Sensing pad, sensing device and monitoring method for monitoring roadbed deformation

Abstract

The invention discloses a sensing pad, a sensing device and a monitoring method for monitoring roadbed deformation. The sensing pad comprises a rigid supporting layer, a micro air gap isolation layer, a gradient stiffness composite matrix and a flexible protection layer which are sequentially arranged from bottom to top. The gradient stiffness composite matrix comprises a plurality of fiber reinforced material layers, and the elastic modulus of each fiber reinforced material layer is increased layer by layer from top to bottom, so that disturbance to a roadbed structure when the sensing pad is laid is reduced, and strain transmission in the sensing pad in a working state is amplified; optical fiber longitudinal and transverse grids are distributed in the gradient stiffness composite matrix; the sensing pad further comprises a multi-mode optical fiber sensing unit; the multi-mode optical fiber sensing unit is embedded in the gradient stiffness composite matrix along the optical fiber longitudinal and transverse grids and used for measuring parameters related to roadbed deformation. The device has the advantages of small device structure disturbance, high coupling reliability and abundant monitoring parameters, and is suitable for three-dimensional deformation long-term monitoring of the roadbed.

Description

Technical Field

[0001] This invention belongs to the field of fiber optic sensing and cold-region roadbed engineering technology, and more specifically, relates to a sensing pad, sensing device and monitoring method for monitoring roadbed deformation. Background Technology

[0002] Permafrost is widely distributed in high-altitude or high-latitude regions. Due to the combined effects of seasonal freezing-thawing cycles, climate warming, vehicle loads, and thermal disturbances during roadbed construction and operation, permafrost roadbeds often experience differential frost heave, thaw settlement, and uneven settlement, which in turn cause problems such as uneven road surfaces, crack propagation, and track geometry instability, seriously threatening the safe operation and service life of roads and railways.

[0003] Currently, technologies used for monitoring deformation of frozen soil subgrades are mainly divided into three categories: traditional geotechnical instruments, electronic sensors, and distributed optical fibers. Traditional geotechnical instruments, such as settlement plates, inclinometers, and temperature probes, provide intuitive and reliable data, but their measurement points are sparse and they can only acquire a single physical quantity, making it difficult to capture three-dimensional deformation processes. Furthermore, these instruments are deeply buried, causing significant construction disturbance and making maintenance difficult, thus unsuitable for long-term, multi-layer monitoring. Electronic sensors, such as electronic strain gauges and accelerometers, while providing real-time performance, suffer from the risk of internal electronic components failing in frequent freeze-thaw cycles at -30°C and below. They also exhibit significant electromagnetic interference, power supply, and waterproofing issues, resulting in high costs for large-scale deployment. Distributed fiber optic sensing has the capabilities of continuous long-distance, high spatial resolution, and multi-parameter monitoring, but it relies on the stable coupling between the fiber and the soil to obtain accurate data. In large-volume roadbeds, the fiber is easily damaged by construction compaction and often slips or breaks under freeze-thaw and freeze-thaw shearing, resulting in signal drift. In addition, the fiber can only provide single information in temperature or strain and lacks the ability to identify the coupling effects of frost heave, thaw settlement, and vehicle load.

[0004] In recent years, researchers have attempted to embed optical fibers into geotextiles or geogrids to form integrated "geo-fiber" monitoring components. However, most of these solutions employ single-layer, uniformly stiff material structures, which cannot simultaneously meet the comprehensive requirements of minimal disturbance to existing roadbeds, multi-depth layered monitoring, high coupling reliability, and high resolution of multi-physical information. Therefore, there is an urgent need for an innovative technology with minimal structural disturbance and multi-modal, multi-depth self-compensating monitoring capabilities to achieve high-precision real-time early warning of differential frost heave-thaw settlement deformation in permafrost roadbeds. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the present invention aims to provide a sensing pad, sensing device, and monitoring method for monitoring roadbed deformation, thereby solving the problems of large roadbed disturbance and low resolution of multiple physical information in existing distributed optical fiber monitoring technology for frozen soil roadbed deformation.

[0006] Based on the above objectives, a first aspect of the present invention provides a sensing pad for monitoring roadbed deformation. The sensing pad comprises, from bottom to top, a rigid support layer, a micro-air gap isolation layer, a gradient stiffness composite matrix, and a flexible protective layer. The gradient stiffness composite matrix comprises multiple layers of fiber-reinforced material, with the elastic modulus of each layer increasing progressively from top to bottom to reduce disturbance to the roadbed structure during pad installation and amplify strain transmission within the sensing pad under operating conditions. A fiber optic grid is distributed within the gradient stiffness composite matrix. The sensing pad further comprises a multimode fiber optic sensing unit embedded within the gradient stiffness composite matrix along the fiber optic grid for measuring parameters related to roadbed deformation.

[0007] Preferably, the multimode fiber optic sensing unit includes a working fiber and a reference relaxation fiber; the working fiber includes a distributed acoustic sensing fiber, a distributed temperature sensing fiber, and a fiber Bragg grating array; the distance between the reference relaxation fiber and the working fiber is less than a distance threshold, the two are in the same layer and on the same thermal path, and the reference relaxation fiber is used to provide a temperature compensation signal to compensate for the temperature drift of the working fiber.

[0008] Preferably, both the working optical fiber and the reference relaxation optical fiber have a plurality of micro-anchoring protrusions spaced apart on the outer sheath of their outer surfaces. The micro-anchoring protrusions are used to enhance the interfacial mechanical coupling between the optical fiber and the soil. The material of the micro-anchoring protrusions includes thermoplastic elastomers, resin composite materials, or resin mineral composite materials. And / or, the height of the micro-anchoring protrusions is 0.2 mm to 1.0 mm.

