A method and system for monitoring the verticality of the full length of a pile based on distributed optical fiber

Abstract

The present application relates to the pile foundation engineering monitoring technical field, provide a kind of based on distributed optical fiber's pile foundation full length perpendicularity monitoring method and system, comprising: through the multi-path optical fiber deployed on the surface of pile body, the transmission of pulsed laser source and the reception of backscattering signal are carried out, and backscattering signal is converted into phase change;Phase change is mapped as time-frequency joint distribution, and time-frequency diagram is obtained, and bending angle prediction value is obtained by depth residual network;Phase change is mapped as pile body curvature, and bending angle prediction value is used to correct pile body curvature, and the perpendicularity deviation angle is obtained by integrating the corrected pile body curvature.Full length perpendicularity is continuously monitored with high precision, and the monitoring blind area problem is solved.

Description

Technical Field

[0001] This invention belongs to the field of pile foundation engineering monitoring technology, and in particular relates to a method and system for monitoring the verticality of the entire length of pile foundations based on distributed optical fiber. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] During the construction of building projects, it is essential to strictly control the verticality of the pile body. The verticality of the pile body is a core quality indicator in pile foundation construction, which directly affects the safety, structural stability, and service life of the project.

[0004] Traditional methods for monitoring the verticality of pile foundations often focus on point-based or localized monitoring using total stations, lidar, and ultrasonic sensors, such as vibration monitoring schemes in soft soil pile foundation construction. However, these methods do not address the issue of continuous monitoring along the entire pile length.

[0005] Point-based measurement methods can only obtain the local tilt angle of the monitoring ring installation position, and cannot achieve continuous monitoring of the entire pile length. Moreover, the spatial resolution is limited by the number of sensors. Furthermore, the mechanical adjustment of the extension rod to adapt to different pile diameters still relies on manual operation and physical contact measurement, which cannot capture minute deformations and has weak resistance to electromagnetic interference. Summary of the Invention

[0006] To address the technical problems mentioned above, this invention provides a method and system for monitoring the verticality of a pile foundation along its entire length based on distributed optical fiber. Optical fiber sensors are deployed along the pile to capture phase changes in the optical signal caused by minute deformations of the pile. A deep residual network is used to predict the pile's bending angle, and the predicted bending angle is used to correct the pile's curvature, thereby accurately calculating the verticality deviation angle. This enables continuous, high-precision monitoring of the verticality along the entire length, solving the problem of monitoring blind spots.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: The first aspect of this invention provides a method for monitoring the verticality of a pile foundation along its entire length based on distributed optical fibers, comprising: A pulsed laser source is emitted and a backscattered Rayleigh signal is received by a multipath optical fiber deployed on the surface of the pile, and the backscattered Rayleigh signal is converted into a phase change. The phase change is mapped to a joint time-frequency distribution to obtain a time-frequency map. The bending angle is then predicted by a deep residual network. The phase change is mapped to the pile curvature, and the pile curvature is corrected by the predicted bending angle. The corrected pile curvature is then integrated to obtain the verticality deviation angle.

[0008] Furthermore, the corrected pile curvature is: ; Where: z is the spatial coordinate, and t is time. This is the predicted value for the bending angle. The curvature of the pile body before correction is α, and the fusion coefficient is α.

[0009] Furthermore, the step of mapping the phase change to the pile curvature includes: after converting the phase change into fiber axial strain, converting the fiber axial strain into pile curvature according to the conjugate beam principle.

[0010] Furthermore, it also includes obtaining the displacement field by iteratively solving the multibody coupled dynamic equations based on the modified pile curvature.

[0011] Furthermore, the multipath optical fiber includes a main sensing optical fiber arranged along the axial direction and auxiliary optical fibers spirally wound at the top and bottom of the pile.

[0012] Furthermore, the multi-path optical fiber deployment process consists of two parts: helical-axial composite wiring and strain coupling enhancement structure.

[0013] Furthermore, the phase change is mapped to a joint time-frequency distribution by performing a Fourier transform on the piecewise signal using a window function.

