Distributed optical fiber sound sensing highway abnormal vibration real-time early warning system

Abstract

The invention discloses a distributed optical fiber acoustic sensing highway abnormal vibration real-time early warning system, and relates to the technical field of highway monitoring. A vibration analysis and mechanism triggering module calculates the speed of vibration amplitude changing along with time, if the change speed exceeds a set threshold value, abnormal fluctuation is marked, the maximum deviation and the minimum deviation of a vibration signal are calculated, and the maximum deviation and the minimum deviation of the vibration signal are calculated; if the deviation values exceed a preset range, an early warning program is started, an early warning strategy output module constructs a material flow state stability index based on the speed of the vibration amplitude changing along with time, creates a construction area uniformity index in combination with a vibration propagation track, analyzes the coordination degree between the vibration behavior and the road construction state according to a regulation and control coordination coefficient, and outputs early warning information. And a corresponding vibration early warning strategy is generated based on the analysis result. The early warning system realizes accurate and operable abnormal vibration real-time early warning through distributed optical fiber sensing, dynamic feature extraction, multi-dimensional index construction and intelligent model analysis.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of highway monitoring, in particular to a distributed optical fiber acoustic sensing highway abnormal vibration real-time early warning system. BACKGROUND

[0002] Highway is a special road system for fast-flowing traffic, usually with high design standards and fast traffic speed. With the increase of traffic volume, especially the frequent passing of large vehicles on the highway, abnormal vibration and noise problems become important factors for road safety and maintenance. Efficient monitoring and early warning systems are crucial for improving road safety and preventing potential accidents.

[0003] Early warning system is a system that can issue warnings in advance of or at the early stage of an event. This system usually relies on sensor data, which is analyzed and processed to provide real-time feedback to relevant personnel or systems to take appropriate measures. In the highway, real-time early warning system can help monitor the structural health status of the road, traffic flow, etc. Through abnormal vibration signals, it can quickly alarm and reduce the possibility of accidents, improving road safety.

[0004] The prior art has the following defects:

[0005] 1. The existing early warning system uses fixed threshold to judge vibration anomaly, only focuses on whether the vibration amplitude exceeds the static set value, and ignores the dynamic characteristics of vibration signals changing over time (such as change rate) and the path anomaly of vibration in the process of spatial propagation (such as propagation trajectory deviation or energy attenuation anomaly), resulting in insufficient sensitivity to early weak but potentially harmful abnormal vibration (such as underground cavity collapse, structure damage initiation, heavy vehicle illegal driving, etc.).

[0006] 2. The early warning strategy often directly outputs simple alarms (such as "abnormal vibration, please check") based on experience rules, lacks quantitative evaluation of vibration source stability (such as roadbed material flow state characteristics) and propagation environment uniformity (such as construction area structure consistency), and cannot accurately associate vibration behavior with actual construction state of highway (such as maintenance stage, construction interference, etc.), resulting in poor pertinence and operability of early warning results, making it difficult to effectively guide subsequent emergency handling or maintenance decision.

[0007] Therefore, the present application proposes a distributed optical fiber acoustic sensing highway abnormal vibration real-time early warning system, which realizes accurate and operable real-time early warning of abnormal vibration through distributed optical fiber sensing, dynamic feature extraction, multi-dimensional index construction and intelligent model analysis. SUMMARY

[0008] The invention aims to provide a distributed optical fiber acoustic sensing highway abnormal vibration real-time early warning system to solve the problems in the background art.

[0009] To achieve the above-mentioned purpose, the invention provides the following technical solution: a distributed optical fiber acoustic sensing highway abnormal vibration real-time early warning system, comprising a data acquisition module, a vibration analysis and mechanism triggering module, and an early warning strategy output module.

[0010] The data acquisition module: real-time acquisition of road surface vibration data by optical fiber sensor, including vibration signal amplitude fluctuation curve and vibration transmission trajectory.

[0011] The vibration analysis and mechanism triggering module: change rate analysis of the vibration signal amplitude fluctuation curve, calculation of the rate of change of the vibration amplitude with time, if the change rate exceeds the set threshold, it is marked as abnormal fluctuation, through real-time analysis of the vibration transmission trajectory, identification of abnormal path or deviation in the signal propagation process, and calculation of the maximum deviation and minimum deviation of the vibration signal, if these deviation values exceed the preset range, the early warning program is started.

[0012] The early warning strategy output module: based on the rate of change of the vibration amplitude with time, a material flow state stability index is constructed, the stability of the vibration signal in different sections of the highway is evaluated, the structural stability of the vibration source is reflected, combined with the vibration propagation trajectory, a construction area uniformity index is created, whether there is unevenness in the vibration signal propagation process is evaluated, the flow state stability index and the construction area uniformity index are taken as input data, and a pre-trained deep learning model is imported, a regulation and coordination coefficient is calculated, the coordination degree between the vibration behavior and the highway construction state is analyzed according to the regulation and coordination coefficient, and the corresponding vibration early warning strategy is generated based on the analysis result.

[0013] In a preferred embodiment, the vibration analysis and mechanism triggering module performs denoising processing on the original vibration signal amplitude fluctuation curve collected by the optical fiber sensor, and then segments the continuous sampling points according to a fixed time window to form a discrete time series data set.

[0014] The time series is traversed in sequence, for each time, the amplitude of the previous adjacent time is extracted, and the vibration amplitude change rate at that time is calculated by difference operation.

[0015] The distribution characteristics of the vibration amplitude change rate are statistically calculated based on historical normal working condition data in advance, and the change rate R i of each point calculated in real time is compared with the threshold T R point by point.

[0016] If any R i >T R , the time is marked as an abnormal fluctuation point, and the corresponding time stamp t is recorded.i , the rate of change value R i , and the amplitude sequence of the front and rear periods.

[0017] In a preferred embodiment, the vibration analysis and mechanism triggering module binds the collected vibration signals with the spatial positions of the corresponding sensors based on the spatial layout positions of the optical fiber sensors, to form a mapping dataset of vibration signals-spatial positions;

[0018] Through spatio-temporal correlation analysis of multi-sensor data, the propagation trajectory L of the vibration signal is reconstructed, and based on normal working condition data, the real-time reconstructed propagation trajectory L is spatially aligned with the ideal trajectory model and the deviation is calculated.