[0009] Preferably, the gradient stiffness composite matrix comprises three fiber-reinforced material layers, with the elastic moduli of each fiber-reinforced material layer from top to bottom being 0.2 GPa to 0.4 GPa, 0.6 GPa to 0.8 GPa, and 1.0 GPa to 1.5 GPa, respectively.

[0010] Preferably, the material of the micro-gap isolation layer includes: polyurethane, polyester rubber, polyether block amide, ethylene propylene diene monomer (EPDM) rubber, silicone rubber, or EVA plastic; and / or, the structure of the micro-gap isolation layer is a honeycomb structure or a three-dimensional warp-knitted spacer fabric structure with an open porosity of 60% to 80%; and / or, the thickness of the micro-gap isolation layer is 0.2 mm to 1.0 mm; and / or, the equivalent compressive modulus of the micro-gap isolation layer in the compressive strain range of 5% to 10% is 0.05 MPa to 0.2 MPa.

[0011] A second aspect of the present invention provides a sensing device for monitoring roadbed deformation, the sensing device comprising a plurality of sensing pads as described above for monitoring roadbed deformation; the roadbed consists of a shallow surface layer, an active layer, and a frozen soil layer in a vertical direction from top to bottom, and the plurality of sensing pads are distributed in different layers of the roadbed; the sensing pads distributed in the shallow surface layer are used to monitor the thermal-load coupling deformation of the shallow surface layer; the sensing pads distributed in the active layer are used to monitor the ice lens frost heave deformation of the active layer; and the sensing pads distributed in the frozen soil layer are used to monitor the thaw settlement-differential settlement at the top interface of the frozen soil.

[0012] A third aspect of the present invention provides a method for monitoring roadbed deformation, the method comprising: acquiring strain rate signals, acoustic emission signals, high-frequency vibration signals, temperature field signals, and point strain signals measured by a sensing pad; wherein the sensing pad is a sensing pad for monitoring roadbed deformation as described above; compensating the point strain signals according to the temperature field signals, soil moisture content, and soil porosity to obtain compensated point strain; inputting the compensated point strain, the strain rate signals, the acoustic emission signals, the high-frequency vibration signals, and the temperature field signals into a preset prediction network model, and outputting the frost heave-thaw settlement amount and confidence level of the foundation.

[0013] Preferably, the compensated point strain is:

[0014] ε corr =ε raw -k T ·ΔT-k ω ·Δω-k n ·Δn

[0015] Where, ε corr For the compensated point strain, ε raw Let k be the strain signal at the specified point. T k is the temperature sensitivity coefficient. ω k is the moisture content sensitivity coefficient. n ΔT is the porosity sensitivity coefficient, Δω is the deviation of the temperature field signal from the local reference value, Δω is the deviation of the soil moisture content from the local reference value, and Δn is the deviation of the soil porosity from the local reference value.

[0016] Preferably, the method further includes: acquiring a reference fiber optic signal measured by a sensing pad; using the reference fiber optic signal to perform first-order linear correction on the temperature drift of the strain rate signal, the acoustic emission signal, the high-frequency vibration signal, the temperature field signal, and the point strain signal; compensating the point strain signal according to the temperature field signal, the soil moisture content, and the soil porosity, specifically including: compensating the point strain signal according to the first-order linearly corrected temperature field signal, the soil moisture content, and the soil porosity to obtain the compensated point strain; inputting the compensated point strain, the strain rate signal, the acoustic emission signal, the high-frequency vibration signal, and the temperature field signal into a preset prediction network model, specifically including: inputting the compensated point strain, the first-order linearly corrected strain rate signal, the first-order linearly corrected acoustic emission signal, the first-order linearly corrected high-frequency vibration signal, and the first-order linearly corrected temperature field signal into the preset prediction network model.

[0017] Preferably, the method further includes: when the product of the freeze-thaw rate and the confidence level is greater than β i ·σ i When (t), a warning signal is issued; β i σ is a multiple of the threshold. i (t) represents the standard deviation of the prediction residuals of the layer in which the sensing pad is located within the foundation.

[0018] Compared with the prior art, the advantages of the present invention include:

[0019] (1) A sensing pad for monitoring roadbed deformation is provided, consisting of a rigid support layer, a micro-air gap isolation layer, a gradient stiffness composite matrix, and a flexible protective layer arranged sequentially from bottom to top. The gradient stiffness composite matrix with a gradually changing modulus avoids the local hard insertion effect of the sensing pad on the existing roadbed and amplifies the small strain of the soil, significantly improving the overall durability and signal sensitivity of the sensing pad. The flexible protective layer absorbs the impact of construction rolling and provides primary protection. The micro-air gap isolation layer can release vertical shear in freeze-thaw cycles and maintain long-term fiber-soil coupling. The rigid support layer provides bottom tensile and shear strength to ensure the overall stability of the sensing pad.

[0020] (2) A sensing device for monitoring roadbed deformation is provided, wherein sensing pads are respectively set in the shallow layer, active layer and frozen soil layer of the roadbed. The sensing pads distributed in the shallow layer monitor the heat-load coupling deformation of the shallow layer, the sensing pads distributed in the active layer monitor the frost heave deformation of the active layer by ice lens, and the sensing pads distributed in the frozen soil layer monitor the thaw settlement-differential settlement of the frozen soil top interface. Through this layered layout, the sensing device can simultaneously acquire the temperature, strain and acoustic information of the roadbed at different depths, revealing the dynamic evolution of the vertical deformation gradient and the freeze-thaw process.

