A method and system for evaluating the stability of a loess slope of a mountainous highway based on GNSS

Abstract

The application provides a loess slope stability evaluation method and system based on GNSS, and relates to the technical field of loess slope stability evaluation.The application first divides the slope area into grids, obtains geological feature data of each grid, divides super nodes according to the geological feature data, determines monitoring points and monitors soil moisture content to analyze the soil moisture content at different depths.Secondly, the original displacement is monitored by using GNSS, the displacement is corrected based on the Fredlund unsaturated soil shear strength theory to obtain the actual displacement, the influence of cumulative rainfall on the soil strength parameter is analyzed, and the soil strength parameter is corrected in real time, the corrected soil strength parameter and the actual displacement are coupled, the safety factor is calculated, and the early warning state is judged according to the rainfall sensitivity coefficient.Finally, the stability of the slope is comprehensively judged according to the early warning state.The method improves the accuracy of the slope stability evaluation through comprehensive analysis of multiple parameters.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of loess slope stability evaluation, in particular to a loess slope stability evaluation method and system for mountainous highway based on GNSS. BACKGROUND

[0002] The stability of loess slopes on mountainous highways is an important factor in ensuring traffic safety and infrastructure durability, especially during rainfall. The soil properties in loess regions make them highly susceptible to changes when water penetrates, increasing the risk of slope landslides. Rainfall not only significantly increases soil moisture content but also causes pore water pressure to rise, thereby weakening the soil's shear strength and leading to slope instability. This phenomenon is particularly evident in mountainous areas, where complex terrain and variable weather conditions make the impact of rainfall on slope stability more pronounced. Furthermore, with the increasing frequency of extreme weather events, changes in rainfall patterns exacerbate the safety risks of loess slopes. Before rainfall, the soil is usually in a relatively dry state and has some stability; however, after rainfall, the rapid penetration of water and saturation of the soil can cause the slope to quickly lose its original stability. This dynamic change makes monitoring and evaluating the stability of slopes before and after rainfall particularly important, and timely understanding of soil conditions and slope changes can help take effective preventive measures to reduce landslide risks and ensure safe travel on mountainous highways.

[0003] Traditional monitoring methods such as soil sampling and manual observation have problems such as insufficient accuracy, poor real-time performance, and lagging response to dynamic changes, especially during rainfall events. These methods are difficult to accurately predict potential landslide risks. Loess soil undergoes significant physical property changes before and after rainfall, leading to increased pore water pressure and reduced soil shear strength, which can trigger landslides. Therefore, how to efficiently and accurately monitor and evaluate the stability of slopes during rainfall has become a pressing technical challenge.

[0004] The above information disclosed in the background section is only intended to strengthen the understanding of the background of the present disclosure, and therefore it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0005] The present application aims to provide a loess slope stability evaluation method and system for mountainous highway based on GNSS to solve the problems raised in the background.

[0006] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0007] A loess slope stability evaluation method for mountainous highway based on GNSS, the specific steps include:

[0008] Step 1: divide the slope area into a plurality of grids, and obtain the geological feature data of each grid, divide all the grids into a plurality of super nodes according to the geological feature data, set a monitoring point in each super node, obtain the water content of the soil at the monitoring point, and analyze the water content of the soil at different depths;

[0009] Step 2: obtain the original displacement of the monitoring point through GNSS, analyze the water content of the soil at the monitoring point based on the Fredlund unsaturated soil shear strength theory, to obtain the displacement correction amount, and correct the original displacement according to the displacement correction amount to obtain the actual displacement of the monitoring point;

[0010] Step 3: obtain the cumulative rainfall and soil strength parameters of the position of the monitoring point, analyze the influence of the cumulative rainfall on the soil strength parameters, and correct the soil strength parameters in real time according to the cumulative rainfall;

[0011] Step 4: coupling the real-time corrected soil strength parameters of the monitoring point with the actual displacement to determine the safety factor of the monitoring point, and determining the rainfall sensitivity factor of the monitoring point through the cumulative rainfall of the monitoring point, determining the early warning state of each monitoring point according to the rainfall sensitivity factor and the safety factor, and determining the stability of the slope according to the early warning state.

[0012] Further, according to the geological feature data, all the grids are divided into a plurality of super nodes, which specifically includes:

[0013] The geological feature data includes soil type, slope, void ratio, and permeability coefficient, the slope area is divided into a plurality of grids, each grid is regarded as a node, and the geological feature data corresponding to each node forms a node geological feature vector, which is represented as:

[0014] wherein, is the geological feature vector, st is the soil type, S is the slope, Vr is the void ratio, Pc is the permeability coefficient, and HWc is the historical water content, i.e. the average water content of the soil surface corresponding to the node in the 24 hours before the start of the evaluation time;

[0015] The slope, void ratio, permeability coefficient and historical water content are subjected to maximum-minimum normalization, and the soil type is converted into One-Hot encoding;

[0016] The geological feature difference between each node is calculated, and the formula is:

[0017]

[0018] wherein, is the geological feature difference between the a-th node and the b-th node, is the weight of the k-th geological feature data, the normalized value of the kth geological feature data of the ath node, the normalized value of the kth geological feature data of the bth node, k is the index of the geological feature data type other than the soil type, and k ∈ [1, 4], is a soil type difference penalty coefficient, is an indicator function, 1 if the soil types of the ath node and the bth node are different, otherwise 0, a and b are indexes of the nodes, and ;

[0019] Each node is regarded as an independent cluster, and an initial cluster set is generated, denoted as , where C represents the initial cluster set, and N represents the total number of nodes; all nodes are traversed, and the geological feature difference between the nodes is calculated, if the geological feature difference is less than a preset difference threshold, and the two nodes are not merged, then the clusters to which the nodes belong are merged, and the iteration is repeated until no new merging occurs, and each merged cluster is regarded as a super node.

[0020] Further, the analysis of the water content at different depths of the soil includes:

[0021] The geometric center of each super node is regarded as a monitoring point, sampling points are set at multiple depths below the monitoring point, and the water content of the soil at the sampling points is obtained, the relationship between the water content and the depth at the sampling points is analyzed, and the following expression is constructed to represent the water content at different depths of the monitoring point, the expression is as follows:

[0022]

[0023] , wherein represents the water content at a vertical downward position z meters below the monitoring point at the current time, represents the surface water content of the monitoring point at the current time, z is the depth, the unit is meters, and t represents the current time, is an exponential decay rate.

[0024] Further, the actual displacement of each monitoring point is obtained, which specifically includes:

[0025] The original displacement of each monitoring point is obtained through GNSS, and the displacement correction amount is calculated based on the Fredlund unsaturated soil shear strength theory, and the calculation formula is as follows:

[0026]

[0027] , wherein is the displacement correction amount at the current time, represents the average value of the water content at different depths of the monitoring point at the time , and is the plastic limit water content of loess. is a saturated water content, is a displacement correction coefficient, is a time variable, and , represents 24 hours;

[0028]

[0029] wherein L is a water content influence depth, is a depth correction term coefficient;

[0030] The original displacement of GNSS monitoring is corrected according to the displacement correction amount, and the calculation formula is:

[0031]

[0032] wherein, is the actual displacement of the monitoring point after correction at the current time, is the original displacement of the monitoring point obtained by GNSS.

