A method for testing roughness coefficient of slope structure surface

Abstract

The application discloses a kind of slope structure surface roughness coefficient test methods, including structure surface pretreatment, rebound detection, profile curve drawing and roughness measurement etc. core steps.By dividing measurement area and marking, average rebound value is obtained using a rebound apparatus, surface profile is drawn in combination with a profilograph, and roughness coefficient JRC is read using a multi-range roughness scale and electronic subdivision technique n ; Rock wall strength, normal stress and other parameters can also be obtained using a point load apparatus, a vibrating wire strain gauge and other devices, and the shear strength is calculated in combination with the Barton-Bandis model. The results are dynamically corrected by integrating the data from multiple measurement areas and correlating environmental parameters. This method improves the testing accuracy and efficiency, enables data integration and full-cycle management, and provides a reliable basis for slope stability assessment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of geotechnical engineering slope stability evaluation, and particularly relates to a method for testing roughness coefficient of slope structural plane. BACKGROUND

[0002] The roughness coefficient of the slope structural plane is a key parameter for evaluating the stability of the rock slope, and directly affects the shear strength of the structural plane, the permeability characteristics and the determination of the slope instability risk. The current testing methods for the roughness coefficient of the structural plane have many technical limitations: the traditional manual measurement method relies on manual experience for judgment, and the error is significant in complex terrain areas; the laboratory direct shear test has high precision, but the sampling process will damage the original morphology of the structural plane, and the test period is too long to meet the real-time needs of the engineering. The existing testing methods also have the problem of insufficient data representativeness, and the single-point detection or single-direction scanning cannot fully reflect the heterogeneous characteristics of the spatial distribution of the structural plane, and the discrete test data lack effective spatial integration means. In terms of parameter correlation, the existing technologies obtain the key parameters JRC, JCS, ψ b and other parameters in a fragmented manner, and do not fully consider the real-time influence of dynamic factors such as environmental temperature and humidity, rainfall, etc. on the test results. In addition, the current testing system lacks a full life cycle management mechanism, and it is difficult to provide continuous and effective benchmark data support for slope maintenance. These technical defects seriously restrict the accuracy and engineering applicability of the slope stability evaluation. SUMMARY

[0003] The purpose of the present application is to provide a method and system for testing the roughness coefficient of the slope structural plane, which has the advantages of improving the test precision, maintaining the original morphology of the structural plane, and realizing the dynamic correlation of multi-dimensional parameters.

[0004] In order to solve the above technical problems, the present application solves the problems by the following technical scheme: a method for testing the roughness coefficient of the slope structural plane, comprising the following steps:

[0005] The structural plane preprocessing step: the surface of the slope structural plane is treated to remove impurities, loose debris and coverings, and several measurement areas are divided according to the spatial distribution characteristics of the structural plane, and a unique spatial identifier is set for each measurement area;

[0006] The structural plane rebound detection step: several measurement points are distributed in each measurement area, and the rebound detector is used to detect the rebound of the several measurement points, and the average rebound value R m of the measurement area is calculated according to the rebound value obtained by the rebound test;

[0007] The profile curve drawing step is configured with a profile curve instrument, and the surface profile curve graph between different detection points in the measurement area is drawn by the profile curve instrument;

[0008] The roughness measurement procedure involves configuring a multi-range roughness gauge with several scale lines, the first scale line being the starting line. The reading method for the multi-range roughness gauge is as follows: At least two peaks of the surface profile curve are tangent to the starting line, and the reading is based on the length L of the surface profile curve of the structural surface. n And read the JRC corresponding to the scale line where the deepest valley is located in the surface profile curve. n The value, that is, the measured length is L n Roughness coefficient JRC of the profile curve n value.

[0009] By adopting the above technical solution, pre-treating the slope structure surface to remove impurities, delineate and mark the test areas, the interference of surface coverings and loose rock debris on the test can be eliminated, ensuring that the test areas are representative and their spatial locations are traceable; the average rebound value R of the test area is obtained through rebound testing. m This provides basic data for subsequent parameter calculations; the surface profile is drawn using a profile curve analyzer and read from the JRC using a multi-range roughness ruler. n This method enables direct and targeted measurement of the surface roughness coefficient, solving the problems of low accuracy and strong subjectivity in traditional manual measurement, and providing reliable basic parameters for slope stability assessment.

[0010] The present invention is further configured to include the following steps:

[0011] The steps for calculating rock wall strength include: setting up a point load cell; selecting a predetermined number of rock blocks with the same geological conditions as the structural surface in the test area; conducting tests using the point load cell; and recording the point load strength index as I. s (50), Rock wall strength JCS n The calculation formula is JCS n =22.82·I s (50);

[0012] Steps for calculating the basic friction angle of a structural surface, including the basic friction angle ψ of a slope structural surface. b The calculation formula is:

[0013] ψ b =0.414R m +9.273;

[0014] The steps for obtaining the normal stress on the structural surface involve arranging several sets of vibrating wire strain gauges on both sides of the structural surface, continuously monitoring the data at a preset frequency for a preset time, and obtaining the average value of the monitored values ​​as σ. n ;

[0015] The steps for calculating the shear strength of the structural plane are as follows: The shear strength τ is calculated according to the Barton-Bandis model, using the following formula:

[0016] τ=σ n tan{JRC n •lg(JCS n / σ n )+ψ b},

[0017] Among them, JCS n To evaluate the rock wall strength of the structural plane; JRC n To evaluate the surface roughness coefficient of the rock mass along the evaluation direction on the structural surface; ψ b To evaluate the basic friction angle of the structural surface; R m σ represents the average surface springback of the structural surface; n This represents the normal stress of the structural surface used for trial calculations.

[0018] By adopting the above technical solution, adding a rock wall strength calculation step, and obtaining JCS using a point load instrument, n Basic friction angle calculation steps, based on R m The quantitative formula was established; the normal stress acquisition step was performed through continuous monitoring using a vibrating wire strain gauge; and the shear strength calculation step was performed by applying the Barton-Bandis model, thus realizing the calculation from the roughness coefficient JRC. n The entire parameter chain is correlated to the shear strength τ. Each parameter is quantitatively coupled through measured data, avoiding the shortcomings of traditional shear strength calculations that rely on empirically determined parameters. This significantly improves the accuracy and engineering applicability of shear strength assessment, providing a more comprehensive theoretical basis for determining slope instability risk.

[0019] The present invention is further configured such that the structural surface springback detection step includes a sub-step:

[0020] Measurement point distribution optimization sub-step: Based on the surface texture features of the structural surface, measurement points are arranged in the measurement area using equidistant grids or random sampling to ensure that the measurement points cover different texture areas of the measurement area;

[0021] Outlier stratification removal sub-step: First, remove rebound values ​​that exceed the preset standard deviation range. Then, remove the maximum and minimum values ​​of a preset proportion from the remaining values. The remaining values ​​are used to calculate the average rebound value R. m .

[0022] By adopting the above technical solution, the measurement point distribution optimization sub-step, based on texture features, uses equidistant grids or random sampling to ensure that the measurement points cover different texture areas of the test area, avoiding the problem of insufficient representativeness in traditional single-point testing; through the outlier hierarchical removal sub-step, data exceeding the standard deviation are first removed, and then extreme values ​​are removed, effectively filtering out abnormal rebound values ​​caused by local defects or operational errors, so that the calculated average rebound value R is... m This more closely resembles the actual surface properties of the structural surface, providing a basis for the subsequent basic friction angle ψ.b The calculation provides more reliable basic data and reduces parameter transfer errors.

[0023] The present invention is further configured such that the contour curve drawing step further includes a sub-step:

[0024] Multi-directional scanning sub-step: Draw contour curves along the main direction and the perpendicular direction of the structural plane, and obtain at least three parallel curves in each direction;

[0025] Curve smoothing sub-step: Digitally denoise the drawn contour curve to remove abnormal fluctuations caused by probe jumps and retain the real surface texture features;

[0026] The electronic precision measurement sub-step digitizes and stores the surface profile curve obtained from the curve smoothing sub-step, and then uses an electronic roughness ruler adapted to a multi-range roughness ruler for measurement.