[0021] (3) A method for monitoring roadbed deformation is provided. The method utilizes strain rate signal, acoustic emission signal, high-frequency vibration signal, temperature field signal and point strain signal measured by the sensing pad, combined with field parameters such as water content and porosity of the monitoring area, and uses the built-in multi-parameter adaptive algorithm to correct residual errors. Then, it inputs the prediction network model to predict the frost heave or thaw settlement at each depth and gives the corresponding confidence level, so as to realize real-time risk management of permafrost roadbed. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the structure of a sensing pad for monitoring roadbed deformation provided in an embodiment of the present invention.

[0023] Figure 2 for Figure 1 The diagram shows the structure of the optical fiber in the sensing pad.

[0024] Figure 3 This is a cross-sectional schematic diagram of a sensing device for monitoring roadbed deformation provided in an embodiment of the present invention.

[0025] Figure 4 A flowchart of a roadbed deformation monitoring method provided in an embodiment of the present invention.

[0026] Explanation of reference numerals in the attached figures: 1 is the rigid support layer, 2 is the micro-air gap isolation layer, 3 is the multimode fiber optic sensing unit, 31 is the fiber body, 32 is the fiber core coating layer, 33 is the outer sheath, 34 is the micro-anchoring protrusion, 4 is the gradient stiffness composite matrix, 5 is the flexible protective layer, 6 is the shallow surface layer, 7 is the active layer, 8 is the permafrost layer, 9 is the sensing pad, 10 is the multimode data acquisition device, and 11 is the wireless transmission module. Detailed Implementation

[0027] In view of the shortcomings of the prior art, the inventors of this invention, through long-term research and extensive practice, have proposed the technical solution of this invention. The following will further explain and illustrate this technical solution, its implementation process, and its principles.

[0028] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0029] Furthermore, in the description of this invention, it should be understood that the terms "upper," "lower," "inner," "outer," "horizontal," "vertical," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0030] In the description of this specification, the references to terms such as "an embodiment," "a particular embodiment," or "the embodiment" indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0031] Figure 1 This is a schematic diagram of the structure of a sensing pad for monitoring roadbed deformation provided in an embodiment of the present invention. (See also...) Figure 1 The sensing pad for monitoring roadbed deformation comprises, from bottom to top, a rigid support layer 1, a micro-air gap isolation layer 2, a gradient stiffness composite matrix 4, and a flexible protective layer 5. The gradient stiffness composite matrix 4 consists of multiple layers of fiber-reinforced material, with the elastic modulus of each layer increasing progressively from top to bottom. This reduces disturbance to the roadbed structure during pad installation and amplifies strain transmission within the sensing pad under operating conditions. A fiber optic grid is distributed within the gradient stiffness composite matrix 4. The sensing pad also includes multimode fiber optic sensing units 3; these units are embedded within the gradient stiffness composite matrix 4 along the fiber optic grid and are used to measure parameters related to roadbed deformation. This sensing pad is suitable for monitoring differential frost heave-thaw settlement deformation of roadbeds in permafrost regions.

[0032] The rigid support layer 1 is used to provide tensile and shear strength at the bottom. Preferably, the material of the rigid support layer 1 still has the advantages of high in-plane tensile stiffness / shear strength, freeze-thaw resistance and corrosion resistance in low temperature environment (≤-40℃), and can reliably bond with the gradient stiffness composite matrix and maintain interlocking with the soil. For example, the rigid support layer 1 is composed of high modulus glass fiber grid, basalt fiber fabric, high modulus fiber composite material, and high-strength polymer geogrid.

[0033] The micro-gap isolation layer 2 is used to mitigate vertical shear and maintain fiber-soil coupling. Preferably, the material of the micro-gap isolation layer includes: polyurethane, polyester rubber, polyether block amide, EPDM rubber, silicone rubber, or EVA plastic; and / or, the structure of the micro-gap isolation layer is a honeycomb structure or a three-dimensional warp-knitted spacer fabric structure with an open porosity of 60% to 80%; and / or, the thickness of the micro-gap isolation layer is 0.2 mm to 1.0 mm; and / or, the equivalent compressive modulus of the micro-gap isolation layer is 0.05 MPa to 0.2 MPa in the compressive strain range of 5% to 10%.

[0034] The micro-gap isolation layer 2 is a compressible honeycomb elastic material. Compressible honeycomb elastic material refers to a polymer or rubber sheet / layer with a regular or quasi-regular honeycomb / mesh structure that maintains elastic compressibility in a low-temperature environment (≤-40℃). Examples include open-cell TPU, TPEE / PEBA elastomer honeycomb grid, EPDM open-cell rubber honeycomb sheet, silicone rubber open-cell honeycomb sheet, EVA microporous honeycomb sheet, three-dimensional warp-knitted spacer fabric, open-cell elastomer foam, elastomer grid / honeycomb core, etc.

[0035] Preferably, the gradient stiffness composite matrix 4 is formed by hot-pressing multiple layers of fiber-reinforced material. The fiber optic mesh can be prepared in the following ways: by forming microgrooves in the mold and inserting optical fibers before hot pressing; or by forming a warp-knitted mesh skeleton + fiber guide tape heat-sealing; or by arranging longitudinal and transverse serpentine lines in the upper and lower layers to achieve mesh coverage, and by using layered staggering or microtube bridging treatment at the intersections to reduce microbending loss and wear.

[0036] In a preferred embodiment, the gradient stiffness composite matrix 4 includes three fiber-reinforced material layers, with the elastic moduli of each fiber-reinforced material layer from top to bottom being 0.2 GPa to 0.4 GPa, 0.6 GPa to 0.8 GPa, and 1.0 GPa to 1.5 GPa, respectively.