[0019] In a preferred embodiment, the vibration analysis and mechanism triggering module compares the calculated local maximum deviation and local minimum deviation with the preset maximum deviation threshold and minimum deviation threshold, and if the real-time calculated local maximum deviation is greater than the maximum deviation threshold or the local minimum deviation is greater than the minimum deviation threshold, it is determined that the propagation trajectory is abnormal, and the abnormal type, key node position and deviation value are recorded.

[0020] In a preferred embodiment, the logic of the real-time reconstructed propagation trajectory L and the ideal trajectory model is spatially aligned and the deviation is calculated is:

[0021] For the nodes on the trajectory, the Euclidean distance between the actual spatial coordinates and the ideal trajectory corresponding node coordinates is calculated to obtain the local deviation value Δd k ;

[0022] The maximum value Δd k and the minimum value Δd max are extracted from all local deviation values Δd min , and the mean value Δd mean of the overall deviation is calculated to comprehensively evaluate the propagation stability.

[0023] In a preferred embodiment, the pre-warning strategy output module standardizes the material flow state stability index I flow and the construction area uniformity index I uniform to a value in the interval [0, 1];

[0024] An optimized deep neural network model is used, the input layer of the deep neural network model receives X, the non-linear features are extracted through the hidden layer, and the regulation and coordination coefficient C is generated by the output layer;

[0025] According to the value of C, the coordination level is divided:

[0026] If C is in [0.8, 1.0], it indicates high coordination; if C is in [0.5, 0.8), it indicates moderate coordination; if C is in [0.2, 0.5), it indicates low coordination; and if C is in [0, 0.2), it indicates extremely low coordination.

[0027] In a preferred embodiment, the construction and calculation logic of the material flow state stability index is as follows:

[0028] A sequence of rate of change of vibration amplitude over time is received, and the corresponding highway space segment identifier is synchronously associated with the sequence;

[0029] Statistical analysis is performed on the rate sequence to extract feature parameters, including rate mean, rate standard deviation, and proportion of mutation points, and a material flow state stability index is generated through hierarchical weighting logic.

[0030] In a preferred embodiment, the construction and calculation logic of the construction area uniformity index is as follows:

[0031] Vibration transmission trajectory feature data, including point sequences, are received, and multi-dimensional analysis is performed on the propagation trajectory and point parameters to extract path linearity deviation, local propagation speed dispersion, and amplitude decay stability, and a construction area uniformity index is generated through hierarchical logic.

[0032] In a preferred embodiment, the data acquisition module captures the fluctuation curve of vibration amplitude over time captured by the optical fiber sensor, reflecting the dynamic fluctuation of vibration intensity and frequency.

[0033] In a preferred embodiment, the data acquisition module monitors the transmission path of the vibration signal at each point on the road, focuses on the time delay and propagation pattern of the signal during the propagation process in different road segments, and extracts the time and space features of the key points.

[0034] In the above technical solution, the technical effects and advantages provided by the present application are as follows:

[0035] The application realizes continuous and high-precision collection of pavement vibration data (including amplitude fluctuation curve and propagation trajectory) through a distributed optical fiber sensor, combines dynamic change rate analysis and spatial trajectory offset detection, can more sensitively identify early abnormal vibration (such as small but rapidly changing vibration or abnormal propagation path), and solves the problem of high missed detection rate of the traditional static threshold method; by constructing a "material flow state stability index" and a "construction area uniformity index", the structural stability of the vibration source and the uniformity of the propagation environment are quantified, a "regulation and coordination coefficient" is calculated by combining a pre-trained deep learning model, precise correlation analysis of vibration behavior and highway construction state is realized, the generated early warning strategy is more targeted (such as distinguishing roadbed damage early warning, construction interference early warning, etc.), the accuracy and practicality of early warning are significantly improved, and more reliable technical support is provided for real-time safety monitoring and intelligent operation and maintenance of expressways. BRIEF DESCRIPTION OF DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments described in the present application, and other drawings can also be obtained by those skilled in the art based on these drawings.

[0037] Figure 1 The figure is a schematic diagram of the early warning system of the present application. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0039] Embodiment:

[0040] Please refer to Figure 1 The present embodiment provides a distributed optical fiber acoustic sensing expressway abnormal vibration real-time early warning system, which comprises a data acquisition module, a vibration analysis and mechanism triggering module, and an early warning strategy output module.

[0041] Data acquisition module: Real-time collection of road vibration data through optical fiber sensors. The collected data can be divided into two categories: 1. Vibration signal amplitude fluctuation curve: Collecting the fluctuation curve of the vibration amplitude captured by the optical fiber sensor over time, reflecting the dynamic fluctuation of vibration intensity and frequency. 2. Vibration transmission trajectory: Monitoring the transmission path of vibration signals at various points on the road, paying special attention to the time delay and propagation pattern during signal propagation in different road sections, extracting the time and space characteristics of key points, and sending the collected data to the vibration analysis and mechanism triggering module and the early warning strategy output module.

[0042] Vibration analysis and mechanism triggering module: Change rate analysis of the vibration signal amplitude fluctuation curve, calculating the rate of change of the vibration amplitude over time. If the change rate exceeds the set threshold, it is marked as abnormal fluctuation. Through real-time analysis of the vibration transmission trajectory, identify abnormal paths or deviations in the signal propagation process, and calculate the maximum and minimum deviations of the vibration signal. If these deviation values exceed the preset range, start the further warning program. The rate of change of the vibration amplitude over time is sent to the early warning strategy output module.

[0043] Early warning strategy output module: Based on the rate of change of the vibration amplitude over time, construct a material flow state stability index to evaluate the stability of the vibration signal in different sections of the highway, reflecting the structural stability of the vibration source. Combined with the vibration propagation trajectory, create a construction area uniformity index to evaluate whether there is unevenness in the vibration signal propagation process, further evaluate the flatness and quality of road construction. The flow state stability index and the construction area uniformity index are used as input data to import a deep learning model that has been trained for a long time. Calculate the regulation and coordination coefficient. Analyze the coordination degree between vibration behavior and highway construction state based on the regulation and coordination coefficient, and generate the corresponding vibration early warning strategy based on the analysis results.

[0044] The present application realizes continuous and high-precision collection of road vibration data (including amplitude fluctuation curve and propagation trajectory) through distributed optical fiber sensors, combines dynamic change rate analysis and spatial trajectory deviation detection, and can more sensitively identify early abnormal vibration (such as small but rapidly changing vibration or abnormal propagation path), solving the problem of high missed detection rate of traditional static threshold method. By constructing the "material flow state stability index" and the "construction area uniformity index", the structural stability of the vibration source and the uniformity of the propagation environment are quantified, combined with the pre-trained deep learning model to calculate the "regulation and coordination coefficient", realizing the accurate correlation analysis of vibration behavior and highway construction state, and generating early warning strategies that are more targeted (such as distinguishing between roadbed damage warning and construction interference warning), significantly improving the accuracy and practicality of early warning, and providing more reliable technical support for real-time safety monitoring and intelligent operation and maintenance of highways.