[0033] Further, the real-time correction of the soil strength parameter specifically includes:

[0034] An automatic rain gauge is installed at each monitoring point to obtain the 24-hour cumulative rainfall at each monitoring point. When the rain gauge detects , the sampling frequency is set to Hz, and when the rain gauge detects , the sampling frequency is increased, specifically by the following formula:

[0035]

[0036] wherein f is the increased sampling frequency, is the hourly rainfall intensity, is the influence coefficient;

[0037] The soil strength parameter includes effective cohesion, effective internal friction angle and pore water pressure. The influence of cumulative rainfall on the soil strength parameter is analyzed and corrected, and the formula is expressed as:

[0038]

[0039] wherein, is the effective cohesion under the influence of cumulative rainfall, is the effective cohesion in the dry state, is the 24-hour cumulative rainfall, is the influence coefficient of the effective cohesion;

[0040]

[0041] wherein, is the internal friction angle under the influence of cumulative rainfall, is the internal friction angle in dry state, is the influence coefficient of the internal friction angle, and the internal friction angle °;

[0042]

[0043] wherein, is the pore water pressure, is the pore water pressure weight, is the influence coefficient of the pore water pressure.

[0044] Further, the determining the safety factor specifically comprises:

[0045] determining a unit soil strip in each monitoring point, recording the soil strip volume of the unit soil strip to obtain the unit soil strip self-weight data, and coupling the real-time corrected soil strength parameter and the actual displacement of each monitoring point to calculate the safety factor of each monitoring point, and the calculation formula is:

[0046]

[0047] wherein, is the safety factor at the current time, is the normal stress of the unit soil strip at the current time, i.e. the vertical stress of the unit soil strip sliding surface, is the unit soil strip self-weight at the monitoring point, is the inclination of the unit soil strip sliding surface at the current time at the monitoring point, i.e. the slope angle of the bottom sliding surface of the slope where the unit soil strip is located, is the slope angle sine value, is the corrected actual displacement of the monitoring point at the current time;

[0048]

[0049] wherein, is the loess unit weight, is the unit soil strip volume.

[0050] Further, the determining the rainfall sensitivity coefficient specifically comprises:

[0051] analyzing the cumulative rainfall data to obtain the rainfall sensitivity coefficient, and the calculation formula is:

[0052]

[0053] wherein, represents the rainfall sensitivity coefficient at the current time, is the displacement correction amount of the monitoring point at the moment, is a time variable, and ∈[ t represents the current moment, is 24 hours, is the influence coefficient of the cumulative rainfall in the rainfall sensitivity coefficient, is the initial shear modulus of the soil, is the shear modulus of the soil at the current moment, is the cumulative rainfall in the previous 24 hours at the current moment.

[0054] Further, the determination of the stability of the slope according to the early warning state specifically comprises:

[0055] when , only the safety factor of the monitoring point at the current moment is analyzed, if the safety factor of the monitoring point is higher than the preset safety threshold, it is determined that the monitoring point is in a safe state at the current moment, if the safety factor of the monitoring point is lower than the preset safety threshold, it is determined that the monitoring point is in an abnormal state at the current moment, and a red early warning needs to be issued;

[0056] if and , a yellow early warning is issued, if and lasts for more than 2 hours, and , an orange early warning is issued, if and , a red early warning is issued, respectively are rainfall sensitivity coefficient thresholds, and , respectively are safety factor thresholds, and , is the safety factor at the current moment;

[0057] if there are two or more monitoring points issuing red early warnings, it is determined that the slope is in an unstable state at the current moment, if there are half or more monitoring points issuing orange early warnings, it is determined that the slope is in an unstable state at the current moment, if there are two-thirds of the monitoring points issuing yellow early warnings, it is determined that the slope is in an unstable state at the current moment, and the rest of the cases are all determined that the slope is in a stable state.

[0058] The application further provides a loess slope stability evaluation system for mountainous highway based on GNSS, which is used to realize the loess slope stability evaluation method for mountainous highway based on GNSS, and comprises:

[0059] The regional division module is configured to divide the slope region into a plurality of grids, obtain geological feature data of each grid, divide all the grids into a plurality of super nodes according to the geological feature data, set a monitoring point in each super node, and obtain the water content of soil at the monitoring point to analyze the water content of soil at different depths.

[0060] The displacement correction module is configured to obtain the original displacement of the monitoring point through GNSS, analyze the water content of soil at the monitoring point based on the Fredlund unsaturated soil shear strength theory to obtain a displacement correction amount, and correct the original displacement according to the displacement correction amount to obtain the actual displacement of the monitoring point.

[0061] The rainfall and soil parameter coupling module is configured to obtain the cumulative rainfall and soil strength parameter at the location of the monitoring point, analyze the influence of the cumulative rainfall on the soil strength parameter, and correct the soil strength parameter in real time according to the cumulative rainfall.

[0062] The safety evaluation and early warning decision module is configured to couple the real-time corrected soil strength parameter and the actual displacement of the monitoring point to determine the safety factor of the monitoring point, determine the rainfall sensitivity factor of the monitoring point through the cumulative rainfall of the monitoring point, determine the early warning state of each monitoring point according to the rainfall sensitivity factor and the safety factor, and determine the stability of the slope according to the early warning state.

[0063] In the above technical solution, the present application provides the following technical effects and advantages:

[0064] The present application realizes high-precision dynamic evaluation of the stability of loess slopes of mountainous highways by combining GNSS monitoring technology and the Fredlund unsaturated soil shear strength theory. By dividing the slope region into grids and super nodes, combining multi-source data such as soil water content and cumulative rainfall, obtaining soil water content at different depths and displacement correction models, the soil strength parameter and actual displacement can be corrected in real time. This multi-parameter coupling analysis method significantly improves the accuracy of slope stability evaluation, especially under rainfall conditions, it can timely capture the dynamic changes of soil properties, thereby effectively predicting potential landslide risks.

[0065] In addition, the scheme realizes a multi-level early warning function by introducing a dual evaluation mechanism of rainfall sensitivity factor and safety factor. According to different threshold combinations, the system can issue yellow, orange or red warnings, thereby providing scientific basis for decision makers. This hierarchical early warning mechanism not only improves the pertinence and timeliness of early warning, but also comprehensively judges the overall stability of the slope according to the distribution of monitoring points, providing reliable technical support for the safe operation of mountainous highways. Overall, this scheme effectively solves the problems of insufficient precision and poor real-time performance of traditional methods through real-time monitoring, dynamic correction and intelligent early warning. BRIEF DESCRIPTION OF DRAWINGS

[0066] Figure 1 The whole method flowchart of the present application is shown in the figure;

[0067] Figure 2 The system structure diagram of the present application is shown in the figure;

[0068] Figure 3 The effective cohesion and safety factor relationship diagram under the influence of the cumulative rainfall of the present application is shown in the figure;

[0069] Figure 4 The pore water pressure and safety factor relationship diagram of the present application is shown in the figure;

[0070] Figure 5 The internal friction and safety factor relationship diagram under the influence of the cumulative rainfall of the present application is shown in the figure;

[0071] Figure 6 The actual displacement and safety factor relationship diagram of the present application is shown in the figure;

[0072] Figure 7 The 24-hour cumulative rainfall and safety factor relationship diagram of the present application is shown in the figure. DETAILED DESCRIPTION

[0073] In order to make the purpose, technical scheme and advantages of the present application more clear and explicit, the present application is further described in detail below in combination with specific embodiments.