[0014] A second aspect of the present invention provides a pile foundation full-length verticality monitoring system based on distributed optical fiber, comprising: The signal acquisition module is configured to: transmit pulsed laser source and receive backscattered Rayleigh signal through multipath optical fiber deployed on the surface of pile body, and convert backscattered Rayleigh signal into phase change; The prediction module is configured to: map the phase change to a joint time-frequency distribution to obtain a time-frequency map, and obtain the predicted bending angle value through a deep residual network; The calculation module is configured to: map the phase change to the pile curvature, correct the pile curvature with the predicted bending angle, and integrate the corrected pile curvature to obtain the verticality deviation angle.

[0015] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps described above in a distributed optical fiber-based full-length verticality monitoring of a pile foundation.

[0016] A fourth aspect of the present invention provides a computer device including a computer-readable storage medium, a processor, and a computer program stored on the computer-readable storage medium and executable on the processor, wherein the processor executes the program to implement the steps described above in a distributed optical fiber-based full-length verticality monitoring system for pile foundations.

[0017] Compared with the prior art, the beneficial effects of the present invention are: This invention deploys fiber optic sensors along the pile body to capture the phase changes of the optical signal caused by minute deformations of the pile body. It predicts the bending angle of the pile body through a deep residual network and corrects the curvature of the pile body using the predicted bending angle to accurately calculate the verticality deviation angle. This enables continuous and high-precision monitoring of verticality along the entire length, solving the problem of monitoring blind spots.

[0018] This invention has strong anti-interference capabilities, and the passive characteristics of fiber optic sensing enable it to work stably in environments with strong electromagnetic fields and high humidity, ensuring the accuracy of monitoring results. Attached Figure Description

[0019] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0020] Figure 1 This is a flowchart of a method for monitoring the verticality of a pile foundation along its entire length based on distributed optical fiber, according to Embodiment 1 of the present invention. Figure 2 This is a flowchart of time-frequency domain signal demodulation according to Embodiment 1 of the present invention; Figure 3 This is a schematic diagram of the spiral-axial composite wiring according to Embodiment 1 of the present invention; Figure 4 This is a schematic diagram of the structure of a computer device according to Embodiment 4 of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0022] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0023] Example 1 This embodiment provides a method for monitoring the verticality of pile foundations along their entire length based on distributed optical fibers.

[0024] Distributed fiber optic sensing technology is a continuous distributed sensing technology that uses optical fiber as both the transmission medium and the sensing element. It can achieve continuous distributed measurement of temperature, strain, sound waves, etc., over a distance of hundreds of kilometers with high density. It has been widely used in many fields such as perimeter security, seismic wave detection, power transmission line monitoring, and underwater cable monitoring.

[0025] This embodiment provides a method for monitoring the verticality of pile foundations along their entire length based on distributed optical fibers, enabling construction quality control and long-term health monitoring of pile foundation projects such as buildings, bridges, and offshore platforms.

[0026] This embodiment provides a method for monitoring the verticality of pile foundations along their entire length based on distributed optical fiber, including a signal modulation algorithm and a verticality inversion model, specifically comprising the following steps: Step 1: Hardware deployment.

[0027] like Figure 3 As shown, according to engineering requirements, a three-dimensional strain sensing network is formed by arranging multi-path optical fibers on the surface of the pile. Specifically, the main sensing optical fiber 1 (with a spacing of 0.5m) is arranged along the axial direction (i.e., vertically on the pile surface), while the auxiliary optical fiber 2, which is spirally wound, is added in key bending moment areas (such as the pile top and pile bottom) to form a three-dimensional strain sensing network, thereby realizing continuous monitoring of verticality along the entire length and solving the problem of monitoring blind spots.

[0028] In this embodiment, the multi-path fiber optic deployment process consists of two parts: helical-axial composite cabling and strain-coupled enhancement structure.

[0029] Among them, the spiral-axial composite wiring involves grooving grooves on the surface of the pile body, with the groove cross-section being V-shaped, embedding optical fibers inside, and filling them with "nano-SiO2 epoxy resin".

[0030] Among them, the strain coupling enhancement structure uses a polyurethane-quartz sand composite coating to wrap the optical fiber. Through the material elastic modulus gradient design, the micro-strain of the pile (με level) is amplified into a phase change (≥0.1rad) that can be measured by the optical fiber.

[0031] like Figure 1 As shown, a light transmitter and receiver are installed at the top of the pile to emit a pulsed laser source and receive the backscattered Rayleigh signal. The laser transmitter emits a pulsed laser source, and the receiver receives the backscattered Rayleigh signal, with a sampling interval of 0.1m.