[0045] Specifically, the specific workflow of the early warning system is as follows:

[0046] S1: Real-time vibration data collection and dynamic monitoring

[0047] Deploy distributed optical fiber sensing systems on key sections of the highway, collect real-time road vibration data through optical fiber sensors, and divide them into two categories:

[0048] One is the vibration signal amplitude fluctuation curve: collect the fluctuation curve of the vibration amplitude captured by the optical fiber sensor over time, reflecting the dynamic fluctuation of vibration intensity and frequency.

[0049] One is the vibration transmission trajectory: monitor the transmission path of the vibration signal at each point on the road, and pay special attention to the time delay and propagation pattern of the signal during transmission in different sections, and extract the time and space characteristics of key points.

[0050] S2: Abnormal vibration identification and response triggering mechanism

[0051] Vibration change rate analysis: analyze the change rate of the vibration signal amplitude fluctuation curve, calculate the rate of change of the vibration amplitude over time, and if the change rate exceeds the set threshold, the system will be marked as abnormal fluctuation.

[0052] Vibration propagation path comparison and trend analysis: through real-time analysis of the vibration transmission trajectory, identify abnormal paths or deviations in the signal propagation process, and calculate the maximum and minimum deviations of the vibration signal. If these deviation values exceed the preset range, start the further warning program.

[0053] S3: Vibration pattern prediction and intelligent control

[0054] Material flow stability index construction: based on the rate of change of vibration amplitude over time, construct the material flow stability index to evaluate the stability of vibration signals in different sections of the highway, reflecting the structural stability of the vibration source.

[0055] Construction area uniformity index analysis: combined with the vibration propagation trajectory, create the construction area uniformity index to evaluate whether there is unevenness in the vibration signal propagation process, and further evaluate the flatness and quality of road construction.

[0056] Deep learning model optimization prediction: input the flow stability index and construction area uniformity index as input data into the deep learning model optimized through long-term training, calculate the control coordination coefficient. Analyze the coordination degree between vibration behavior and highway construction state based on the control coordination coefficient, and generate the corresponding vibration warning strategy based on the analysis results.

[0057] The data acquisition module collects real-time road vibration data through optical fiber sensors. The collected data is divided into two categories: one is the vibration signal amplitude fluctuation curve, which reflects the dynamic fluctuation of vibration intensity and frequency by collecting the fluctuation curve of the vibration amplitude captured by the optical fiber sensor over time. The other is the vibration transmission trajectory, which monitors the transmission path of the vibration signal at each point on the road, especially the time delay and propagation pattern of the signal during transmission in different road sections. The time and space characteristics of key points are extracted, and the collected data is sent to the vibration analysis and mechanism triggering module and the early warning strategy output module.

[0058] The data acquisition module is the front-end core of the distributed optical fiber acoustic sensing highway abnormal vibration real-time early warning system. Its function is to capture multi-dimensional physical information of road vibration in real time through the optical fiber sensor network and convert it into structured data stream for analysis. The acquisition objects of this module are clearly divided into two categories of key data: vibration signal amplitude fluctuation curve (reflecting the time-domain dynamic characteristics of vibration intensity and frequency) and vibration transmission trajectory (describing the path characteristics and transmission rules of vibration energy in the process of spatial propagation). After synchronous acquisition, preprocessing and feature extraction, the two types of data are sent to the vibration analysis and mechanism triggering module (for dynamic identification of abnormal fluctuations) and the early warning strategy output module (for construction of stability index), respectively. The acquisition and processing process needs to strictly ensure the spatio-temporal synchronization and feature integrity.

[0059] The core objective of this type of data is to record the continuous fluctuation characteristics of the vibration amplitude over time, and its dynamic characteristics (such as sudden peak value, frequency shift) can be directly related to the intensity and occurrence time of local disturbance events on the road surface (such as vehicle sudden braking, pothole impact):

[0060] Optical fiber sensors (such as distributed acoustic sensing (DAS) systems) are continuously laid along the highway lane below or beside the guardrail, with a spatial resolution of meters (e.g., one virtual sensing unit per 1 meter). Through the interference measurement principle, the small changes in the refractive index of the optical fiber caused by external vibration are converted into optical phase modulation signals, which are then demodulated to output the vibration amplitude at the corresponding position. Each virtual sensing unit captures the vibration signal at its monitoring point in real time, outputs the original electrical signal (usually in volts or digital quantization values), and forms a discretized amplitude sequence after analog-to-digital conversion (ADC). For each virtual sensing unit, the system continuously collects vibration amplitudes at a fixed sampling frequency (such as 1 kHz or higher, ensuring that high-frequency vibration components can be captured), generating a time series data set in chronological order where t n is the nth sampling time (time stamp accurate to the millisecond level), A nis the amplitude value (physical dimension after calibration, such as mV or normalized unit) corresponding to this moment. The continuous sampling points are smoothed (for example, moving average or low-pass filtering) by a sliding window (such as a 1-second window). High-frequency noise interference (such as slight jitter caused by environmental wind noise) is eliminated, and finally the amplitude fluctuation curve f(t) reflecting the dynamic fluctuation of vibration intensity and frequency is generated (i.e. A n With t n continuous representation) changes.

[0061] The amplitude fluctuation curve of each sensing unit is accompanied by its spatial coordinate information (determined by physical positioning when the optical fiber is laid out, such as kilometer stake number + lateral offset), ensuring that the specific occurrence position of the vibration event can be associated in subsequent analysis. At the same time, the data collection clock of all sensing units is synchronized through GPS or optical fiber time synchronization protocol (time synchronization accuracy better than 1 microsecond), ensuring the time sequence comparability of multi-point data (for example, used for subsequent calculation of vibration propagation time delay).

