Expressway underground cavity real-time monitoring system based on distributed optical fiber sensing

Abstract

The application discloses a highway underground cavity real-time monitoring system based on distributed optical fiber sensing, relates to the technical field of highway underground cavity monitoring, and comprises the following steps: determining a highway to be detected, monitoring vibration information when a vehicle passes by by using multiple optical fiber sensors, monitoring vehicle information, obtaining vehicle vibration information data, performing vibration screening processing, performing vibration information preprocessing, obtaining standard vibration information data, performing vibration energy analysis, performing distribution deviation analysis on the monitoring positions of each optical fiber sensor, obtaining distribution deviation data, and performing underground cavity determination analysis on the monitoring positions of each optical fiber sensor according to the distribution deviation data. The application is used to solve the problem that the existing highway underground cavity monitoring technology cannot accurately monitor highway underground cavities in real time according to the distribution of vibration signals of multiple monitoring positions when monitoring highway underground cavities based on optical fiber sensing.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of highway underground cavity monitoring, in particular to a highway underground cavity real-time monitoring system based on distributed optical fiber sensing. BACKGROUND

[0002] Highway underground cavity monitoring technology refers to a comprehensive technical system that, for the subgrade and underground within a certain depth range of expressways and ordinary highways, identifies the formation, location, scale, shape and development trend of underground cavities through non-destructive testing, sensing and data analysis, and provides data support for road safety warning and maintenance.

[0003] The existing highway underground cavity monitoring technology uses methods such as geological radar and high-density electrical method to monitor highway underground cavities. These methods are mostly single inspection type, i.e., they continue to monitor specific locations of expressways on demand or periodically, and cannot monitor underground cavities in real time, i.e., they cannot capture the dynamic expansion process of underground cavities, and there is a risk that underground cavities may develop rapidly between two detections without being discovered. For example, the patent application with publication number CN112505789A discloses an underground cavity monitoring device and method based on microwave radar. This scheme monitors underground cavities using microwave radar and cannot monitor expressways in real time. The existing highway underground cavity monitoring technology often uses the size difference of vibration amplitudes to determine whether there is an underground cavity when monitoring highway underground cavities based on optical fiber sensing. However, the amplitude of the vibration signal of optical fiber sensing is easily affected by vehicle load, temperature changes, road settlement and other disturbances. Different vehicle types, speeds and loads produce significantly different vibration amplitudes on the subgrade. If the threshold for judgment is set too low, normal vibrations caused by heavy trucks may be misjudged as anomalies caused by cavities. If the threshold for judgment is set too high, early cavities may not be discovered in time. Therefore, the existing highway underground cavity monitoring technology cannot accurately monitor highway underground cavities in real time based on the distribution of vibration signals from multiple monitoring locations. SUMMARY

[0004] The present application aims to at least solve one of the technical problems in the prior art, by determining the expressway to be detected, monitoring the vibration information when the vehicle passes through by using multiple optical fiber sensors, and monitoring the vehicle information to obtain vehicle vibration information data; and performing vibration screening processing, and vibration information preprocessing to obtain standard vibration information data; and performing vibration energy analysis, and distributing deviation analysis on the monitoring position of each optical fiber sensor to obtain distribution deviation data; and performing underground cavity determination analysis on the monitoring position of each optical fiber sensor according to the distribution deviation data; to solve the problem that the existing highway underground cavity monitoring technology cannot accurately monitor the highway underground cavity in real time according to the distribution of vibration signals of multiple monitoring positions when monitoring the highway underground cavity based on optical fiber sensing.

[0005] To achieve the above-mentioned purpose, the present application provides an expressway underground cavity real-time monitoring system based on distributed optical fiber sensing, comprising an information acquisition module, an information processing module, a deviation analysis module, and a cavity determination module.

[0006] The information acquisition module is used to determine the expressway to be detected, monitor the vibration information when the vehicle passes through by using multiple optical fiber sensors, and monitor the vehicle information to obtain vehicle vibration information data.

[0007] The information processing module is used to perform vibration screening processing on the vehicle vibration information data, and vibration information preprocessing to obtain standard vibration information data.

[0008] The deviation analysis module comprises an energy analysis unit and a distribution analysis unit, the energy analysis unit performs vibration energy analysis according to the standard vibration information data, and the distribution analysis unit is used to perform distribution deviation analysis on the monitoring position of each optical fiber sensor to obtain distribution deviation data.

[0009] The cavity determination module performs underground cavity determination analysis on the monitoring position of each optical fiber sensor according to the distribution deviation data.

[0010] Further, the information acquisition module is configured with an information acquisition strategy, and the information acquisition strategy comprises:

[0011] For a section of expressway to be monitored, denoted as a monitoring section, the optical fiber used for monitoring is arranged continuously underground and longitudinally along the monitoring section, multiple optical fiber sensors used for collecting vibration signals are uniformly arranged along the arranged optical fiber used for monitoring, and the monitoring section is sequentially denoted as monitoring unit 1 to monitoring unit n along the traveling direction, where n is the total number of monitoring units.

[0012] Any one monitoring unit is denoted as monitoring unit i, where i∈[1, n], and any vehicle is denoted as a first vehicle.

[0013] The monitoring unit i collects the corresponding amplitude in real time and arranges them in time sequence, denoted as amplitude time sequence, and repeats the collection of amplitude time sequence of all monitoring units.

[0014] Further, the information collection strategy further includes:

[0015] When the first vehicle passes the position of the monitoring unit i on the monitoring section, the monitoring unit i collects the corresponding amplitude when the first vehicle passes, and arranges them in time sequence, denoted as vibration information i of the first vehicle;

[0016] and obtains the vehicle speed of the first vehicle when passing the monitoring unit i and the license plate of the first vehicle, denoted as vehicle information i of the first vehicle;

[0017] The corresponding vibration information and vehicle information of the first vehicle when passing the monitoring unit are collected in turn. If the first vehicle completely passes the monitoring unit 1 to the monitoring unit n, the first vehicle is denoted as an effective vehicle; the vibration information 1 to the vibration information n corresponding to the first vehicle are denoted as monitoring vibration information of the first vehicle, and the vehicle information 1 to the vehicle information n corresponding to the first vehicle are denoted as monitoring vehicle information of the first vehicle;

[0018] The monitoring vibration information of the first vehicle and the monitoring vehicle information of the first vehicle are denoted as vehicle vibration information data of the first vehicle.