[0074] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present application should be understood as the general meaning understood by those skilled in the art to which the present application belongs. The "first", "second" and similar words used in the present application do not represent any order, quantity or importance, but are only used to distinguish different components. "Include" or "contain" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connection" or "connection" and similar words are not limited to physical or mechanical connection, but can include electrical connection, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to represent relative positional relationship, when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0075] Embodiment:

[0076] Please refer to Figures 1 to 7 The present application provides a technical scheme:

[0077] A mountainous highway loess slope stability evaluation method based on GNSS, the specific steps include:

[0078] Step 1: divide the slope area into a plurality of grids, and obtain the geological feature data of each grid, divide all the grids into a plurality of super nodes according to the geological feature data, set a monitoring point in each super node, and obtain the water content of the soil at the monitoring point to analyze the water content of the soil at different depths;

[0079] In this embodiment, dividing all the grids into a plurality of super nodes according to the geological feature data specifically includes:

[0080] The geological feature data includes soil type, slope, void ratio, permeability coefficient, the slope area is divided into a plurality of grids, each grid is regarded as a node, and the geological feature data corresponding to each node forms a node geological feature vector, which is represented as:

[0081] wherein, is the geological feature vector, st is the soil type, S is the slope, Vr is the void ratio, Pc is the permeability coefficient, and HWc is the historical water content, i.e., the average water content of the soil surface of the node in the previous 24 hours at the current time.

[0082] Soil samples at different positions of the slope can be obtained through geological drilling or artificial exploration pits, and a portable soil classifier such as a laser particle size analyzer can be used to quickly analyze the soil composition; the soil type can also be determined by laboratory analysis of the soil samples. The slope data can be collected by a GNSS receiver to obtain high-precision three-dimensional coordinates of the grid nodes, and the local slope can be calculated by the height difference and horizontal distance between adjacent nodes; or a total station or theodolite can be used to measure the slope angle, which is suitable for small-scale supplementary verification. The void ratio data are tested by the cutting ring method on the soil samples. The permeability coefficient can also be measured by a double-ring infiltration test; or it can be obtained by laboratory testing of soil samples using a variable head permeameter or a constant head permeameter. Soil water content sensors are arranged in each grid, and data are recorded automatically every 1 hour, and the average value of the previous 24 hours at the start evaluation time is calculated as the historical water content.

[0083] The slope, void ratio, permeability coefficient, and historical water content are subjected to maximum-minimum normalization, the soil type is converted to One-Hot encoding, and the soil type is divided into three types, namely clay, silt, and sand, with clay corresponding to the code 100, silt corresponding to the code 010, and sand corresponding to the code 001.

[0084] The geological feature difference between the nodes is calculated, and the formula is:

[0085]

[0086] wherein, is the geological feature difference between the a th node and the b th node, is the weight of the kth geological feature data of the ath node, is the normalized value of the kth geological feature data of the ath node, is the normalized value of the kth geological feature data of the bth node, where the geological feature data is not included in the summation, k is the index of the geological feature data other than soil type, and k ∈ [1, 4], is the soil type difference penalty coefficient, is the indicator function, which is 1 if the soil types of the ath node and the bth node are different, and 0 otherwise, a and b are the indices of the nodes, and ;

[0087] Each node is considered as an independent cluster, and an initial cluster set is generated, denoted as where C represents the initial cluster set, and N represents the total number of nodes; all nodes are traversed, and the geological feature difference between nodes is calculated, if the geological feature difference is less than the preset difference threshold, and the two nodes are not merged, then the clusters to which they belong are merged, and the iteration is repeated until no new merging occurs, and each merged cluster is considered as a super node.

[0088] Here, the first need to analyze the slope divided into regions, so here is through the historical geological features of the slope data is divided, so you can consult the existing geological map, geological report and related literature to obtain, or through the relevant database to query. By dividing the slope area into several grids and obtaining the geological feature data of each grid, this method can realize the fine analysis of the geological characteristics of the slope area. Such subdivision makes each node more accurately reflect the local geological conditions, providing a reliable data basis for subsequent stability evaluation. Dividing all grids into multiple super nodes can effectively integrate the node information of similar geological features, reduce the computational complexity, and improve the efficiency of subsequent analysis. The setting of super nodes makes it possible to perform calculations based on smaller node sets when evaluating stability, reducing the difficulty of data processing. By setting monitoring points in each super node, the water content data of different depths of soil can be obtained. This data not only reflects the water content of surface and deep soil, but also helps to establish a more accurate vertical distribution model of water content, providing important water data support for slope stability analysis. Calculating the geological feature difference between nodes, combined with methods such as max-min normalization and One-Hot encoding, makes different soil types and features quantifiable. This method ensures that the comparison between nodes is fair and reliable, helping to form effective super nodes in subsequent clustering and analysis.

[0089] The formula for calculating the difference in geological characteristics between nodes combines multiple geological characteristic data (slope, porosity ratio, permeability coefficient, and historical water content) to provide a comprehensive measure of the difference in geological characteristics. This multi-dimensional consideration helps to fully reflect the geological differences between nodes and avoids the bias that may be caused by a single characteristic. The maximum-minimum normalization method is used to process different characteristics with uniform dimensions, ensuring that each characteristic is compared on the same scale. Slope is generally more important, especially in steep slope areas, where the inclination angle has a greater impact on stability, so it is recommended to give it a higher weight. The porosity ratio affects the bearing capacity and permeability of the soil, so it should also be given a medium to high weight. The permeability coefficient has a significant impact on water flow and slope stability, so the weight can be set to medium. The historical water content can provide important background information for stability, but compared to other factors, it has less impact on regional division, so it is recommended to give it the lowest weight.

[0090] In this embodiment, analyzing the water content at different depths of the soil specifically includes:

[0091] The geometric center of each super node is regarded as a monitoring point, sampling points are set at multiple depths below the monitoring point, and the water content of the soil at the sampling points is obtained. The relationship between the water content and the depth at the sampling points is analyzed to construct the following expression to represent the water content at different depths of the monitoring point:

[0092]

[0093] wherein, represents the water content at a vertical position z meters below the monitoring point at the current time, represents the surface water content of the monitoring point at the current time, z is the depth in meters, and t represents the current time, is the exponential decay rate;

[0094] The exponential decay rate reflects the tendency of water to gradually weaken due to soil adsorption, root absorption, evaporation, or drainage during vertical infiltration, the greater the value, the weaker the soil's ability to retain water or the smaller the resistance to downward infiltration, and the water content decreases rapidly with depth, the smaller the value, the more gradual the change in water content with depth, indicating that water can be more evenly distributed to the deep layer. The exponential decay rate is obtained by substituting the measured water content of the soil, the depth, and the surface water content at the sampling points into the above formula and fitting by the least squares method.

[0095] Step 2: Obtain the original displacement of the monitoring point through GNSS, analyze the water content of the soil at the monitoring point based on the Fredlund unsaturated soil shear strength theory to obtain a displacement correction amount, and correct the original displacement according to the displacement correction amount to obtain the actual displacement of the monitoring point;

[0096] In this embodiment, the actual displacement of each monitoring point specifically includes:

[0097] The original displacement of each monitoring point is obtained by GNSS, and the displacement correction amount is calculated based on the Fredlund unsaturated soil shear strength theory, and the calculation formula is as follows:

[0098]

[0099] wherein, is the displacement correction amount at the current time, represents the average value of the water content at different depths of the monitoring point at time, is the plastic limit water content of loess, and the plastic limit water content of most loess is between 15% and 25%, so the value can be taken as 18%, or it can be obtained by consulting relevant materials of the geographical location of the slope; is the saturated water content, which can be determined by saturation test (such as capillary saturation method) according to the specification, is the displacement correction amount coefficient, which can be obtained by fitting historical data, is a time variable, and , represents 24 hours;

[0100]

[0101] wherein, L is the water content influence depth, is the depth correction term coefficient. Here, the water content influence depth represents the limit depth of the influence of water content on displacement change, and the water content of deeper soil no longer affects the displacement change of the monitoring point after exceeding this depth. For a general loess slope, the value range of L is [5, 8], and the unit is meter, because the plastic collapse zone of loess is usually concentrated in [0, 6] meters, and this range is significantly affected by rainfall infiltration and evaporation.

