Layered compaction and stability control construction method for high-filled (steep) embankment of expressway

Abstract

The invention discloses an expressway high-filled (steep) embankment layered compaction and stability control construction method, and relates to the technical field of geotechnical engineering.The expressway high-filled (steep) embankment layered compaction and stability control construction method comprises the steps that S1, intelligent filler recognition is conducted, a vehicle-mounted near-infrared spectrometer is combined with a laser particle analyzer, and a grading curve and the fine particle content are analyzed in real time; s2, dynamic optimization of compaction parameters, and matching of vibration frequency / rolling times based on filler characteristics; s3-S4, slope compaction compensation is carried out, a hydraulic compaction plate carries out self-adaptive angle adjustment, and a pressure sensor array triggers edge area hydraulic compensation; s5, monitoring a porosity cloud picture, inverting porosity distribution by a dielectric sensor, and supplementing pressure by an automatic navigation road roller when the porosity distribution exceeds a threshold value; and S6, strain abrupt change is blocked, and when the strain abrupt change rate monitored by the FBG optical fiber exceeds a set threshold value, nano silicate slurry is injected in a second level for reinforcement. According to the method, through intelligent sensing network chain construction, a man-machine collaborative decision-making mechanism and risk blocking capability, the problems of edge compaction defects, out-of-control deep deformation, low working procedure collaborative efficiency and the like of the high-filled abrupt slope embankment are solved.

Description

Technical Field

[0001] The invention relates to the technical field of geotechnical engineering, and in particular to a construction method for layered compaction and stability control of high-fill (steep) embankments of expressways. Background Art

[0002] As highways extend into mountainous and hilly areas, routes often need to cross complex terrain such as deep ditches and steep slopes. This requires connecting high and low roadbeds with high-fill embankments or steep-slope embankments to reduce the scale and cost of bridge and tunnel projects. High-fill (steep) embankments, due to their large fill height, steep base slope, heavy deadweight load, and complex stress distribution, are prone to uneven foundation settlement, embankment sliding along the base or within, and slope collapse, directly impacting highway operational safety. Mountainous areas experience large volumes of abandoned earthwork. Using excavated earthwork for high-fill embankment construction can reduce land occupation and environmental damage, but issues such as large variations in filler properties and difficulty in compaction must be addressed.

[0003] Traditional layered compaction techniques are prone to uneven compaction and loose slopes on steep slopes. Furthermore, the lack of dynamic monitoring makes it difficult to accurately monitor construction quality in real time. However, the development of numerical analysis, intelligent monitoring, and new compaction equipment has provided support for the refined design and construction of high-fill (steep) embankments.

[0004] There are still problems in the existing technology that need to be solved urgently: the filler properties do not match the compaction process. The fillers in mountainous areas are mostly crushed stone soil, blocky stone soil, weathered rock, etc., with large gradation differences and unstable fine particle content. Traditional compaction parameters (such as number of rolling times and moisture content) are difficult to universalize, which can easily lead to excessive internal porosity and insufficient strength. When filling steep embankments, the filler close to the slope is prone to insufficient edge compaction due to limited mechanical operating space, forming a sliding weak surface. Summary of the Invention

[0005] In order to solve the above technical problems, a construction method for layered compaction and stability control of high-fill (steep) roads on highways is provided. This technical solution solves the problem of mismatch between the above-mentioned filler characteristics and compaction technology.

[0006] In order to achieve the above objects, the technical solution adopted by the present invention is:

[0007] The construction method of layered compaction and stability control of high-fill (steep) embankments on expressways includes:

[0008] S1. Use a vehicle-mounted near-infrared spectrometer in conjunction with a laser particle size analyzer to detect the gradation distribution curve, peak crushed stone particle size, and fine particle content of each layer of filler in real time.

[0009] S2. Dynamically match the optimal compaction parameter combination based on filler identification results;

[0010] S3. Install a retractable hydraulic compaction plate on the side of the embankment slope, with its working surface forming a predetermined angle with the slope;

[0011] S4. Using a pressure sensor array, the compaction degree of the edge area is monitored in real time. When the compaction degree difference is greater than a predetermined ratio, the hydraulic compensation mechanism is activated to apply a predetermined additional pressure.

[0012] S5. Use a distributed dielectric constant sensor to measure the dielectric properties of the filler and inversely calculate the porosity distribution cloud map; when the local porosity is greater than a predetermined ratio, the location is automatically marked and the fixed-point pressure replenishment program is triggered;

[0013] S6. After each filling height is reached, FBG fiber grating sensors are implanted to monitor horizontal strain and settlement rate. If the strain mutation rate is greater than a predetermined threshold, construction is immediately stopped and nano-silicate slurry is injected for reinforcement.

[0014] Preferably, the S1 specifically includes:

[0015] The vehicle-mounted platform was modified to include a shock-resistant testing cabin behind the roller's cab, which houses a near-infrared spectrometer and a laser particle size analyzer. In-situ calibration was performed, with standard sand samples used to calibrate the instrument before each shift, establishing a localized filler database.

[0016] The real-time inspection process involves filler paving: after the bulldozer spreads the filler, the roller drives over the new layer at a predetermined speed; spectral scanning: a near-infrared light source is incident on the filler surface at a predetermined angle to collect the reflected spectrum; particle size analysis: a laser beam penetrates the filler surface and scatters, and a receiver captures the diffraction pattern; and data fusion: a weighted algorithm is used to integrate the spectral and laser data.

[0017] Extract gradation characteristic values ​​and generate gradation distribution curves based on the ASTM D6913 standard. Classify the fused data by particle size into large particle groups, coarse particle groups, and fine particle groups; and automatically fit continuous distribution curves.

[0018] Gravel particle size peak location, identification of maximum frequency particle size, calculation of fine particle content, and inversion of near-infrared characteristic bands.