[0019] Further, the information processing module is configured with an information processing strategy, and the information processing strategy includes:

[0020] For the vibration information i of the first vehicle; the latest segment without any vehicle passing and continuous not less than the first time length before the first vehicle passing is intercepted from the amplitude time sequence of the monitoring unit i, denoted as reference amplitude segment, wherein the first time length is T1;

[0021] Any amplitude in the reference amplitude segment is denoted as BA, the reference amplitude segment is arranged in size order, and the median is obtained, denoted as MA;

[0022] The absolute difference value of BA and MA is calculated, denoted as absolute deviation corresponding to BA; the absolute deviation corresponding to all amplitudes in the reference amplitude segment is repeatedly calculated to obtain the corresponding absolute deviation sequence;

[0023] The absolute deviation sequence is arranged in size order, and the median is obtained, denoted as MC; for BA, if the absolute deviation corresponding to BA is greater than (k1*MC), BA is determined as an abnormal amplitude, otherwise BA is determined as a normal amplitude, wherein k1 is a set proportion coefficient;

[0024] All amplitudes in the reference amplitude segment are repeatedly determined, and all abnormal amplitudes are removed to obtain the first amplitude sequence.

[0025] Further, the information processing strategy further comprises:

[0026] The first amplitude sequence is evenly divided into three segments, sequentially recorded as the first amplitude segment, the second amplitude segment and the third amplitude segment according to the time sequence;

[0027] The average values of the first amplitude segment, the second amplitude segment and the third amplitude segment are calculated respectively, sequentially recorded as CP1, CP2 and CP3; and the standard deviations of the first amplitude segment, the second amplitude segment and the third amplitude segment are calculated respectively, sequentially recorded as CB1, CB2 and CB3;

[0028] The weights of the first amplitude segment, the second amplitude segment and the third amplitude segment are sequentially set as Q1, Q2 and Q3, Q1<Q2<Q3, Q1+Q2+Q3=1; the weighted mean QP and the weighted standard deviation QB are calculated respectively, wherein QP=Q1*CP1+Q2*CP2+Q3*CP3, QB=Q1*CB1+Q2*CB2+Q3*CB3; (QP+k2*QB) is calculated, recorded as the dynamic threshold of the vibration information i, wherein k2 is a set proportion coefficient.

[0029] Further, the information processing strategy further comprises:

[0030] The average value of the maximum k3% amplitude in the vibration information i is calculated, recorded as the maximum average value; if the maximum average value of the vibration information i is less than the corresponding dynamic threshold, the vibration information i is recorded as invalid information; otherwise, it is marked as normal information, wherein k3% is a set proportion;

[0031] The vibration information 1 to the vibration information n corresponding to the first vehicle are repeatedly judged, if there is invalid information in the vibration information 1 to the vibration information n corresponding to the first vehicle, the vehicle vibration information data corresponding to the first vehicle is marked as invalid; otherwise, it is marked as suspicious;

[0032] If the vehicle vibration information data corresponding to the first vehicle is marked as suspicious, all vehicle speed magnitudes in the vehicle information 1 to the vehicle information n corresponding to the first vehicle are obtained, recorded as the vehicle speed set, the coefficient of variation of the vehicle speed set is calculated, recorded as the speed variation coefficient CV, if CV>k4, the vehicle vibration information data corresponding to the first vehicle is marked as invalid, otherwise, the vehicle vibration information data corresponding to the first vehicle is marked as valid, wherein k4 is a set threshold;

[0033] The valid vehicle vibration information data of the valid vehicle is repeatedly collected; if the vehicle vibration information data corresponding to the first vehicle is marked as valid; the vibration information 1 to the vibration information n corresponding to the first vehicle are sequentially subjected to Fourier transform, sequentially obtaining the vibration frequency domain signal 1 to the vibration frequency domain signal n, recorded as the standard vibration information data.

[0034] Further, the energy analysis unit is configured with an energy analysis strategy, the energy analysis strategy comprising:

[0035] For the vibration frequency domain signal i, the frequency point corresponding to each amplitude is obtained, and all the frequency points are arranged in ascending order, denoted as a frequency point sequence;

[0036] The size of the sliding window is set to k5, and the sliding step is k6. From the beginning of the frequency point sequence, the sliding window is sequentially moved backward, and each time the sliding window is sequentially recorded as window 1 to window n1 according to the sliding order, where n1 is the total number of windows, and any window is recorded as window j, j∈[1, n1];

[0037] Based on the vibration frequency domain signal i, the amplitudes corresponding to all the frequency points in the window j are obtained, and the squares of all the corresponding amplitudes are calculated and summed, denoted as the spectral energy of the window j. The spectral energies of the window 1 to the window n1 are repeatedly calculated, and all the spectral energies are arranged in ascending order, denoted as a window energy sequence.

[0038] Further, the distribution analysis unit is configured with a distribution analysis strategy, the distribution analysis strategy comprising:

[0039] Take the spectral energy of the first k7% of the window energy sequence, and record the corresponding window as a stable window; record the frequency points in the stable window as stable frequency points, obtain the stable frequency points in all stable windows, and remove duplicates to obtain a stable frequency point set, wherein k7% is a set proportion;

[0040] Based on the vibration frequency domain signal i, the amplitudes corresponding to the stable frequency point set points are obtained, and the squares of all the corresponding amplitudes are calculated, denoted as stable frequency point energies, and the average WP and the standard deviation WB of all the stable frequency point energies are calculated; (WP+k8*WB) is recorded as the basic energy threshold WY; wherein k8 is a set proportion coefficient;

[0041] The square of the amplitude corresponding to each frequency point in the frequency point sequence is calculated, denoted as the corresponding frequency point energy. The frequency point with a frequency point energy greater than (k9*WY) is recorded as a high amplitude point, the maximum high amplitude point is obtained, and the corresponding frequency is recorded as the maximum distribution point of the vibration frequency domain signal i; wherein k9 is a set proportion coefficient;

[0042] Repeat the maximum distribution point of the vibration frequency domain signal 1 to the vibration frequency domain signal n, and sequentially record them as the maximum distribution point 1 to the maximum distribution point n in order.

[0043] Further, the distribution analysis strategy further comprises:

[0044] For the maximum distribution point i, the maximum distribution point i is excluded from the maximum distribution point 1 to the maximum distribution point n, and linear fitting is performed on the remaining maximum distribution points to obtain a corresponding fitting line, and a fitting value of the maximum distribution point i in the corresponding fitting line is obtained, and a difference value of the maximum distribution point i and the corresponding fitting value is calculated, which is recorded as a distribution deviation;

[0045] The distribution deviations of the maximum distribution points 1 to n are repeatedly calculated, and are sequentially recorded as distribution deviation 1 to distribution deviation n, and are marked as distribution deviation data.

[0046] Further, the cavity determination module is configured with a cavity determination strategy, and the cavity determination strategy comprises:

[0047] If the corresponding distribution deviation i of the maximum distribution point i is greater than e0, it is judged that the position of the corresponding monitoring unit i is suspected to have an underground cavity, wherein e0 is a set threshold value;

[0048] And the position of the monitoring unit i is repeatedly judged according to the collected effective vehicle vibration information data of the effective vehicle, and if the position of the monitoring unit i is judged to be suspected to have an underground cavity for e1 times in succession, it is judged that the position of the monitoring unit i is determined to have an underground cavity, and a warning is performed, wherein e1 is a set number;

[0049] The monitoring units 1 to n are repeatedly judged, and corresponding warnings are performed.