[0062] Collection and processing of vibration transmission trajectory:

[0063] This type of data focuses on the transmission path characteristics of the vibration signal between the monitoring points on the road. By analyzing the time difference of the signal arriving at different positions, the energy attenuation mode and the spatial propagation direction, key information such as the position of the vibration source and the uniformity of the propagation medium is extracted. Its processing logic needs to focus on the two dimensions of "time delay" and "propagation mode":

[0064] When a point on the road is disturbed (such as vehicle tire impact) to produce vibration, the vibration will propagate in the form of elastic wave along the road structure (such as asphalt layer, base layer) to the surrounding, and multiple virtual sensing units in the optical fiber sensor network will detect the vibration signal successively. The system records the moment t . (j represents the number of the jth sensing unit) when each sensing unit first detects the effective vibration signal through a high-precision time synchronization mechanism (such as the aforementioned GPS / optical fiber time synchronization), and combines the spatial coordinates (x . ,y . 3 of the unit to form a "signal arrival time-spatial position" data set (M is the total number of sensing units participating in analysis).

[0065] Taking the most possible initial position of the vibration source (such as the excitation point preliminarily estimated by grid search or least squares method) as a reference, the time delay At .k of the vibration signal arriving at different sensing units is calculated k -t .(j, k are any two sensing units). According to the theoretical propagation velocity v of vibration in the pavement medium (which can be pre-determined by calibration experiment, for example, the compression wave velocity of common asphalt pavement is about 1000-1500 m / s), the time delay is converted into the spatial propagation distance difference .k = v · At .k . Through the geometric constraints of multiple sets of 0 .k , Ax .k , Ay .k 3 (Ax .k = x k - x . , Ay .k = y k - y . ), the actual propagation trajectory L of the vibration signal is reconstructed (for example, the propagation path curve is fitted by using the triangular positioning method or the ray tracing algorithm).

[0066] On the reconstructed propagation trajectory, representative key points (such as the starting point of the propagation path (approximate position of the vibration source), the relay turning point (such as encountering the layered interface of the roadbed), and the terminal point (position of the remote sensing unit)) are selected, and their time and space characteristic parameters are extracted:

[0067] Time characteristics: signal arrival time t ke； of the key point, cumulative propagation time t <um relative to the initial point.

[0068] Spatial characteristics: geographical coordinates 0 ke； x, y ke； 3 of the key point, spatial interval As of the adjacent key points.

[0069] Propagation mode characteristics: local propagation velocity v lo<al = As / At (reflecting the uniformity of the medium, if v lo<al significantly deviates from the theoretical value v, there may be unevenness or hidden defects in the roadbed).

[0070] Energy characteristics (optional): if the sensor supports amplitude attenuation monitoring, the vibration amplitude A ke； of the key point is extracted, and the attenuation rate a = A k - A . 3 / As between adjacent points is calculated (used to assist in judging the abnormal energy loss in the propagation process).

[0071] In the data collection phase, the system will establish a "typical propagation pattern library" based on historical normal operating condition data (such as the propagation trajectory template of a road in good condition without construction interference). The real-time collected propagation trajectory data will be quickly compared with the template (for example, similarity is calculated by dynamic time warping (DTW) algorithm), and if significant deviation is found (such as unexpected corner in the propagation path, local propagation speed drops), it will be marked as "potential abnormal trajectory" and further analyzed in the subsequent module.

[0072] After the above processing of the two types of collected data (amplitude fluctuation curve and transmission trajectory), they need to be transmitted in real time to the downstream module in a standardized format:

[0073] Vibration signal amplitude fluctuation curve: sent in the form of time series data packet, containing sensor unit number, spatial coordinates, sampling time stamp, amplitude sequence {(t n ,A n )}, and optional pre-processing parameters (such as filter window size, calibration coefficient);

[0074] Vibration transmission trajectory: sent in the form of path feature data packet, containing key point sequence {0x ke； ,y ke； ,t ke； ,v lo<al ,A ke； 3}, reconstructed propagation trajectory geometric parameters (such as path curvature, total propagation time), abnormality screening label (normal / potential abnormal). The transmission of the two types of data needs to strictly ensure the timing consistency (for example, by associating the amplitude and propagation information of the same vibration event through a unified time stamp), and through a high-speed communication network (such as optical fiber Ethernet or 5G private network) to realize low-latency (millisecond level) transmission, to ensure that the vibration analysis and mechanism triggering module can obtain the original data in time, and the early warning strategy output module can receive the key parameters for subsequent analysis.

[0075] Suppose a fiber sensor network on a certain section of a highway detects a vibration event at a certain time (t = 12:00:00.000) - virtual sensor unit S1 (coordinates (0, 0)) first detects a vibration with amplitude A1 = 50mV at t1 = 12:00:00.000, adjacent unit S2 (coordinates (10, 0)) detects amplitude A A = 45mV at t A = 12:00:00.003, S3 (coordinates (20, 0)) detects amplitude A D = 40mV at t D = 12:00:00.006. The system first constructs the amplitude fluctuation curve of S1 (records the 50mV peak at t1 and the subsequent decay process), and calculates the time delay Δt 1A= 0.003 s, At 1D = 0.006 s, combined with the theoretical wave speed v = 1200 m / s, resulting in Ad 1A = 1200 x 0.003 = 3.6 m (close to the actual spatial interval of 10 m in the local propagation section, verifying the reasonableness), the reconstructed propagation trajectory shows that the vibration propagates along a straight line, but the local speed v lo<al = (20-10) / (0.006-0.003) = 3333 m / s (significantly higher than the theoretical value), the system marks this section as "abnormal propagation mode" and sends the amplitude curve (the amplitude of S1-S3 changes with time) and the trajectory characteristics (key point coordinates, speed deviation) to the downstream module in synchronization. Through the above process, the data acquisition module realizes the full-dimensional perception of the "time-domain intensity dynamics" and "spatial propagation path" of the road vibration, laying a data foundation for the system's high-precision anomaly recognition and intelligent early warning.

[0076] The vibration analysis and mechanism triggering module analyzes the rate of change of the vibration signal amplitude fluctuation curve, calculates the rate of change of the vibration amplitude with time, and marks it as an abnormal fluctuation if the rate of change exceeds the set threshold. Through real-time analysis of the vibration transmission trajectory, abnormal paths or deviations in the signal propagation process are identified, and the maximum and minimum deviations of the vibration signal are calculated. If these deviation values exceed the preset range, further warning procedures are started, and the rate of change of the vibration amplitude with time is sent to the warning strategy output module.