[0019] Preferably, the S2 specifically includes:

[0020] Dynamic switching of excitation modes: when the fine particle content is less than a set ratio, the impact rolling mode is activated, using high-energy impact; when the fine particle content is greater than or equal to a set ratio, the vibration and static pressure modes are switched, using medium and low excitation forces; when the gradation unevenness coefficient is greater than a set threshold, variable frequency vibration is added, using high-frequency vibration and low-frequency vibration;

[0021] Control formula:

[0022]

[0023] Where F is the exciting force, C is the percentage of fine particle content;

[0024] Intelligent decision-making on the number of compaction passes. The basic number of compaction passes is based on the preset benchmark value of the filler type, and the dynamic correction factor is calculated:

[0025]

[0026] Where N is the corrected number of rolling passes, which is the final number of rolling passes determined after comprehensive consideration of various factors and is used to guide actual rolling construction to achieve the expected compaction effect; N0 is the benchmark number of passes, which is the initial number of rolling passes preset according to different filler types and provides a benchmark for subsequent corrections; d max is the maximum particle size of the filler, which affects the difficulty of rolling and the number of compaction passes required. The larger the maximum particle size, the more rolling passes may be required to achieve the compaction effect; C is the fine particle content;

[0027] Adaptive control of paving layer thickness: adjust the layer thickness according to the gradation characteristics. If the gradation is uniform, the layer thickness increases; if the gradation is poor, the layer thickness decreases; if the fine particle content suddenly exceeds the set ratio, a transition layer of a set thickness is inserted.

[0028] Collaboratively optimize travel speed, establish a speed-energy transfer model, and set dynamic speed limit rules.

[0029] Preferably, the S3 specifically includes:

[0030] The working surface inclination angle is controlled by an adjustable angle design, which allows the angle to be adjusted in real time through a hydraulic push rod to ensure that the working surface always has a preload angle with the normal direction of the slope.

[0031] The compacting tooth array has wedge-shaped alloy steel teeth arranged in a staggered pattern. The three-stage telescopic mechanism uses three sets of hydraulic cylinders in series to achieve telescopic travel, adapting to the needs of steep slope operations.

[0032] The hydraulic station outputs pressure, and the compaction reaction force is adjusted steplessly through a proportional valve. The integrated piezoelectric ceramic pressure sensor and MEMS gyroscope provide real-time feedback of contact pressure and posture data.

[0033] The roller uses GNSS positioning to determine the slope's position. When the center of the machine is a predetermined distance from the slope shoulder, the retractable mechanism automatically triggers the deployment of the compacting plate. Millimeter-wave radar monitors slope obstacles in real time, providing early warning and shutdown in the event of large rocks.

[0034] Layered compaction is implemented. In the initial compaction stage, pre-compaction is performed with contact pressure. In the main compaction stage, when the difference in compaction degree between the edge area and the middle area is greater than the predetermined ratio, the high-pressure compensation mode is activated. For inter-tooth maintenance, after each multiple rolling operations, the high-frequency micro-vibration mode is activated to automatically remove the embedded filler between the teeth.

[0035] Preferably, the S4 specifically includes:

[0036] Pressure sensor array deployment, a three-dimensional monitoring network architecture, with monitoring sections equidistantly arranged along the slope line, and three-dimensional piezoelectric ceramic sensors buried at different depths from the slope surface in each section. Sensor nodes are networked via a CAN bus, transmitting pressure data in real time at a predetermined sampling rate. Alloy protective baffles are installed on the surface. The compaction degree inversion algorithm is based on a pressure-density mapping model.

[0037] Compaction difference determination mechanism, spatial difference analysis, the edge area is the monitoring area within a predetermined range from the slope, difference calculation is performed, and graded response thresholds are used;

[0038] The working process of the compensation mechanism is that the sensor detects that the difference is greater than the set ratio, the control center generates a compensation instruction, the proportional valve adjusts the oil pressure, the hydraulic cylinder extends the compaction plate, applies additional pressure, and the sensor re-measures the compaction degree. If the difference is reduced to the set ratio, construction continues. Otherwise, the pressure is increased and compensation is performed again.

[0039] Preferably, the S5 specifically includes:

[0040] The sensor array architecture implants multi-frequency dielectric sensors along the cross section of the roadbed, with one layer vertically arranged for each predetermined fill thickness, forming a three-dimensional monitoring network. Adopting an industrial IoT architecture, each sensor node is connected to an edge gateway via an RS485 bus, which uploads data to a cloud platform at a 5Hz frequency.

[0041] Dielectric properties-porosity inversion model, based on Maxwell-Garnett mixed medium theory:

[0042]

[0043] Where, ε eff ε is the measured dielectric constant, which is the dielectric constant of the filler as a whole obtained by measurement. It reflects the polarization characteristics of the filler in the electromagnetic field and can be used to characterize the dielectric properties of the filler. m ε is the dielectric value of the filler matrix, which is the dielectric constant of the filler matrix material. The calibration range is 2.8-4.2. It represents the dielectric properties of the filler matrix itself and provides a basic parameter for calculating porosity. f is the dielectric constant of the pore gas, fixed at 1.0, representing the dielectric constant of the gas in the filler pores. In the inversion model, it is a known fixed value and is used to influence the measured dielectric constant together with the dielectric constant of the filler matrix. φ is the target porosity, which refers to the ratio of the pore volume to the total volume of the filler to be solved. It is a key parameter to be determined in the dielectric property-porosity inversion model, and the porosity is inferred from the measured dielectric constant.

[0044] Eliminate moisture content interference and introduce dual-frequency measurement compensation:

[0045] φ true =φ 1MHz -0.38(φ 100MHz -φ 1MHz )

[0046] Where, φ true is the true porosity, which means the more accurate porosity value after eliminating the interference of water content; φ 1MHz The porosity is measured at a frequency of 1 MHz. At low frequencies, the dielectric properties of the pore gas dominate, and the porosity measured at this time is less affected by the water content. 100MHz This is the porosity measured at a frequency of 100 MHz. High frequencies are sensitive to moisture, and the porosity measured at this time will be interfered by the moisture content.

[0047] The response mechanism for porosity exceeding the standard is adopted, and the abnormality judgment standard is based on the porosity range to determine the status; Beidou grid positioning uses the Beidou-3 grid code to mark abnormal points; the pressure replenishment program is cloud platform detection, which generates pressure replenishment coordinates. The roller navigates to the target point, activates the high-frequency vibratory hammer, applies the established standard pressure for the established time, and re-measures the dielectric value to see whether the porosity is less than or equal to the established ratio. If so, construction continues; if not, secondary pressure replenishment and grouting preparation are carried out.