[0050] The beneficial effects of the present application are: the present application determines the expressway to be detected, utilizes a plurality of optical fiber sensors to monitor the vibration information when a vehicle passes, and monitors vehicle information to obtain vehicle vibration information data; the vehicle vibration information data is subjected to vibration screening processing and vibration information preprocessing to obtain standard vibration information data; vibration energy analysis is performed according to the standard vibration information data, and distribution deviation analysis is performed on the monitoring positions of each optical fiber sensor to obtain distribution deviation data; underground cavity determination analysis is performed on the monitoring positions of each optical fiber sensor according to the distribution deviation data; when the highway underground cavity is monitored based on the optical fiber sensing, the highway underground cavity can be accurately monitored in real time according to the distribution of the vibration signals of a plurality of monitoring positions;

[0051] The present application can effectively remove the noise and extreme value in the reference amplitude segment, improve the accuracy of the dynamic threshold by selecting the reference amplitude segment without vehicle passing, calculating the absolute deviation sequence based on the median, and removing the abnormal amplitude; the amplitude sequence after denoising is divided into three segments, the weighted mean and weighted standard deviation are calculated, and the dynamic threshold is constructed, which can give more weight to the recent period and accurately capture the background vibration characteristics when the vehicle passes; all maximum distribution points are removed point by point, then linear fitting is performed, and the absolute difference of a point relative to the fitting line is calculated as the distribution deviation; the influence of isolated abnormal points on fitting can be suppressed, the abnormal situation of the frequency band distribution of the monitoring unit can be more accurately judged, and the accuracy of the underground cavity monitoring can be improved. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 It is a principle block diagram of the system of the present application;

[0053] Figure 2 It is a step flow chart of the method of the present application;

[0054] Figure 3 It is a flow chart of invalid information screening of the present application;

[0055] Figure 4 It is a structure schematic diagram of the electronic device of the present application. DETAILED DESCRIPTION

[0056] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0057] Embodiment 1, please refer to Figure 1 The present application provides a highway underground cavity real-time monitoring system based on distributed optical fiber sensing, which comprises an information acquisition module, an information processing module, a deviation analysis module and a cavity determination module;

[0058] The information acquisition module is used to determine the highway to be detected, monitor the vibration information when the vehicle passes by using a plurality of optical fiber sensors, and monitor the vehicle information to obtain the vehicle vibration information data;

[0059] The information collection module is configured with an information collection strategy, and the information collection strategy comprises: for a to-be-monitored section of expressway, denoted as a monitoring section, laying an optical fiber for monitoring in the underground of the monitoring section and continuously along the longitudinal direction of the monitoring section, uniformly arranging a plurality of optical fiber sensors for collecting vibration signals along the laid optical fiber for monitoring, and sequentially denoting as monitoring unit 1 to monitoring unit n along the traveling direction of the monitoring section, wherein n is the total number of monitoring units; the optical fiber is generally buried in the lower part of the roadbed, 0.5 meters to 3 meters away from the road surface, and when a cavity is formed underground, the vibration characteristics of the vehicle passing through will change, and these changes will be captured by the optical fiber and converted into an analyzable signal, and then the existence of the cavity is inferred;

[0060] Denote any one monitoring unit as monitoring unit i, wherein i∈[1,n], and denote any one vehicle as a first vehicle;

[0061] Let the monitoring unit i collect the corresponding amplitude in real time and arrange in time sequence, denoted as an amplitude time sequence, repeatedly collect the amplitude time sequence of all monitoring units, and the sampling frequency can be set by itself, but the sampling frequencies of the monitoring units are consistent.

[0062] When the first vehicle passes through the position of the monitoring unit i on the monitoring section, the corresponding amplitude of the first vehicle passing through is collected by the monitoring unit i, that is, the vibration signal, and is arranged in time sequence, denoted as the vibration information i of the first vehicle;

[0063] And the vehicle speed of the first vehicle passing through the monitoring unit i and the license plate of the first vehicle are obtained, denoted as the vehicle information i of the first vehicle; because the same vehicle will produce different vibration signals at different speeds, the speed of the vehicle needs to be collected for subsequent screening, and the license plate is used to identify the first vehicle;

[0064] The corresponding vibration information and vehicle information of the first vehicle passing through the monitoring unit are repeatedly collected in sequence, if the first vehicle completely passes through the monitoring unit 1 to the monitoring unit n, the first vehicle is denoted as an effective vehicle; the vibration information 1 to the vibration information n corresponding to the first vehicle are denoted as the monitoring vibration information of the first vehicle, and the vehicle information 1 to the vehicle information n corresponding to the first vehicle are denoted as the monitoring vehicle information of the first vehicle; if the first vehicle does not completely pass through the monitoring unit 1 to the monitoring unit n due to special reasons, the vibration signal is missing, which is not convenient for cross-unit comparison, fitting and deviation calculation, and then can be discarded for subsequent analysis;

[0065] The monitoring vibration information of the first vehicle and the monitoring vehicle information of the first vehicle are denoted as the vehicle vibration information data of the first vehicle;

[0066] In the specific implementation process, the same vehicle passes through the monitoring unit with the underground cavity and the monitoring unit without the underground cavity at the same speed, and the generated vibration signals are different, when the position of the monitoring unit has no underground cavity, the frequency spectrum energy of the generated vibration signal is mainly concentrated in the low frequency range, and the frequency spectrum energy is low and uniform in the medium frequency and high frequency parts; and when the position of the monitoring unit has the underground cavity, the frequency spectrum energy distribution of the generated vibration signal moves to the high frequency direction, so whether the underground cavity exists below the monitoring unit can be judged by analyzing the frequency spectrum energy distribution of the vibration signal generated when the vehicle passes through each monitoring unit.

[0067] The information processing module is used for vibration screening processing of the vehicle vibration information data, and vibration information preprocessing, to obtain standard vibration information data;

[0068] The information processing module is configured with an information processing strategy, and the information processing strategy includes: referring to Figure 3 As shown in the figure, for the vibration information i of the first vehicle; the first vehicle passes through the latest, and there is no any vehicle passing through, and the continuous piece segment not less than the first time length is intercepted from the amplitude time sequence of the monitoring unit i, and is recorded as the reference amplitude piece segment, wherein the first time length is T1;

[0069] By selecting the continuous piece segment without other vehicles passing through before the vehicle arrives, the background vibration level of the unit in the same environment can be obtained, including natural vibration, foundation noise and sensor noise other than traffic, etc. In the embodiment, T1=1 minute, which can be flexibly set. The latest piece segment without vehicle passing through is selected in order to update the threshold dynamically and reduce the influence of road condition change on the threshold estimation, for example, the road vibration changes with time, temperature and construction.