[0027] When the deepest valley of the surface profile curve lies between two scale lines, the two scale lines are equally divided using an electronic roughness ruler. Based on the subdivided scale lines after the division, the precise JRC is calculated and read. n value.

[0028] By adopting the above technical solutions, the contour curve drawing steps are optimized: multi-directional scanning, obtaining parallel curves along the main direction and perpendicular direction, can reflect the anisotropy of surface roughness and avoid the one-sidedness of single-direction scanning; curve smoothing and digital noise reduction can eliminate abnormal fluctuations caused by probe jumps, retain true texture features, and solve the problem of noise interference in the original curve; electronic precision measurement, with electronic roughness ruler subdivided scale, when the valley is between two scale lines, JRC is achieved through subdivision calculation. n The accurate reading of the value reduces the error of traditional manual reading to more than 1 / 10, significantly improving the accuracy of roughness coefficient measurement.

[0029] The present invention is further configured such that: the contour curve drawing step is equipped with a contour validity verification module, the contour validity verification module being used to compare JRC curves in different directions within a unified survey area. n Differences, when JRCs in different directions within the same survey area n When the difference between values ​​exceeds a preset threshold, the contour curve of the measurement area is re-acquired.

[0030] By adopting the above technical solution and comparing JRC data from different directions within the same survey area using the contour validity verification module, nWhen the difference exceeds the preset threshold, the profile curve is re-acquired. This can promptly detect abnormal data caused by scanning direction deviation, instrument failure, or local abrupt changes in the structural surface, ensuring the consistency and reliability of the roughness coefficient test results in the same test area, and avoiding misleading subsequent shear strength calculations and slope assessments due to data deviations.

[0031] The invention is further configured to include a data integration step, comprising a laser scanner for integrating test data from several test areas into complete structural surface data, including:

[0032] The steps for setting benchmark points are as follows: Select at least three non-collinear benchmark points on the slope structure surface, record the spatial coordinates of each benchmark point and mark them, and the data of each survey area should contain the association information of at least one benchmark point.

[0033] Data preprocessing sub-steps: Digitally process the surface profile curves obtained from the profile curve drawing step for each test area, extracting the curve coordinate information; simultaneously organize the average rebound values ​​R obtained from the structural surface rebound detection step for each test area. m JRC obtained from the roughness measurement procedure n This is associated with the spatial location of the corresponding survey area;

[0034] Coordinate matching and stitching sub-step: Based on the reference point coordinates set in the reference point setting sub-step, the numerical contour curve coordinates and R values ​​of each survey area are matched. m Values ​​and JRC n The values ​​are transformed into a unified coordinate system, and the iterative nearest point algorithm is used to match the contour curve edges of adjacent survey areas, so that the contour curves of adjacent survey areas are seamlessly connected, forming a complete comprehensive map of the structural surface contour.

[0035] Data integration and display sub-step: Combine the stitched complete structural surface contour map with the R of each survey area. m Value, JRC n Spatial correlation annotation is performed on the values, and the roughness coefficient and springback value distribution characteristics of different regions of the structural surface are displayed through color gradient or numerical annotation.

[0036] By adopting the above technical solution, multi-area data integration is achieved through laser scanners: Reference point setting ensures that data from each area are correlated under a unified coordinate system, solving the problem of inconsistent spatial locations across dispersed areas; coordinate matching and stitching utilize an iterative nearest-point algorithm to achieve seamless connection of contour curves between adjacent areas, forming a complete overall structural surface contour; data integration and display, through spatial correlation annotation and color gradient display, intuitively presents the roughness coefficients (JRC) of different regions. n Rebound value R mThe distribution characteristics address the shortcomings of traditional scattered data in global analysis, providing visualized global data support for the overall stability assessment of slope structures.

[0037] The present invention is further configured to include a slope detection and maintenance step, wherein the slope detection and maintenance step includes sub-steps:

[0038] The benchmark data storage sub-step is used to provide a comparison benchmark for continuous slope monitoring and maintenance. The integrated display sub-step shows the complete surface contour map of the structural surface after splicing, and the average rebound value R of each test area. m Roughness coefficient JRC n Basic friction angle ψ b The shear strength τ, along with the spatial coordinates of the test area and the test timestamp, is stored in the benchmark database;

[0039] Slope zoning coding sub-steps: Based on the spatial distribution characteristics of the structural surfaces, the slope is zoned and coded, with each zone associated with at least three reference point coordinates and the initial test JRC. n The average value and the τ-average value are used as regional reference units for subsequent maintenance.

[0040] By adopting the above technical solution, the integrated structural surface data, including contour and R, is stored in the reference data storage sub-step. m JRC n ψ b The system uses τ to link spatial coordinates and timestamps, providing a traceable initial benchmark for long-term slope monitoring. Slope zoning coding, which links benchmark points with the average value of initial parameters, divides the slope into independently assessable reference units, facilitating accurate location of monitoring areas in subsequent maintenance. This solves the problems of low efficiency in full-area monitoring and difficulty in locating local risks in traditional slope maintenance, providing a clear regional reference for targeted maintenance.

[0041] The present invention is further configured such that: the slope detection and maintenance step is equipped with a subsequent detection cycle control module, used to determine the time interval for subsequent detections, including:

[0042] Based on the initial shear strength, the slope area is divided into several levels. Areas with higher shear strength have longer testing intervals, while areas with lower shear strength have shorter testing intervals.

[0043] When natural environmental changes occur or slopes show significant deformation, the detection interval for the corresponding area should be shortened, and supplementary detection of the slope should be carried out.

[0044] By adopting the above technical solution, the subsequent inspection cycle control module classifies the slope area according to the initial shear strength τ, shortens the inspection interval for high-risk areas (i.e., low-τ areas) and extends the interval for low-risk areas (i.e., high-τ areas), thus achieving optimized allocation of inspection resources. At the same time, when the natural environment changes or the slope shows significant deformation, shortening the inspection interval of the corresponding area can respond promptly to the impact of sudden working conditions on the structural surface, solving the problem of rigid traditional inspection cycles and difficulty in dealing with dynamic risks, and improving the timeliness and economy of slope maintenance.

[0045] The present invention is further configured such that the slope detection and maintenance steps include:

[0046] Periodic retesting sub-step: Perform rebound testing and roughness retesting on key areas according to a preset cycle, and update R. m JRC n and the value of τ;

[0047] Model iteration sub-step: Based on the newly measured data, revise the shear strength calculation model and optimize the parameter weights under different geological conditions.

[0048] By adopting the above technical solution, the periodic retesting sub-step performs rebound testing and roughness retesting on key areas according to a preset cycle, and updates R... m JRC n The τ value can be used to track the evolution of structural parameters over time. The model iteration sub-steps are based on newly measured data to correct the shear strength calculation model and optimize the parameter weights under different geological conditions. This solves the problem that traditional static models are difficult to adapt to the long-term evolution of structural surfaces, enabling shear strength assessment to dynamically match the actual state of the slope and improving the accuracy and predictive ability of long-term monitoring.

[0049] The present invention is further configured to include an environmental and structural dynamic correction step, which includes:

[0050] Real-time monitoring of environmental parameters: Deploy humidity sensors, rainfall and rock strain gauges in the slope survey area to continuously collect data on slope environmental temperature, humidity, cumulative rainfall and micro-strain of structural surfaces, and establish a time series database of environmental parameters.

[0051] The environmental correlation model construction sub-step analyzes environmental parameters and JRC using machine learning algorithms. n R m ψ b The dynamic correlation pattern, quantifying the JRC when the rainfall increases by a preset threshold. n The attenuation coefficient and the effect of temperature change on R m Correction factor;

[0052] The dynamic correction sub-step for shear strength, based on the real-time environmental monitoring sub-step and the environmental correlation model construction sub-step, corrects the shear strength τ of the structural surface online. The formula is τ'=τ(1+α•ΔT+β•ΔR), where α is the temperature correction coefficient, β is the rainfall correction coefficient, ΔT is the difference between the measured temperature and the initial test temperature, and ΔR is the difference between the cumulative rainfall and the rainfall in the initial test period.