[0102] The original displacement of GNSS monitoring is corrected according to the displacement correction amount, and the calculation formula is:

[0103]

[0104] wherein, is the corrected actual displacement of the monitoring point at the current time, is the original displacement of the monitoring point obtained by GNSS.

[0105] The three-dimensional coordinate change (east, north, and elevation direction) obtained by the GNSS receiver through satellite signal solution is usually output with millimeter-level precision, which can effectively monitor the displacement change of each monitoring point. However, after the loess absorbs water, the montmorillonite mineral expands, causing the GNSS monitoring point to rise (false displacement), or the water content increases, reducing the stiffness of the soil body, and intensifying the settlement of the GNSS antenna tower, i.e., the swelling effect and softening effect. Therefore, by analyzing the water content in the soil and correcting the displacement collected by the GNSS, the actual displacement data can be obtained, thereby improving the accuracy of subsequent analysis.

[0106] By combining GNSS technology with water content at different locations, real-time displacement data of slope monitoring points can be obtained. This accurate displacement monitoring enables engineers to timely discover potential slope instability risks, improving the management ability of slope safety. By using the displacement correction formula, the monitoring data can be dynamically adjusted according to the actual water content of the soil. This dynamic correction mechanism ensures that the monitoring data is more accurate and can better reflect the true soil displacement. By using the Fredlund unsaturated soil shear strength theory, the calculation of displacement correction has a solid theoretical basis, effectively considering the unsaturated characteristics of the soil, and improving the scientificity and credibility of the model. The displacement correction coefficient and the depth correction term coefficient obtained by fitting historical data provide robust parameters for long-term monitoring, which is crucial for continuous slope stability analysis. In the displacement correction formula, represents the change of soil water content over time, is the plastic limit water content, is the saturated water content. By comparing the deviation of the actual water content from these two values, the influence of the water state of the soil on its shear strength and displacement change can be effectively captured, reflecting the direct impact of water on soil deformation and stability. The depth correction term coefficient allows the model to adjust according to the depth characteristics of the soil, considering the changes in physical properties (such as compactness, pore structure, etc.) of the soil at different depths, making the calculation more realistic. By integrating the water content difference within a time period, the cumulative impact of water change on displacement within the observation period can be comprehensively reflected. This design can well capture dynamic changes and adapt to changing environmental conditions. The displacement correction coefficient can be obtained by fitting historical data, which ensures the applicability of the coefficient under specific regional and soil conditions, enhancing the practical value of the model.

[0107] The relationship between and reflects the influence of the water state of the soil on its physical properties. When the water content Moisture content above the plastic limit When the soil is below saturated water content, the plasticity and fluidity of the soil increase, which may lead to greater displacement; When the soil moisture content increases, far exceeding the plastic limit, the fluidity of the soil increases, resulting in an increase in the displacement correction. When the moisture content approaches the saturated moisture content, the soil may show lower shear strength due to reaching a saturated state, and the displacement correction may be relatively small.

[0108] Step 3: Obtain the cumulative rainfall and soil strength parameters at the monitoring point, analyze the impact of the cumulative rainfall on the soil strength parameters, and correct the soil strength parameters in real time based on the cumulative rainfall;

[0109] In this embodiment, the real-time correction of the soil strength parameter specifically includes:

[0110] Install an automatic rain gauge at each monitoring point to obtain the cumulative rainfall of the previous 24 hours at each monitoring point. Set the sampling frequency to Hz, when the rain gauge detects When , the sampling frequency is increased. Specifically, the sampling frequency is increased by the following formula:

[0111]

[0112] Where f is the increased sampling frequency, is the hourly rainfall intensity, is the influence coefficient, which is calibrated through loess penetration test.

[0113] when When the rainfall intensity is low, water infiltration is slow, and soil strength parameters change slowly, so the demand for data update is low, and low-frequency sampling can capture trends. This sampling frequency also avoids data redundancy and resource waste. High rainfall intensity is detected when the soil is in a high-intensity state. Rapid infiltration of rainwater in this situation can cause a sudden increase in pore water pressure and a decrease in effective stress. Loess softens when exposed to water, causing a sharp drop in cohesion and internal friction angle, potentially triggering shallow landslides or localized collapses. Therefore, high-frequency sampling is required to capture transient changes in soil parameters in real time to avoid missing critical states.

[0114] This formula reflects the timeliness requirement of dynamically matching rainfall intensity with slope instability risk. Since the stability of loess slopes is extremely sensitive to rainfall infiltration, soil parameters change slowly during low rainfall, and fixed low-frequency sampling can save resources; while during high rainfall, rapid water infiltration causes a sharp deterioration of strength parameters, and the sampling frequency needs to be increased through linear superposition terms to ensure real-time capture of critical states. The linear model quantifies the increment of sampling frequency required by unit rainfall intensity, balances the data accuracy and system efficiency within the complexity of engineering realization, and is applicable to most rainfall-triggered slope instability warning scenarios.

[0115] The soil strength parameters include effective cohesion, effective internal friction angle, and pore water pressure. The influence of cumulative rainfall on soil strength parameters is analyzed, and the formula for correcting the effective cohesion is expressed as:

[0116]

[0117] wherein, is the effective cohesion under the influence of cumulative rainfall, is the effective cohesion in the dry state, is the 24-hour cumulative rainfall, is the influence coefficient of effective cohesion.

[0118] The progressive weakening effect of cumulative rainfall on the cohesion of loess is quantitatively described by an exponential decay relationship. Since the cementation structure of loess is gradually destroyed after being exposed to water, the cohesion shows a nonlinear decrease with the infiltration of water, and the 24-hour cumulative rainfall reflects the sustained influence of water input. The exponential form can reasonably depict this physical process: when there is no cumulative rainfall, the cohesion remains at its maximum value, which is equal to the effective cohesion in the dry state. As the cumulative rainfall increases, the cohesion decays at an accelerated rate, which is controlled by . The value of can be obtained through indoor loess rainfall infiltration and shear combined tests, simulating different cumulative rainfall, controlling water infiltration to make the sample reach a stable water content distribution, then performing a shear test, and finally fitting the test data to obtain the influence coefficient. This model not only conforms to the mechanical properties of loess softening when exposed to water, but also adapts to the sensitivity differences of different soil types by the parameter , providing a simple and physically meaningful correction method for engineering warning.

[0119] The effective cohesion is the soil's ability to stick together when the effects of pore water pressure are excluded. As the amount of rainfall increases, the water content of the soil also increases. The increase in water often results in a decrease in the soil's ability to stick together, which affects the soil's effective cohesion. As the amount of rainfall increases, the soil can reach a saturated state. The pore water in the saturated soil exerts pressure on the soil particles, reducing the soil's effective stress, which results in a decrease in the effective cohesion. It can also be seen from the formula that, under the influence of the same cumulative rainfall, the greater the effective cohesion in the dry state, the greater the effective cohesion under the influence of the cumulative rainfall; conversely, the effective cohesion under the influence of the cumulative rainfall will be smaller. At the same time, as the cumulative rainfall increases, the effective cohesion under the influence of the cumulative rainfall is decreasing, and the two are inversely related.