[0070] Any amplitude in the reference amplitude piece segment is recorded as BA, the reference amplitude piece segment is arranged in size order, and the median is obtained and recorded as MA;

[0071] The absolute difference value of BA and MA is calculated and recorded as the absolute deviation corresponding to BA. The absolute deviation corresponding to all amplitudes in the reference amplitude piece segment is repeatedly calculated to obtain the corresponding absolute deviation sequence. The median and the corresponding median absolute deviation are not sensitive to single-point pulse noise and short-time interference, can effectively identify and eliminate the burst noise in the reference segment, such as transient impact caused by road construction near the construction and occasional electronic interference; and improve the accuracy of the subsequent dynamic threshold.

[0072] The absolute deviation sequence is arranged in size order, and the median is obtained and recorded as MC; for BA, if the absolute deviation corresponding to BA is greater than (k1*MC), BA is determined as an abnormal amplitude, otherwise BA is determined as a normal amplitude, wherein k1 is a set proportion coefficient; in the embodiment, k1=2.5, which can be flexibly set.

[0073] Repeat the determination of all amplitudes in the reference amplitude segment, and all abnormal amplitudes are removed to obtain a first amplitude sequence.

[0074] The first amplitude sequence is evenly divided into three segments, sequentially recorded as a first amplitude segment, a second amplitude segment and a third amplitude segment in time sequence; the corresponding time distribution is early, medium and recent;

[0075] The average values of the first amplitude segment, the second amplitude segment and the third amplitude segment are calculated respectively, sequentially recorded as CP1, CP2 and CP3; and the standard deviations of the first amplitude segment, the second amplitude segment and the third amplitude segment are calculated respectively, sequentially recorded as CB1, CB2 and CB3;

[0076] The weights of the first amplitude segment, the second amplitude segment and the third amplitude segment are set in sequence as Q1, Q2 and Q3, Q1<Q2<Q3, Q1+Q2+Q3=1; the weighted mean QP and the weighted standard deviation QB are calculated respectively, wherein QP=Q1*CP1+Q2*CP2+Q3*CP3, QB=Q1*CB1+Q2*CB2+Q3*CB3; (QP+k2*QB) is calculated and recorded as the dynamic threshold of vibration information i, wherein k2 is a set proportion coefficient, in this embodiment, Q1=0.2, Q2=0.3, Q3=0.5, k2=3, which can be flexibly set, the sequence is evenly divided into three segments and different weights are assigned, which can make the more recent background vibration have a larger proportion and improve the representativeness of the dynamic threshold.

[0077] The average value of the maximum k3% amplitude in the vibration information i is calculated and recorded as the maximum average value; if the maximum average value of the vibration information i is less than the corresponding dynamic threshold, the vibration information i is recorded as invalid information; otherwise, it is marked as normal information, wherein k3% is a set proportion;

[0078] In this embodiment, k3%=5%; reflects whether a significant signal is generated when the vehicle passes; if this part of the peak value is not higher than the dynamic threshold, it means that the vehicle does not generate enough recognizable vibration signal on the unit; only when the vibration signal amplitude of the detection unit exceeds the threshold, it is determined as normal information, the signal that does not exceed the threshold, i.e. does not exceed the background noise, is determined as invalid information, which is directly discarded and does not enter the subsequent analysis;

[0079] The vibration information 1 to vibration information n corresponding to the first vehicle are repeatedly determined; if there is invalid information in the vibration information 1 to vibration information n corresponding to the first vehicle, the vehicle vibration information data corresponding to the first vehicle is marked as invalid; otherwise, it is marked as suspicious;

[0080] If the vehicle vibration information data corresponding to the first vehicle is marked as suspicious, all vehicle speed values in the vehicle information 1 to the vehicle information n corresponding to the first vehicle are obtained, denoted as a vehicle speed set, a coefficient of variation of the vehicle speed set is calculated, denoted as a speed variation coefficient CV, if CV>k4, the vehicle vibration information data corresponding to the first vehicle is marked as invalid, otherwise the vehicle vibration information data corresponding to the first vehicle is marked as valid, wherein k4 is a set threshold value;

[0081] In the embodiment, k4=0.05, on the highway, the speed of a vehicle should be relatively stable when passing through different units, if the measured speed fluctuates greatly between units, it means that the vehicle may not be the same vehicle, even if it is the same vehicle, the vibration signal difference will be large due to the large speed difference, which is not conducive to subsequent judgment, so it is marked as invalid and not subjected to subsequent analysis and processing.

[0082] The valid vehicle vibration information data of the valid vehicle is repeatedly collected, if the vehicle vibration information data corresponding to the first vehicle is marked as valid, the vibration information 1 to the vibration information n corresponding to the first vehicle are subjected to Fourier transform in sequence, and vibration frequency domain signals 1 to vibration frequency domain signals n are obtained in sequence, denoted as standard vibration information data; that is, the time domain vibration signal, the amplitude changing with time, is converted into a frequency domain signal, the amplitude distributed with frequency; the frequency domain can more directly see the energy concentration area, which is convenient for subsequent determination of underground cavities.

[0083] In the specific implementation process, some vehicles may produce vibration signals consistent with background vibration due to their own weight or speed when passing through the monitoring unit, that is, the vibration signal is hidden in the background vibration, if such vibration signal is used for subsequent analysis, it is easy to be disturbed and misled by the background vibration, and misjudgment is easy to occur, so it is discarded.

[0084] The deviation analysis module includes an energy analysis unit and a distribution analysis unit, the energy analysis unit performs vibration energy analysis according to the standard vibration information data, and the distribution analysis unit is used for distribution deviation analysis on the monitoring positions of each optical fiber sensor to obtain distribution deviation data.

[0085] The energy analysis unit is configured with an energy analysis strategy, and the energy analysis strategy includes: for the vibration frequency domain signal i, the frequency points corresponding to each amplitude are obtained, and all the frequency points are arranged in ascending order, denoted as a frequency point sequence.

[0086] The size of the sliding window is set to k5, the sliding step is k6, and the sliding window is sequentially slid backward from the start of the frequency point sequence. Each sliding window is sequentially recorded as window 1 to window n1, where n1 is the total number of windows. Any window is recorded as window j, j∈[1, n1]; in this embodiment, k5=2000, k6=200, which can be flexibly set;

[0087] Based on the vibration frequency domain signal i, the amplitudes corresponding to all frequency points in window j are obtained, and the squares of all corresponding amplitudes are calculated and summed to obtain the spectral energy of window j. The spectral energies of window 1 to window n1 are repeatedly calculated, and all spectral energies are arranged in ascending order to obtain a window energy sequence. Other methods can also be used to calculate the spectral energy in the window according to the actual application scenario.