[0053] By adopting the above technical solution, and through the real-time monitoring sub-step of environmental parameters in the dynamic correction step of environment and structure, humidity sensors, rainfall sensors, and rock strain gauges are deployed in the slope survey area to continuously collect data on slope environmental temperature, humidity, cumulative rainfall, and structural surface micro-strain, establishing a time-series database of environmental parameters; in the environmental correlation model construction sub-step, machine learning algorithms are used to analyze the relationship between environmental parameters and JRC. n R m ψ b The dynamic correlation pattern, quantifying the JRC when the rainfall increases by a preset threshold. n The attenuation coefficient and the effect of temperature change on R m The correction coefficient; the dynamic correction sub-step for shear strength, based on the real-time environmental monitoring sub-step and the environmental correlation model construction sub-step, corrects the shear strength τ of the structural surface online, using the formula τ'=τ×(1+α×ΔT+β×ΔR), where α is the temperature correction coefficient, β is the rainfall correction coefficient, ΔT is the difference between the measured temperature and the initial test temperature, and ΔR is the difference between the cumulative rainfall and the initial test period rainfall. This solves the problem of result deviation caused by ignoring environmental factors in traditional static assessment, making the slope stability assessment more in line with real-time working conditions and significantly reducing the misjudgment of engineering risks caused by environmental changes.

[0054] The present invention has significant technical effects due to the adoption of the above technical solutions: The method and system for testing the roughness coefficient of slope structural surface provided in this application solves the problems of traditional methods such as damage to structural surface morphology, data dispersion and impact of environmental factors on assessment accuracy by dividing the test area for multi-dimensional detection, adopting digital contour analysis and dynamic parameter correction. It has the advantages of improving test accuracy, maintaining the original morphology of the structural surface and realizing dynamic correlation of multi-dimensional parameters. Attached Figure Description

[0055] Figure 1 This is a schematic diagram illustrating the steps of a method for testing the roughness coefficient of a slope structure surface.

[0056] Figure 2 This is the contour curve of test point number H1-1 in Table 2 of the embodiment;

[0057] Figure 3 This is the contour curve of test point number H1-2 in Table 2 of the embodiment;

[0058] Figure 4 This is the contour curve diagram of test point numbers H1-3 in Table 2 of the embodiment;

[0059] Figure 5 This is the contour curve diagram of test point numbers H1-4 in Table 2 of the embodiment;

[0060] Figure 6 This is the contour curve diagram of test point numbers H1-5 in Table 2 of the embodiment;

[0061] Figure 7 This is the contour curve diagram of test point numbers H1-6 in Table 2 of the embodiment. Detailed Implementation

[0062] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0063] Example:

[0064] Existing methods for testing the surface roughness coefficient of slope structures suffer from several problems, including difficulty in balancing accuracy and efficiency, insufficient data representativeness and integration capabilities, weak parameter correlation, and poor environmental adaptability. Traditional manual measurements have large errors, laboratory direct shear tests are time-consuming and damage the original morphology of the samples, and rebound testing and contour scanning only target local areas, resulting in scattered data that cannot reflect the overall characteristics of the surface structure. Parameter acquisition is disconnected and does not consider the dynamic influence of the environment, and static models are difficult to accurately assess slope stability. Therefore, a testing method that balances accuracy and efficiency needs to be developed. This method can improve parameter reliability through multi-source data integration and dynamic correction mechanisms, solve the data dispersion problem by dividing the test area and establishing spatial markers, and simultaneously acquire surface strength and morphological characteristics by combining rebound testing and contour curve plotting. Using a multi-range roughness ruler can reduce manual reading errors, ultimately forming a systematic testing scheme.

[0065] This application proposes a test method that includes structural surface preprocessing, springback detection, profile curve plotting, and roughness measurement, combined with... Figure 1 The pretreatment step cleans the slope structure surface and divides the test area, with each test area having a unique spatial identifier; the rebound test step arranges test points in the test area and calculates the average rebound value; the profile curve drawing step obtains the surface profile map through the instrument; the roughness measurement step uses a multi-range ruler to read the roughness coefficient corresponding to the profile curve.

[0066] Survey area division refers to dividing the slope into several independent testing units based on the spatial distribution characteristics of the structural surface. For example, each survey area can be a 1-square-meter area, spatially identified by GPS coordinates or QR codes to ensure data traceability. Rebound testing refers to using a rebound hammer to conduct impact tests on the measuring points. For example, 20 measuring points are arranged in each survey area, and the average value is calculated after removing outliers to reflect the overall strength of the structural surface. Contour curve drawing refers to scanning the surface morphology along a preset direction using a contact or laser profilometer. For example, three curves are collected along the main direction and three along the vertical direction to eliminate local interference. Multi-range roughness ruler refers to a measuring tool with graded scale lines. For example, the scale line spacing is 0.5 mm. The starting line is used to align the curve peaks, and the roughness coefficient is determined by the position of the deepest valley, reducing subjective judgment errors.

[0067] After removing surface impurities in the pretreatment step, the structural surface is divided into multiple test areas according to the degree of crack development. For example, the test area is reduced in areas with dense cracks to improve resolution. During rebound testing, equidistant grid points are used, and outlier data is eliminated using mathematical statistics methods to ensure that the average rebound value represents the overall characteristics of the test area. Profile curve plotting uses bidirectional scanning, and digital noise reduction eliminates instrument vibration interference while preserving true texture features. For roughness measurement, the profile curve is aligned with the scale lines of a multi-range ruler, and the JRC value is directly read based on the valley location. n For example, when the valley bottom is located between the 5th and 6th scale lines, an interpolation method is used to calculate the precise value.

[0068] This method achieves systematic data management through test area division and spatial identification, avoiding insufficient representativeness of local data; it combines springback detection with contour curves to simultaneously acquire strength and morphological parameters, improving parameter correlation; and it quantifies the measurement process with multi-range roughness scales, reducing reliance on manual experience. Traditional methods rely on a single detection means, leading to error accumulation, while this solution forms a complete test chain through multi-step collaborative optimization.

[0069] This application solves the problems of data dispersion, large subjective error and parameter disconnect in traditional methods, realizes efficient and accurate testing of the roughness coefficient of slope structure surface, enhances the data integration capability through test area division and spatial identification, provides multi-dimensional parameter support through rebound detection and contour curve drawing, and quantifies the measurement process with multi-range roughness scales, providing reliable basic data for slope stability assessment.

[0070] This application also proposes a method for testing the roughness coefficient of slope structural surfaces: rock wall strength calculation steps, configuring a point load instrument, selecting a predetermined number of rock blocks with the same geological conditions as the structural surface in the test area, conducting tests using the point load instrument, and recording the point load strength index I. s (50), Rock wall strength JCS n The calculation formula is JCS n =22.82*Is (50); Calculation steps for the basic friction angle of the structural surface, the basic friction angle ψ of the slope structural surface b The calculation formula is ψ b =0.414Rm+9.273; The steps for obtaining the normal stress on the structural surface are as follows: several sets of vibrating wire strain gauges are arranged on both sides of the structural surface; data is collected and continuously monitored at a preset frequency for a preset time; the average value of the monitored values ​​is taken as σ. n The steps for calculating the shear strength of the structural plane are as follows: Calculate the shear strength τ according to the Barton-Bandis model, using the formula τ = σ. n tan{JRC n •lg(JCS n / σ n )+ψ b}

[0071] Table 1 shows the rock rebound test records and the basic friction angles of the structural surfaces. R is calculated based on the data from 16 measuring points. m ; through ψ b =0.414Rm+9.273, the basic friction angle ψ of the structural surface is obtained. b .