[0077] The core function of the vibration analysis and mechanism triggering module is to extract dynamic features and detect spatial propagation anomalies from the collected vibration signals, identify potential abnormal vibration events through multi-dimensional analysis, and provide key parameter input for subsequent warning strategies. The processing flow of this module can be divided into two parallel and related analysis submodules: the vibration signal time-domain dynamic feature analysis submodule (for the rate of change of the vibration amplitude with time) and the vibration signal spatial propagation feature analysis submodule (for the abnormal path / shift of the vibration transmission trajectory), ultimately triggering the warning program through threshold judgment logic and passing the key parameters to the warning strategy output module.

[0078] Capture the non-steady-state characteristics of the vibration signal in the time dimension, i.e., the speed of change of the vibration amplitude with time (change rate), such characteristics are usually strongly related to sudden disturbances (such as local road collapse, impact load generated by heavy vehicle sudden braking), and traditional static threshold method is prone to miss such events due to neglecting dynamic process. The specific processing logic is as follows:

[0079] First, the original vibration signal amplitude fluctuation curve collected by the optical fiber sensor is denoised (for example, using sliding average filtering or wavelet threshold denoising to eliminate environmental low-frequency noise interference) to retain effective vibration information; then, the continuous sampling points are segmented according to a fixed time window (for example, 0.1 seconds or adjusted according to the sampling frequency) to form a discrete time series data set where t i is the i-th sampling time, A i is the vibration amplitude corresponding to the time.

[0080] For each time window, the instantaneous change rate of the vibration amplitude is calculated for the continuous sampling points (usually 3-5 adjacent points are taken to balance the sensitivity and noise resistance). The processing logic is as follows: the time series is traversed in sequence, and the amplitude of the previous adjacent time (such as A iI1 ,A iIA , etc.) is extracted for each time t i (i≥2), and the change rate R i of the vibration amplitude of the point is calculated by difference operation or sliding window fitting slope method (unit: amplitude unit / second). For example, if the adjacent two-point difference method is used, then R i =|A i -A iI1 | / (t i -t iI1 ); if the three-point sliding window linear fitting is used, the amplitudes A iI1 ,A i ,A iK1 corresponding to the times t iI1 ,t i ,t iK1 are fitted by least squares method to obtain the fitting straight line slope as the estimated value of R i .

[0081] The distribution characteristics of the vibration amplitude change rate are statistically calculated based on historical normal operating condition data (such as vibration signals of a highway without construction interference and with a perfect road surface) (for example, the 95% or 99% quantile is calculated as the reference threshold T R , or the reasonable rate range of different road sections is dynamically learned by a machine learning model), and the real-time calculated change rate R i of each point is compared with the threshold T R point by point. If there is any R i >T R (i.e., the vibration amplitude at a certain time suddenly changes significantly in a very short time), the time is marked as an “abnormal fluctuation point”, and the corresponding time stamp t i , change rate value R i and amplitude sequence of the previous and subsequent time periods (for subsequent correlation analysis) are recorded.

[0082] When the vibration radiates from the excitation source (such as local disturbance point on the road surface) to the outside, its energy distribution and propagation path should comply with the expected rules of normal road conditions (for example, vibration should propagate along an approximate straight line or a specific attenuation mode in a homogeneous medium). Abnormal paths or deviations may be caused by internal cavities in the roadbed, hidden construction defects (such as un-compacted areas), or external foreign intrusions (such as large rocks falling), which need to be identified through the quantification of the geometric characteristics of the spatial trajectory and the amount of energy deviation. The specific processing logic is as follows:

[0083] Based on the spatial layout position of the optical fiber sensor (the geographical coordinates or relative position relationship of each sensor are known), the collected vibration signal is bound to the spatial position of the corresponding sensor to form a "vibration signal-spatial position" mapping data set Where P . is the spatial coordinates (such as latitude and longitude or planar rectangular coordinates) of the jth sensor, and V . is the amplitude or arrival time of the vibration signal detected by the sensor (if the system supports time synchronization, the arrival time is preferred to more accurately reverse the propagation path). Through the spatio-temporal correlation analysis of multi-sensor data (for example, based on time difference of arrival (TDOA) or amplitude attenuation gradient), the propagation trajectory L of the vibration signal (i.e. the spatial propagation path curve from the excitation source to each sensor) is reconstructed.

[0084] Based on normal working condition data in advance. The real-time reconstructed propagation trajectory L is spatially aligned with the ideal trajectory model and the deviation is calculated:

[0085] Spatial deviation: for the key nodes on the trajectory (such as sampling points at fixed distances on the propagation path), the Euclidean distance (or other geometric distance measures) between the actual spatial coordinates and the corresponding node coordinates of the ideal trajectory is calculated to obtain a series of local deviation values Δd k (k = 1, 2,..., K, K is the total number of trajectory nodes).

[0086] Global deviation statistics: the maximum value Δd k (maximum deviation degree of the propagation path) and the minimum value Δd max (minimum abnormal disturbance influence under the deviation lower limit, or used to assist in judging the dispersion of the deviation distribution) are extracted from all local deviation values Δd min (k = 1, 2,..., K, K is the total number of trajectory nodes). mean or standard deviation (optional) to comprehensively evaluate the propagation stability.

[0087] The maximum deviation threshold T OP,max and the minimum deviation threshold T OP,min allowed are set in advance through historical data statistics or engineering experience (for example, derived from the allowable uneven settlement range of the roadbed in the highway design specification, or determined by the 99th percentile deviation value under normal working conditions). If the real-time calculation of Δdmax >T OP,max (i.e. there is a significant abnormal path deviation, such as the vibration signal deviates from the expected route beyond the allowed range) or Δd min <T OP,min (e.g. a negative deviation anomaly occurs due to an unexpected local energy concentration), the module determines that there is an "abnormal propagation trajectory" and records the abnormal type (deviation / energy attenuation anomaly), the key node position and the deviation value.

[0088] The analysis results are associated by "or" logic (i.e. either of the abnormal determinations triggers the subsequent process): if the vibration amplitude change rate R i exceeds the threshold T R (time domain anomaly) or the maximum l minimum deviation Δd max / Δd min of the vibration transmission trajectory exceeds the preset range (spatial anomaly), the module starts the further warning procedure. At this time, the key parameters obtained by the current analysis (including the change rate R i of the abnormal fluctuation point, the maximum deviation Δd max of the propagation trajectory, the corresponding position, the time of abnormal occurrence, etc.) are packaged and output, and the rate sequence of the change of the vibration amplitude with time (used for subsequent construction of the material flow state stability index) is sent to the warning strategy output module in real time as one of the input parameters.