[0088] The distribution analysis unit configures a distribution analysis strategy, which includes: taking the spectral energy of the first k7% of the window energy sequence, and recording the corresponding window as a stable window; recording the frequency points in the stable window as stable frequency points, obtaining all stable frequency points in the stable window, and removing duplicates to obtain a stable frequency point set, where k7% is a set proportion; in this embodiment, k7%=20. These stable windows are intervals with low and uniform spectral energy distribution. Through them, the energy level of the interval with low and uniform spectral energy distribution, i.e., the background energy level, can be understood, and the interval with concentrated energy distribution can be further inferred.

[0089] Based on the vibration frequency domain signal i, the amplitudes corresponding to the stable frequency point set points are obtained, and the squares of all corresponding amplitudes are calculated to obtain stable frequency point energies. The average WP and standard deviation WB of all stable frequency point energies are calculated. (WP+k8*WB) is recorded as the basic energy threshold WY; where k8 is a set proportion coefficient; in this embodiment, k8=3, which can be flexibly set.

[0090] The square of the amplitude corresponding to each frequency point in the frequency point sequence is calculated, recorded as the corresponding frequency point energy, and the frequency point with a frequency point energy greater than (k9*WY) is recorded as a high amplitude point. The maximum high amplitude point is obtained, and the corresponding frequency is recorded as the maximum distribution point of the vibration frequency domain signal i; where k9 is a set proportion coefficient; k9>1, in this embodiment, k9=1.5, which can be flexibly set to avoid misjudgment of high amplitude points, and (k9*WY) is used to find frequency points with energy significantly higher than the background energy.

[0091] The maximum distribution points of the vibration frequency domain signals 1 to n are repeatedly obtained and sequentially recorded as maximum distribution point 1 to maximum distribution point n.

[0092] For the maximum distribution point i, the maximum distribution point i is excluded from the maximum distribution point 1 to the maximum distribution point n, and a linear fitting is performed on the remaining maximum distribution points to obtain a corresponding fitting line, and a fitting value of the maximum distribution point i in the corresponding fitting line is obtained, and a difference value of the maximum distribution point i and the corresponding fitting value, i.e. the maximum distribution point i minus the corresponding fitting value, is recorded as a distribution deviation; the exclusion point can avoid the influence of the abnormal point on the fitting line, and a more accurate fitting line is obtained;

[0093] The distribution deviations of the maximum distribution points 1 to n are repeatedly calculated, and are sequentially recorded as distribution deviation 1 to distribution deviation n, and are marked as distribution deviation data;

[0094] In the specific implementation process, the maximum distribution points of each monitoring unit are arranged in a spatial order to obtain the maximum distribution point 1 to the maximum distribution point n, i.e. the distribution of the maximum distribution with the position of the monitoring unit; the sequence is used for fitting, which can be used to detect the shift of the maximum distribution point of the monitoring unit, such as the maximum distribution point of the monitoring unit deviating from the overall trend, which may be caused by the underground cavity.

[0095] The cavity determination module determines and analyzes the position of each optical fiber sensor according to the distribution deviation data;

[0096] The cavity determination module is configured with a cavity determination strategy, and the cavity determination strategy includes: if the corresponding distribution deviation i of the maximum distribution point i is greater than e0, it is judged that the position of the corresponding monitoring unit i is suspected to have an underground cavity, wherein e0 is a set threshold; in this embodiment, e0=40, which can be flexibly set according to the actual application scene;

[0097] And according to the collected effective vehicle vibration information data, the position of the monitoring unit i is repeatedly judged, if the position of the monitoring unit i is continuously judged e1 times that the position of the monitoring unit i is suspected to have an underground cavity, it is judged that the position of the monitoring unit i is determined to have an underground cavity, and a warning is given; wherein e1 is a set number, in this embodiment, e1=3, which can be flexibly set;

[0098] The monitoring unit 1 to the monitoring unit n are repeatedly judged, and corresponding warnings are given;

[0099] In the specific implementation process, under normal circumstances, when the position of the monitoring unit has an underground cavity, the frequency spectrum energy distribution of the generated vibration signal will move to the high frequency direction, i.e. if the corresponding distribution deviation i of the maximum distribution point i is greater than e0, if the corresponding distribution deviation i of the maximum distribution point i is less than (-e0), it may be related to other local structure changes, such as settlement, material degradation, which can be further detected.

[0100] Embodiment 2, please refer to Figure 2As shown, the application provides a highway underground cavity real-time monitoring method based on distributed optical fiber sensing, which comprises the following steps:

[0101] Step S1, determine the highway to be detected, use multiple optical fiber sensors to monitor the vibration information when the vehicle passes, and monitor the vehicle information to obtain the vehicle vibration information data; Step S1 comprises the following sub-steps:

[0102] Step S101, for a section of highway to be monitored, denoted as a monitoring section, lay optical fibers for monitoring in the underground of the monitoring section and continuously along the longitudinal direction of the monitoring section, and uniformly set multiple optical fiber sensors for collecting vibration signals along the laid optical fibers for monitoring, and sequentially denoted as monitoring unit 1 to monitoring unit n along the travel direction of the monitoring section, wherein n is the total number of monitoring units;

[0103] Step S102, denote any one monitoring unit as monitoring unit i, wherein i∈[1, n], and denote any one vehicle as the first vehicle;

[0104] Step S103, let the monitoring unit i collect the corresponding amplitude in real time, and arrange in time sequence, denoted as amplitude time sequence, repeat the collection of amplitude time sequence of all monitoring units.

[0105] Step S104, when the first vehicle passes through the position of the monitoring unit i on the monitoring section, use the monitoring unit i to collect the corresponding amplitude when the first vehicle passes, and arrange in time sequence, denoted as the vibration information i of the first vehicle;

[0106] Step S105, and obtain the vehicle speed and the license plate of the first vehicle when the first vehicle passes through the monitoring unit i, denoted as the vehicle information i of the first vehicle;

[0107] Step S106, sequentially repeat the collection of corresponding vibration information and vehicle information when the first vehicle passes through the monitoring unit, if the first vehicle completely passes through the monitoring unit 1 to the monitoring unit n, then the first vehicle is denoted as an effective vehicle; and the vibration information 1 to the vibration information n corresponding to the first vehicle are denoted as the monitoring vibration information of the first vehicle, and the vehicle information 1 to the vehicle information n corresponding to the first vehicle are denoted as the monitoring vehicle information of the first vehicle;

[0108] Step S107, the monitoring vibration information of the first vehicle and the monitoring vehicle information of the first vehicle are denoted as the vehicle vibration information data of the first vehicle.