[0048] Preferably, the S6 specifically includes:

[0049] The FBG sensor network is deployed in a three-dimensional monitoring architecture, with a layer of sensors implanted for each given fill thickness, forming a grid pattern. The sensing units are configured with horizontal strain FBGs and settlement rate FBGs. The protection design includes stainless steel armored optical cables and ceramic bases to prevent shearing. A reference sensor group is set up on the stable bedrock outside the embankment.

[0050] Strain mutation monitoring, based on a mutation rate determination model and a hierarchical response strategy, collects strain data in real time to determine whether the mutation rate is greater than a predetermined threshold. If so, a level 3 alarm is triggered, construction is stopped, personnel are evacuated, and the slurry injection system is activated; if not, monitoring continues.

[0051] Nanosilicate slurry reinforcement: nanosilicate, PVA fiber, active magnesium oxide and ultrafine silica powder are compounded in a predetermined ratio; directional grouting: positioning drilling is performed based on the coordinates of the FBG strain peak point; the grouting section is divided into shallow and deep layers, and grouting is carried out according to the predetermined pressure and slurry volume.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] The present invention proposes to achieve full-section real-time perspective of filler properties, compaction status, pore distribution, and strain evolution through multi-source sensor fusion including near-infrared spectroscopy, laser particle size analyzer, dielectric sensor, and FBG optical fiber, upgrading quality monitoring from "discrete point sampling" to "full-section continuous scanning" to eliminate the risk of missed detection of hidden defects.

[0054] The introduction of a hydraulic compensation mechanism and intelligent judgment mechanism uses a pressure sensor array to dynamically capture compaction differences in edge areas, automatically triggering targeted pressure compensation. Combined with an adjustable angle compaction plate, it precisely matches the slope geometry to achieve targeted energy delivery. This forms a closed loop of "perception-decision-compensation," eliminating the potential slippage risks caused by uneven slope compaction at the root.

[0055] A real-time early warning model for sudden changes in strain rate was established, using FBG optical fibers to capture abnormal strain fluctuations within seconds. A graded response threshold was set to coordinate the start and stop of construction equipment. The rapid setting properties of nano-grout were integrated to achieve automated "locating-grouting-verification" procedures. This shifted landslide prevention and control from post-event emergency response to in-the-moment, second-by-second blocking during the event, significantly improving the inherent safety of the project. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a flow chart of the construction method for layered compaction and stability control of high-fill (steep) embankments on expressways. DETAILED DESCRIPTION

[0057] The following description is intended to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are merely examples, and those skilled in the art may conceive of other obvious variations.

[0058] Reference Figure 1 As shown in the figure, the construction method of layered compaction and stability control of high-fill (steep) embankment of expressway includes:

[0059] S1. Use a vehicle-mounted near-infrared spectrometer in conjunction with a laser particle size analyzer to detect the gradation distribution curve, peak crushed stone particle size, and fine particle content of each layer of filler in real time.

[0060] S2. Dynamically match the optimal compaction parameter combination based on filler identification results;

[0061] S3. Install a retractable hydraulic compaction plate on the side of the embankment slope, with its working surface forming a predetermined angle with the slope;

[0062] S4. Using a pressure sensor array, the compaction degree of the edge area is monitored in real time. When the compaction degree difference is greater than a predetermined ratio, the hydraulic compensation mechanism is activated to apply a predetermined additional pressure.

[0063] S5. Use a distributed dielectric constant sensor to measure the dielectric properties of the filler and inversely calculate the porosity distribution cloud map; when the local porosity is greater than a predetermined ratio, the location is automatically marked and the fixed-point pressure replenishment program is triggered;

[0064] S6. After each filling height is reached, FBG fiber grating sensors are implanted to monitor horizontal strain and settlement rate. If the strain mutation rate is greater than a predetermined threshold, construction is immediately stopped and nano-silicate slurry is injected for reinforcement.

[0065] It should be noted that S1 provides basic filler properties → S2 generates compaction strategy → S3 / S4 performs edge compaction → S5 verifies pore uniformity → S6 monitors structural response → abnormal data is fed back to S2 to optimize parameters.

[0066] Core parameter linkage relationship:

[0067] Monitoring items Control Target Related technology modules Fine particle content (S1) Excitation mode (S2) Frequency conversion vibration anti-scattering Edge compaction (S4) Hydraulic compensation pressure Pressure-displacement dual control (S3) Porosity (S5) Compressed energy High frequency vibration hammer positioning (S5) Horizontal strain (S6) Slurry injection volume Nanosilicate ratio (S6)

[0068] When S5 detects a sudden change in porosity (caused by uneven grading), S2 automatically increases the number of compaction passes, while S3 increases the slope pressure compensation to avoid S6 triggering a strain alarm.

[0069] Three levels of defense: Primary, S2 dynamic parameters + S3 slope compaction to prevent uneven compaction; Intermediate, S5 porosity monitoring + S4 compensation to eliminate local defects; Advanced, S6 strain warning + S6 grouting reinforcement to block structural instability.

[0070] Said S1 specifically includes:

[0071] The vehicle-mounted platform was modified to include a shock-resistant testing cabin behind the roller's cab, which houses a near-infrared spectrometer and a laser particle size analyzer. In-situ calibration was performed, with standard sand samples used to calibrate the instrument before each shift, establishing a localized filler database.

[0072] The real-time inspection process involves filler paving: after the bulldozer spreads the filler, the roller drives over the new layer at a predetermined speed; spectral scanning: a near-infrared light source is incident on the filler surface at a predetermined angle to collect the reflected spectrum; particle size analysis: a laser beam penetrates the filler surface and scatters, and a receiver captures the diffraction pattern; and data fusion: a weighted algorithm is used to integrate the spectral and laser data.

[0073] Extract gradation characteristic values ​​and generate gradation distribution curves based on the ASTM D6913 standard. Classify the fused data by particle size into large particle groups, coarse particle groups, and fine particle groups; and automatically fit continuous distribution curves.

[0074] Gravel particle size peak location, identification of maximum frequency particle size, calculation of fine particle content, and inversion of near-infrared characteristic bands.