[0037] The flexible protective layer 5 is used to absorb road surface loads and prevent construction damage. Preferably, the material of the flexible protective layer 5 retains the advantages of flexibility, tear resistance, and abrasion resistance even at low temperatures (≤-40℃), and is easy to hot-press or adhesively bond with the gradient stiffness composite substrate 4. The flexible protective layer 5 is made, for example, of low-temperature resistant polyester woven fabric, ultra-high molecular weight polyethylene, or PTFE-coated fiberglass cloth.

[0038] In a preferred embodiment, the multimode fiber optic sensing unit 3 includes a working fiber and a reference relaxation fiber. The working fiber includes a distributed acoustic sensing fiber (DAS), a distributed temperature sensing fiber (DTS), and a fiber Bragg grating (FBG) array. The distance between the reference relaxation fiber and the working fiber is less than a distance threshold; they are in the same layer and on the same thermal path. The reference relaxation fiber is used to provide a temperature compensation signal to compensate for the temperature drift of the working fiber. The reference relaxation fiber is in a near-zero strain state.

[0039] The DAS channel outputs coherent scattering signals to extract features such as strain rate, acoustic emission, and high-frequency vibration; the DTS channel provides a continuous temperature field along the line; the FBG channel acquires the center wavelength of the grating—strain-type FBGs are deployed at the interface between the differential settlement sensitive area and the ice-rich layer to obtain point strain, and temperature-type FBGs are configured as local temperature / decoupling references when necessary.

[0040] Both the working fiber and the reference relaxation fiber are led out from the edge of the housing via quick-connect fittings to interface with external data acquisition-demodulation devices. The quick-connect fittings are waterproof quick-connect / outgoing cable assemblies, such as field-assembleable fiber optic quick connectors, industrial-grade waterproof quick-plug fiber optic connectors, and flange-through-cabin cable assemblies with waterproof glands and stress relief.

[0041] Preferably, the reference relaxation fiber is located in the edge cavity of the pad and is fixed by a low-modulus, low-shrinkage, low-temperature / freeze-thaw resistant material that is chemically compatible with the fiber coating (PI / acrylate), such as low-modulus silicone, polyurethane gel or silicone gel, to maintain a zero-strain reference.

[0042] Preferably, the center wavelength of the fiber Bragg grating array is 1525nm–1575nm (preferably 1545nm–1565nm or 1565nm–1565.7nm) or located in the C / L band of 1525nm–1565nm, and the axial spacing between adjacent gratings is 5cm–10cm. Based on the strain / temperature range, a wavelength interval ≥ (single grating bandwidth + expected maximum drift + safety margin) is set between the center wavelength of a single grating and adjacent gratings to avoid spectral overlap.

[0043] In a preferred embodiment, the outer sheaths of both the working optical fiber and the reference relaxation optical fiber are provided with a plurality of micro-anchoring protrusions spaced apart. These micro-anchoring protrusions enhance the interfacial mechanical coupling between the optical fiber and the soil. The materials of the micro-anchoring protrusions include thermoplastic elastomers, resin composite materials, or resin-mineral composite materials. And / or, the height of the micro-anchoring protrusions is 0.2 mm to 1.0 mm. More preferably, the height of the micro-anchoring protrusions is 0.3 mm to 0.6 mm, for example, 0.4 mm or 0.5 mm; the spacing between adjacent micro-anchoring protrusions is 50 mm to 150 mm.

[0044] In this embodiment, the optical fiber is covered with micro-anchoring bumps or a flexible reinforcing sleeve, forming a mechanical lock with the surrounding soil to prevent slippage or breakage during freeze-thaw shearing and construction disturbances. The material of the micro-anchoring bumps remains firmly attached to the outer sheath / substrate of the optical fiber even in low-temperature environments (≤-40℃) and freeze-thaw cycles, providing repeatable mechanical interlocking and maintaining the process's mass-producibility. The micro-anchoring bumps are formed, for example, from thermoplastic elastomers or resin composites, selected from TPU, TPEE, SEBS-TPE, TPV, PA12 / PA11, PETG, EVA, copolyamide, copolyester hot melt adhesive, or resin composite layers containing mineral particles (e.g., basalt sand / alumina / silica sand); they are fabricated through molding, hot-pressing, ultrasonic hot riveting, screen coating / spray curing, co-extrusion, or 3D microstructure printing.

[0045] In this embodiment of the invention, the sensing pad is sheet-like and can be spliced ​​side-by-side according to the width of the roadbed. From top to bottom, the sensing pad consists of a flexible protective layer, a gradient stiffness composite matrix, a micro-gap isolation layer, and a rigid support layer. The flexible protective layer absorbs the impact of construction compaction and provides primary protection; the gradient stiffness composite matrix amplifies the strain signal while reducing disturbance to the original roadbed structure; the micro-gap isolation layer can slowly release vertical shear during freeze-thaw cycles and maintain long-term fiber-soil coupling; the rigid support layer provides tensile and shear strength at the bottom, ensuring the overall stability of the pad.

[0046] This invention provides a sensing pad for monitoring roadbed deformation, specifically designed for monitoring differential frost heave and thaw settlement in roadbeds of highways and railways in permafrost regions. The sensing pad embeds various types of optical fibers, including distributed acoustic sensing fibers, distributed temperature sensing fibers, and fiber Bragg grating arrays, into a multi-layered geosynthetic material with gradient stiffness. Through a flexible-rigid gradient, micro-gaps isolation, and micro-anchoring coupling design, it achieves efficient coupling with the roadbed soil without significantly disturbing the existing structure. Combined with a reference relaxation fiber for hardware-level compensation, it can continuously acquire temperature, strain, vibration, and acoustic information at different depths, and differentiate frost heave, thaw settlement, cracks, and load effects in real time. Compared to existing technologies, this invention has significant advantages in spatial coverage, coupling reliability, error self-compensation, and ease of deployment, and can be widely used for long-term health monitoring and operation and maintenance management of infrastructure in permafrost regions.