[0072] Table 1: Rock Rebound Test Record Sheet

[0073]

[0074] A point load cell is a portable testing device used to determine the point load strength of rock. Specifically, it can be implemented using a hydraulic loading device in conjunction with a pressure sensor. The failure load value is obtained by fracturing a rock block, and this value is used to calculate the rock wall strength. A vibrating wire strain gauge is a sensor that measures strain based on the change in the frequency of wire vibration. Specifically, it can be implemented using a steel wire structure encapsulated in a waterproof shell. Continuous data of the normal stress on the structural surface is obtained through frequency signal conversion. The Barton-Bandis model is a shear strength calculation model based on structural surface roughness, rock wall strength, and friction angle. Specifically, it can be implemented using JRC... n JCS n ψ b Substituting into the formula achieves parameter coupling calculation, where the basic friction angle ψ of the slope structural surface is... b The calculation formula comes from the investigation of unstable rock masses on environmental slopes.

[0075] In the rock wall strength calculation step, the strength parameters of the rock blocks in the field are directly obtained through point load tests. For example, 5-8 rock samples from the same layer as the structural surface are selected for fracturing tests. The point load strength index is converted into the equivalent uniaxial compressive strength to establish consistency with the strength of the structural surface in the field. In the basic friction angle calculation step, the statistical average value of the rebound value in the test area is used as the basis. For example, R is calculated using data from 16 test points after removing outliers. mSubstituting the values ​​into the linear regression formula eliminates the bias of empirical values. The normal stress acquisition step uses a vibrating wire strain gauge array; for example, three sets of sensors are placed at a distance of 50cm on both sides of the structural surface, continuously collecting data for 7 days at a frequency of once per hour. The influence of instantaneous load fluctuations is eliminated through averaging. The shear strength calculation step uses the measured JRC... n JCS n ψ b With continuous monitoring σ n Input the Barton-Bandis model, and for example, determine the shear strength curves under different normal stresses through iterative calculations to achieve dynamic parameter correlation.

[0076] Existing methods for rock wall strength often employ laboratory uniaxial compressive strength testing, which requires destructive sampling and differs from the strength of the in-situ structural surface. This proposed solution, however, directly correlates in-situ rock mass properties through in-situ load testing. Traditional basic friction angles often rely on empirical values ​​or standard rock sample tests, neglecting the influence of the structural surface condition. This solution establishes a quantitative relationship based on rebound values, improving parameter accuracy. Conventional normal stress monitoring typically uses single static measurements; this solution uses vibrating wire strain gauges to continuously acquire data, effectively reflecting dynamic load changes. Existing shear strength models suffer from scattered parameter sources and lack coupling; this solution integrates field-measured JRC data. n JCS n ψ b and σ n This enables multi-parameter collaborative computation.

[0077] This application addresses the problems of mismatch between rock wall strength and on-site structural surface strength, deviations in empirically determined basic friction angle values, and incomplete static monitoring data of normal stress, which lead to errors in shear strength calculation. By quantitatively correlating in-situ load testing with rebound values, the authenticity of rock wall strength and friction angle parameters is ensured. Continuous normal stress monitoring eliminates instantaneous load interference and improves data reliability. A multi-parameter coupled calculation model enhances the accuracy of shear strength prediction, providing accurate basic data for slope stability assessment.

[0078] The surface springback testing step also includes a measurement point distribution optimization sub-step and an outlier stratification removal sub-step. The measurement point distribution optimization sub-step arranges measurement points within the test area using an equidistant grid or random sampling method based on the surface texture features of the surface, ensuring that the measurement points cover different texture areas within the test area. The outlier stratification removal sub-step first removes springback values ​​exceeding a preset standard deviation range, and then removes the maximum and minimum values ​​from the remaining values ​​according to a preset proportion. The remaining values ​​are used to calculate the average springback value R. m .

[0079] The measurement point distribution optimization sub-step refers to planning the arrangement of measurement points based on the spatial variation characteristics of the surface texture of the structural surface. Specifically, it can be achieved using the equidistant grid method or the random sampling method. For example, for structural surfaces with striped textures, the equidistant grid method can be used to evenly distribute points along the texture direction; for randomly distributed fragmented textures, the random sampling method can be used to generate random coordinate points within the measurement area. This step avoids data deviation caused by measurement points being concentrated in a single feature area by covering different texture areas.

[0080] The outlier stratification removal sub-step refers to the use of a phased data cleaning strategy, which can be achieved by combining standard deviation screening and extreme value removal. For example, the preset standard deviation range can be ±2 times the average value, and the measurement point data exceeding this range are judged as outliers; the preset proportion can be 5%-10% of the remaining data. This step effectively eliminates interference data caused by instrument misoperation or local rock mass defects through the stratified removal mechanism.

[0081] The measurement point distribution optimization sub-step analyzes the spatial distribution pattern of the surface texture of the structural surface and dynamically adjusts the measurement point layout strategy. For structural surfaces with obvious directionality, such as those with parallel joints or layered structures, an equidistant grid method is used to arrange measurement point columns along the main texture direction, while auxiliary measurement points are arranged at intervals in the vertical direction. When the surface texture is irregularly distributed, a random sampling algorithm is used to generate measurement point coordinates to ensure that each texture sub-region is covered by measurement points. The outlier stratification removal sub-step first calculates the standard deviation of the rebound values ​​of all measurement points, automatically filtering out abnormally high or low values ​​caused by the probe not perpendicularly contacting the rock surface or local rock fragmentation. Then, the remaining data is sorted, and extreme data is removed according to a preset ratio, such as removing the highest and lowest 5% of the measurement point data. Finally, the average rebound value R is calculated using the middle 90% of the effective data. m .

[0082] Traditional springback testing often uses fixed-interval sampling points or manual selection based on experience, which easily leads to sampling points being concentrated in flat areas of the surface, failing to reflect the true roughness of the structural surface. This solution uses texture features for adaptive sampling, significantly improving the spatial representativeness of the sampling points. Existing outlier processing often uses simple arithmetic averaging or single threshold removal, which cannot effectively identify complex outliers. This solution employs a layered removal mechanism, first eliminating obvious outliers and then eliminating statistical biases, thus improving the R-value. m The calculated values ​​are closer to the actual rock mass condition.

[0083] This application effectively solves the problem of insufficient spatial representativeness of springback test data for structural surfaces, ensuring that the test points cover different texture areas and avoiding overall data distortion caused by oversampling of local areas. At the same time, through a layered data cleaning mechanism, it significantly improves the reliability of the springback value calculation results, providing accurate basic parameters for subsequent shear strength calculation. While maintaining testing efficiency, this scheme systematically improves data quality and overcomes the technical defects of traditional methods, such as strong subjectivity in manual point selection and rough handling of abnormal data.

[0084] The contour curve drawing process also includes the following sub-steps: a multi-directional scanning sub-step, which draws contour curves along the main direction and perpendicular direction of the structural surface, acquiring at least three parallel curves in each direction; a curve smoothing sub-step, which performs digital noise reduction on the drawn contour curves, removing abnormal fluctuations caused by probe jumps and preserving the true surface texture features; an electronic precision measurement sub-step, which digitizes and stores the surface contour curve obtained from the curve smoothing sub-step, and performs measurement using an electronic roughness ruler adapted to multi-range roughness rulers; when the deepest valley of the surface contour curve is between two scale lines, the two scale lines are equally divided using the electronic roughness ruler, and the precise JRC is calculated and read based on the subdivided scale lines after equal division. n value.

[0085] The multi-directional scanning sub-step refers to acquiring contour data along the main extension direction of the structural surface and in directions perpendicular to it. This can be achieved by moving the probe along a preset path using a mechanical or laser profilometer. By covering curves with different orientations, the anisotropic characteristics of the structural surface are obtained. The curve smoothing sub-step involves filtering the acquired raw contour data. This can be done using wavelet transform algorithms or moving average methods to eliminate noise interference caused by sensor jitter or surface debris, preserving the true undulation shape. The electronic precision measurement sub-step involves converting the contour curve into a digital signal for storage and analysis. This can be achieved by using a high-precision analog-to-digital converter to convert analog signals into pixel coordinate data, combined with the scale recognition algorithm of an electronic roughness ruler for automatic measurement. The equal division of the subdivision scale lines refers to equally dividing the adjacent scale intervals when the valley position lies between two standard scale lines. This can be done using linear interpolation or quadratic curve fitting to calculate the JRC at the intermediate position. n value.