[0075] It should be noted that the seismic test cabin has been modified to use a three-point suspended shock-absorbing base (damping coefficient 0.25) to isolate the roller vibration; a double-layer insulation cabin (outer layer 304 stainless steel, inner layer ceramic fiber), with a temperature control of 25±3℃ to ensure instrument accuracy; the instrument integration includes a near-infrared spectrometer (wavelength range 900-2500nm, resolution ±2nm) and a laser particle size analyzer (range 0.01-3000μm), with the optical paths of the two instruments coaxially designed;

[0076] In-situ calibration: The calibration process includes: standard sand sample implantation (ASTM C77820-30 silica sand) to establish a local reflectance benchmark; background noise elimination, collecting dark current data 100 times per second to compensate for ambient light interference; a filler database, storing >200 regional filler characteristics to match local geological conditions.

[0077] Real-time gradation detection process:

[0078] Filler paving → Inspection chamber: Bulldozer paving thickness 30cm±5cm;

[0079] Detection chamber → Near-infrared spectrometer: emits near-infrared light with an incident angle of 45°

[0080] Near-infrared spectrometer → data processing unit: collects reflectance spectra (1400-1900nm characteristic band)

[0081] Detection chamber → Laser particle size analyzer: emits 650nm laser beam

[0082] Laser particle size analyzer → data processing unit: obtain diffraction pattern (scattering angle 0.1°-40°)

[0083] Data processing unit → gradation output: weighted fusion algorithm W = 0.6*spectrum+0.4*laser

[0084] Data fusion formula:

[0085] Where D 融合 is the fused data, which is the comprehensive result obtained by calculating the near-infrared absorbance and laser diffraction intensity distribution, and is used to characterize certain characteristics of the sample; α is the near-infrared weight, which is 0.6, indicating the proportion of near-infrared absorbance in data fusion, emphasizing the importance of near-infrared absorbance in comprehensive evaluation; I NIR Characteristic band absorbance is the absorbance value of the sample in the near-infrared characteristic band, which reflects the degree of absorption of the sample to the light in a specific near-infrared band and is related to the physical and chemical properties of the sample (such as moisture content, composition, etc.); I min The minimum absorbance value is used as a reference benchmark to normalize the absorbance so that different absorbance values ​​can be compared and integrated on a relative scale; I maxis the maximum absorbance value, which is also used to normalize the absorbance. min Together they determine the relative range of absorbance; β is the laser weight, which is set to 0.4 and represents the proportion of laser diffraction intensity distribution in data fusion, reflecting the role of laser diffraction data in comprehensive evaluation; S LDA is the laser diffraction intensity distribution, which is the light intensity distribution data of the sample after laser diffraction, and contains information such as the sample particle size distribution. The particle size characteristics of the sample can be inferred by analyzing the light intensity distribution; ∑S is the sum of all light intensity distributions, which is used to normalize the laser diffraction intensity distribution so that different light intensity distributions are comparable and unified when fused.

[0086] Grading characteristic value extraction method, grading standard:

[0087] Particle size group Particle size range (mm) Detection methods focus Giant particle group 60-200 Laser diffraction (accuracy ±3%) Coarse-grained group 2-60 Near infrared + laser fusion Fine-grained group 0.075-2 Near-infrared spectroscopy (dominant)

[0088] Curve fitting: Generate continuous grading curve based on cubic spline interpolation method, correlation coefficient R 2 ≥0.97

[0089] The peak location of the gravel particle size is the particle size corresponding to the zero first-order derivative of the laser diffraction pattern.

[0090] The S2 specifically includes:

[0091] Dynamic switching of excitation modes: when the fine particle content is less than a set ratio, the impact rolling mode is activated, using high-energy impact; when the fine particle content is greater than or equal to a set ratio, the vibration and static pressure modes are switched, using medium and low excitation forces; when the gradation unevenness coefficient is greater than a set threshold, variable frequency vibration is added, using high-frequency vibration and low-frequency vibration;

[0092] Control formula:

[0093]

[0094] Where F is the exciting force, C is the percentage of fine particle content;

[0095] Intelligent decision-making on the number of compaction passes. The basic number of compaction passes is based on the preset benchmark value of the filler type, and the dynamic correction factor is calculated:

[0096]

[0097] Where N is the corrected number of rolling passes, which is the final number of rolling passes determined after comprehensive consideration of various factors and is used to guide actual rolling construction to achieve the expected compaction effect; N0 is the benchmark number of passes, which is the initial number of rolling passes preset according to different filler types and provides a benchmark for subsequent corrections; d maxis the maximum particle size of the filler, which affects the difficulty of rolling and the number of compaction passes required. The larger the maximum particle size, the more rolling passes may be required to achieve the compaction effect; C is the fine particle content;

[0098] Adaptive control of paving layer thickness: adjust the layer thickness according to the gradation characteristics. If the gradation is uniform, the layer thickness increases; if the gradation is poor, the layer thickness decreases; if the fine particle content suddenly exceeds the set ratio, a transition layer of a set thickness is inserted.

[0099] Collaboratively optimize travel speed, establish a speed-energy transfer model, and set dynamic speed limit rules.

[0100] It should be noted that the setting rules for the benchmark number N0 are:

[0101] Filler type N0N0 (times) Applicable working case gravel soil 6 Maximum particle size 60mm, Cu=8 Sandstone gravel 8 Angular particles account for more than 65% Weathered shale material 4 Fine particle content 28%, moisture content 9%

[0102] Adaptive control of paving layer thickness, layer thickness adjustment strategy:

[0103] Grading characteristics Layer thickness control range Theoretical basis <![CDATA[Uniformity coefficient C u ≤ 5]]> +20％ Reduce interlayer interface effects <![CDATA[C u ≥10 and C c =1-3]]> -15％ Prevent coarse particles from concentrating to form overhead structures Fine particle content mutation ΔC>8% / layer Insert a 10cm transition layer Eliminate shear weak points

[0104] For the transition layer design, the material requirement is that the gradation is between the upper and lower layers, and Dmax≤40mm; the compaction standard is to increase the compaction degree by 5% compared with the upper and lower layers.

[0105] The S3 specifically includes:

[0106] The working surface inclination angle is controlled by an adjustable angle design, which allows the angle to be adjusted in real time through a hydraulic push rod to ensure that the working surface always has a preload angle with the normal direction of the slope.