[0120] The formula for correcting the effective internal friction angle is:

[0121]

[0122] wherein, is the internal friction angle under the influence of the cumulative rainfall, is the internal friction angle in the dry state, is the influence coefficient of the internal friction angle, and the internal friction angle °.

[0123] The internal friction angle is an important component of the shear strength of soil, reflecting the friction between soil particles. Rainfall causes the soil to become wet, which can cause changes in the contact mode and structure between soil particles, thereby affecting the internal friction angle. When the water content of the soil increases, water can form lubrication between soil particles, which can cause the friction to decrease and the internal friction angle to decrease. Especially when the rainfall is large, the flowability of the soil increases, causing the friction angle to decrease. The linear relationship used to describe the decay of the internal friction angle with rainfall is because the internal friction angle is mainly determined by the friction and biting action between soil particles, and rainfall infiltration gradually weakens this mechanical property by lubricating the particle contact surface and dissolving the cementing material, and the degradation process is closer to a linear response than an exponential decay of the cohesion. The lower limit value (internal friction angle ≥ 15°) is based on engineering experience to ensure that there is still some residual shear strength even under high rainfall conditions, and to avoid excessive correction that would result in overly conservative stability calculations. The influence coefficient The influence coefficient needs to be calibrated through direct shear or triaxial tests of saturated samples to reflect the sensitivity of the friction characteristics of a specific loess to changes in water content. As can be seen from the formula, as the cumulative rainfall increases, the corresponding internal friction angle decreases; conversely, the internal friction angle increases. Similarly, under the same cumulative rainfall, the greater the internal friction angle in the dry state, the greater the internal friction angle under the influence of the cumulative rainfall, and the two are positively related.

[0124] The formula for correcting the pore water pressure is expressed as:

[0125]

[0126] wherein, is the pore water pressure, is the pore water pressure weight, is the influence coefficient of the pore water pressure.

[0127] The pore water pressure reflects the water pressure in the soil pores. By correcting the pore water pressure, especially when the rainfall exceeds 50 mm, the influence of rainfall on the soil can be better reflected. The increase of the pore water pressure will significantly reduce the effective stress of the soil, further affecting the stability of the soil. In the analysis of slope stability, the dynamic correction of the soil strength parameters can help engineers to evaluate the safety state of the slope in real time, especially during rainfall. Through the real-time updated strength parameters, the potential instability risk can be better identified, so that necessary preventive measures can be taken. The influence coefficient of the pore water pressure reflects the growth rate of the pore water pressure with the super-threshold rainfall, which can be calibrated through saturated and unsaturated seepage tests combined with pore water pressure monitoring.

[0128] When rainfall occurs, water will penetrate into the soil, causing an increase in pore water pressure. This elevated pressure will affect the effective stress of the soil, thereby changing the strength characteristics of the soil. When the rainfall exceeds a certain threshold (such as 50 mm), the pore water pressure reaches a critical point, which may cause the soil to lose bearing capacity, thereby affecting the stability. The formula shows that the influence of rainfall on pore water pressure is different in different ranges. The increase of the pore water pressure will cause the decrease of the effective stress (principle of effective stress), thereby reducing the soil strength and further triggering the risk of landslides and other geological disasters. Through the analysis of the correction formula of the soil strength parameters, it can be found that with the increase of rainfall, the influence on the original soil strength parameters will increase, the greater the 24-hour cumulative rainfall, the smaller the corresponding effective cohesion and internal friction angle, showing an inverse correlation. The greater the rainfall, the greater the pore water pressure, showing a positive correlation.

[0129] Step 4: Coupling the real-time corrected soil strength parameters of the monitoring points with the actual displacement to determine the safety factor of the monitoring points, and then determining the rainfall sensitivity factor of the monitoring points through the cumulative rainfall of the monitoring points, determining the warning state of each monitoring point according to the rainfall sensitivity factor and the safety factor, and determining the stability of the slope according to the warning state.

[0130] In this embodiment, determining the safety factor specifically includes:

[0131] In each monitoring point, the unit soil strip is determined, the soil strip volume of the unit soil strip is recorded, the self-weight data of the unit soil strip is obtained, the real-time corrected soil strength parameter of each monitoring point is coupled with the actual displacement to calculate the safety factor of each monitoring point, and the calculation formula is:

[0132]

[0133] wherein, the safety factor at the current time, the normal stress of the unit soil strip at the monitoring point at the current time, that is, the vertical stress of the sliding surface of the unit soil strip, the self-weight of the unit soil strip at the monitoring point, the inclination of the sliding surface of the unit soil strip at the monitoring point at the current time, that is, the slope angle of the bottom sliding surface of the slope where the unit soil strip is located, the sine value of the slope angle, the corrected actual displacement of the monitoring point at the current time;

[0134]

[0135] wherein, the unit weight of loess, the volume of the unit soil strip.

[0136] The normal stress data of the slope can be obtained by soil slope analysis or in-situ testing (such as drilling sampling) method, and the unit soil strip can be randomly sampled at the monitoring point to obtain the self-weight and volume data of the unit soil strip; the unit weight data of loess can be obtained by laboratory testing, such as specific gravity bottle method, drying method, etc., or through the geological report of the region and related literature; the slope angle of the slope can be measured directly on site by measuring tools such as level, measuring angle, or can be obtained through the geological report of the region and related literature. If there is an existing topographic map, the slope angle information can be extracted from the topographic map. The sliding surface of the slope where the unit soil strip is located can be determined by establishing a slope model.

[0137] The safety factor is a key indicator in slope stability analysis, which reflects the ratio between the external load (such as self-weight, pore water pressure caused by rainfall, etc.) and the shear strength of the slope. Specifically, the larger the safety factor, the better the stability of the slope, and the smaller the risk of instability. The effective cohesion and internal friction angle, two parameters that affect the shear strength of the soil, are the main resistance when the soil is under pressure. The normal stress is the external pressure that the soil bears, and the pore water pressure will reduce the effective stress Thus affecting the shear strength of the soil. The slope angle affects the gravity component experienced by the soil, the steeper the slope, the greater the gravity component, which can lead to greater instability risk. Rainfall increases the pore water pressure, changing the effective stress of the soil, thus affecting the safety factor. Parameters positively correlated with the safety factor include: effective cohesion, normal stress, and internal friction angle, an increase in effective cohesion directly increases the safety factor, enhancing the shear strength of the soil and reducing the risk of instability; an increase in normal stress (under constant pore water pressure) will result in an increase in effective stress, thus increasing the safety factor; an increase in internal friction angle increases the safety factor, indicating an increase in soil friction and an increase in anti-sliding ability. Parameters negatively correlated with the safety factor include: pore water pressure, unit weight, slope angle, and rainfall, an increase in pore water pressure will reduce the effective stress, thus reducing the safety factor and increasing the risk of instability; an increase in unit weight and slope angle will cause the safety factor to decrease, as the gravity component increases, which can lead to greater sliding force; an increase in rainfall will usually result in an increase in pore water pressure, thus reducing the safety factor.

[0138] By coupling the real-time corrected soil strength parameters (such as effective cohesion, internal friction angle, and pore water pressure) with the actual displacement, the safety factor can be dynamically calculated. This real-time evaluation enables the monitoring system to timely reflect the safety state of the slope and quickly respond to potential risks. The calculation of the safety factor takes into account various factors such as normal stress, unit weight of the soil strip, slope angle, etc., making the evaluation results more comprehensive and accurate, and better reflecting the actual condition of the slope.