[0086] During the contour curve drawing process, scanning paths are first arranged along the main direction of the structural surface and its perpendicular direction. For example, a laser probe is used to move along a straight trajectory with a fixed step size. At least three parallel curves are collected in each direction to cover different texture areas. The collected raw contour data undergoes digital noise reduction processing. For example, by setting a threshold to filter out abrupt jumps, the fluctuating curves that reflect the true surface features are retained. The processed contour curve is converted into a digital image with coordinate information, stored in a database, and interfaced with the scale recognition module of the electronic roughness ruler. When measuring the deepest valley position, if it falls between two scale lines, the system automatically divides the adjacent scale interval into several equal parts. For example, a 1 mm interval is subdivided into 0.1 mm scales. The precise JRC is determined by calculating the proportional relationship of the subdivided position of the valley. n value.

[0087] Traditional methods scan the contour curve only in a single direction, which cannot reflect the anisotropic roughness distribution of the structural surface. This solution, however, obtains multi-dimensional data through multi-directional scanning, improving the comprehensiveness of parameter characterization. Existing manual measurements rely on visual judgment of the valley bottom corresponding to the scale line, with an error of ±0.5 scale. This solution achieves precise readings at the 0.1 scale level through electronic subdivision of the scale. In addition, existing profilometers often generate noise interference due to probe vibration. This solution effectively eliminates abnormal fluctuations through digital noise reduction, making the contour curve closer to the real surface morphology.

[0088] This application solves the data partiality problem caused by single-direction contour scanning. It improves the accuracy of roughness coefficient measurement through multi-directional curve acquisition and digital noise reduction processing. The electronic subdivision scale technology breaks through the minimum scale limitation of traditional measuring tools, enabling JRC... n The accuracy of the readings is improved by an order of magnitude, providing a reliable data foundation for subsequent shear strength calculations. At the same time, the digital profile data storage and processing method supports the rapid integration and analysis of data from multiple survey areas, enhancing the systematic nature of slope stability assessment.

[0089] The contour curve drawing process includes a contour validity verification module, which is used to compare JRC curves in different directions within the same survey area. n Differences, when JRCs in different directions within the same survey area n When the difference between values ​​exceeds a preset threshold, the contour curve of the measurement area is re-acquired.

[0090] The contour validity verification module is a functional unit used to verify the consistency of contour curve measurement results in different directions. It can be implemented using software algorithms or hardware logic circuits, by comparing JRC measurements in different directions. n Value differences determine data reliability; the preset threshold refers to the allowable differences in JRC values. nThe upper limit of the difference can be set as a percentage or an absolute value. For example, data resampling can be triggered when the difference exceeds 5%.

[0091] During the contour curve drawing process, contour curves are obtained along both the main direction and the perpendicular direction of the structural surface, and the corresponding JRC is calculated. n The contour validity verification module automatically compares the JRC values ​​in two directions. n If the difference exceeds a preset threshold, it is determined that the current contour curve has a measurement deviation or local structural anomaly, triggering a contour curve re-acquisition process. For example, when the preset threshold is 8%, if the JRC in both directions... n If the values ​​are 12.5 and 14.2 respectively, and the difference reaches 13.6%, the system will automatically mark the data of the test area as abnormal and start the retest.

[0092] Traditional methods only draw contour curves along a single direction, which cannot identify data deviations caused by improper selection of measurement direction or anisotropy of local structural surfaces. This solution effectively avoids shear strength calculation errors caused by distortion of single-direction measurement results through multi-directional data cross-validation and difference threshold control.

[0093] This application solves the problem of insufficient data representativeness caused by the single measurement direction of the profile curve in the prior art, and ensures that the roughness coefficient measurement results of different directions are consistent, thereby improving the accuracy of the shear strength calculation model. At the same time, through the dynamic judgment mechanism of preset threshold, unnecessary repeated measurements are avoided while ensuring data reliability, thus optimizing the testing efficiency.

[0094] Table 2 shows the statistical results of the roughness coefficient of a certain test area, with the corresponding map number and accompanying description. Figures 2 to 7 .

[0095] Table 2 Statistical Table of Roughness Coefficient Results for Test Area

[0096]

[0097] A data integration step for a slope surface roughness coefficient testing method includes a laser scanner to integrate test data from several test areas into complete overall surface data. The steps include setting benchmark points, data preprocessing, coordinate matching and splicing, and data integration and display. Setting benchmark points involves selecting at least three non-collinear benchmark points on the slope surface, recording and marking the spatial coordinates of each benchmark point, and ensuring that each test area's data includes at least one benchmark point's association information. Data preprocessing involves digitizing the surface contour curves obtained from the contour curve drawing step for each test area, extracting the curve's coordinate information, and simultaneously organizing the average rebound value obtained from the surface rebound detection step and the roughness coefficient obtained from the roughness measurement step for each test area, associating them with the corresponding spatial locations of the test areas. The coordinate matching and splicing step involves transforming the numerical contour curve coordinates, average rebound value, and roughness coefficient of each test area to a unified coordinate system based on the benchmark point coordinates, and using an iterative nearest-point algorithm to match the contour curve edges of adjacent test areas, ensuring that the edges of adjacent test areas are aligned. The seamless connection of the contour curves forms a complete comprehensive map of the structural surface contour. The data integration and display sub-steps include spatially associating the spliced ​​complete comprehensive map of the structural surface contour with the average rebound value and roughness coefficient of each survey area. The distribution characteristics of the roughness coefficient and rebound value of different areas of the structural surface are displayed through color gradient or numerical annotation. The reference point refers to a reference point with a fixed spatial position on the slope structure surface. Specifically, coordinates can be collected using a total station or GPS positioning equipment to establish the spatial correlation of data from each survey area. Digital processing refers to converting the contour curves into digital signals containing coordinate information. This can be achieved using image vectorization software for easy computer storage and analysis. The iterative nearest point algorithm is an optimization algorithm based on point cloud registration. Specifically, high-precision matching can be achieved by calculating the geometric feature points of the contour curves of adjacent survey areas to eliminate splicing errors at the boundaries of the survey areas. Color gradient annotation refers to using color level changes to reflect the differences in parameter distribution. This can be achieved using geographic information system software to intuitively display the spatial variation trend of performance parameters in each area of ​​the structural surface.

[0098] Three reference points are selected on the slope structure surface to form a spatial triangle network. When data is collected by a laser scanner for each survey area, at least one reference point coordinate must be included. After digital conversion, the contour curve of the survey area generates point cloud data containing X, Y, and Z coordinates. The average rebound value and roughness coefficient are added as attribute fields to the corresponding survey area dataset. During the coordinate transformation process, an affine transformation model is used to align the local coordinate system of each survey area to the global coordinate system. The iterative nearest point algorithm continuously adjusts the rotation and translation matrix to make the overlapping area of ​​the contour curves of adjacent survey areas reach the minimum distance convergence. The stitched comprehensive map displays the roughness coefficient distribution in the form of a heat map, and the rebound value of each survey area is marked in the form of a bubble chart.

[0099] Traditional methods can only obtain isolated data from local survey areas, and the lack of spatial coordinate correlation makes it impossible to construct an overall structural surface model. This solution establishes a spatial reference system through benchmark points and combines coordinate transformation and iterative registration algorithms to achieve accurate splicing of data from multiple survey areas, solving the problem of difficulty in integrating scattered data. In existing technologies, manual splicing of contour curves results in misalignment and gaps, while this solution uses digital processing and automatic registration algorithms to control splicing errors within the millimeter range, significantly improving the geometric accuracy of the overall structural surface model. This application realizes the spatial integration and visualization of test data from multiple survey areas, accurately reflecting the spatial distribution law of structural surface roughness and resilience. The comprehensive structural surface model constructed based on a unified coordinate system can provide global data support for slope stability analysis, assisting engineers in quickly identifying high-risk areas. The automated data processing process reduces the error risk caused by manual intervention and improves the efficiency and reliability of slope detection and assessment.