[0107] The compacting tooth array has wedge-shaped alloy steel teeth arranged in a staggered pattern. The three-stage telescopic mechanism uses three sets of hydraulic cylinders in series to achieve telescopic travel, adapting to the needs of steep slope operations.

[0108] The hydraulic station outputs pressure, and the compaction reaction force is adjusted steplessly through a proportional valve. The integrated piezoelectric ceramic pressure sensor and MEMS gyroscope provide real-time feedback of contact pressure and posture data.

[0109] The roller uses GNSS positioning to determine the slope's position. When the center of the machine is a predetermined distance from the slope shoulder, the retractable mechanism automatically triggers the deployment of the compacting plate. Millimeter-wave radar monitors slope obstacles in real time, providing early warning and shutdown in the event of large rocks.

[0110] Layered compaction is implemented. In the initial compaction stage, pre-compaction is performed with contact pressure. In the main compaction stage, when the difference in compaction degree between the edge area and the middle area is greater than the predetermined ratio, the high-pressure compensation mode is activated. For inter-tooth maintenance, after each multiple rolling operations, the high-frequency micro-vibration mode is activated to automatically remove the embedded filler between the teeth.

[0111] It should be noted that the three-stage telescopic compaction mechanism is as follows: base hydraulic cylinder → stroke 200mm → intermediate hydraulic cylinder → stroke 250mm → final hydraulic cylinder → stroke 250mm → compaction plate;

[0112] The total extension length is 700mm (adapting to the slope ratio of 1:0.75-1:1.25), the synchronous control accuracy, the displacement difference of each cylinder is ≤±1.5mm (to avoid eccentric loading).

[0113] The wedge-shaped alloy steel tooth array is made of WC-10Co cemented carbide with a Rockwell hardness of HRA ≥ 89. The tooth dimensions are 40mm in base width and 60mm in height, capable of crushing fillers with a particle size of ≤ 80mm. The tooth arrangement is a 15° staggered diamond array to eliminate compaction blind spots. The surface treatment is a CrN nano-coating, reducing the friction coefficient to 0.15.

[0114] Pressure-attitude dual feedback system, piezoelectric ceramic sensor: range 0-5MPa, linearity 0.5%, real-time monitoring of contact pressure at the implant tooth root; MEMS attitude: three-axis accelerometer + gyroscope, vibration spectrum analysis to identify filler overhead.

[0115] Three-stage pressure control strategy: initial pressure: 0.5-0.8MPa, to eliminate initial gaps in the ply (speed ≤ 1.5km / h); main pressure: 1.2-1.8MPa; final pressure: 0.3-0.5MPa, to eliminate wheel marks (vibration frequency reduced to 15Hz).

[0116] Inter-tooth maintenance technology, high-frequency micro-vibration blockage removal, excitation parameters are 200Hz / 0.3mm amplitude, energy consumption is <0.5kWh / time, blockage identification logic: pressure sensor fluctuation >15% → trigger blockage removal program; accelerometer spectrum has a peak at 100-150Hz → judged as filler jam.

[0117] The S4 specifically includes:

[0118] Pressure sensor array deployment, a three-dimensional monitoring network architecture, with monitoring sections equidistantly arranged along the slope line, and three-dimensional piezoelectric ceramic sensors buried at different depths from the slope surface in each section. Sensor nodes are networked via a CAN bus, transmitting pressure data in real time at a predetermined sampling rate. Alloy protective baffles are installed on the surface. The compaction degree inversion algorithm is based on a pressure-density mapping model.

[0119] Compaction difference determination mechanism, spatial difference analysis, the edge area is the monitoring area within a predetermined range from the slope, difference calculation is performed, and graded response thresholds are used;

[0120] The working process of the compensation mechanism is that the sensor detects that the difference is greater than the set ratio, the control center generates a compensation instruction, the proportional valve adjusts the oil pressure, the hydraulic cylinder extends the compaction plate, applies additional pressure, and the sensor re-measures the compaction degree. If the difference is reduced to the set ratio, construction continues. Otherwise, the pressure is increased and compensation is performed again.

[0121] It should be noted that the sensor array deployment:

[0122] Layout section spacing: 10m (increased to 5m on steep slopes), covering the full slope length fluctuation characteristics;

[0123] Single-section sensor depth: 0.3m (shallow layer) / 1.2m (middle layer) / 2.0m (deep layer), capturing compaction transmission attenuation in layers;

[0124] Three-axis piezoelectric ceramic sensor: range 0-5MPa, temperature drift ±0.1% / ℃, simultaneous monitoring of normal and tangential stress;

[0125] Protective partition: tungsten carbide alloy (hardness HRA92), thickness 8mm, resistant to 300kN impact load.

[0126] Data transmission system, sensor node → CAN bus hub → edge computing gateway → control center; sampling rate is 500Hz in the main compaction phase and 10Hz in the intermittent phase; anti-interference design is twisted pair shield + RS485 relay (transmission error rate <10 -6 ).

[0127] Compaction difference (ΔK) graded response:

[0128] Difference interval Response Level Disposal measures / ΔK ≤5％ 5％< ΔK ≤10％ 10％< ΔK ≤15％ / ΔK >15％

[0129] Compensation agency workflow:

[0130] Sensor array → Control center: Real-time transmission of ΔK data → Decision module: Generates compensation command if ΔK>10% → Proportional valve: Outputs oil pressure PID adjustment signal → Hydraulic cylinder: Drives compaction plate to extend → Compaction plate: Apply additional pressure (0.5-2.0MPa) → Soil layer: Targeted pressure compensation;

[0131] Sensor array → Control center: re-measure compaction degree; Control center → determine whether ΔK is ≤5%. If so, continue construction; if not, perform secondary compensation (pressure +0.3MPa).

[0132] The S5 specifically includes:

[0133] The sensor array architecture implants multi-frequency dielectric sensors along the cross section of the roadbed, with one layer vertically arranged for each predetermined fill thickness, forming a three-dimensional monitoring network. Adopting an industrial IoT architecture, each sensor node is connected to an edge gateway via an RS485 bus, which uploads data to a cloud platform at a 5Hz frequency.