[0032] The edge computing terminal (data processing terminal) deploys a lightweight ResNet-18 model, outputting a full-length verticality cloud map every 30 seconds, and triggering an early warning when the deviation exceeds 0.1%.

[0033] Step 2: Undersampling optimization.

[0034] Undersampling optimization reduces the sampling rate from the traditional 1GHz to 200MHz using bandpass filtering and quadrature demodulation techniques, reducing the data volume by 80% while maintaining phase resolution (error <0.05°). High-precision phase resolution at low sampling rates is achieved through bandpass filtering and quadrature demodulation.

[0035] Bandpass sampling theorem: If the signal frequency band is located in [fL, fH], the sampling rate fs satisfies fs ≥ 2B (B = fH - fL), which can avoid spectral aliasing. By designing a bandpass filter to suppress out-of-band noise, the sampling rate can be reduced from 1 GHz to 200 MHz, and the amount of data can be reduced by 80%.

[0036] Quadrature demodulation technology: It uses I / Q demodulation to downconvert high-frequency signals to baseband and combines Hilbert transform to extract the phase information of complex signals.

[0037] The phase resolution error is:

[0038] Where N is the number of sampling points, the accuracy requirements are met by improving the signal-to-noise ratio (SNR) and optimizing the number of sampling points.

[0039] The core function of the phase resolution error ΔΦ extracted by orthogonal demodulation is to quantify the ability to detect minute phase changes and directly determine the measurement accuracy of key parameters such as strain and vibration.

[0040] (1) Define sensitivity.

[0041] Physical meaning: ΔΦ is the smallest resolvable phase change (unit: degree or radian), characterizing the detection limit for weak vibrations or deformations.

[0042] Example: If ΔΦ < 0.05°, it means that phase fluctuations as small as 0.05° can be identified, corresponding to a strain resolution of about 0.4με (micro-strain).

[0043] (2) Ensure the reliability of dynamic phase change ΔΦ(z,t).

[0044] ΔΦ(z,t) is the phase difference between position z and time t along the fiber, used to invert strain and vibration events.

[0045] If ΔΦ is too large (e.g., >0.05°), the signal-to-noise ratio (SNR) of ΔΦ(z,t) will decrease, which will affect the accuracy of subsequent calculations.

[0046] (3) Limit the upper limit of the accuracy of key parameters.

[0047] Phase error is passed to downstream parameters through differentiation and integration operations: Accumulated errors can lead to severe distortion of the perpendicularity deviation angle θ(z) or displacement ΔD(z).

[0048] Step 3: Obtain the phase change ΔΦ(z) ).

[0049] ΔΦ(z The expression represents the phase change along the fiber length z at time t. The process for obtaining the Rayleigh scattering signal demodulation based on the Φ-OTDR system is as follows: Laser pulse injection and scattering: A narrow-pulse laser pulse is emitted into the optical fiber, and Rayleigh scattered light along the fiber returns to the detector.

[0050] Coherent detection: The backscattered Rayleigh light is interfered with the reference light by an interferometer (such as a Mach-Zehnder interferometer) to form a time-domain signal.

[0051] Phase demodulation: Phase demodulation algorithms (such as phase difference method and phase tracking algorithm) are used to extract the phase information of the scattered signal. When the pile body is subjected to stress and deformation, the optical fiber is strained, resulting in an optical path difference, which manifests as a phase change ΔΦ(z). ).

[0052] Step 4, Time-Frequency Domain Signal Demodulation Flowchart, the specific process is as follows: Figure 2 As shown, it includes: (1) Input the original phase signal ΔΦ(z) By improving MEEMD decomposition, a noise-reduced signal is output.

[0053] In the time domain, an improved MEEMD decomposition algorithm is used, and adaptive white noise optimization is introduced to separate environmental vibration and pile deformation signals.

[0054] By introducing an auxiliary signal set of paired positive and negative white noise, the mode mixing problem of traditional EMD is suppressed. The statistical characteristics of white noise (zero mean and uniform power spectrum) are adaptively adjusted by adjusting the injection intensity to ensure that the decomposed intrinsic mode functions (IMFs) more accurately reflect the essential characteristics of the signal.

[0055] Adaptive white noise optimization: The time-varying VARMA model is used to dynamically model the decomposed IMF. By real-time estimation of the white noise covariance matrix, the separation efficiency of environmental vibration noise and pile deformation signal is optimized, achieving a theoretical lower limit of ≥30% improvement in signal-to-noise ratio.