[0109] Step S2, perform vibration screening processing on the vehicle vibration information data, and perform vibration information preprocessing to obtain standard vibration information data; Step S2 comprises the following sub-steps:

[0110] Step S201, for the vibration information i of the first vehicle; the latest, and without any vehicle passing, and continuous not less than the first time length of the segment of the amplitude time sequence of the monitoring unit i before the first vehicle passing is intercepted, recorded as the reference amplitude segment, wherein the first time length is T1;

[0111] Step S202, any one amplitude in the reference amplitude segment is recorded as BA, the reference amplitude segment is arranged in order of size, and the median is obtained, recorded as MA;

[0112] Step S203, the absolute difference value of BA and MA is calculated, recorded as the absolute deviation corresponding to BA; the absolute deviation corresponding to all amplitudes in the reference amplitude segment is repeatedly calculated, and the corresponding absolute deviation sequence is obtained;

[0113] Step S204, the absolute deviation sequence is arranged in order of size, and the median is obtained, recorded as MC; for BA, if the absolute deviation corresponding to BA is greater than (k1*MC), BA is determined as an abnormal amplitude, otherwise BA is determined as a normal amplitude, wherein k1 is a set proportion coefficient;

[0114] Step S205, all amplitudes in the reference amplitude segment are repeatedly determined, and all abnormal amplitudes are removed, and the first amplitude sequence is obtained.

[0115] Step S206, the first amplitude sequence is evenly divided into three segments, and sequentially recorded as the first amplitude segment, the second amplitude segment and the third amplitude segment in time sequence;

[0116] Step S207, the average values of the first amplitude segment, the second amplitude segment and the third amplitude segment are calculated respectively, sequentially recorded as CP1, CP2 and CP3; and the standard deviations of the first amplitude segment, the second amplitude segment and the third amplitude segment are calculated respectively, sequentially recorded as CB1, CB2 and CB3;

[0117] Step S208, the weights of the first amplitude segment, the second amplitude segment and the third amplitude segment are sequentially set as Q1, Q2 and Q3, Q1<Q2<Q3, Q1+Q2+Q3=1; the weighted mean QP and the weighted standard deviation QB are calculated respectively, wherein QP=Q1*CP1+Q2*CP2+Q3*CP3, QB=Q1*CB1+Q2*CB2+Q3*CB3; (QP+k2*QB) is calculated, recorded as the dynamic threshold of the vibration information i, wherein k2 is a set proportion coefficient.

[0118] Step S209, the average value of the maximum k3% amplitude in the vibration information i is calculated, recorded as the maximum average value; if the maximum average value of the vibration information i is less than the corresponding dynamic threshold, the vibration information i is recorded as invalid information; otherwise, it is marked as normal information, wherein k3% is a set proportion;

[0119] Step S210, continue to judge the vibration information 1 to the vibration information n corresponding to the first vehicle, if there is invalid information in the vibration information 1 to the vibration information n corresponding to the first vehicle, mark the vehicle vibration information data corresponding to the first vehicle as invalid; otherwise, mark as suspicious;

[0120] Step S211, if the vehicle vibration information data corresponding to the first vehicle is marked as suspicious, get all vehicle speed sizes in the vehicle information 1 to the vehicle information n corresponding to the first vehicle, denoted as vehicle speed set, calculate the coefficient of variation of the vehicle speed set, denoted as speed variation coefficient CV, if CV>k4, mark the vehicle vibration information data corresponding to the first vehicle as invalid, otherwise mark the vehicle vibration information data corresponding to the first vehicle as valid, wherein k4 is a set threshold value;

[0121] Step S212, repeat collecting valid vehicle vibration information data of valid vehicles; if the vehicle vibration information data corresponding to the first vehicle is marked as valid; Fourier transform is performed on the vibration information 1 to the vibration information n corresponding to the first vehicle in turn, and vibration frequency domain signals 1 to vibration frequency domain signals n are obtained in turn, denoted as standard vibration information data.

[0122] Step S3, perform vibration energy analysis according to the standard vibration information data, and perform distribution deviation analysis on the monitoring position of each optical fiber sensor to obtain distribution deviation data; step S3 includes the following substeps:

[0123] Step S301, for the vibration frequency domain signal i, get the frequency point corresponding to each amplitude, and arrange all the frequency points in ascending order, denoted as frequency point sequence;

[0124] Step S302, set the size of the sliding window as k5 and the sliding step as k6, and slide from the start of the frequency point sequence to the back in turn, and arrange the sliding window of each sliding in turn as window 1 to window n1 according to the sliding order, wherein n1 is the total number of windows, and any window is denoted as window j, j∈[1, n1];

[0125] Step S303, based on the vibration frequency domain signal i, get the amplitudes corresponding to all the frequency points in the window j, and calculate the square of all the corresponding amplitudes respectively, and sum them up, denoted as the spectral energy of the window j, repeat the calculation of the spectral energy of the window 1 to the window n1, and arrange all the spectral energies in ascending order, denoted as window energy sequence.

[0126] Step S304, take the spectrum energy of the first k7% of the window energy sequence, and mark the corresponding window as a stable window; mark the frequency points in the stable window as stable frequency points, obtain the stable frequency points in all stable windows, and remove duplicates to obtain a stable frequency point set, where k7% is a set proportion;

[0127] Step S305, based on the vibration frequency domain signal i, obtain the amplitude corresponding to the stable frequency point set, and calculate the square of all corresponding amplitudes to obtain stable frequency point energy, and calculate the average WP and standard deviation WB of all stable frequency point energy; mark (WP+k8*WB) as the basic energy threshold WY; where k8 is a set proportion coefficient;

[0128] Step S306, calculate the square of the amplitude corresponding to each frequency point in the frequency point sequence, mark it as the corresponding frequency point energy, and mark the frequency point with a frequency point energy greater than (k9*WY) as a high amplitude point, obtain the maximum high amplitude point, and mark the corresponding frequency as the maximum distribution point of the vibration frequency domain signal i; where k9 is a set proportion coefficient;

[0129] Step S307, repeat the maximum distribution points of vibration frequency domain signal 1 to vibration frequency domain signal n, and mark them in order as maximum distribution point 1 to maximum distribution point n.

[0130] Step S308, for the maximum distribution point i, remove the maximum distribution point i from the maximum distribution point 1 to the maximum distribution point n, and linearly fit the remaining maximum distribution points to obtain the corresponding fitting line, and obtain the fitting value of the maximum distribution point i in the corresponding fitting line, and calculate the difference between the maximum distribution point i and the corresponding fitting value, mark it as the distribution deviation;

[0131] Step S309, repeat the calculation of the distribution deviation of the maximum distribution point 1 to the maximum distribution point n, and mark them in order as distribution deviation 1 to distribution deviation n, and mark them as distribution deviation data.