[0134] Dielectric properties-porosity inversion model, based on Maxwell-Garnett mixed medium theory:

[0135]

[0136] Where, ε eff ε is the measured dielectric constant, which is the dielectric constant of the filler as a whole obtained by measurement. It reflects the polarization characteristics of the filler in the electromagnetic field and can be used to characterize the dielectric properties of the filler. m ε is the dielectric value of the filler matrix, which is the dielectric constant of the filler matrix material. The calibration range is 2.8-4.2. It represents the dielectric properties of the filler matrix itself and provides a basic parameter for calculating porosity. f is the dielectric constant of the pore gas, fixed at 1.0, representing the dielectric constant of the gas in the filler pores. In the inversion model, it is a known fixed value and is used to influence the measured dielectric constant together with the dielectric constant of the filler matrix. φ is the target porosity, which refers to the ratio of the pore volume to the total volume of the filler to be solved. It is a key parameter to be determined in the dielectric property-porosity inversion model, and the porosity is inferred from the measured dielectric constant.

[0137] Eliminate moisture content interference and introduce dual-frequency measurement compensation:

[0138] φ true =φ 1MHz -0.38(φ 100MHz -φ 1MHz )

[0139] Where, φ true is the true porosity, which means the more accurate porosity value after eliminating the interference of water content; φ 1MHz is the porosity measured at a frequency of 1 MHz. At low frequencies, the dielectric properties of the pore gas dominate, and the porosity measured at this time is less affected by the water content; φ 100MHz This is the porosity measured at a frequency of 100 MHz. High frequencies are sensitive to moisture, and the porosity measured at this time will be interfered by the moisture content.

[0140] The response mechanism for porosity exceeding the standard is adopted, and the abnormality judgment standard is based on the porosity range to determine the status; Beidou grid positioning uses the Beidou-3 grid code to mark abnormal points; the pressure replenishment program is cloud platform detection, which generates pressure replenishment coordinates. The roller navigates to the target point, activates the high-frequency vibratory hammer, applies the established standard pressure for the established time, and re-measures the dielectric value to see whether the porosity is less than or equal to the established ratio. If so, construction continues; if not, secondary pressure replenishment and grouting preparation are carried out.

[0141] It should be noted that the sensor array is deployed with a horizontal spacing of 5m (encrypted to 2.5m in the center of the roadbed), covering the porosity distribution of the entire cross-section; the vertical layer spacing is one layer every 0.6m of fill, matching the compaction layer thickness; multi-frequency dielectric sensors, dual-frequency measurement (1MHz / 100MHz), synchronously obtain matrix and moisture content information; the protection level is IP68, the compressive strength is ≥50MPa, and it can resist rolling impact.

[0142] Dual-frequency measurement principle: 1MHz low frequency: electromagnetic waves have a large penetration depth (about 2m), mainly reflecting the distribution of pore gas; 100MHz high frequency: sensitive to the polarization of water molecules (dielectric constant ≈ 80), capturing moisture content disturbances.

[0143] Response to porosity exceeding the standard, abnormal judgment standard: if the porosity range is φ≤15%, the status is judged to be excellent, and normal construction is carried out; if the porosity range is 15%<φ≤18%, the status is judged to be warning, and the monitoring frequency is increased to 10Hz; if the porosity range is 18%<φ≤22%, the status is judged to exceed the standard, and Beidou positioning pressure replenishment is triggered; if the porosity range is φ>22%, the status is judged to be a serious defect, and grouting preparation + excavation review is carried out.

[0144] The workflow of the compaction procedure is as follows: cloud platform → roller: send compaction coordinates → navigation system: automatic path planning; roller → compaction mechanism: locate to the target grid → high-frequency vibratory hammer: activate (35Hz / 1.5MPa); compaction mechanism → sensor: remeasure dielectric value → cloud platform: upload new porosity; if the new porosity is ≤18%, then cloud platform → roller: continue construction; otherwise, cloud platform → grouting unit: start the preparatory procedure.

[0145] The S6 specifically includes:

[0146] The FBG sensor network is deployed in a three-dimensional monitoring architecture, with a layer of sensors implanted for each given fill thickness, forming a grid pattern. The sensing units are configured with horizontal strain FBGs and settlement rate FBGs. The protection design includes stainless steel armored optical cables and ceramic bases to prevent shearing. A reference sensor group is set up on the stable bedrock outside the embankment.

[0147] Strain mutation monitoring, based on a mutation rate determination model and a hierarchical response strategy, collects strain data in real time to determine whether the mutation rate is greater than a predetermined threshold. If so, a level 3 alarm is triggered, construction is stopped, personnel are evacuated, and the slurry injection system is activated; if not, monitoring continues.

[0148] Nanosilicate slurry reinforcement: nanosilicate, PVA fiber, active magnesium oxide and ultrafine silica powder are compounded in a predetermined ratio; directional grouting: positioning drilling is performed based on the coordinates of the FBG strain peak point; the grouting section is divided into shallow and deep layers, and grouting is carried out according to the predetermined pressure and slurry volume.

[0149] It should be noted that the optical fiber sensor network is laid out in three dimensions:

[0150] The elevation layering interval is one layer per 1.0m of fill, matching the compaction settlement sensitive areas; the density of the T-shaped grid is 5m×5m (the core area is densely packed to 3m×3m), covering the deformation transmission path;

[0151] The sensor type is a dual-function FBG array. The horizontal strain FBG has a range of ±5000με and an accuracy of 1με; the sedimentation rate FBG has a resolution of 0.01mm / min.

[0152] The protection system is a 316L stainless steel armored optical cable (compression resistance ≥ 80MPa) and a zirconia ceramic base (shear resistance 150kN) to resist rolling impact and soil shear;

[0153] The benchmark group is set up at a stable bedrock ≥50m away from the embankment, with a burial depth of 3m, to eliminate temperature / vibration interference (drift <0.5%).

[0154] Tiered response strategy:

[0155] Level 1 warning: The mutation rate threshold is 50με / min, the sampling rate is increased to 100Hz, and the surrounding sensor review is started;

[0156] Level 2 alarm: The mutation rate threshold is 100με / min, with sound and light alarms, and construction speed is reduced by 50%;

[0157] Level 3 emergency: The mutation rate threshold is 200με / min, immediately shut down and evacuate, and activate the grouting system preparation.