[0047] Based on the same inventive concept, the present invention provides a sensing device for monitoring roadbed deformation. The sensing device includes multiple sensing pads for monitoring roadbed deformation. These sensing pads are the same as those used in the above embodiments for monitoring roadbed deformation. Their structure will not be described in detail here.

[0048] The roadbed consists of a shallow surface layer, an active layer, and a frozen soil layer, arranged vertically from top to bottom. Multiple sensor pads are distributed within different layers of the roadbed. The sensor pads distributed in the shallow surface layer are used to monitor the thermal-load coupled deformation of the shallow surface layer; the sensor pads distributed in the active layer are used to monitor the frost heave deformation of the active layer caused by ice lenses; and the sensor pads distributed in the frozen soil layer are used to monitor the thaw settlement and differential settlement at the top interface of the frozen soil.

[0049] In this embodiment, to achieve accurate characterization of different physical processes, the sensing pads can be stacked in multiple layers or laid in multiple layers. Preferably, the sensing device has three sensing pads, distributed in the shallow layer, active layer, and permafrost layer, respectively. The shallow layer monitors the freeze-thaw cycle and vehicle load coupled deformation at the subgrade structure layer-fill interface; the middle layer is located at the bottom interface of the active layer and is used to track ice lens growth and frost heave; the deep layer is placed at the top interface of the ice-rich permafrost and is used to observe the evolution of thaw settlement and differential settlement. Through this layered layout, the sensing device can simultaneously acquire temperature, strain, and acoustic information of the subgrade at different depths, revealing the dynamic evolution of the vertical deformation gradient and freeze-thaw process.

[0050] In newly constructed roadbeds or existing permafrost roadbeds, sensing pads can be laid synchronously or directionally inserted to the target depth according to the design. In newly constructed roadbeds, the sensing pads are laid synchronously with the layered filling. After the base is leveled, the pad body is first laid out and tensioned for correction, then backfilled with fill material and mechanically compacted to ensure tight coupling between the pad and the soil. When laying multiple layers, the pads are implemented layer by layer according to the designed burial depth. The quick-connect interfaces on both sides of the pad body can be spliced ​​on-site to adapt to roadbeds of different widths and structural forms. In the scenario of existing roadbed renovation, the sensing pads or single optical fibers can be embedded into the target depth through drilling or directional insertion methods, combined with high-performance backfill material to achieve low-disturbance and high-efficiency system installation.

[0051] Based on the same inventive concept, this invention provides a method for monitoring roadbed deformation, see reference. Figure 4 This method includes steps S100-S300.

[0052] Step S100: Collect strain rate signal, acoustic emission signal, high-frequency vibration signal, temperature field signal and point strain signal measured by the sensing pad; the sensing pad is the sensing pad used to monitor roadbed deformation in the above embodiment.

[0053] Distributed acoustic sensing fiber optics measures strain rate signals, acoustic emission signals, and high-frequency vibration signals; distributed temperature sensing fiber optics measures temperature field signals; and fiber Bragg grating arrays measure point strain signals. Preferably, the DAS sampling frequency is 5kHz to 10kHz; and the DTS sampling spacing is 0.25m to 0.5m.

[0054] Step S200: The point strain signal is compensated based on the temperature field signal, the soil moisture content and the soil porosity to obtain the compensated point strain.

[0055] In a preferred embodiment, the compensated point strain is:

[0056] ε corr =ε raw -k T ·ΔT-k ω ·Δω-k n ·Δn

[0057] Where, ε corr For the compensated point strain, ε raw For point strain signals, k T k is the temperature sensitivity coefficient. ω k is the moisture content sensitivity coefficient. n Let be the porosity sensitivity coefficient, ΔT be the deviation of the temperature field signal from the local reference value, Δω be the deviation of the foundation moisture content from the local reference value, and Δn be the deviation of the foundation porosity from the local reference value. Preferably, kT k ω k n A sliding window least squares adaptive update is used, with a window width of, for example, 24h. The soil moisture content ω is obtained through direct measurement; the soil porosity n is calculated from the on-site dry density and particle density.

[0058] Step S300: Input the compensated point strain, strain rate signal, acoustic emission signal, high-frequency vibration signal and temperature field signal into the preset prediction network model, and output the frost heave-thaw settlement amount and confidence level of the foundation.

[0059] The prediction network model, for example, employs a segmented LSTM-Attention network. The output of the prediction network model is the frost heave-thaw settlement Δh for each layer i (shallow layer, active layer, or permafrost layer). i (t) and the corresponding confidence level ρ i (t).

[0060] Preferably, the prediction network model is divided into three time scales: 0h to 24h, 1d to 7d, and 7d to 30d, to capture short-term frost heave, weekly frost heave, and monthly thaw settlement, respectively.

[0061] In a preferred embodiment, before step S200, the method further includes: acquiring a reference fiber optic signal measured by the sensing pad, and using the reference fiber optic signal to perform first-order linear correction on the temperature drift of the strain rate signal, acoustic emission signal, high-frequency vibration signal, temperature field signal, and point strain signal. In step S200, the point strain signal is compensated based on the first-order linearly corrected temperature field signal, soil moisture content, and soil porosity to obtain the compensated point strain. In step S300, the compensated point strain, the first-order linearly corrected strain rate signal, the first-order linearly corrected acoustic emission signal, the first-order linearly corrected high-frequency vibration signal, and the first-order linearly corrected temperature field signal are input into a preset prediction network model.