[0132] Step S4, according to the distribution deviation data, the monitoring position of each optical fiber sensor is analyzed and judged for underground cavity; Step S4 includes the following sub-steps:

[0133] Step S401, for the maximum distribution point i, if the corresponding distribution deviation i is greater than e0, it is determined that the position of the corresponding monitoring unit i is suspected to have an underground cavity, where e0 is a set threshold;

[0134] Step S402, and according to the collected effective vehicle vibration information data, the position of the monitoring unit i is repeatedly judged, if the position of the monitoring unit i is continuously judged e1 times to have an underground cavity, it is determined that the position of the monitoring unit i has an underground cavity, and a warning is given, where e1 is a set number;

[0135] Step S403, repeat the judgment on monitoring unit 1 to monitoring unit n, and carry out the corresponding early warning.

[0136] Embodiment 3, please refer to Figure 4 As shown, Figure 4 An example of a structural diagram of an electronic device, which can include: processor, communication interface, memory and communication bus, wherein the processor, communication interface, memory through the communication bus complete mutual communication. The memory stores computer readable instructions, and the processor can call the instructions in the memory, when the computer readable instructions are executed by the processor, run as in the steps of the method for real-time monitoring of highway underground cavity based on distributed optical fiber sensing, to realize the following functions: determine the highway to be detected, use a plurality of optical fiber sensors to monitor the vibration information when the vehicle passes, and monitor the vehicle information, get the vehicle vibration information data; vibration screening processing is carried out on the vehicle vibration information data, and vibration information preprocessing is carried out, to obtain standard vibration information data; according to the standard vibration information data, vibration energy analysis is carried out, and the distribution deviation of the monitoring position of each optical fiber sensor is analyzed, to obtain the distribution deviation data; according to the distribution deviation data, the monitoring position of each optical fiber sensor is analyzed.

[0137] In addition, the logical instructions in the memory described above can be realized in the form of software functional units and sold or used as independent products when used, which can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or say the part that contributes to the prior art or part of the technical solutions can be embodied in the form of software products, and the computer software product is stored in a storage medium, including a plurality of instructions to make a computer device (may be a personal computer, server, or network equipment, etc.) execute all or part of the steps of the method described in various embodiments of the present application. The foregoing storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and various program code storage media.

[0138] In embodiment 4, the application further provides a computer readable storage medium, and the application provides a storage medium, which stores a computer program, and the computer program is executed by a processor to run the steps in the method for monitoring underground cavities of expressways in real time based on distributed optical fiber sensing, so as to realize the following functions: determining an expressway to be detected, monitoring vibration information when a vehicle passes by using a plurality of optical fiber sensors, and monitoring vehicle information to obtain vehicle vibration information data; performing vibration screening processing on the vehicle vibration information data, and performing vibration information preprocessing to obtain standard vibration information data; performing vibration energy analysis according to the standard vibration information data, and performing distribution deviation analysis on the monitoring positions of each optical fiber sensor to obtain distribution deviation data; and performing underground cavity determination analysis on the monitoring positions of each optical fiber sensor according to the distribution deviation data.

[0139] Through the description of the above embodiments, the embodiments of the application can be provided as a method, a system or a computer program product. Based on such understanding, the above technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the method described in each embodiment or some parts of the embodiment.

[0140] In the embodiments provided by the present application, it should be understood that the disclosed system or method can be implemented in other ways. The above described embodiments are merely illustrative. For example, the division of the modules or units is merely a logical function division, and there can be another division manner in actual implementation. For example, a plurality of modules or units can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the shown or discussed modules can be indirect coupling or communication connection through some communication interfaces. The coupling or communication connection can be electrical, mechanical or in other forms.

[0141] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, but not limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A real-time monitoring system for underground cavities in highways based on distributed optical fiber sensing, characterized in that, It includes an information acquisition module, an information processing module, a deviation analysis module, and a void detection module; The information acquisition module is used to determine the highway to be detected, use multiple fiber optic sensors to monitor the vibration information of vehicles passing by, and monitor vehicle information to obtain vehicle vibration information data. The information processing module is used to perform vibration screening on vehicle vibration information data and to perform vibration information preprocessing to obtain standard vibration information data. The deviation analysis module includes an energy analysis unit and a distribution analysis unit. The energy analysis unit performs vibration energy analysis based on standard vibration information data, and the distribution analysis unit performs distribution deviation analysis on the monitoring position of each fiber optic sensor to obtain distribution deviation data. The cavity detection module performs underground cavity detection analysis on the monitoring location of each fiber optic sensor based on the distribution deviation data. The distributed analysis unit is configured with a distributed analysis strategy, which includes: Take the first k7% of the spectral energy of the window energy sequence and mark the corresponding window as the stable window; mark the frequency points in the stable window as the stable frequency points, obtain the stable frequency points in all stable windows, remove duplicates, and obtain the set of stable frequency points, where k7% is the set ratio; Based on the vibration frequency domain signal i, the amplitude corresponding to the set point of the stable frequency point is obtained, and the square of all corresponding amplitudes is calculated and denoted as the stable frequency point energy. The average value WP and standard deviation WB of all stable frequency point energies are calculated. (WP+k8*WB) is denoted as the basic energy threshold WY; where k8 is the set scaling factor. Calculate the square of the amplitude corresponding to each frequency point in the frequency point sequence, and record it as the corresponding frequency point energy. Record the frequency points with frequency point energy greater than (k9*WY) as high amplitude points, obtain the maximum high amplitude point, and record the corresponding frequency as the maximum distribution point of the vibration frequency domain signal i; where k9 is the set scaling factor. Repeatedly acquire the maximum distribution points of vibration frequency domain signal 1 to vibration frequency domain signal n, and record them sequentially as maximum distribution point 1 to maximum distribution point n; For the maximum distribution point i, remove the maximum distribution point i from the maximum distribution points 1 to n, and perform linear fitting on the remaining maximum distribution points to obtain the corresponding fitting line. Obtain the fitting value of the maximum distribution point i in the corresponding fitting line, and calculate the difference between the maximum distribution point i and the corresponding fitting value, which is denoted as the distribution deviation. Repeatedly calculate the distribution deviations from the maximum distribution point 1 to the maximum distribution point n, and record them sequentially as distribution deviation 1 to distribution deviation n, marking them as distribution deviation data.

2. The real-time monitoring system for underground cavities in highways based on distributed optical fiber sensing according to claim 1, characterized in that, The information collection module is configured with information collection strategies, which include: For a section of highway to be monitored, it is called the monitoring section. Under the monitoring section, the optical fiber used for monitoring is continuously laid in the longitudinal direction of the monitoring section. Multiple optical fiber sensors for collecting vibration signals are evenly set along the laid monitoring optical fiber. They are sequentially called monitoring unit 1 to monitoring unit n along the monitoring section in the direction of travel, where n is the total number of monitoring units. Let any monitoring unit be denoted as monitoring unit i, where i∈[1,n], and let any vehicle be denoted as the first vehicle; The monitoring unit i collects the corresponding amplitude in real time and arranges them in chronological order, recording them as an amplitude time series. The amplitude time series of all monitoring units are collected repeatedly.