[0158] Nanocomposite slurry compounding solution:

[0159] Nanosilicate: 60%, fills micro cracks and generates CSH gel, particle size 50nm, specific surface area ≥180m 2 / g;

[0160] PVA fiber: 1.5%, three-dimensional network crack resistance (bridging effect), length 12mm, tensile strength 1.2GPa;

[0161] Active magnesium oxide: 8%, compensates for shrinkage, improves interface bonding, activity index ≥85%;

[0162] Ultrafine silica fume: 30.5%, promotes volcanic ash reaction and increases density, D50 = 5μm, SiO2 content > 92%;

[0163] The initial setting time is 25min; the compressive strength at 7d is 42MPa; the permeability coefficient is 10 -9 cm / s.

[0164] In summary, the advantages of the present invention are:

[0165] Breaking through the time and space limitations of traditional manual sampling detection, the system uses near-infrared-laser particle size analysis to analyze filler grading in real time, a dielectric sensing network to scan the porosity distribution across the board, and FBG optical fiber to capture deep strain in seconds, thereby building a full-dimensional perspective capability covering filler characteristics, compaction status, and structural response, completely eliminating blind spots for hidden defects.

[0166] Decision instructions are dynamically generated based on sensor data, with compaction parameters automatically matching filler lithology and adaptively adjusting vibration frequency and number of passes. A hydraulic compensation mechanism targets edge compaction discrepancies, while an adjustable-angle compaction plate overcomes slope compaction bottlenecks. Porosity cloud maps guide and compensate for compaction, enabling centimeter-level defect location and treatment. This creates a closed loop of "perception-analysis-execution," eradicating biases inherent in human experience.

[0167] A graded response mechanism for sudden strain rate changes has been established, with FBG optical fibers capturing abnormal strain rates in real time. This system also enables the coordinated emergency stop of construction equipment and the evacuation of personnel. Furthermore, nano-slurry directional rapid-setting reinforcement ensures the immediate prevention of dangerous situations. This upgrades landslide prevention and control from passive rescue to active immunization.

[0168] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions merely illustrate the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A construction method for layered compaction and stability control of high-fill (steep) embankments on expressways, characterized in that: Including: S1. By using a vehicle-mounted near-infrared spectrometer in combination with a laser particle size analyzer, the grading distribution curve, the peak gravel particle size, and the fine particle content of each layer of filler are detected in real time; S2. Based on the filler identification results, the optimal compaction parameter combination is dynamically matched; S3. A telescopic hydraulic compaction plate is installed on the side of the embankment slope, and its working surface forms a certain angle with the slope; S4. Through a pressure sensor array, the compaction degree of the edge area is monitored in real time. When the compaction degree difference is greater than a certain ratio, the hydraulic compensation mechanism is activated to apply a certain additional pressure; S5. The dielectric properties of the filler are measured by using a distributed dielectric constant sensor to inversely calculate the porosity distribution cloud map; when the local porosity is greater than a certain ratio, the position is automatically marked and the fixed-point compaction supplement program is triggered; S6. After filling a certain height, FBG fiber Bragg grating sensors are implanted to monitor the horizontal strain and the settlement rate; if the strain mutation rate is greater than a certain threshold, the construction is immediately stopped and nano-silicate slurry is injected for reinforcement.

2. The highway high-fill (steep) embankment layered compaction and stability control construction method according to claim 1 is characterized in that: The specific content of S6 includes: FBG sensor network layout, three-dimensional monitoring structure. For elevation stratification, a layer of sensors is implanted every time a certain thickness is filled, forming a grid in the shape of a Chinese character "tian"; sensing unit, setting horizontal strain FBG and settlement rate FBG, and the protection design includes a stainless steel armored optical cable and a ceramic base for anti-shearing; reference group setting, setting a reference sensor group at a stable bedrock outside the embankment.

3. The highway high-fill (steep) embankment layered compaction and stability control construction method according to claim 2 is characterized in that: The specific content of S6 includes: Strain mutation monitoring, based on a mutation rate determination model, adopting a hierarchical response strategy, collecting strain data in real time, determining whether the mutation rate is greater than a certain threshold. If so, a three-level alarm is triggered, the construction is stopped and the personnel are evacuated, and the slurry injection system is started; if not, continue to monitor; Nano-silicate slurry reinforcement, compounding nano-silicate, PVA fiber, active magnesium oxide, and ultrafine silica powder in a certain ratio; directional grouting, based on the coordinates of the FBG strain peak point, positioning drilling is carried out; the grouting section is divided into shallow layer and deep layer, and grouting is carried out according to a certain pressure and slurry volume.

4. The highway high-fill (steep) embankment layered compaction and stability control construction method according to claim 3 is characterized in that: The specific content of S5 includes: Sensor array architecture, implanting multi-frequency dielectric sensors along the roadbed cross-section, arranging a layer every time a certain thickness is filled in the vertical direction to form a three-dimensional monitoring network; adopting an industrial Internet of Things architecture, each sensor node is connected to the edge gateway through an RS485 bus, and the gateway uploads data to the cloud platform at a frequency of 5Hz; Dielectric property-porosity inversion model, based on the Maxwell-Garnett mixed medium theory: Where, ε eff ε is the measured dielectric constant, which is the dielectric constant of the filler as a whole obtained by measurement. It reflects the polarization characteristics of the filler in the electromagnetic field and can be used to characterize the dielectric properties of the filler. m ε is the dielectric value of the filler matrix, which is the dielectric constant of the filler matrix material. The calibration range is 2.8-4.

2. It represents the dielectric properties of the filler matrix itself and provides a basic parameter for calculating porosity. f is the dielectric constant of the pore gas, fixed at 1.0, representing the dielectric constant of the gas in the filler pores. In the inversion model, it is a known fixed value and is used to influence the measured dielectric constant together with the dielectric constant of the filler matrix. φ is the target porosity, which refers to the ratio of the pore volume to the total volume of the filler to be solved. It is a key parameter to be determined in the dielectric property-porosity inversion model, and the porosity is inferred from the measured dielectric constant. Elimination of moisture content interference, introducing dual-frequency measurement compensation: f true =φ 1MHz -0.38(φ 100MHz -f 1MHz ) Where, φ true is the true porosity, which means the more accurate porosity value after eliminating the interference of water content; φ 1MHz The porosity is measured at a frequency of 1 MHz. At low frequencies, the dielectric properties of the pore gas dominate, and the porosity measured at this time is less affected by the water content. 100MHz This is the porosity measured at a frequency of 100 MHz. High frequencies are sensitive to moisture, and the porosity measured at this time will be interfered by the moisture content.