[0062] Specifically, the reference fiber signal is the temperature change ΔT measured by the reference relaxation fiber. ref Based on ΔT ref First-order temperature drift correction is performed on the working fiber optic signal. Specifically, the FBG strain channel subtracts the temperature term according to the temperature-strain decoupling formula, and the DAS channel is based on ΔT... ref Low-frequency phase / distance calibration drift relocking is performed using the DTS channel with ΔT. ref The source power is differentially corrected with the baseline without offsetting the measured temperature; the correction coefficient uses a piecewise linear model, for example, when |ΔT ref |>10℃ Automatic switching of calibration segment.

[0063] In a preferred embodiment, when the product of frost heave-thaw settlement and reliability is |Δh i(t)·ρ i (t)|greater than β i ·σ i When (t), an early warning signal is issued (e.g., a tiered early warning signal or a general early warning signal); β i σ is a multiple of the threshold. i (t) represents the standard deviation of the prediction residuals of the sensing pad within the foundation layer. Preferably, 3 ≤ β i ≤4.

[0064] Furthermore, monitoring results can be uploaded to the cloud platform in real time via 4G / 5G wireless to dynamically update the health spectrum of regional roadbed groups, enabling the operation and maintenance department to quickly assess and intervene.

[0065] The present invention provides a sensing pad, sensing device, and monitoring method for monitoring roadbed deformation. Through integrated structural design, multi-depth layered deployment, and intelligent data analysis, it achieves long-term, stable, and high-precision monitoring of differential roadbed deformation in the complex environment of permafrost regions. Compared with existing monitoring schemes, the present invention significantly improves spatial coverage and coupling stability, reduces temperature-humidity drift errors, and is more convenient to construct and maintain, possessing broad engineering application prospects. Compared with existing technologies, the present invention can achieve the following beneficial effects:

[0066] (1) Structural integration and gradient stiffness design. In the sensing pad, the flexible protective layer, gradient stiffness composite matrix, micro-air gap isolation layer and rigid support layer are integrated and hot-pressed. The gradient stiffness composite matrix with gradually changing modulus avoids the local hard insertion effect on the existing roadbed and amplifies the small strain of the soil, which significantly improves the overall durability and signal sensitivity.

[0067] (2) Multimodal collaborative monitoring. Three types of optical fibers, namely DAS, DTS, and FBG, are embedded in the same sensing pad to achieve simultaneous acquisition of vibration / acoustics, temperature, and high-precision point strain. The complementarity of multiple physical quantities enables the system to distinguish between ice lensing, melt sinking and vehicle load effects, greatly expanding the diagnostic dimensions.

[0068] (3) High coupling stability and signal amplification. Micro-anchoring protrusions lock the optical fiber into the fine-grained soil, and the honeycomb isolation strip releases vertical shear, maintaining the long-term coupling coefficient above 0.8. The higher modulus layer under the gradient stiffness layer amplifies the strain of the upper soft region a second time, ensuring data accuracy.

[0069] (4) Error self-compensation and intelligent early warning. The reference relaxation fiber provides a hardware-level temperature drift benchmark; multi-parameter algorithm-level correction such as water content and porosity further eliminates environmental drift; the prediction network model outputs predicted values ​​and credibility on hourly, daily and monthly scales, and automatically issues tiered or overall early warnings when the threshold is exceeded.

[0070] (5) Multi-depth three-dimensional monitoring. The sensing pad can be layered. Stacking The system monitors heat-load coupling deformation in the shallow layer, frost heave through ice lenses in the middle layer, and thaw settlement of ice-rich permafrost in the deep layer. The vertical gradient information is consistent, accurately reflecting the entire process of freeze-thaw evolution and differential settlement.

[0071] (6) Standardized and modular construction with low disturbance adaptability. The sensor pad can be rolled up for transportation and quickly deployed on site; the lateral quick connectors enable multi-piece splicing. The new roadbed can be laid synchronously without extending the construction period; the existing roadbed can be drilled or directionally inserted for installation, with minimal construction disturbance and strong adaptability.

[0072] (7) Strong durability and environmental adaptability. The material system is resistant to low temperature of -40℃, salt corrosion, and freeze-thaw cycles; the fiber optic sheath and grid are both UV resistant and chemical corrosion resistant. The sensing pad and sensing device can operate continuously and stably for more than ten years in extreme permafrost environments.

[0073] (8) Economy and scalability. The width of a single pad is 0.8m to 1.5m, and it can be connected horizontally or vertically as needed; multimodal integration reduces the number of external lines and connectors, reducing maintenance costs. The data interface is open, supporting seamless integration with existing monitoring cloud platforms and building information modeling (BIM) systems, facilitating future expansion and upgrades.

[0074] The technical solution of the present invention will be further described in detail below with reference to several preferred embodiments and accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. Test methods in the following embodiments that do not specify specific conditions are generally performed under conventional conditions.

[0075] See Figure 3 The diagram shows a cross-section of sensing devices arranged in the roadbed for monitoring roadbed deformation. The highway or railway roadbed is divided into a shallow surface layer 6, an active layer 7, and a frozen soil layer 8 from top to bottom. Sensing pads 9 are evenly distributed in the shallow surface layer 6, active layer 7, and frozen soil layer 8. An additional multimodal data acquisition device 10 and a wireless transmission module 11 are provided. The multimodal data acquisition device 10 is installed in the shoulder or manhole and connected to the multimodal fiber optic sensing unit 3 inside the sensing pad 9 via a pre-buried optical cable. The wireless transmission module 11 is integrated with the multimodal data acquisition device 10 and can upload monitoring data to the dispatch center or cloud platform in real time.