[0100] It also includes slope inspection and maintenance steps, which include sub-steps: a benchmark data storage sub-step to provide a comparison benchmark for continuous slope inspection and maintenance, and a data integration and display sub-step to present a complete structural surface profile map after splicing, and the average rebound value R of each test area. m Roughness coefficient JRC n Basic friction angle ψ b The shear strength τ is associated with the spatial coordinates of the test area and the test timestamp, and stored in the benchmark database; the slope zoning coding sub-step divides the slope into zones based on the spatial distribution characteristics of the structural surfaces, with each zone associated with at least 3 benchmark coordinates and the initial test JRC. n The average value and the τ-average value are used as regional reference units for subsequent maintenance.

[0101] A benchmark database refers to a database that stores parameters such as the comprehensive profile of the structural surface, rebound value, roughness coefficient, friction angle, and shear strength, and associates them with spatial coordinates and test time. It can be implemented using a relational database or a time-series database. It is used to provide historical data comparison benchmarks for subsequent inspection and maintenance. Slope zoning coding refers to dividing the slope into multiple independent areas based on the spatial distribution characteristics of the structural surface and assigning them unique codes. It can be implemented using a geographic information system combined with a three-dimensional coordinate grid. Each zone establishes a regional reference unit through benchmark point coordinates and initial parameters, which facilitates rapid positioning and data traceability during subsequent maintenance.

[0102] The benchmark data storage sub-step associates and stores the composite surface profile map of the structural surface with the mechanical parameters of each test area, forming a benchmark database containing spatial location, timestamps, and multi-dimensional parameters. For example, after the initial inspection, a complete composite surface profile map is generated through coordinate matching and stitching sub-steps, and the R values ​​of each test area are stored. m JRC n ψb The τ value is calculated through a data integration and display sub-step. This data is associated with the corresponding three-dimensional coordinates and detection time, and stored in the benchmark database. The slope zoning and coding sub-step further divides the slope into multiple independent units. Each unit's spatial range is determined by at least three benchmark coordinates, and the initial test JRC is used to... n The average value and the τ-average value serve as the baseline parameters for this unit. During subsequent maintenance, the baseline data for the corresponding region can be quickly retrieved through the partition coding and compared and analyzed in conjunction with the newly measured parameters.

[0103] Traditional methods lack a benchmark database and dynamic retesting mechanism, making it impossible to track the changing trends of structural surface parameters over time. For example, in existing technologies, detection data is mostly stored in isolation and is not associated with spatial coordinates and timestamps, making it difficult to compare historical data. This solution stores multi-dimensional parameters and spatial information in a benchmark database and establishes a regional data reference system by combining slope zoning coding. This allows subsequent detection data to be directly correlated with the initial benchmark in time and space, effectively supporting the full life cycle management of slopes.

[0104] This application solves the problems of scattered storage of slope detection data and lack of historical comparison benchmarks. It realizes the dynamic correlation between structural surface parameters and spatial location and detection time. The benchmark database provides a unified data reference for subsequent re-measurement, while the slope zoning code improves maintenance efficiency through regional data management. The combination of the two can accurately identify areas with abnormal structural surface parameters, providing a data foundation for dynamic assessment of slope stability.

[0105] Furthermore, it proposes that the slope inspection and maintenance steps be equipped with a subsequent inspection cycle control module to determine the time interval for subsequent inspections. This includes dividing the slope area into several levels based on the initial shear strength, with longer inspection intervals for areas with higher shear strength and shorter intervals for areas with lower shear strength; when natural environmental changes occur or the slope shows significant deformation, the inspection interval for the corresponding area is shortened to conduct supplementary inspections of the slope.

[0106] The subsequent inspection cycle control module refers to an automated decision-making unit that dynamically adjusts the inspection frequency based on the risk level of the slope area. Specifically, it can be implemented using a preset shear strength threshold range division algorithm combined with a real-time environmental monitoring data triggering mechanism. For example, shear strength can be divided into three levels: high, medium, and low, each corresponding to a different preset inspection cycle. This module solves the problem of rigid inspection cycles in existing technologies through dynamic hierarchical management. Natural environmental changes refer to slope anomalies caused by external factors such as rainfall, temperature fluctuations, or seismic activity. This can be monitored in real time using humidity sensors, displacement gauges, and meteorological data interfaces. For example, an early warning is triggered when the daily rainfall exceeds twice the historical average. This feature is used to identify the impact of sudden environmental factors on slope stability and avoid monitoring delays caused by sudden environmental changes. Significant deformation refers to physical deformation such as crack expansion or local collapse on the slope surface. This can be achieved by periodically generating a three-dimensional morphological comparison model using a laser scanner and automatically identifying it using a preset deformation threshold. This feature provides an objective basis for shortening the inspection interval by quantifying the degree of deformation.

[0107] During implementation, the slope is first divided into different risk level zones based on the initial shear strength test results. For example, areas with shear strength higher than 50 MPa are classified as low-risk zones, and the testing cycle can be set to 12 months; medium-risk zones with shear strength between 30-50 MPa have a testing cycle of 6 months; and high-risk zones with shear strength lower than 30 MPa have a testing cycle of 3 months. At the same time, temperature and humidity sensors and displacement monitoring devices are deployed on the slope. When continuous rainfall reaches 200 mm or a single-day temperature drop exceeds 15 degrees Celsius, the testing cycle of the affected area is automatically shortened to 50% of the original cycle. For deformation areas with cumulative displacement exceeding 5 mm, a supplementary testing procedure is immediately initiated. Through the above-mentioned classification and triggering mechanism, the optimal allocation of testing resources is achieved.

[0108] Existing methods typically employ a fixed cycle for uniform testing of all areas, such as a comprehensive test every 6 months. This approach fails to differentiate the actual needs of different risk areas, resulting in insufficient testing in high-risk areas or wasted resources in low-risk areas. In contrast, this solution establishes a differentiated management mechanism through shear strength grading and dynamically adjusts the testing frequency based on real-time environmental data, ensuring that the monitoring density matches the degree of slope risk.

[0109] This application enables precise monitoring based on the actual risk level of different areas of the slope. While ensuring the monitoring density in high-risk areas, it reduces the detection cost in low-risk areas. When encountering extreme weather or sudden deformation, it can automatically shorten the detection interval to capture changes in structural parameters in a timely manner, avoiding slope instability accidents caused by monitoring delays. In addition, the dynamic adjustment mechanism can reduce the need for manual intervention and improve the level of intelligence in slope maintenance and management.

[0110] This application further proposes slope detection and maintenance steps, including a periodic re-measurement sub-step and a model iteration sub-step. The periodic re-measurement sub-step performs rebound testing and roughness re-measurement on key areas according to a preset cycle, and updates R... m JRC n and τ value; the model iteration sub-steps revise the shear strength calculation model based on the newly measured data and optimize the parameter weights under different geological conditions.

[0111] The periodic re-measurement sub-step refers to the process of periodically repeating the detection of key areas of the slope. This can be achieved by using preset time intervals or trigger-based detection mechanisms, such as automatically triggering detection every quarter or starting detection based on abnormal sensor data. This step solves the problem that traditional one-time detection cannot track dynamic changes in parameters by continuously acquiring the latest data. The model iteration sub-step refers to the process of optimizing the parameters of the shear strength calculation model. This can be achieved by using regression analysis or machine learning algorithms, such as adjusting the weight coefficients of different geological factors in the model using the gradient descent method. This step solves the problem that static models cannot adapt to complex geological conditions by dynamically correcting the model parameters.