[0139] In this embodiment: the volume of the unit soil strip is set to 2.5 , is set to 30°, is set to 18, is set to 0.05, is set to 0.01, is set to 0.02, is set to 0.15, is set to 0.1, is set to 1.2, 25 sets of soil strength parameters, normal stress, actual displacement, and 24-hour cumulative rainfall data are collected, and the specific measurement data is shown in the following table:

[0140] Table 1: Measurement data statistics

[0141]

[0142] Referring to Figures 3 to 7, and the data in the above table show that as the effective cohesion under the influence of cumulative rainfall increases, the safety factor will also increase. The greater the difference between the normal stress and the pore water pressure, the greater the safety factor. The larger the internal friction angle, the greater the safety factor. As the actual displacement increases, the safety factor will decrease. As the 24-hour rainfall increases, the safety factor will also decrease. The internal friction angle is set to be no less than 15°. Even if the soil is fully saturated under extreme rainfall conditions, the residual internal friction angle of loess will usually not be lower than 15°. This is also to prevent the unreasonable situation of the friction angle returning to zero due to excessive correction, and to ensure that the calculation retains the most basic shear resistance of the soil. What needs to be explained in the table is that the normal stress improves the shear resistance through the effective stress, and the effective stress improves the shear resistance through It is reflected that the normal stress theory is positively correlated with the safety factor. However, the root cause analysis found that, for example, in the first set of data, the normal stress is 50 and the pore water pressure is 0. , the 25th set of data, the normal stress is 170, the pore water pressure is 15.74, , although the normal stress increased by 240%, From 0 to 15.74kPa (only 9.3% of the effective stress ( ) still increased significantly (from 50 to 154.26 kPa), but the safety factor decreased, indicating that It is not the main reason and needs to be analyzed in combination with other parameters. Analyzing the effect of rainfall on strength parameters, the increase of 24-hour cumulative rainfall shows that the effective cohesion under the influence of cumulative rainfall is exponentially attenuating, the internal friction angle under the influence of cumulative rainfall is also linearly attenuating, the pore water pressure is nonlinearly increasing, and the displacement is also cumulatively increasing. Comparing the first and fifth groups, decreased by 86% (from 24.56 to 3.43 kPa), A decrease of 59% (from 0.661 to 0.268), even though ( ) increases, the numerator Still significantly reduced, and The attenuation is much greater than Therefore, although the increase in normal stress alone will increase the safety factor, in actual slopes, the degradation of strength parameters and the increase in pore water pressure caused by rainfall infiltration dominate the downward trend of the safety factor, resulting in the positive effect of normal stress being masked. Similarly, even if the pore water pressure is 0 in the first nine sets of data, the safety factor decreases due to changes in other strength parameters. Figure 4 The safety factor fitting curve also shows a downward trend.

[0143] In this embodiment, determining the rainfall sensitivity coefficient specifically includes:

[0144] The accumulated rainfall data is analyzed to obtain the rainfall sensitivity coefficient, and the calculation formula is:

[0145]

[0146] wherein, represents the rainfall sensitivity coefficient at the current time, is the displacement correction amount of the monitoring point at the time, is a time variable, and ∈[ , t] t represents the current time, is 24 hours, is the influence coefficient of the accumulated rainfall in the rainfall sensitivity coefficient, is the initial shear modulus of the soil, is the shear modulus of the soil at the current time, is the accumulated rainfall in the previous 24 hours at the current time. The soil shear modulus data can be obtained by consulting relevant literature. Many studies provide shear modulus data for different soil types under specific conditions, especially in literature related to geological engineering and civil engineering. It can also be obtained through laboratory tests, penetration tests, and in-situ shear wave velocity tests.

[0147] By correcting the displacement of each monitoring point, the soil deformation caused by rainfall can be effectively reflected. The relationship between the cumulative displacement and the rainfall is direct, that is, the more the rainfall, the higher the pore water pressure, and the lower the effective stress of the soil, resulting in an increase in displacement. Therefore, the total sum of the cumulative displacement provides a basis for the rainfall sensitivity coefficient. By introducing the influence coefficient, the influence of rainfall on the sensitivity coefficient can be adjusted, which is reasonable. Under different soil conditions and different environmental conditions, the influence of rainfall on the soil may be different, so an adjustment coefficient is used to reflect this difference, improving the applicability of the formula. It can be obtained by fitting the historical data using the least squares method. This ratio takes into account the changes in the mechanical properties of the soil during rainfall. Rainfall can cause the shear modulus of the soil to decrease, reflecting changes in the strength and deformation capacity of the soil. By calculating the ratio of the initial modulus to the current modulus, the impact of rainfall on the mechanical properties of the soil can be quantified, thereby affecting the value of the rainfall sensitivity coefficient. This consideration is scientific because the strength and stiffness of the soil change with changes in humidity and pore water pressure. By accumulating rainfall data over multiple time periods, the cumulative effect of rainfall on soil displacement can be reflected, consistent with the time-dependent nature of rainfall effects. The effects of rainfall are usually lagging, and the consideration of cumulative displacement can more comprehensively reflect the effects of rainfall on the soil at different times. Overall, the rainfall sensitivity coefficient formula, through the introduction of multiple relevant factors, achieves a quantitative analysis of the impact of rainfall on the soil. The formula reasonably combines displacement, rainfall, influence coefficient, and shear modulus changes to form a comprehensive evaluation system, which helps to dynamically monitor and assess the stability of the soil under rainfall conditions.

[0148] Rainfall Sensitivity Coefficient Mainly used to quantify the degree of influence of rainfall on soil stability and displacement changes. It comprehensively considers the displacement changes , rainfall and changes in the shear modulus of the soil . The numerical value can reflect the sensitivity of the soil under rainfall conditions, and the higher the value, the stronger the response of the soil to rainfall, which may lead to greater displacement or instability risk. Conversely, a lower value indicates that the soil is more stable under rainfall conditions. By calculating the rainfall sensitivity coefficient, a quantitative basis can be provided for the safety assessment of the soil, helping engineers and researchers assess the risk under specific rainfall conditions and develop appropriate prevention and reinforcement measures to improve the safety and stability of the soil. The increase in rainfall will cause an increase in pore water pressure, resulting in a decrease in effective stress of the soil and an increase in displacement. As an influence coefficient, it can adjust the impact of rainfall on displacement, reflecting the different effects of rainfall on soil types, soil layer structures, and geological conditions. The shear modulus of the soil changes due to rainfall. Wetting and the rise in pore water pressure caused by rainfall usually leads to a decrease in shear modulus, thereby affecting the shear strength of the soil. As the amount of rainfall increases, , the displacement change will also generally increase, indicating that the more rainfall, the greater the displacement change. In the formula, is the denominator, and the rainfall sensitivity coefficient will decrease as the amount of rainfall increases (if other factors remain unchanged), reflecting that the impact of rainfall on the soil is adjusted. This formula essentially measures the deformation efficiency per unit of rainfall, i.e. , if a large amount of rainfall only causes a small deformation ( ), indicating that the slope has strong anti-seepage ability and even a small amount of rainfall will cause large deformation ( ), indicating that the slope is extremely sensitive. The characteristic that the rainfall increases and the rainfall sensitivity coefficient decreases just reveals the cumulative damage effect of the slope: under heavy rainfall The decrease indicates that the slope has consumed part of the rainfall energy through early deformation, which is consistent with the strain hardening characteristics of loess. As the soil becomes moist due to rainfall, the shear modulus usually decreases, causing the shear strength of the soil to decrease, thus affecting the displacement change. Therefore, The reduction will lead to increase.