[0076] The flexible protective layer of the sensing pad is made of polyester filament woven fabric resistant to -40℃, with a thickness of approximately 1mm. The gradient stiffness composite matrix has a total thickness of 8mm to 12mm and is integrally formed by hot pressing three layers of glass fiber / basalt fiber reinforced resin with progressively increasing elastic moduli: the upper layer has a modulus of 0.3GPa, the middle layer 0.7GPa, and the lower layer 1.3GPa. An optical fiber mesh is pre-embedded both longitudinally and laterally within the gradient stiffness composite matrix: a longitudinal bandwidth of 50mm and a transverse bandwidth of 20mm, with a spacing of 10cm to 20cm depending on the monitoring resolution. The optical fiber mesh includes DAS, DTS, and FBG arrays at key locations, and a reference fiber in a relaxed state is laid parallel to it for hardware-level temperature and humidity drift compensation. A polyurethane honeycomb micro-gaps isolation strip with a thickness of approximately 0.5mm is inserted between the gradient stiffness composite matrix and the rigid support layer; the honeycomb pores can undergo 2% to 5% controlled compression during vertical loads and freeze-thaw shearing, thus mitigating shear strain and maintaining fiber-soil bonding. The rigid support layer uses basalt fiber grid with an elastic modulus of 30GPa, which can provide high tensile and shear strength to the pad, ensuring overall stability during handling, laying and service life.

[0077] See Figure 2 The structure of the optical fiber in the sensing pad is shown. The optical fiber body 31 is a standard single-mode silica optical fiber; the core coating layer 32 is made of low-temperature resistant polyimide with a thickness of about 20 μm; micro-anchoring bumps 34 with a height of 0.3 mm to 0.6 mm are uniformly distributed on the outer side of the coating layer. The bump material is hot-melt polyolefin particles, which are partially melted into the gradient stiffness composite matrix during hot pressing, and then mechanically locked with fine-grained soil after rolling. The coupling coefficient can be maintained above 0.8 for a long time; the outermost layer is the outer sheath 33, which is a flexible sheath made of high-density polyethylene or fluorinated polymer, providing waterproof, wear-resistant and chemical corrosion-resistant protection. The DAS optical fiber is laid in a single loop along the entire length of the pad to capture strain rate and acoustic events; the DTS optical fiber is arranged in a dual loop with a temperature sampling interval of 0.25 m to 0.5 m; the FBG is bonded in a linear array at a grid pitch of 5 cm to 10 cm at the grid intersections to monitor highly sensitive local strain and temperature.

[0078] For newly constructed roadbeds, the laying process is as follows: After the base is finely prepared using a grader at the designed depth, the deepest layer of sensing pads is laid first. The optical cable is then pulled to the road shoulder through a pre-reserved conduit. Subsequently, layers of fill material are added and mechanically compacted. Once the next designed depth is reached, the middle and upper layers of sensing pads are laid. Quick-connect fittings on the side edges of the pads are spliced ​​on-site to ensure the continuity and sealing of the optical fiber and the sheath.

[0079] For existing roadbeds, the modification process is as follows: A φ120mm rotary drill is used to reach the target depth. After cleaning the hole, a coiled sensing pad or a single optical fiber is inserted, followed by the injection of self-compacting fine-grained sand-polyurethane grout. The grout expands and solidifies, forming a secondary anchor on the hole wall, while simultaneously filling micro-gaps to ensure coupling. The optical cable is then connected to the multi-modal data acquisition equipment via a newly installed or existing cable tray. The equipment then uploads the data to the cloud via a wireless transmission module.

[0080] Figure 3 The system continuously acquires DAS acoustic-vibration waveforms at a sampling rate of 5kHz to 10kHz, polls DTS temperature at 1-hour intervals, and simultaneously reads FBG wavelength drift. The acquisition software first uses real-time temperature information from a reference fiber to perform hardware-level temperature drift compensation on all working fibers. Then, combining field parameters such as water content and porosity of the monitoring area, it further corrects residual errors using a built-in multi-parameter adaptive algorithm. The multimodal time-series data, after dual compensation, is segmented and input into an LSTM-Attention network: an hourly model captures transient responses to thermal expansion and contraction, a daily model tracks rapid ice lens growth, and a monthly model assesses the trend of thaw settlement accumulation. The network outputs frost heave or thaw settlement at each depth and provides corresponding confidence levels. The system compares the results with historical health bands; when any layer or overall deformation exceeds a set threshold, an early warning is automatically triggered and pushed to the operation and maintenance center via a wireless interface, achieving real-time risk management of permafrost subgrades. This system has the advantages of small structural disturbance, high coupling reliability, rich monitoring parameters, and strong error self-compensation capability. It can be widely used for long-term three-dimensional deformation monitoring and intelligent operation and maintenance of infrastructure such as highways, railways, and airport pavements in permafrost areas.

[0081] It should be understood that the above embodiments are merely illustrative of the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made in accordance with the spirit and essence of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A sensing pad for monitoring roadbed deformation, characterized in that, The sensing pad comprises, from bottom to top, a rigid support layer, a micro-air gap isolation layer, a gradient stiffness composite matrix, and a flexible protective layer; The gradient stiffness composite matrix includes multiple layers of fiber-reinforced material, with the elastic modulus of each layer increasing progressively from top to bottom, in order to reduce the disturbance to the roadbed structure during the laying of the sensing pad and amplify the strain transmission within the sensing pad under working conditions; the gradient stiffness composite matrix contains a longitudinal and transverse fiber grid. The sensing pad also includes a multimode fiber optic sensing unit; the multimode fiber optic sensing unit is embedded inside the gradient stiffness composite matrix along the fiber optic grid and is used to measure parameters related to roadbed deformation.