3. The real-time monitoring system for underground cavities in highways based on distributed optical fiber sensing according to claim 2, characterized in that, Information collection strategies also include: When the first vehicle passes the location of monitoring unit i on the monitored road section, the amplitude value corresponding to the passing of the first vehicle is collected by monitoring unit i and arranged in chronological order, and recorded as the vibration information i of the first vehicle. And obtain the vehicle speed and license plate of the first vehicle when it passes through monitoring unit i, and record them as vehicle information i of the first vehicle; The vibration information and vehicle information corresponding to the first vehicle passing through the monitoring unit are collected in sequence. If the first vehicle passes through monitoring unit 1 to monitoring unit n completely, the first vehicle is recorded as a valid vehicle; and the vibration information 1 to vibration information n corresponding to the first vehicle is recorded as the monitoring vibration information of the first vehicle, and the vehicle information 1 to vehicle information n corresponding to the first vehicle is recorded as the monitoring vehicle information of the first vehicle. The vibration information of the first vehicle and the vehicle information of the first vehicle are recorded as the vehicle vibration information data of the first vehicle.

4. The real-time monitoring system for underground cavities in highways based on distributed optical fiber sensing according to claim 3, characterized in that, The information processing module is configured with information processing strategies, which include: For the vibration information i of the first vehicle; extract the closest segment before the first vehicle passes through the amplitude time series of monitoring unit i, where no other vehicle passes through, and the segment is continuous and not less than the first time length, and record it as the reference amplitude segment, where the first time length is T1; Let any one of the amplitudes in the reference amplitude segment be denoted as BA. Arrange the reference amplitude segments in order of size and obtain the median, which is denoted as MA. Calculate the absolute difference between BA and MA, and denote it as the absolute deviation corresponding to BA; repeat the calculation of the absolute deviation corresponding to all amplitudes in the reference amplitude segment to obtain the corresponding absolute deviation sequence; Arrange the absolute deviation sequence in order of magnitude and obtain the median, denoted as MC; for BA, if the absolute deviation corresponding to BA is greater than (k1*MC), then BA is determined to be an abnormal amplitude, otherwise BA is determined to be a normal amplitude, where k1 is the set scaling factor; Repeat the judgment process for all amplitudes in the reference amplitude segment and remove all abnormal amplitudes to obtain the first amplitude sequence.

5. The real-time monitoring system for underground cavities in highways based on distributed optical fiber sensing according to claim 4, characterized in that, Information processing strategies also include: The first amplitude sequence is evenly divided into three segments, which are denoted as the first amplitude segment, the second amplitude segment, and the third amplitude segment in chronological order. Calculate the average values ​​of the first amplitude segment, the second amplitude segment, and the third amplitude segment respectively, and denote them as CP1, CP2, and CP3 in order; and calculate the standard deviations of the first amplitude segment, the second amplitude segment, and the third amplitude segment respectively, and denote them as CB1, CB2, and CB3 in order. Set the weights of the first amplitude segment, the second amplitude segment, and the third amplitude segment as Q1, Q2, and Q3 in sequence, where Q1 < Q2 < Q3 and Q1 + Q2 + Q3 = 1; calculate the weighted mean QP and the weighted standard deviation QB respectively, where QP = Q1 * CP1 + Q2 * CP2 + Q3 * CP3 and QB = Q1 * CB1 + Q2 * CB2 + Q3 * CB3; calculate (QP + k2 * QB), which is denoted as the dynamic threshold of vibration information i, where k2 is the set proportionality coefficient.

6. The real-time monitoring system for underground cavities in highways based on distributed optical fiber sensing according to claim 5, characterized in that, The information processing strategy further includes: Calculate the average value of the amplitudes of the largest k3% in the vibration information i, which is denoted as the maximum mean value; if the maximum mean value of the vibration information i is less than the corresponding dynamic threshold, then mark the vibration information i as invalid information; otherwise, mark it as normal information, where k3% is the set ratio. Repeat the judgment on the vibration information 1 to the vibration information n corresponding to the first vehicle. If there is invalid information in the vibration information 1 to the vibration information n corresponding to the first vehicle, then mark the vehicle vibration information data corresponding to the first vehicle as invalid; otherwise, mark it as suspected. If the vehicle vibration information data corresponding to the first vehicle is marked as suspected, then obtain all the vehicle speed magnitudes in the vehicle information 1 to the vehicle information n corresponding to the first vehicle, which is denoted as the vehicle speed set, and calculate the coefficient of variation of the vehicle speed set, which is denoted as the speed coefficient of variation CV. If CV > k4, then mark the vehicle vibration information data corresponding to the first vehicle as failed; otherwise, mark the vehicle vibration information data corresponding to the first vehicle as valid, where k4 is the set threshold. Repeat collecting the effective vehicle vibration information data of the valid vehicles; if the vehicle vibration information data corresponding to the first vehicle is marked as valid; then perform Fourier transform on the vibration information 1 to the vibration information n corresponding to the first vehicle in sequence, and obtain the vibration frequency domain signals 1 to the vibration frequency domain signals n in sequence, which are denoted as the standard vibration information data.

7. The real-time monitoring system for underground cavities in highways based on distributed optical fiber sensing according to claim 6, characterized in that, The energy analysis unit is configured with an energy analysis strategy, and the energy analysis strategy includes: For the vibration frequency domain signal i, obtain the frequency points corresponding to each amplitude, and arrange all the frequency points in ascending order, which is denoted as the frequency point sequence. Set the sliding window size as k5 and the sliding step size as k6, and slide backward from the start of the frequency point sequence in sequence. Denote each sliding window in the sliding order as window 1 to window n1 in sequence, where n1 is the total number of windows, and denote any one window as window j, j ∈ [1, n1]. Based on the vibration frequency domain signal i, obtain the amplitudes corresponding to all the frequency points in window j, and calculate the squares of all the corresponding amplitudes respectively, and sum them up, which is denoted as the spectral energy of window j. Repeat the calculation of the spectral energy of window 1 to window n1, and arrange all the spectral energies in ascending order, which is denoted as the window energy sequence.

8. The real-time monitoring system for underground cavities in highways based on distributed optical fiber sensing according to claim 7, characterized in that, The cavity determination module is configured with a cavity determination strategy, and the cavity determination strategy includes: For the maximum distribution point i, if the corresponding distribution deviation i is greater than e0, then it is judged that there may be an underground cavity at the position of the corresponding monitoring unit i, where e0 is the set threshold. The location of monitoring unit i is repeatedly determined based on the valid vehicle vibration information data of the collected vehicles. If the location of monitoring unit i is suspected to have an underground cavity for e1 consecutive times, the location of monitoring unit i is confirmed to have an underground cavity, and an early warning is issued. Here, e1 is the number of units set. Repeatedly judge monitoring units 1 to n and issue corresponding warnings.

Images