[0056] Signal model: Where s(t) is the target signal and n(t) is the environmental noise. The residual noise was separated into independent modal components after MEEMD decomposition.

[0057] (2) Input the processed noise reduction signal into the STFT time spectrum and output the time-frequency diagram.

[0058] STFT time-frequency analysis: By using a window function to perform a Fourier transform on the piecewise signal, the non-stationary signal is mapped to a joint time-frequency distribution, providing a high-resolution input for bending angle prediction. Its mathematical form is: ; Where w(t) is the window function, optimizing the window length and overlap rate to balance time-frequency resolution; X(t,f) is the joint time-frequency distribution (complex matrix); x(τ) is the input time-domain signal (the phase change ΔΦ(z) after noise reduction). w(t-τ) is a window function that determines the local spectral characteristics and suppresses spectral leakage; Δt is the time length of the window function; f is the frequency, and j is the imaginary sign; It is a complex exponential function, representing a rotating phase factor with frequency f.

[0059] (3) Input the time-frequency graph into the ResNet-18 model and output the predicted bending angle. .

[0060] A deep residual network (ResNet-18) is constructed, which takes the short-time Fourier transform (STFT) time spectrum as input and outputs the predicted bending angle. It is trained using experimental data.

[0061] A frequency domain feature extraction framework based on a deep residual network (ResNet-18) is constructed, which combines the time-frequency representation capability of short-time Fourier transform (STFT).

[0062] The ResNet-18 architecture can alleviate the gradient vanishing problem through residual connections. The network input is the amplitude-phase joint feature of the spectrum when STFT is used, and the output layer adopts a fully connected structure to map to the bending angle prediction value.

[0063] The training data for ResNet-18 includes 2000 sets of simulation data (based on the finite element mechanical model) and 500 sets of experimental data to ensure generalization ability.

[0064] The loss function design of ResNet-18 adopts a composite loss function of mean squared error (MSE) and angle error: L=α·MSE(ypred,ytrue)+β·arccos(cos(θpred-θtrue)); In this context, α and β are weighting coefficients, optimizing the absolute error and directional consistency of the angle prediction. The MSE term represents the mean square error, where ypred is the model-predicted bending angle value (scalar or vector), and ytrue is the true label (e.g., inclinometer measurement or simulation baseline). The MSE term penalizes the absolute deviation between the predicted and true values ​​to ensure numerical accuracy. θpred is the model-predicted angle direction (unit: radians or degrees); θtrue is the true angle direction. The angle error term ensures directional consistency (e.g., the pile's tilt direction is "eastward" or "westward") and avoids periodic angle errors (e.g., 359° differs from 1° by 2°, but direct subtraction results in an incorrect difference of 358°).

[0065] The time-domain and frequency-domain processing are optimized in tandem. Time-domain noise reduction provides a high signal-to-noise ratio input for frequency-domain analysis. The target signal separated by MEEMD has more significant frequency-domain features after passing through STFT, which reduces the risk of overfitting in deep learning models.

[0066] Frequency domain features guide the adjustment of time domain parameters. The prediction error output by ResNet can be fed back to the adaptive noise injection module of MEEMD to dynamically adjust the white noise intensity and the number of decomposition layers, thereby achieving iterative optimization.

[0067] Step 5: Verticality inversion model.

[0068] The verticality inversion model employs a multi-physics coupling model, establishing a pile-soil-fiber multi-body coupled finite element model, and considering the Φ-OTDR phase change ΔΦ(z) The mapping is given by the pile curvature κ(z). The perpendicularity deviation angle θ(z) is obtained by integration. )=∫κ(z )dz.

[0069] (1) Phase change ΔΦ(z) ) and strain ε(z) ) relationship.

[0070] ΔΦ(z ) and fiber axial strain ε(z) The relationship is linear, and the formula is:

[0071] Where: n is the refractive index of the fiber core; λ is the laser wavelength; L eff The effective operating length (related to pulse width and fiber characteristics).

[0072] This formula can be used to calculate ΔΦ(z) ) is converted into strain ε(z) distributed along the pile body. ).

[0073] (2) Strain ε(z) ) to curvature κ(z) The mapping of ).