2. The sensing pad for monitoring roadbed deformation according to claim 1, characterized in that, The multimode fiber optic sensing unit includes a working fiber and a reference relaxation fiber; The working optical fiber includes a distributed acoustic sensing fiber, a distributed temperature sensing fiber, and a fiber Bragg grating array. The distance between the reference relaxation fiber and the working fiber is less than a distance threshold. They are in the same layer and on the same thermal path. The reference relaxation fiber is used to provide a temperature compensation signal to compensate for the temperature drift of the working fiber.

3. The sensing pad for monitoring roadbed deformation according to claim 2, characterized in that, Both the working optical fiber and the reference relaxation optical fiber have a number of micro-anchoring protrusions spaced apart on the outer sheath of their outer surfaces. These micro-anchoring protrusions are used to enhance the mechanical coupling between the optical fiber and the soil interface. The material of the micro-anchoring protrusions includes thermoplastic elastomers, resin composites, or resin mineral composites; and / or, the height of the micro-anchoring protrusions is 0.2 mm to 1.0 mm.

4. The sensing pad for monitoring roadbed deformation according to claim 1, characterized in that, The gradient stiffness composite matrix comprises three fiber-reinforced material layers, with the elastic moduli of each fiber-reinforced material layer from top to bottom being 0.2 GPa to 0.4 GPa, 0.6 GPa to 0.8 GPa, and 1.0 GPa to 1.5 GPa, respectively.

5. The sensing pad for monitoring roadbed deformation according to claim 1, characterized in that, The materials of the micro-gap isolation layer include: polyurethane, polyester rubber, polyether block amide, EPDM rubber, silicone rubber, or EVA plastic; And / or, the structure of the micro-air gap isolation layer is a honeycomb structure with an open porosity of 60% to 80% or a three-dimensional warp-knitted spacer fabric structure; And / or, the thickness of the micro-air gap isolation layer is 0.2 mm to 1.0 mm; And / or, the equivalent compressive modulus of the micro-air gap isolation layer is 0.05MPa to 0.2MPa in the compressive strain range of 5% to 10%.

6. A sensing device for monitoring roadbed deformation, characterized in that, The sensing device includes a plurality of sensing pads for monitoring roadbed deformation as described in any one of claims 1-5; The roadbed consists of a shallow layer, an active layer, and a frozen soil layer in the vertical direction from top to bottom, with multiple sensing pads distributed in different layers of the roadbed. Sensor pads distributed in the shallow layer are used to monitor the thermal-load coupled deformation of the shallow layer; sensor pads distributed in the active layer are used to monitor the frost heave deformation of the active layer by ice lenses; and sensor pads distributed in the frozen soil layer are used to monitor the thaw settlement-differential settlement at the top interface of the frozen soil.

7. A method for monitoring roadbed deformation, characterized in that the method... include: The sensor pad acquires strain rate signals, acoustic emission signals, high-frequency vibration signals, temperature field signals, and point strain signals measured by the sensor pad; the sensor pad is the sensor pad for monitoring roadbed deformation as described in any one of claims 1-5. The point strain signal is compensated based on the temperature field signal, the soil moisture content and the soil porosity to obtain the compensated point strain. The compensated point strain, the strain rate signal, the acoustic emission signal, the high-frequency vibration signal, and the temperature field signal are input into a preset prediction network model, which outputs the frost heave-thaw settlement amount and the confidence level of the foundation.

8. The method for monitoring roadbed deformation according to claim 7, characterized in that, The compensated point strain is: e corr =e raw -k T ·ΔT-k ω ·Dω-k n ·Δn Where, ε corr For the compensated point strain, ε raw The point strain signal is given, kT is the temperature sensitivity coefficient, and k ω k is the moisture content sensitivity coefficient. n ΔT is the porosity sensitivity coefficient, Δω is the deviation of the temperature field signal from the local reference value, Δω is the deviation of the soil moisture content from the local reference value, and Δn is the deviation of the soil porosity from the local reference value.

9. The method for monitoring roadbed deformation according to claim 7, characterized in that, The method further includes: acquiring a reference fiber optic signal measured by the sensing pad, and using the reference fiber optic signal to perform first-order linear correction on the temperature drift of the strain rate signal, the acoustic emission signal, the high-frequency vibration signal, the temperature field signal, and the point strain signal; The point strain signal is compensated based on the temperature field signal, soil moisture content, and soil porosity. Specifically, the point strain signal is compensated based on the temperature field signal after first-order linear correction, soil moisture content, and soil porosity to obtain the compensated point strain. The compensated point strain, the strain rate signal, the acoustic emission signal, the high-frequency vibration signal, and the temperature field signal are input into a preset prediction network model. Specifically, this includes inputting the compensated point strain, the strain rate signal after first-order linear correction, the acoustic emission signal after first-order linear correction, the high-frequency vibration signal after first-order linear correction, and the temperature field signal after first-order linear correction into a preset prediction network model.

10. The method for monitoring roadbed deformation according to claim 7, characterized in that, The method further includes: when the product of the frost heave-thaw settlement amount and the confidence level is greater than β i ·σ i When (t), a warning signal is issued; β i σ is a multiple of the threshold. i (t) represents the standard deviation of the prediction residuals of the layer in which the sensing pad is located within the foundation.

Images