[0089] Suppose that the fiber optic sensors of a certain highway section monitor that at a certain time, the vibration amplitude increases from 5 mV to 30 mV in 0.2 seconds (change rate R i = (30-5) / 0.2 = 125 mV / s), and the 99% quantile threshold T R of the historical normal data of the section = 80 mV / s, then R i >T R the time domain anomaly flag is triggered; at the same time, through the vibration propagation trajectory display reconstructed by the multi-sensor data, the signal that should have propagated along a straight line to the downstream sensor actually deviated by 1.2 meters (exceeding the preset T OP,max = 0.8 meters), so the spatial anomaly determination is established. The module outputs the related parameters of the two abnormal events and transmits the change rate sequence to the warning strategy output module to provide the basis for subsequent evaluation of the roadbed material stability. Through the above processing flow, the vibration analysis and mechanism triggering module realizes multi-dimensional and high-sensitivity detection of abnormal vibrations, solves the missed detection problem caused by the traditional method of relying only on a static amplitude threshold, and provides key dynamic and spatial feature parameters for the accurate formulation of subsequent warning strategies.

[0090] The early warning strategy output module constructs a material flow state stability index based on the rate of change of vibration amplitude over time, evaluates the stability of vibration signals at different sections of the highway, and reflects the structural stability of the vibration source. In combination with the vibration propagation trajectory, a construction area uniformity index is created to evaluate whether there is unevenness in the propagation process of the vibration signal, further evaluating the flatness and quality of road construction. The flow state stability index and the construction area uniformity index are input into a deep learning model that has been trained for a long time to optimize, and a regulation coordination coefficient is calculated. The coordination degree between vibration behavior and highway construction state is analyzed according to the regulation coordination coefficient, and the corresponding vibration early warning strategy is generated based on the analysis results.

[0091] The early warning strategy output module is the decision center of the distributed optical fiber acoustic sensing highway abnormal vibration real-time early warning system, and its core function is to convert the dynamic characteristics (rate of change of vibration amplitude over time, vibration propagation trajectory) provided by the vibration analysis and mechanism triggering module into early warning strategies that can guide engineering practice. This module generates a hierarchical and directional early warning strategy by constructing a material flow state stability index (evaluating the vibration stability and structural health status of different sections of the highway) and a construction area uniformity index (quantifying the spatial uniformity and construction quality in the vibration propagation process), and introducing a regulation coordination coefficient (representing the matching degree of vibration behavior and highway construction state). Its processing flow covers four key steps: construction and calculation of the material flow state stability index, generation and analysis of the construction area uniformity index, input integration of the deep learning model and calculation of the coordination coefficient, and generation of the early warning strategy based on the coordination coefficient. Each step must strictly follow the engineering logic and data feature association rules.

[0092] Construction and calculation of the material flow state stability index:

[0093] This index aims to quantify the dynamic characteristics of the change of vibration amplitude over time (i.e., the time-domain stability of the vibration signal) to reflect the material flow state characteristics (such as compactness, elastic modulus) and structural integrity of the highway at a specific section. The more stable the rate of change of vibration amplitude (mainly low-frequency small fluctuations), the more stable the material flow state (such as well-compacted asphalt layer or intact base layer); on the contrary, the sudden rate (high-frequency large fluctuations) may indicate loose material (such as roadbed cavity), structural damage (such as base layer cracking), or external impact (such as heavy vehicle illegal unloading).

[0094] Receive the rate sequence of change of vibration amplitude over time output by the vibration analysis and mechanism triggering module (R i is the rate of change of vibration amplitude in the i-th time window, unit: amplitude unit / second; T is the total number of time windows), and the spatial section identifier of the highway corresponding to the sequence (such as kilometer post number + lateral interval) is synchronously associated.

[0095] For the rate sequence {Ri Statistical analysis is performed to extract key characteristic parameters. Mean μ r : reflects the overall level of vibration rate in this paragraph Benchmark value for judging vibration activity; standard deviation σ R : measures the degree of dispersion of rate fluctuations High σ R Indicates that the rate fluctuates violently (may correspond to uneven internal stress of the material). Mutation point ratio P mu[ : The proportion of sampling points whose rate exceeds the dynamic threshold T R (derived from historical normal data statistics, such as 99% quantile) High proportion indicates frequent sudden disturbance (such as local material loosening).

[0096] Based on the above characteristics, the material flow state stability index I flow (value range 0-1, the closer to 1, the higher the stability) is generated by hierarchical weighting logic: if μ R is lower than the low activity threshold T _,low (such as 5mV / s), σ R is lower than the dispersion threshold T `,low (such as 3mV / s), and P mu[ is lower than the mutation threshold T a,low (such as 5%), it is determined as "high stability", I flow is assigned a value of 0.9-1.0. If μ R is in the medium activity interval (T _,low < μ R < T _,bicb ), σ R and P mu[ are moderately high (but not supercritical), it is determined as "medium stability", I flow is assigned a value of 0.5-0.8. If μ R exceeds the high activity threshold T _,bicb (such as 20mV / s), or σ R and P mu[ exceed the critical value (such as σ R >10mV / s or P mu[ >20%), it is determined as "low stability", I flow is assigned a value of 0-0.4. Note: thresholds T _,low / bicb , T `,low / bicb , etc. are determined by historical normal operating data (such as vibration rate statistical distribution of highway without construction disturbance and perfect road surface), for example, take 95% or 99% quantile as the boundary.

[0097] Generation and analysis of construction area uniformity index:

[0098] The index indirectly reflects the flatness of the road construction area (such as road paving uniformity), the compactness of the base (such as the quality of filler layer compaction), and hidden defects (such as cavities, soft interlayers) by analyzing the spatial uniformity of the vibration signal during propagation (including the straightness of the propagation trajectory, the consistency of the time delay, and the stability of the energy attenuation). The more uniform the vibration propagation (closer to a straight line, smaller time difference between each point, and stable amplitude attenuation), the higher the construction quality. On the contrary, abnormal deviation (curved trajectory), time delay dispersion (uneven propagation speed), or amplitude mutation (energy concentration / loss) may indicate construction defects.

[0099] The vibration transmission trajectory feature data output by the vibration analysis and mechanism triggering module is received, including the key point sequence (K is the number of key points; x . ,y . is the spatial coordinate; t . is the signal arrival time; v lo<al is the local propagation speed; A . is the vibration amplitude).

