold B. The specific logic is to reserve buffer time for the data. The proportion coefficient k (i.e., B = k * A) needs to be determined comprehensively based on the slope engineering grade, the importance of the protected object and the noise level of the monitoring data. Usually, the value of k is between 1.15 and 1.25. For example, for a first-level slope, k can be 1.2. Then when A=1.20, B=1.44 (about 1.4).

[0103] The critical state threshold C is the theoretical limit value of the stability coefficient F, which is typically 1.0 in limit equilibrium theory. When F < 1.0, the slope is in a limit equilibrium state or an unstable state. Therefore, in the risk assessment logic of this invention, the critical state threshold C is set to 1.0 as the critical point between moderate risk (slope tending towards instability) and high risk (slope on the verge of or already instability).

[0104] In an embodiment of the present invention, the graded control measures in step S5 include the following steps:

[0105] For stable conditions, routine monitoring will be implemented, with a monitoring frequency of once per day.

[0106] For mild risks, enhanced monitoring will be implemented, with the monitoring frequency increased to twice a day, and on-site inspections will be strengthened.

[0107] For moderate risks, engineering intervention measures will be implemented, the monitoring frequency will be increased to 4 times / day, and surface drainage and local load reduction measures will be initiated.

[0108] For high-risk situations, implement emergency response measures, including immediate traffic control, personnel evacuation, and slope counter-pressure measures.

[0109] In embodiments of the present invention, the engineering intervention measures further include the following steps: implementing graded load reduction at the rear edge of the slope, controlling the load reduction rate and volume, repairing and improving the surface drainage system to ensure smooth drainage, and implementing limited counterpressure in the toe area of ​​the slope to improve anti-sliding capacity.

[0110] Emergency response also includes the following steps: immediately activate the emergency response mechanism, establish an on-site command center, implement traffic control, close off dangerous areas, organize the emergency evacuation of people in the affected areas, and implement rapid counterpressure and emergency support at the foot of the slope.

[0111] In an embodiment of the present invention, the graded control measures in step S5 also include control effect feedback optimization. By monitoring the slope response data after the implementation of the control measures, the effectiveness of the control measures is evaluated based on the slope response. Based on the evaluation results, the prediction model parameters and control strategies are dynamically corrected, the control experience database is updated, and the system's adaptive capability is improved.

[0112] Specifically, the control effect feedback optimization includes: automatically adding newly collected monitoring data (such as deformation rate and stability coefficient F) after the implementation of control measures as new samples to the historical dataset used to train the adaptive deep hybrid prediction model; periodically (e.g., weekly or monthly) fine-tuning the weights of the deep autoregressive prediction network based on the added new data; and recalculating the contribution weights using the latest error data, so that the prediction model can adapt to the new dynamics of the slope after engineering control.

[0113] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various modifications and variations. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0114] It should be understood that the specific order or hierarchy of steps in the disclosed process is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process may be rearranged without departing from the scope of this disclosure. The appended method claims provide elements of various steps in an exemplary order and are not intended to limit the scope to the specific order or hierarchy described.

[0115] In the detailed description above, various features are combined together in a single embodiment to simplify this disclosure. This approach to disclosure should not be construed as reflecting an intention that embodiments of the claimed subject matter require more features than are explicitly stated in each claim. Rather, as reflected in the appended claims, the invention is presented with fewer features than all of the features in a single disclosed embodiment. Therefore, the appended claims are hereby explicitly incorporated into the detailed description, with each claim representing a separate preferred embodiment of the invention.

[0116] Those skilled in the art will also understand that the various illustrative logic blocks, modules, circuits, and algorithm steps described in conjunction with the embodiments herein can be implemented as electronic hardware, computer software, or a combination thereof. To clearly illustrate the interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps described above are generally described in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system. Those skilled in the art can implement the described functionality in alternative ways for each specific application; however, such implementation decisions should not be construed as departing from the scope of this disclosure.

[0117] The steps of the methods or algorithms described in conjunction with the embodiments herein can be directly embodied in hardware, software modules executed by a processor, or a combination thereof. The software modules can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium well known in the art. An exemplary storage medium is connected to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. The ASIC can reside in a user terminal. Alternatively, the processor and storage medium can exist as discrete components in the user terminal.

[0118] For software implementation, the techniques described in this application can be implemented using modules (e.g., procedures, functions, etc.) that perform the functions described in this application. This software code can be stored in memory units and executed by a processor. The memory units can be implemented within the processor or outside the processor; in the latter case, they are communicatively coupled to the processor via various means, as is well known in the art.

[0119] The foregoing description includes examples of one or more embodiments. It is certainly impossible to describe all possible combinations of components or methods in order to describe the above embodiments, but those skilled in the art will recognize that further combinations and arrangements of the various embodiments are possible. Therefore, the embodiments described herein are intended to cover all such changes, modifications, and variations that fall within the scope of the appended claims. Furthermore, the term "comprising" as used in the specification or claims is interpreted in a manner similar to the term "including," as interpreted when used as a conjunction in the claims. Additionally, the use of any term "or" in the specification of the claims is intended to mean "non-exclusive or."