[0149] In this embodiment, determining the stability of the slope according to the warning state specifically includes:

[0150] when When the safety factor of the monitoring point at the current moment is analyzed, if the safety factor of the monitoring point is higher than the preset safety threshold, the monitoring point is judged to be in a safe state at the current moment; if the safety factor of the monitoring point is lower than the preset safety threshold, the monitoring point is judged to be in an abnormal state at the current moment and a red warning needs to be issued;

[0151] like and , a yellow warning is issued and the monitoring frequency should be increased. and lasts for more than 2 hours, and , an orange warning is issued, the outer lanes should be closed, and sandbags should be piled at the foot of the slope. and , a red alert is issued, and all traffic should be closed and evacuated urgently. are rainfall sensitivity coefficient thresholds, and , are the safety factor thresholds, and , is the safety factor at the current moment;

[0152] If two or more monitoring points issue a red warning, the slope is judged to be in an unstable state at the current moment; if half or more of the monitoring points issue an orange warning, the slope is judged to be in an unstable state at the current moment; if two-thirds of the monitoring points issue a yellow warning, the slope is judged to be in an unstable state at the current moment; in all other cases, the slope is judged to be in a stable state at the current moment.

[0153] The above thresholds can be set by referring to the rainfall sensitivity coefficient and safety factor when landslide events occurred and when no landslide events occurred in historical data. For example, 、 、 , 、 、 .

[0154] According to the relationship between rainfall sensitivity coefficient and safety factor, different warning levels (red, orange, yellow) are set, which allows multi-level risk assessment based on actual rainfall conditions and soil safety conditions. This step-by-step warning mechanism can timely reflect the state changes of the soil and provide a basis for decision-making. The rationality of this warning logic is mainly reflected in its multi-level and dynamic evaluation mechanism. By combining the rainfall sensitivity coefficient (Sr) and the safety factor (FS), and setting a progressive threshold ( and The system can distinguish different risk levels: from low-risk yellow alerts to high-risk red alerts, covering the full range from potential threats to emergencies. For example, when Sr > 0.8 and FS < 1, a red alert is triggered directly, which conforms to the engineering principle that the shear strength of loess slopes drops sharply under heavy rainfall. Requiring an orange alert to last for at least two hours avoids misjudgments caused by brief rainfall interference. This design considers both the cumulative effect of rainfall (via Sr) and the real-time mechanical state of the soil (via FS), making the warning results more scientific and reliable.

[0155] Its advantages are: first, dynamic adaptability, through the historical landslide data calibration threshold (such as =0.8 (corresponding to extreme rainfall conditions), ensuring that the threshold matches local geological conditions. Second, redundant judgments are implemented. Instead of relying solely on single-point data, the system's fault tolerance is enhanced by linking multiple monitoring points (for example, two red alerts or half of the orange alerts trigger an instability determination). Third, a tiered response is implemented: a yellow alert prompts attention, an orange alert requires enhanced monitoring, and a red alert calls for urgent action, forming a stepped emergency response chain. This logic is particularly suitable for areas significantly affected by rainfall, such as loess slopes, providing early warning of soil strength parameter deterioration, buying valuable time for emergency response.

[0156] See also Figure 2 The present invention further provides a GNSS-based mountain highway loess slope stability assessment system, wherein the GNSS-based mountain highway loess slope stability assessment system is used to implement the above-mentioned GNSS-based mountain highway loess slope stability assessment method, comprising:

[0157] The regional division module is used to divide the slope area into several grids and obtain the geological characteristic data of each grid. Based on the geological characteristic data, all grids are divided into multiple super nodes. A monitoring point is set in each super node to obtain the soil moisture content at the monitoring point to analyze the soil moisture content at different depths.

[0158] The displacement correction module is configured to acquire the original displacement of the monitoring point through GNSS, analyze the water content of the soil at the monitoring point based on the Fredlund unsaturated soil shear strength theory, acquire a displacement correction amount, and correct the original displacement according to the displacement correction amount to acquire the actual displacement of the monitoring point.

[0159] The rainfall and soil parameter coupling module is configured to acquire the cumulative rainfall and the soil strength parameter at the position of the monitoring point, analyze the influence of the cumulative rainfall on the soil strength parameter, and correct the soil strength parameter in real time according to the cumulative rainfall.

[0160] The safety evaluation and early warning decision module is configured to couple the real-time corrected soil strength parameter and the actual displacement of the monitoring point to determine the safety factor of the monitoring point, determine the rainfall sensitivity factor of the monitoring point through the cumulative rainfall of the monitoring point, determine the early warning state of each monitoring point according to the rainfall sensitivity factor and the safety factor, and determine the stability of the slope according to the early warning state.

[0161] The above formulas are all dimensionless numerical calculations, the formulas are obtained by software simulation of a large amount of data to obtain a formula of the most recent real situation, and the preset parameters in the formulas are set by a person skilled in the art according to the actual situation.

[0162] The above embodiments can be realized wholly or partially by software, hardware, firmware or any combination thereof. When realized by software, the above embodiments can be realized in the form of a computer program product wholly or partially. Those skilled in the art can realize that the units and algorithm steps of the examples described in connection with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized by hardware or software methods depends on the specific application and design constraints of the technical solutions.

[0163] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, which can be located in one place or distributed on multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiments.

[0164] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the present application, which should be covered within the protection scope of the present application.