[0074] Curvature κ(z) ) and strain ε(z) The conversion is based on the beam bending theory in mechanics of materials: assuming the pile is an elastic beam, according to the principle of conjugate beams, the pile curvature κ(z) ) and strain ε(z) The relationship is: Where y is the distance from the optical fiber to the neutral axis of the pile. This method is applicable to single-core optical fibers and directly calculates the local curvature through distributed strain, but the position of the neutral axis must be known.

[0075] In multibody coupling models (such as pile-soil-fiber systems or flexible structure dynamics), the local curvature κ is usually geometrically related to the bending angle θ. Combining beam theory and continuum mechanics, the corrected curvature parameter can be expressed as: ; in: The instantaneous bending angle (spatial coordinates z, time t) predicted by ResNet-18. Curvature calculated in the traditional finite element model ; α This is the fusion coefficient (dynamically adjusted through experiments or adaptive algorithms).

[0076] (3) Multibody coupling dynamic equations. Solve the dynamic equations iteratively and output the displacement field u and stress distribution.

[0077] Consider the multibody coupled dynamic equations of the pile-soil-fiber system: ; in: Indicates pile displacement; EI: bending stiffness; F soil : Soil reaction force (modeled using Py or Tz curves); F fiber : Fiber optic sensing inversion force (and phase change ΔΦ(z)) (Related); ρ represents the mass density of the pile material (unit: kg / m³). 3 ( ), which refers to the mass of a pile per unit volume.

[0078] In a pile-soil-fiber coupled system, stress distribution refers to the spatiotemporal evolution of stress (such as axial stress and bending stress) within the cross-section of the pile at different depths z and times t.

[0079] Main stress components: Bending stress σbending (z,t): caused by the pile curvature κ (z,t), which dominates the cross-sectional stress when the pile undergoes lateral deformation; Axial stress σaxial (z,t): caused by the pile axial strain ϵaxial (z,t) (such as pile top load or soil friction).

[0080] Stress distribution is used to assess the structural safety of piles (such as concrete cracking and steel yielding) and to locate potential failure sections.

[0081] Key formula chain: curvature (Beam Theory); Bending strain (y: distance from the neutral axis of the cross section); Bending stress (Hooke's Law); Axial stress (axial strain); Total stress σ(z,t)=σbending(z,t)+σaxial(z,t) (superposition principle); The calculation steps of stress distribution are as follows: Step : separating transverse displacement u_transverse from displacement field u(z,t) and axial displacement u_axial ; Step : obtaining curvature κ(z,t) by taking the second derivative of the transverse displacement; Step : determining the distance from the neutral axis y∈[-R,R] according to the cross-sectional shape (e.g., radius R of a circular pile); Step : substituting the elastic modulus E of the material to calculate the bending stress and axial stress; Step : superimposing stress components and outputting the total stress distribution σ(z,y,t) (the cross-sectional position y needs to be considered).

[0082] In this embodiment, the displacement field u(z) obtained by finite element solution is fed back to the soil reaction model. The purpose of feeding back to the soil reaction model is to realize the dynamic coupling calculation of pile-soil interaction (nonlinear, typically n<1), where Ksoil is the soil stiffness coefficient, u is the displacement of the pile shaft, n is a dimensionless parameter describing the nonlinear degree of the relationship between soil reaction and displacement, generally with a value range of 0<n<1 (the specific value is determined by the soil type and stress state). Necessity of dynamic feedback: the soil stiffness will soften (e.g., sand) or harden (e.g., clay) as the displacement increases, and Fsoil needs to be updated in real time to reflect the actual soil-pile interaction.

[0083] The fusion coefficient α is adjusted by back propagation to make κ(z,t) more adapted to actual working conditions.

[0084] (4) Calculate the verticality deviation angle θ(z )= . Wherein, L represents the pile length.

[0085] (5) Calculate the offset ΔD(z ). The horizontal displacement corresponding to the verticality deviation angle is: . It is required to include the real-time displacement curve and the historical cumulative displacement trend.

[0086] (6) Verification step: compare the prediction error (e.g., root mean square error of displacement RMSE) of the traditional finite element model and the ResNet correction model to verify the effectiveness of curvature correction.

[0087] Step 6: Output the verticality inversion model.

[0088] Step 601, Geometric deformation parameters of the pile body: curvature distribution κ(z) ), perpendicularity deviation angle θ (z ), offset ΔD(z) ).