Claims

1. A method for slope stability evaluation and control based on multi-factor deformation analysis, characterized in that, Includes the following steps: S1 acquires various monitoring parameters of the slope, including surface displacement, deep displacement, soil volumetric water content, groundwater level, and rainfall. S2, Based on the monitoring parameters, the adaptive deep hybrid prediction model is used to predict the slope deformation trend and obtain the deformation trend probability prediction result; S3, in sync with the prediction step, the slope stability is evaluated in real time using a dynamic stability evaluation model based on the monitoring parameters, and the slope stability coefficient F is obtained; S4. Based on the probability prediction results of deformation trends and the slope stability coefficient F, risk assessment is conducted to determine the current risk level of the slope. S5, based on the current risk level, triggers corresponding graded control measures to proactively control slope stability.

2. The slope stability evaluation and control method based on multi-factor deformation analysis according to claim 1, characterized in that, The construction of the adaptive deep fusion prediction model in step S2 includes the following steps: A Gaussian filter prediction model was constructed to extract multi-scale features from monitoring data and separate deformation feature components of different frequencies. A deep autoregressive prediction network model is constructed, integrating an attention mechanism to receive raw monitoring data and feature components after Gaussian filtering. The Gaussian filter prediction model and the deep autoregressive prediction network model are combined in parallel to form the initial adaptive deep hybrid prediction model.

3. The slope stability evaluation and control method based on multi-factor deformation analysis according to claim 2, characterized in that, Step S2, which uses an initial adaptive depth fusion prediction model to predict the slope deformation trend, includes the following steps: The latest time-series monitoring data is simultaneously input into the Gaussian filter prediction model and the deep autoregressive prediction network model to obtain two prediction values ​​respectively. Calculate the prediction errors of the two prediction models, and dynamically allocate the contribution weights of the two prediction models using an inverse error ratio algorithm; The two predicted values ​​are weighted and fused based on the assigned weights to generate the final probability prediction result.

4. The slope stability evaluation and control method based on multi-factor deformation analysis according to claim 1, characterized in that, The construction of the stability dynamic evaluation model in step S3 includes the following steps: Based on real-time monitored soil volumetric water content data, the matrix suction of unsaturated soil is calculated using a soil-water characteristic curve model, and the suction stress is calculated based on the matrix suction. A dynamic stability evaluation model considering changes in absorption stress was established based on the actual geological model of the slope, and the slope stability coefficient F was solved using an iterative method.

5. The slope stability evaluation and control method based on multi-factor deformation analysis according to claim 4, characterized in that, The stress absorption calculation in step S32 includes the following steps: The soil volumetric water content at different depths is obtained from the time domain reflectometer embedded in the slope soil. The effective saturation is calculated based on the volumetric water content to reflect the saturation state of the soil. The matric suction of unsaturated soil is obtained by multiplying the matric suction of unsaturated soil with the effective saturation obtained based on the volumetric water content using the soil-water characteristic curve model.

6. The slope stability evaluation and control method based on multi-factor deformation analysis according to claim 5, characterized in that, The solution for the slope stability coefficient F includes the following steps: By spatially superimposing and comparing the weak zones identified through geological exploration with the abrupt displacement zones identified through deep displacement monitoring, the shape and location of the potential sliding surface of the slope can be determined. The soil on the sliding surface is considered as a whole, and the forces acting on it are analyzed, including sliding force, anti-sliding force, cohesion and friction. The stability coefficient F of the slope at the current moment is obtained through iterative calculation.

7. The slope stability evaluation and control method based on multi-factor deformation analysis according to claim 6, characterized in that, The risk assessment in step S4 includes the following steps: A four-level risk classification system is established, including stable status, low risk, medium risk, and high risk. Risk level assessment is performed based on both the slope stability coefficient F and the predicted deformation trend probability.

8. The slope stability evaluation and control method based on multi-factor deformation analysis according to claim 7, characterized in that, The risk level assessment under the dual conditions includes: When the stability coefficient F ≥ the safety standard threshold A and the deformation trend is stable, it is determined to be a stable state; When the warning activation threshold B ≤ stability coefficient F < safety standard threshold A or deformation prediction shows slight acceleration, it is judged as a slight risk; When the critical state threshold C ≤ stability coefficient F < early warning activation threshold B or deformation prediction shows a significant acceleration, it is judged as a moderate risk. When the stability coefficient F is less than the critical state threshold C or the deformation prediction shows a sharp acceleration, it is judged as high risk.

9. The slope stability evaluation and control method based on multi-factor deformation analysis according to claim 8, characterized in that, The graded control measures in step S5 include the following steps: For stable conditions, routine monitoring will be implemented, with a monitoring frequency of once per day. For mild risks, enhanced monitoring will be implemented, with the monitoring frequency increased to twice a day, and on-site inspections will be strengthened. For moderate risks, engineering intervention measures will be implemented, the monitoring frequency will be increased to 4 times / day, and surface drainage and local load reduction measures will be initiated. For high-risk situations, implement emergency response measures, including immediate traffic control, personnel evacuation, and slope counter-pressure measures.

10. A slope stability evaluation and control method based on multi-factor deformation analysis according to claim 9, characterized in that, The engineering intervention measures also include implementing graded load reduction at the rear edge of the slope, controlling the rate and volume of load reduction, repairing and improving the surface drainage system to ensure smooth drainage, and implementing limited counterpressure in the toe area of ​​the slope to improve anti-sliding capacity.

Images