[0100] The propagation trajectory and key point parameters are analyzed in multiple dimensions to extract the following core features:

[0101] Path straightness deviation Δ line : The least squares method is used to fit the straight line of the spatial distribution of the key points, and the average perpendicular distance of each key point to the fitted straight line is calculated (reflecting the degree of deviation of the trajectory from the ideal straight line, d . is the distance of the jth key point to the straight line).

[0102] Local propagation speed dispersion σ j : The standard deviation of the local speed v lo<al of each key point is calculated (the average local speed is , high dispersion indicates poor uniformity of the propagation medium (such as uneven layer compaction of the roadbed).

[0103] Amplitude attenuation stability Δ l : If the sensor supports amplitude monitoring, the amplitude attenuation rate α . = 0A . -A .K1 3 / Δs . (Δs . is the spatial interval between adjacent points), and the standard deviation σ m of the attenuation rate is extracted (high σ m indicates unstable energy loss, which may correspond to local cavities or soft layers).

[0104] Based on the above features, the construction area uniformity index I uniform(0-1, the closer to 1, the higher the uniformity). If Δ line below the linearity threshold T O,line (0.5m, for example), σ j below the velocity dispersion threshold T `,j (100m / s, for example), and σ m below the attenuation dispersion threshold T `,m (0.05dB / m, for example), it is determined to be "high uniformity", I uniform assigned a value of 0.9-1.0. If Δ line and σ j are in the medium deviation interval (such as Δ line [0.5, 1.5]m, σ j [100, 300]m / s), but not exceeding the serious defect threshold, it is determined to be "medium uniformity", I uniform assigned a value of 0.5-0.8. If Δ line exceeds the high deviation threshold T O,bicb (3m, for example), or σ j and σ m are significantly high (such as σ j >500m / s or σ m >0.2dB / m), it is determined to be "low uniformity", I uniform assigned a value of 0-0.4. Note: The thresholds T O,line , T `,j are determined by the vibration propagation trajectory statistics of normal construction sections (such as the uniform propagation characteristics of newly paved asphalt pavement).

[0105] Calculation of coordination coefficient:

[0106] The coefficient integrates the material flow stability index I flow and the construction area uniformity index I uniform through a deep learning model, quantifying the coordination degree of vibration dynamic characteristics and actual construction state (such as maintenance stage, construction disturbance intensity). The closer the coordination coefficient is to 1, the better the match between vibration behavior (such as stable low-frequency fluctuation) and construction state (such as compacted pavement); the lower the coefficient (such as close to 0), the higher the correlation between vibration anomaly (such as high-frequency mutation) and construction defect (such as un-compacted base). The processing logic is as follows:

[0107] Standardize I flow and I uniform to values within the [0, 1] interval (if the original value is already in this range, input directly), and add optional auxiliary features (such as the spatial position of the vibration event, the construction activity record corresponding to the time stamp (such as whether it is in the paving period, the rolling period)), forming a two-dimensional / three-dimensional input vector X = [I flow , I uniform, auxiliary features].

[0108] A pre-optimized deep neural network (DNN) or long short-term memory network (LSTM, if considering time-dependent) is used. The input layer of the model receives X, extracts nonlinear features through hidden layers (such as 3-5 fully connected layers with ReLU activation function), and generates the regulation coordination coefficient C (value range 0-1) at the output layer. The training data of the model comes from the historical labeled data set (containing known vibration index, construction state label and expert rated coordination level), and the loss function (such as mean square error or cross entropy loss) is optimized by back propagation algorithm.

[0109] According to the value of C, the coordination level is divided:

[0110] C∈[0.8,1.0]:High coordination (vibration behavior and construction state completely match, such as stable vibration of normal traffic road);

[0111] C∈[0.5,0.8):Moderate coordination (slight mismatch, such as slight vibration fluctuation caused by construction interference);

[0112] C∈[0.2,0.5):Low coordination (significant mismatch, such as abnormal vibration caused by local material loosening);

[0113] C∈[0,0.2):Extremely low coordination (serious mismatch, such as severe vibration anomaly caused by hidden defects (cavities, weak layers)).

[0114] Based on the coordination coefficient, the early warning strategy is generated (directional hierarchical warning, according to the level of regulation coordination coefficient C, combined with the spatial position of vibration event and construction background information (such as whether it is in maintenance period, whether there is construction machinery operation nearby), the hierarchical and directional early warning strategy is generated, which guides the engineering personnel to accurately locate the problem and take targeted measures, and the strategy grading rules are:

[0115] Extremely low coordination (C<0.2): Trigger "emergency warning" (red), prompt "suspected hidden structural defects (such as cavities, weak layers), need to immediately block the road section and carry out geological radar scanning and structure detection";

[0116] Low coordination (0.2≤C<0.5): Trigger "high risk warning" (orange), prompt "local material stability anomaly (such as roadbed loosening, base cracking), suggest to limit heavy vehicle traffic and arrange roadbed reinforcement";

[0117] Moderate coordination (0.5≤C<0.8): Trigger "general warning" (yellow), prompt "slight construction interference or environmental factors (such as small rockfall, local uneven paving), suggest to strengthen patrol and record vibration events";

[0118] High coordination (C≥0.8): no early warning triggered (green), marked as "normal vibration behavior".

[0119] The early warning level, the associated highway section coordinates, the vibration event timestamp, and the key parameters (such as I flow , I uniform , and C value) are pushed to the relevant responsible departments (such as the maintenance center and the construction supervision unit) in real time through a visualization platform (such as a GIS map superimposed with early warning hotspots) or a mobile terminal (such as a patrol personnel APP), along with disposal suggestions (such as detection range and traffic restriction measures).

[0120] If the monitoring data of a highway K50+200-K50+300 section shows that the material flow stability index I flow = 0.3 (low stability, corresponding to high-frequency and large vibration rate), the construction area uniformity index I uniform = 0.2 (low uniformity, corresponding to serious deviation of propagation trajectory and high local velocity dispersion), and the regulation and coordination coefficient C calculated by the deep learning model is 0.15 (extremely low coordination), the module will generate an "emergency early warning" to indicate that there may be a roadbed cavity or hidden weak layer in this area, and suggest that the 100-meter section be immediately closed and a geological radar be used to detect the roadbed integrity. At the same time, the coordinates (K50+200-K50+300), the vibration event time (current time), and the key parameters are pushed to the maintenance center and the road administration department.

[0121] In the description of the present specification, the description of the terms "one embodiment", "example", "specific example", and the like means that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In the present specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0122] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details and do not limit the present invention to the specific embodiments. Obviously, many modifications and changes can be made according to the content of the present specification. The present specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can well understand and utilize the present invention. The present invention is limited only by the claims and their entire scope and equivalents.