[0089] Step 602, Physical field distribution parameters.

[0090] Strain field distribution ε(z) ): The raw strain data monitored by optical fiber, after temperature compensation, is the net strain value, with a spatial resolution of ±0.5m.

[0091] Stress field distribution σ(z) ): Local stress calculated based on the elastic modulus (E) and Poisson's ratio (ν) of the pile material, outputting the extreme values ​​of tensile / compressive stress and the location of the critical section.

[0092] Extreme values ​​of tensile / compressive stress: Maximum tensile stress: σmax = σaxial + E·R·κmax (occurs at the edge of the section where y = -R); Maximum compressive stress: σmin = σaxial - E·R·κmax (occurs at y = R). Critical section location report: Mark the depth and time when σmax exceeds the limit.

[0093] Pile-soil interaction forces: The horizontal resistance (py curve parameters) and frictional force (tz curve parameters) of the soil around the pile obtained by inversion are used to evaluate the soil constraint effect.

[0094] Step 603: Environmental compensation and disturbance correction.

[0095] Temperature compensation results: Temperature field distribution data separated by dual-core optical fibers are combined with differential algorithms to eliminate spurious strain caused by thermal expansion.

[0096] Noise suppression metrics: Signal-to-noise ratio (SNR) after dynamic filtering, and vibration interference spectrum analysis results (such as the frequency range of interference from construction machinery).

[0097] Step 604: Model Validation and Error Analysis.

[0098] Benchmark comparison data: the difference Δθ(z) between the measured verticality of the traditional inclinometer and the model inversion result, output error distribution map and root mean square error (RMSE).

[0099] Confidence assessment: based on the model uncertainty interval (e.g., 95% confidence band) from Monte Carlo simulation, or the posterior probability distribution derived from Bayesian inversion.

[0100] Step 605: Visualization and Diagnostic Information.

[0101] 3D Deformation Cloud Map: Dynamic 3D rendering of the spatial deformation of the pile, supporting layered display of curvature and displacement at different depths.

[0102] Real-time warning signal: Alarm information triggered by a preset threshold (e.g., ΔD(z)>10 mm / m), including the offset direction and emergency handling suggestions.

[0103] Step 606: Output auxiliary parameters.

[0104] Soil layering characteristics: The soil layer elastic modulus (Es) and cohesion (c) optimized synchronously during the inversion process are used to verify the rationality of the soil-pile coupling model.

[0105] Dynamic time series data: verticality change rate (dθ / dt) and periodic vibration response (such as frequency-amplitude spectrum) under high-frequency monitoring.

[0106] The output of the verticality inversion model must encompass four core dimensions: geometric deformation parameters (curvature, deviation angle, displacement), physical field data (strain, stress), environmental correction results (temperature, noise), and validation metrics (error analysis, confidence level). It should also provide visualization and early warning support. In practical applications, dynamic calibration (such as static load tests) is needed to continuously optimize model parameters, and multi-sensor fusion (such as inclinometers and strain gauges) should be used to improve the reliability of the output results.

[0107] This embodiment provides a method for monitoring the verticality of a pile foundation along its entire length based on distributed optical fiber. Fiber optic sensors are deployed along the pile body, and Φ-OTDR technology is used to capture the phase changes of the optical signal caused by minute deformations of the pile body. By inverting the bending angle of the pile body through time-domain reflectometry data, continuous monitoring of the verticality along the entire length can be achieved, solving the problem of monitoring blind spots.

[0108] This embodiment provides a method for monitoring the verticality of pile foundations along their entire length based on distributed optical fibers. It supports monitoring throughout the entire life cycle, continuously recording data from the pile driving stage, and supports construction correction and service life health assessment.

[0109] This embodiment provides a method for monitoring the verticality of pile foundations along their entire length based on distributed optical fiber. It has strong anti-interference capabilities, and the passive characteristics of optical fiber sensing enable it to work stably in environments with strong electromagnetic fields and high humidity, ensuring the accuracy of monitoring results.

[0110] Example 2 This embodiment provides a pile foundation full-length verticality monitoring system based on distributed optical fiber, which specifically includes: The signal acquisition module is configured to: transmit pulsed laser source and receive backscattered Rayleigh signal through multipath optical fiber deployed on the surface of pile body, and convert backscattered Rayleigh signal into phase change; The prediction module is configured to: map the phase change to a joint time-frequency distribution to obtain a time-frequency map, and obtain the predicted bending angle value through a deep residual network; The calculation module is configured to: map the phase change to the pile curvature, correct the pile curvature with the predicted bending angle, and integrate the corrected pile curvature to obtain the verticality deviation angle.