[0112] In the periodic retesting sub-step, the selection of key areas can be based on the shear strength grading results of the initial test. For example, areas with τ values ​​below a preset threshold are marked as high-priority test areas. The test cycle can be set according to the slope stability level, for example, high-risk areas are retested monthly, and low-risk areas are retested every six months. The retesting data is updated to the database in real time via a wireless transmission module and compared with historical data to form a time series. The model iteration sub-step analyzes R under different geological conditions by establishing a parameter correlation matrix. m JRC n Nonlinear relationship with τ value, for example, the effect of increased humidity on JRC in argillaceous rock layers. n The attenuation correction factor, in the granite layer, strengthens temperature on R m The influence weights are determined, and a new calculation coefficient table is generated after model optimization for use in the next round of shear strength calculation.

[0113] The preset cycle can be dynamically adjusted according to seasonal changes. For example, the detection interval during the rainy season is shortened to 50% of the regular cycle. The parameter weight optimization can adopt the random forest algorithm, and lithology, humidity, and historical deformation can be used as feature variables to input the model for training. The updated shear strength calculation model can generate a visual parameter heat map to help engineers quickly identify weak areas.

[0114] Traditional methods employ fixed detection cycles and lack model update mechanisms, resulting in insufficient monitoring frequency in high-risk areas or waste of resources in low-risk areas. This solution achieves optimized allocation of detection resources and continuous improvement in calculation accuracy through dynamic cycle adjustment and model parameter iteration. In existing technologies, the shear strength model parameter weights are fixed and cannot reflect the differences in geological characteristics of different rock strata. This solution achieves adaptive optimization of parameter weights through machine learning algorithms.

[0115] This application addresses the issues of rigid inspection cycles and poor model adaptability in slope maintenance, achieving precise matching between inspection frequency and risk level. It improves the applicability of the shear strength calculation model in different geological environments. Through continuous data updates and model iterations, it can effectively track the time-sensitive changes in structural parameters, providing dynamic and reliable data support for slope maintenance decisions.

[0116] This application further proposes a dynamic correction step for environmental and structural properties, which includes a real-time environmental parameter monitoring sub-step, an environmental correlation model construction sub-step, and a dynamic shear strength correction sub-step. The real-time environmental parameter monitoring sub-step involves deploying humidity sensors, rainfall sensors, and rock strain gauges in the slope survey area to continuously collect data on slope environmental temperature, humidity, cumulative rainfall, and structural surface micro-strain, establishing a time-series database of environmental parameters. The environmental correlation model construction sub-step analyzes the relationship between environmental parameters and JRC (Junior Shear Strength Ratio) using machine learning algorithms. n R m ψ b The dynamic correlation pattern, quantifying the JRC when the rainfall increases by a preset threshold. n The attenuation coefficient and the effect of temperature change on R m The correction coefficient; the dynamic correction sub-step for shear strength is based on the real-time environmental monitoring sub-step and the environmental correlation model construction sub-step, and corrects the shear strength τ of the structural surface online. The formula is: τ'=τ×(1+α×ΔT+β×ΔR), where α is the temperature correction coefficient, β is the rainfall correction coefficient, ΔT is the difference between the measured temperature and the initial test temperature, and ΔR is the difference between the cumulative rainfall and the rainfall in the initial test period.

[0117] A humidity sensor is a device used to monitor the air humidity around a slope structure. Specifically, a capacitive humidity sensor can be used, which reflects humidity by detecting changes in the dielectric constant. This is used to capture changes in rock mass water content caused by rainfall infiltration. A rock mass strain gauge is a miniature sensor embedded inside the structural surface. Specifically, a vibrating wire strain gauge can be used, which calculates the micro-strain of the rock mass by measuring changes in the vibration frequency of the steel wire. This is used to monitor the deformation response of the structural surface to environmental influences. A machine learning algorithm is a method of building predictive models through data training. Specifically, a random forest algorithm can be used, which analyzes the nonlinear relationship between environmental parameters and roughness coefficient by constructing multiple decision trees. This is used to discover the relationship between temperature, rainfall, and JRC (Jaw Roughness Coefficient).n The inherent law of attenuation, the attenuation coefficient is a parameter that quantifies the impact of environmental factors on roughness, and can be specifically calculated using the gradient descent method. This is achieved by fitting historical data to the rainfall increment and JRC (Junior Ratio Controlled Roughness). n The proportional relationship of the decrease is used to dynamically correct the shear strength calculation model. The temperature correction coefficient is a parameter characterizing the effect of temperature change on the rebound value, which can be determined through multiple regression analysis based on R under different temperature conditions. m Correction relationships are established based on measured values ​​to eliminate the interference of temperature fluctuations on the rebound test results.

[0118] Humidity sensors and rock strain gauges deployed in the slope monitoring area collect environmental data at preset frequencies, such as recording temperature, humidity, and micro-strain values ​​every 10 minutes, forming a continuous monitoring dataset. When rainfall occurs, the humidity sensors detect an increase in humidity, while the rock strain gauges capture changes in micro-strain values ​​caused by water absorption and expansion of the structural surface. After preprocessing, this data is input into a machine learning model, which analyzes historical rainfall data in relation to JRC (Junior Ratio Controlled Reduction). n The correspondence between values ​​was established, and the JRC was adjusted for every 10mm increase in rainfall. n An empirical coefficient of 0.15 was used to attenuate the rock mass. Simultaneously, a temperature sensor recorded the thermal expansion and contraction effect of the rock mass caused by diurnal temperature differences. Combined with rebound hammer test results at different temperatures, the effect of R on temperature increase of 1°C was determined. m The correction factor of 0.3 is used to calculate the shear strength. The real-time temperature difference ΔT and the cumulative rainfall difference ΔR are substituted into the correction formula. For example, when the measured temperature is 5℃ higher than the initial test and the rainfall is 50mm more, the shear strength τ' will be dynamically adjusted based on the original calculation result by coefficients α=0.02 and β=-0.003, thereby eliminating the calculation deviation caused by environmental factors.

[0119] Traditional methods for calculating shear strength only use static environmental parameters and do not consider the effects of continuous rainfall or temperature changes on the physical properties of structural surfaces. For example, in routine testing, if heavy rainfall occurs, the JRC measured under dry conditions is still used. n The previous method overestimated the shear strength of the structural surface by approximately 12%-18%. This new approach, however, uses a sensor network to capture real-time environmental changes and combines this with a dynamic correction model built through machine learning. This improves the accuracy of the shear strength calculation to over 95% match the actual on-site conditions. Furthermore, existing technologies rely on empirical formulas to correlate environmental and mechanical parameters, while this approach, through data-driven quantitative analysis, can accurately identify the impact of rainfall infiltration on JRC (Jump Reduction Capacity). n The critical threshold for attenuation, for example, triggering JRC when cumulative rainfall exceeds 80mm. n Automatic correction of values.

[0120] This application realizes the dynamic response of the shear strength calculation model to environmental factors, solving the problem of accuracy degradation faced by traditional static models in complex environments. By establishing a quantitative relationship between environmental parameters and roughness coefficients, it can accurately reflect the JRC caused by rainfall. n By combining real-time monitoring data with online correction algorithms to study the impact of attenuation and temperature fluctuations on rebound values, the system can automatically adapt to current environmental conditions in slope stability analysis, avoiding safety assessment errors caused by parameter lag. This scheme further provides dynamic early warning capabilities for slope maintenance. For example, when the corrected shear strength τ' is lower than the safety threshold, the system can automatically trigger an early warning signal to guide engineers to take reinforcement measures.