Claims

1. A distributed fiber optic acoustic sensing system for real-time early warning of abnormal vibrations on highways, characterized in that: It includes a data acquisition module, a vibration analysis and mechanism triggering module, and an early warning strategy output module; Data acquisition module: Real-time acquisition of road vibration data through fiber optic sensors. The acquired data includes vibration signal amplitude fluctuation curves and vibration transmission trajectories. Vibration analysis and mechanism triggering module: Analyzes the rate of change of vibration signal amplitude fluctuation curve, calculates the rate of change of vibration amplitude with time, and marks abnormal fluctuation if the rate of change exceeds the set threshold. Through real-time analysis of vibration transmission trajectory, it identifies abnormal paths or deviations in the signal propagation process and calculates the maximum and minimum deviations of vibration signal. If these deviation values ​​exceed the preset range, an early warning program is activated. Early warning strategy output module: Based on the rate of change of vibration amplitude over time, a material flow stability index is constructed to assess the stability of vibration signals in different sections of the highway, reflecting the structural stability of the vibration source. Combined with the vibration propagation trajectory, a construction area uniformity index is created to assess whether there is non-uniformity in the vibration signal propagation process. The flow stability index and the construction area uniformity index are used as input data and imported into a pre-trained deep learning model to calculate the control coordination coefficient. Based on the control coordination coefficient, the degree of coordination between vibration behavior and highway construction status is analyzed, and corresponding vibration early warning strategies are generated based on the analysis results.

2. The distributed fiber optic acoustic sensing real-time early warning system for abnormal vibrations on highways according to claim 1, characterized in that: The vibration analysis and mechanism triggering module performs noise reduction processing on the amplitude fluctuation curve of the original vibration signal collected by the fiber optic sensor, and then segments the continuous sampling points according to a fixed time window to form a discrete time series dataset. The time series is traversed sequentially. For each moment, the amplitude of its preceding adjacent moment is extracted, and the rate of change of vibration amplitude at that moment is calculated by difference operation. The distribution characteristics of vibration amplitude change rate are statistically analyzed in advance based on historical normal operating condition data, and the change rate R at each point is calculated in real time. i With threshold T R Perform point-by-point comparison; If there exists any R i >T R If so, mark that moment as an abnormal fluctuation point and record its corresponding timestamp t. i Rate of change R i And the amplitude sequence of the preceding and following time periods.

3. The distributed fiber optic acoustic sensing real-time early warning system for abnormal vibrations on highways according to claim 2, characterized in that: The vibration analysis and mechanism triggering module binds the collected vibration signals with the corresponding spatial positions of the optical fiber sensors based on the spatial deployment location of the optical fiber sensors, forming a vibration signal-spatial position mapping dataset. By analyzing the spatiotemporal correlation of multi-sensor data, the propagation trajectory L of the vibration signal is reconstructed. Based on normal operating condition data, the real-time reconstructed propagation trajectory L is spatially aligned and the deviation is calculated with the ideal trajectory model.

4. A distributed fiber optic acoustic sensing system for real-time early warning of abnormal vibrations on highways according to claim 2, characterized in that: The vibration analysis and mechanism triggering module compares the calculated maximum and minimum local deviation values ​​with the preset maximum and minimum deviation thresholds. If the real-time calculated maximum local deviation value is greater than the maximum deviation threshold or the minimum local deviation value is greater than the minimum deviation threshold, it is determined that the propagation trajectory is abnormal, and the abnormality type, key node location and deviation value are recorded.

5. A distributed fiber optic acoustic sensing system for real-time early warning of abnormal vibrations on highways according to claim 4, characterized in that: The logic for spatial alignment and deviation calculation between the real-time reconstructed propagation trajectory L and the ideal trajectory model is as follows: For each node on the trajectory, the Euclidean distance between its measured spatial coordinates and the corresponding node coordinates on the ideal trajectory is calculated to obtain the local deviation value Δd. k ; From all local deviation values ​​Δd k Extract the maximum value Δd max and minimum value Δd min Simultaneously calculate the mean value Δd of the overall deviation. mean To comprehensively assess the stability of transmission.

6. A distributed fiber optic acoustic sensing real-time early warning system for abnormal vibrations on highways according to any one of claims 1-5, characterized in that: The early warning strategy output module outputs the material flow stability index I. flow With construction area uniformity index I uniform Standardized to values ​​within the interval [0,1]; A pre-optimized deep neural network model is used. The input layer of the deep neural network model receives X, the hidden layer extracts nonlinear features, and the output layer generates the regulation and coordination coefficient C. Coordination levels are determined based on the value of C: If C∈[0.8,1.0], it indicates high coordination; if C∈[0.5,0.8), it indicates moderate coordination; if C∈[0.2,0.5), it indicates low coordination; if C∈[0,0.2), it indicates very low coordination.

7. A distributed fiber optic acoustic sensing system for real-time early warning of abnormal vibrations on highways according to claim 6, characterized in that: The construction and calculation logic of the material flow stability index is as follows: Receive the rate sequence of vibration amplitude changing over time, and synchronously associate the sequence with the corresponding highway spatial segment identifier; Statistical analysis is performed on the rate sequence to extract characteristic parameters, including the rate mean, rate standard deviation, and percentage of abrupt change points. A material flow stability index is then generated using hierarchical weighted logic.

8. A distributed fiber optic acoustic sensing system for real-time early warning of abnormal vibrations on highways according to claim 6, characterized in that: The construction and calculation logic of the uniformity index of the construction area is as follows: The system receives vibration transmission trajectory characteristic data, including point sequences, performs multi-dimensional analysis on the propagation trajectory and point parameters, extracts path linearity deviation, local propagation velocity dispersion, and amplitude attenuation stability, and generates a construction area uniformity index through hierarchical logic.

9. A distributed fiber optic acoustic sensing system for real-time early warning of abnormal vibrations on highways according to claim 1, characterized in that: The data acquisition module collects the vibration amplitude fluctuation curve over time captured by the fiber optic sensor, reflecting the dynamic fluctuations of vibration intensity and frequency.

10. A distributed fiber optic acoustic sensing system for real-time early warning of abnormal vibrations on highways according to claim 1, characterized in that: The data acquisition module monitors the transmission path of vibration signals at various points on the road, focuses on the time delay and propagation mode of the signal during its propagation in different road sections, and extracts the temporal and spatial characteristics of key points.

Images