Claims

1. A method for evaluating the stability of loess slopes of mountainous highways based on GNSS, characterized in that, The specific steps include: Step 1: divide the slope area into a plurality of grids, and obtain the geological feature data of each grid, divide all the grids into a plurality of super nodes according to the geological feature data, set a monitoring point in each super node, obtain the water content of the soil at the monitoring point, and analyze the water content of the soil at different depths; Step 2: obtain the original displacement of the monitoring point through GNSS, analyze the water content of the soil at the monitoring point based on the Fredlund unsaturated soil shear strength theory, obtain the displacement correction amount, and correct the original displacement according to the displacement correction amount to obtain the actual displacement of the monitoring point; Step 3: obtain the cumulative rainfall and soil strength parameters at the position of the monitoring point, analyze the influence of the cumulative rainfall on the soil strength parameters, and correct the soil strength parameters in real time according to the cumulative rainfall; Step 4: couple the real-time corrected soil strength parameters of the monitoring point with the actual displacement to determine the safety factor of the monitoring point, determine the rainfall sensitivity factor of the monitoring point through the cumulative rainfall of the monitoring point, determine the early warning state of each monitoring point according to the rainfall sensitivity factor and the safety factor, and determine the stability of the slope according to the early warning state; Obtain the unit soil strip from each monitoring point one by one, record the soil strip volume of the unit soil strip to obtain the unit soil strip self-weight data, couple the real-time corrected soil strength parameters of each monitoring point with the actual displacement to calculate the safety factor of each monitoring point, and the calculation formula is: wherein, is the safety factor at the current time, is the normal stress of the unit soil strip at the monitoring point at the current time, i.e. the vertical stress of the sliding surface of the unit soil strip, is the self-weight of the unit soil strip at the monitoring point, is the inclination angle of the sliding surface of the unit soil strip at the monitoring point at the current time, i.e. the slope angle of the bottom sliding surface of the slope where the unit soil strip is located, is the sine value of the slope angle, is the actual displacement of the monitoring point at the current time after correction. wherein, is the heavy loess, is the volume of the unit soil bar; Analyze the cumulative rainfall data to obtain the rainfall sensitivity factor, and the calculation formula is: wherein, represents the rainfall sensitivity coefficient at the current time, is the displacement correction amount of the monitoring point at the current time, is a time variable, and ∈[ , t], t represents the current time, is 24 hours, is the influence coefficient of the cumulative rainfall in the rainfall sensitivity coefficient, is the soil shear modulus at the initial time, is the soil shear modulus at the current time, is the cumulative rainfall in the previous 24 hours at the current time. 2.The method according to claim 1, wherein, Divide all the grids into a plurality of super nodes according to the geological feature data, which specifically includes: The geological feature data includes soil type, slope, porosity ratio, permeability coefficient and historical water content, a slope region is divided into a plurality of grids, each grid is regarded as a node, and geological feature data corresponding to each node forms a node geological feature vector, which is represented as: wherein, is the geological feature vector, st is the soil type, S is the slope, Vr is the void ratio, Pc is the permeability coefficient, and HWc is the historical water content, i.e., the average water content of the soil surface corresponding to the node in the previous 24 hours. Perform maximum-minimum normalization on the slope, porosity ratio, permeability coefficient and historical water content, and convert the soil type to One-Hot encoding; Calculate the difference value of the geological features between nodes, and the formula is: wherein, is a difference of geological features between the a-th node and the b-th node, is a weight of the k-th geological feature data, is a normalized value of the k-th geological feature data of the a-th node, is a normalized value of the k-th geological feature data of the b-th node, k is an index of other geological feature data types except soil type, and k ∈ [1, 4], is a soil type difference penalty coefficient, is an indicator function, 1 if the soil types of the a-th node and the b-th node are different, otherwise 0, a, b are indices of nodes, and ; Each node is regarded as an independent cluster, and an initial cluster set is generated, denoted as wherein C represents the initial cluster set, and N represents the total number of nodes; all nodes are traversed, and the geological feature difference between nodes is calculated; if the geological feature difference is less than a preset difference threshold value, and the two nodes have not been merged, the clusters to which the nodes belong are merged, and iteration is repeated until no new merging occurs, and each merged cluster is regarded as a super node. 3.The mountainous highway loess slope stability evaluation method based on GNSS according to claim 2, characterized in that, Analyze the water content of the soil at different depths, which specifically includes: Take the geometric center of each super node as a monitoring point, set sampling points at multiple depths below the monitoring point, and obtain the water content of the soil at the sampling points, analyze the relationship between the water content and the depth at the sampling points, and construct the following expression to represent the water content of the monitoring point at different depths, which is as follows: wherein, represents the water content at the monitoring point at the current time, at a vertical downward z meter position, represents the water content at the monitoring point at the current time, at a vertical downward z meter position, is the exponential decay rate.

4. The method according to claim 3, wherein, Obtain the actual displacement of each monitoring point, which specifically includes: Obtain the original displacement of each monitoring point through GNSS, calculate the displacement correction amount based on the Fredlund unsaturated soil shear strength theory, and the calculation formula is as follows: wherein, is a displacement correction amount at a current time, represents a mean value of the water content of the monitoring point at a current time, is a plastic limit water content of the loess, is a saturated water content, is a displacement correction amount coefficient, is a time variable, and , represents 24 hours; wherein L is the moisture content influence depth, is the depth correction term coefficient, represents the moisture content at the vertical downward z meters position of the monitoring point at the current time. Correct the original displacement monitored by GNSS according to the displacement correction amount, and the calculation formula is: wherein, is the actual displacement of the monitoring point at the current time after correction, is the original displacement of the monitoring point acquired by GNSS. 5.The mountainous highway loess slope stability evaluation method based on GNSS according to claim 1, characterized in that, Correct the soil strength parameters in real time, which specifically includes: An automatic rain gauge is installed at each monitoring point to obtain 24-hour accumulated rainfall at each monitoring point. In the process of collecting the accumulated rainfall of the monitoring point in the past 24 hours based on the automatic rain gauge, when the automatic rain gauge detects , the sampling frequency is set to Hz, and when the automatic rain gauge detects , the sampling frequency is increased, specifically by the following formula: where f is the improved sampling frequency, is the hourly rainfall intensity, is the influence coefficient; The soil strength parameters include effective cohesion, effective internal friction angle and pore water pressure, the influence of the cumulative rainfall on the soil strength parameters is analyzed, and the soil strength parameters are corrected, and the formula is expressed as: wherein, is the effective cohesion under the influence of accumulated rainfall, is the effective cohesion in the dry state, is the 24-hour accumulated rainfall, is the influence coefficient of the effective cohesion; wherein, is the internal friction angle under the influence of accumulated rainfall, is the internal friction angle in dry state, is the influence coefficient of the internal friction angle, the internal friction angle °; wherein, Pwpis the pore water pressure, Pwpw is the pore water pressure weight, Pwpf is the influence coefficient of the pore water pressure. 6.The method according to claim 1, wherein, Determine the stability of the slope according to the early warning state, which specifically includes: When the safety factor of the monitoring point is higher than the preset safety threshold, it is determined that the monitoring point is in a safe state at the current time, and if the safety factor of the monitoring point is lower than the preset safety threshold, it is determined that the monitoring point is in an abnormal state at the current time, and a red early warning needs to be issued. If and , a yellow warning is issued, if and lasts for more than 2 hours, and , an orange warning is issued, if and , a red warning is issued, are respectively rainfall sensitivity coefficient thresholds, and , are respectively safety coefficient thresholds, and , is the safety coefficient at the current time. If two or more monitoring points issue red early warning, it is judged that the slope is in an unstable state at the current time, if half or more than half of the monitoring points issue orange early warning, it is judged that the slope is in an unstable state at the current time, if two-thirds of the monitoring points issue yellow early warning, it is judged that the slope is in an unstable state at the current time, and the rest are judged that the slope is in a stable state at the current time.

7. A GNSS-based evaluation system for loess slope stability of mountainous highways, characterized in that, The mountainous highway loess slope stability evaluation system based on GNSS is used to realize the mountainous highway loess slope stability evaluation method based on GNSS, and comprises: The regional division module is used for dividing the slope region into a plurality of grids, and obtaining the geological feature data of each grid, dividing all the grids into a plurality of super nodes according to the geological feature data, setting a monitoring point in each super node, obtaining the water content of the soil at the monitoring point, and analyzing the water content of the soil at different depths; The displacement correction module is used for obtaining the original displacement of the monitoring point through GNSS, analyzing the water content of the soil at the monitoring point based on the Fredlund unsaturated soil shear strength theory, obtaining the displacement correction amount, and correcting the original displacement according to the displacement correction amount to obtain the actual displacement of the monitoring point; The rainfall and soil parameter coupling module is used for obtaining the cumulative rainfall and soil strength parameter of the position of the monitoring point, analyzing the influence of the cumulative rainfall on the soil strength parameter, and correcting the soil strength parameter in real time according to the cumulative rainfall; The safety evaluation and early warning decision module is used for coupling the real-time corrected soil strength parameter of the monitoring point with the actual displacement to determine the safety factor of the monitoring point, determining the rainfall sensitivity factor of the monitoring point through the cumulative rainfall of the monitoring point, determining the early warning state of each monitoring point according to the rainfall sensitivity factor and the safety factor, and determining the stability of the slope according to the early warning state.

Images