Claims

1. A method for testing the roughness coefficient of a slope structural surface, characterized in that, Includes the following steps: Structural surface pretreatment steps: The slope structural surface is treated to remove impurities, loose rock debris and covering materials. Several survey areas are divided according to the spatial distribution characteristics of the structural surface, and each survey area is assigned a unique spatial identifier. The structural surface rebound testing procedure involves distributing several testing points in each testing area, performing rebound tests on these points using a rebound hammer, and calculating the average rebound value R for the testing area based on the rebound values ​​obtained from the rebound tests. m ; The contour curve drawing process involves using a contour curve analyzer to draw the surface contour curve between different detection points within the test area. The roughness measurement procedure involves configuring a multi-range roughness gauge with several scale lines, the first scale line being the starting line. The reading method for the multi-range roughness gauge is as follows: At least two peaks of the surface profile curve are tangent to the starting line, and the reading is based on the length L of the surface profile curve of the structural surface. n And read the JRC corresponding to the scale line where the deepest valley is located in the surface profile curve. n The value, that is, the measured length is L n Roughness coefficient JRC of the profile curve n value; The data integration step involves a laser scanner used to integrate test data from several test areas into complete structural surface data, including: The steps for setting benchmark points are as follows: Select at least three non-collinear benchmark points on the slope structure surface, record the spatial coordinates of each benchmark point and mark them, and the data of each survey area should contain the association information of at least one benchmark point. Data preprocessing sub-steps: Digitally process the surface profile curves obtained from the profile curve drawing step for each test area, extracting the curve coordinate information; simultaneously organize the average rebound values ​​R obtained from the structural surface rebound detection step for each test area. m JRC obtained from the roughness measurement procedure n This is associated with the spatial location of the corresponding survey area; Coordinate matching and stitching sub-step: Based on the reference point coordinates set in the reference point setting sub-step, the numerical contour curve coordinates and R values ​​of each survey area are matched. m Values ​​and JRC n The values ​​are transformed into a unified coordinate system, and the iterative nearest point algorithm is used to match the contour curve edges of adjacent survey areas, so that the contour curves of adjacent survey areas are seamlessly connected, forming a complete comprehensive map of the structural surface contour. Data integration and display sub-step: Combine the stitched complete structural surface contour map with the R of each survey area. m Value, JRC n Spatial correlation annotation is performed on the values, and the roughness coefficient and springback value distribution characteristics of different regions of the structural surface are displayed through color gradient or numerical annotation.

2. The method for testing the surface roughness coefficient of a slope structure according to claim 1, characterized in that, It also includes the following steps: The steps for calculating rock wall strength include: setting up a point load cell, selecting a predetermined number of rock blocks of the same geological type as the structural surface in the test area, conducting tests using the point load cell, and recording the point load strength index as I. s (50), Rock wall strength JCS n The calculation formula is JCS n =22.82·I s (50); Steps for calculating the basic friction angle of a structural surface, including the basic friction angle ψ of a slope structural surface. b The calculation formula is: ψ b =0.414R m +9.273; The steps for obtaining the normal stress on the structural surface involve arranging several sets of vibrating wire strain gauges on both sides of the structural surface, continuously monitoring at a preset frequency for a preset time, and collecting data. The average value of the monitored values ​​is then taken as σ. n ; The steps for calculating the shear strength of the structural plane are as follows: The shear strength τ is calculated according to the Barton-Bandis model, using the following formula: τ=σ n tan{JRC n •lg(JCS n / s n )+ψ b }, Among them, JCS n To evaluate the rock wall strength of the structural plane; JRC n To evaluate the surface roughness coefficient of the rock mass along the evaluation direction on the structural surface; ψ b To evaluate the basic friction angle of the structural surface; R m σ represents the average surface springback of the structural surface; n This represents the normal stress of the structural surface used for trial calculations.

3. The method for testing the surface roughness coefficient of a slope structure according to claim 1, characterized in that, The structural surface springback detection step further includes the following sub-steps: Measurement point distribution optimization sub-step: Based on the surface texture features of the structural surface, measurement points are arranged in the measurement area using equidistant grids or random sampling to ensure that the measurement points cover different texture areas of the measurement area; Outlier stratification removal sub-step: First, remove rebound values ​​that exceed the preset standard deviation range. Then, remove the maximum and minimum values ​​of a preset proportion from the remaining values. The remaining values ​​are used to calculate the average rebound value R. m .

4. The method for testing the surface roughness coefficient of a slope structure according to claim 1, characterized in that, The contour curve drawing step also includes sub-steps: Multi-directional scanning sub-step: Draw contour curves along the main direction and the perpendicular direction of the structural plane, and obtain at least three parallel curves in each direction; Curve smoothing sub-step: Digitally denoise the drawn contour curve to remove abnormal fluctuations caused by probe jumps and retain the real surface texture features; The electronic precision measurement sub-step digitizes and stores the surface profile curve obtained from the curve smoothing sub-step, and then uses an electronic roughness ruler adapted to a multi-range roughness ruler for measurement. When the deepest valley of the surface profile curve lies between two scale lines, the two scale lines are equally divided using an electronic roughness ruler. Based on the subdivided scale lines after the division, the precise JRC is calculated and read. n value.

5. The method for testing the surface roughness coefficient of a slope structure according to claim 4, characterized in that, The contour curve drawing step is equipped with a contour validity verification module, which is used to compare JRC curves in different directions within the same survey area. n Differences, when JRCs in different directions within the same survey area n When the difference between values ​​exceeds a preset threshold, the contour curve of the measurement area is re-acquired.

6. The method for testing the surface roughness coefficient of a slope structure according to claim 1, characterized in that, It also includes slope inspection and maintenance steps, which include sub-steps: The benchmark data storage sub-step is used to provide a comparison benchmark for continuous slope monitoring and maintenance. The integrated display sub-step shows the complete surface contour map of the structural surface after splicing, and the average rebound value R of each test area. m Roughness coefficient JRC n Basic friction angle ψ b The shear strength τ, along with the spatial coordinates of the test area and the test timestamp, is stored in the benchmark database; Slope zoning coding sub-steps: Based on the spatial distribution characteristics of the structural surfaces, the slope is zoned and coded, with each zone associated with at least three reference point coordinates and the initial test JRC. n The average value and the τ-average value are used as regional reference units for subsequent maintenance.

7. The method for testing the surface roughness coefficient of a slope structure according to claim 6, characterized in that, The slope inspection and maintenance steps are equipped with a subsequent inspection cycle control module to determine the time interval for subsequent inspections, including: Based on the initial shear strength, the slope area is divided into several levels. Areas with higher shear strength have longer testing intervals, while areas with lower shear strength have shorter testing intervals. When natural environmental changes occur or slopes show significant deformation, the detection interval for the corresponding area should be shortened, and supplementary detection of the slope should be carried out.

8. The method for testing the surface roughness coefficient of a slope structure according to claim 7, characterized in that, The slope inspection and maintenance steps also include: Periodic retesting sub-step: Perform rebound testing and roughness retesting on key areas according to a preset cycle, and update R. m JRC n and the value of τ; Model iteration sub-step: Based on the newly measured data, revise the shear strength calculation model and optimize the parameter weights under different geological conditions.

9. The method for testing the surface roughness coefficient of a slope structure according to claim 1, characterized in that, It also includes a dynamic environmental and structural correction step, which includes: Real-time monitoring of environmental parameters: Deploy humidity sensors, rainfall and rock strain gauges in the slope survey area to continuously collect data on slope environmental temperature, humidity, cumulative rainfall and micro-strain of structural surfaces, and establish a time series database of environmental parameters. The environmental correlation model construction sub-step analyzes environmental parameters and JRC using machine learning algorithms. n R m ψ b The dynamic correlation pattern, quantifying the JRC when the rainfall increases by a preset threshold. n The attenuation coefficient and the effect of temperature change on R m Correction factor; The dynamic correction sub-step for shear strength, based on the real-time environmental monitoring sub-step and the environmental correlation model construction sub-step, corrects the shear strength τ of the structural surface online. The formula is τ'=τ(1+α•ΔT+β•ΔR), where α is the temperature correction coefficient, β is the rainfall correction coefficient, ΔT is the difference between the measured temperature and the initial test temperature, and ΔR is the difference between the cumulative rainfall and the rainfall in the initial test period.

Images