5. The highway high-fill (steep) embankment layered compaction and stability control construction method according to claim 4 is characterized in that: The specific content of S5 includes: Porosity exceeding standard response mechanism, the abnormal determination standard is to determine the state according to the porosity range; Beidou grid positioning, using the Beidou No. 3 grid code to mark the abnormal points; the compaction supplement program is detected by the cloud platform, generating the compaction coordinates, the road roller navigates to the target point, activates the high-frequency vibration hammer, applies a certain standard pressure and lasts for a certain time,复测 the dielectric value, whether the porosity is less than or equal to a certain ratio. If so, continue the construction. If not, carry out secondary compaction and grouting preparation.

6. The highway high-fill (steep) embankment layered compaction and stability control construction method according to claim 5 is characterized in that: The specific content of S2 includes: Dynamic switching of excitation modes: when the fine particle content is less than a set ratio, the impact rolling mode is activated, using high-energy impact; when the fine particle content is greater than or equal to a set ratio, the vibration and static pressure modes are switched, using medium and low excitation forces; when the gradation unevenness coefficient is greater than a set threshold, variable frequency vibration is added, using high-frequency vibration and low-frequency vibration; Control formula: Where F is the exciting force, C is the percentage of fine particle content; Intelligent decision-making on the number of compaction passes. The basic number of compaction passes is based on the preset benchmark value of the filler type, and the dynamic correction factor is calculated: Where N is the corrected number of rolling passes, which is the final number of rolling passes determined after comprehensive consideration of various factors and is used to guide actual rolling construction to achieve the expected compaction effect; N0 is the benchmark number of passes, which is the initial number of rolling passes preset according to different filler types and provides a benchmark for subsequent corrections; d max is the maximum particle size of the filler, which affects the difficulty of rolling and the required number of compaction passes. The larger the maximum particle size, the more rolling passes may be required to achieve the compaction effect; C is the fine particle content.

7. The method for layered compaction and stability control of highway high-fill (steep) embankment according to claim 6, characterized in that: The S2 specifically includes: Adaptive control of paving layer thickness: adjust the layer thickness according to the gradation characteristics. If the gradation is uniform, the layer thickness increases; if the gradation is poor, the layer thickness decreases; if the fine particle content suddenly exceeds the set ratio, a transition layer of a set thickness is inserted. Collaboratively optimize travel speed, establish a speed-energy transfer model, and set dynamic speed limit rules.

8. The highway high-fill (steep) embankment layered compaction and stability control construction method according to claim 7 is characterized in that: Said S1 specifically includes: The vehicle-mounted platform was modified to include a shock-resistant testing cabin behind the roller's cab, which houses a near-infrared spectrometer and a laser particle size analyzer. In-situ calibration was performed, with standard sand samples used to calibrate the instrument before each shift, establishing a localized filler database. The real-time inspection process involves filler paving: after the bulldozer spreads the filler, the roller drives over the new layer at a predetermined speed; spectral scanning: a near-infrared light source is incident on the filler surface at a predetermined angle to collect the reflected spectrum; particle size analysis: a laser beam penetrates the filler surface and scatters, and a receiver captures the diffraction pattern; and data fusion: a weighted algorithm is used to integrate the spectral and laser data. Extract gradation characteristic values ​​and generate gradation distribution curves based on the ASTM D6913 standard. Classify the fused data by particle size into large particle groups, coarse particle groups, and fine particle groups; and automatically fit continuous distribution curves. Gravel particle size peak location, identification of maximum frequency particle size, calculation of fine particle content, and inversion of near-infrared characteristic bands.

9. The highway high-fill (steep) embankment layered compaction and stability control construction method according to claim 8, characterized in that: The S4 specifically includes: Pressure sensor array deployment, a three-dimensional monitoring network architecture, with monitoring sections equidistantly arranged along the slope line, and three-dimensional piezoelectric ceramic sensors buried at different depths from the slope surface in each section. Sensor nodes are networked via a CAN bus, transmitting pressure data in real time at a predetermined sampling rate. Alloy protective baffles are installed on the surface. The compaction degree inversion algorithm is based on a pressure-density mapping model. Compaction difference determination mechanism, spatial difference analysis, the edge area is the monitoring area within a predetermined range from the slope, difference calculation is performed, and graded response thresholds are used; The working process of the compensation mechanism is that the sensor detects that the difference is greater than the set ratio, the control center generates a compensation instruction, the proportional valve adjusts the oil pressure, the hydraulic cylinder extends the compaction plate, applies additional pressure, and the sensor re-measures the compaction degree. If the difference is reduced to the set ratio, construction continues. Otherwise, the pressure is increased and compensation is performed again.

10. The highway high-fill (steep) embankment layered compaction and stability control construction method according to claim 9, characterized in that: The S3 specifically includes: The working surface inclination angle is controlled by an adjustable angle design, which allows the angle to be adjusted in real time through a hydraulic push rod to ensure that the working surface always has a preload angle with the normal direction of the slope. The compacting tooth array has wedge-shaped alloy steel teeth arranged in a staggered pattern. The three-stage telescopic mechanism uses three sets of hydraulic cylinders in series to achieve telescopic travel, adapting to the needs of steep slope operations. The hydraulic station outputs pressure, and the compaction reaction force is adjusted steplessly through a proportional valve. The integrated piezoelectric ceramic pressure sensor and MEMS gyroscope provide real-time feedback of contact pressure and posture data. The roller uses GNSS positioning to determine the slope's position. When the center of the machine is a predetermined distance from the slope shoulder, the retractable mechanism automatically triggers the deployment of the compacting plate. Millimeter-wave radar monitors slope obstacles in real time, providing early warning and shutdown in the event of large rocks. Layered compaction is implemented. In the initial compaction stage, pre-compaction is performed with contact pressure. In the main compaction stage, when the difference in compaction degree between the edge area and the middle area is greater than the predetermined ratio, the high-pressure compensation mode is activated. For inter-tooth maintenance, after each multiple rolling operations, the high-frequency micro-vibration mode is activated to automatically remove the embedded filler between the teeth.

Images