[0111] It should be noted that each module in this embodiment corresponds one-to-one with each step in Embodiment 1, and their specific implementation processes are the same, so they will not be repeated here.

[0112] Example 3 This embodiment provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in Embodiment 1 above for monitoring the verticality of a pile foundation along its entire length.

[0113] Example 4 This embodiment provides a computer device, such as... Figure 4 As shown, the system includes a computer-readable storage medium 1003, a processor 1001, a communication interface 1002, and a computer program stored on the computer-readable storage medium 1003 and executable on the processor 1001. The processor 1001, communication interface 1002, and computer-readable storage medium 1003 can be connected via a bus or other means. The communication interface 1002 is used to receive and transmit data. When the processor 1001 executes the program, it implements the steps described in Embodiment 1 above for monitoring the verticality of a pile foundation along its entire length using distributed optical fiber.

[0114] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0115] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for monitoring the verticality of pile foundations along their entire length based on distributed optical fibers, characterized in that, include: A pulsed laser source is emitted and a backscattered Rayleigh signal is received by a multipath optical fiber deployed on the surface of the pile, and the backscattered Rayleigh signal is converted into a phase change. The phase change is mapped to a joint time-frequency distribution to obtain a time-frequency map. The bending angle is then predicted by a deep residual network. The phase change is mapped to the pile curvature, and the pile curvature is corrected by the predicted bending angle. The corrected pile curvature is then integrated to obtain the verticality deviation angle. The step of mapping the phase change to pile curvature includes: converting the phase change into fiber axial strain, and then, based on the conjugate beam principle, converting the fiber axial strain into pile curvature; the corrected pile curvature is: ； Where: z is the spatial coordinate, and t is time. This is the predicted value for the bending angle. The curvature of the pile body before correction is α, and the fusion coefficient is α.

2. The method for monitoring the verticality of a pile foundation along its entire length based on distributed optical fiber as described in claim 1, characterized in that, Also includes: Based on the corrected pile curvature, the displacement field is obtained by iteratively solving the multibody coupled dynamic equations.

3. The method for monitoring the verticality of a pile foundation along its entire length based on distributed optical fiber as described in claim 1, characterized in that, The multipath optical fiber includes a main sensing optical fiber arranged along the axial direction and auxiliary optical fibers spirally wound at the top and bottom of the pile.

4. The method for monitoring the verticality of a pile foundation along its entire length based on distributed optical fiber as described in claim 1, characterized in that, The multi-path fiber deployment process consists of two parts: helical-axial composite wiring and strain-coupled enhancement structure.

5. The method for monitoring the verticality of a pile foundation along its entire length based on distributed optical fiber as described in claim 1, characterized in that, By using a window function to perform a Fourier transform on the segmented signal, the phase change is mapped to a joint time-frequency distribution.

6. A pile foundation full-length verticality monitoring system based on distributed optical fiber, characterized in that, include: The signal acquisition module is configured to: transmit pulsed laser source and receive backscattered Rayleigh signal through multipath optical fiber deployed on the surface of pile body, and convert backscattered Rayleigh signal into phase change; The prediction module is configured to: map the phase change to a joint time-frequency distribution to obtain a time-frequency map, and obtain the predicted bending angle value through a deep residual network; The calculation module is configured to: map the phase change to the pile curvature, correct the pile curvature with the predicted bending angle, and integrate the corrected pile curvature to obtain the verticality deviation angle; The step of mapping the phase change to pile curvature includes: converting the phase change into fiber axial strain, and then, based on the conjugate beam principle, converting the fiber axial strain into pile curvature; the corrected pile curvature is: ； Where: z is the spatial coordinate, and t is time. This is the predicted value for the bending angle. The curvature of the pile body before correction is α, and the fusion coefficient is α.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps in the full-length verticality monitoring of pile foundations based on distributed optical fiber as described in any one of claims 1-5.

8. A computer device comprising a computer-readable storage medium, a processor, and a computer program stored on the computer-readable storage medium and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the full-length verticality monitoring of pile foundations based on distributed optical fiber as described in any one of claims 1-5.

Images