An anti-interference method for drainage pipeline sensing signals based on distributed optical fiber sensing

Abstract

The present invention relates to the field of optical fiber sensing technology, and particularly relates to an anti-interference method for drainage pipe sensing signals based on distributed optical fiber sensing. By using acquisition software to collect the original signal data of the effective monitoring section of the drainage pipe in real time, strain rate data is obtained; based on a filter, real-time integration and filtering conversion of the strain rate data are performed to obtain strain data, and then Hilbert transform is carried out to form a sequence of analytic signals; the vibration amplitude of the analytic signal is calculated continuously in short windows, the vibration amplitude distribution is statistically analyzed, and abnormally large values on the channel are obtained; based on the abnormally large values, the vibration waterfall diagram is binarized, and the Hough transform is used to detect the traffic signal part; the data in the time period of the interference signal is removed from the original data to obtain the signal in the non-interfered time period, that is, the water flow signal. This method converts the DAS signal into image data, identifies and removes the areas significantly interfered by traffic signals, so that the remaining part can better reflect the intensity of the real water flow vibration signal.

Description

Technical Field

[0001] The present invention relates to the technical field of optical fiber sensing, and particularly relates to an anti-interference method for drainage pipe sensing signals based on distributed optical fiber sensing. Background Art

[0002] Urban drainage pipes are often located on both sides of roads or directly under roads, and the vibrations generated by frequent traffic flows on the roads will interfere with the signal detection inside the pipes themselves.

[0003] In the prior art, for example, the "Online Monitoring System and Method for Drainage Pipes Based on OFDR Distributed Optical Fibers" disclosed in the Chinese invention patent application (publication number CN108731743A) describes arranging 7 optical fibers inside the drainage pipes and attaching them to the inner wall (3 for strain + 3 for strain compensation + 1 for top temperature measurement), and calculating the fluid flow rate, pipe fullness, siltation, and defects. This technology uses multiple groups of optical fibers (adding strain compensation optical fibers) and uses the differences in vibration perception of physically different optical fibers to resist interference.

[0004] Since each DAS device has a limited detection distance in the above technology, multiple strain sensing optical cables need to be arranged inside the pipes, so the length of the water pipes that can be detected will be greatly shortened; the optical cables that need to be arranged need to be fixed at specific positions in the pipes, and epoxy resin is required for fixation in the upper section of the water pipes, which poses certain difficulties for construction, especially for pipes with relatively small diameters, as it is not easy for construction workers to enter. Summary of the Invention

[0005] The purpose of the present invention is to provide an anti-interference method for drainage pipe sensing signals based on distributed optical fiber sensing, aiming to reduce the interference of traffic vibration signals through post-signal processing.

[0006] To achieve the above purpose, the present invention provides an anti-interference method for drainage pipe sensing signals based on distributed optical fiber sensing, including the following steps:

[0007] Set an appropriate sampling rate, and collect the original signal data of the effective monitoring section of the drainage pipe in real time through acquisition software to obtain strain rate data;

[0008] Based on a filter, perform real-time integration and filtering conversion on the strain rate data to obtain strain data;

[0009] Perform Hilbert transform on the strain data and form a sequence of analytical signals;

[0010] Use a continuous short window to calculate the vibration amplitude of the analytical signal, and statistically analyze the vibration amplitude distribution to obtain abnormally large values on the channel;

[0011] Based on the abnormally large values, the vibration waterfall Figure 2Value quantization, using the Hough transform to detect the traffic signal part;

[0012] Remove the data of the time period with interference signals from the original data to obtain the signal of the non-interfered time period, that is, the water flow signal.

[0013] Among them, the sampling rate is at least twice the maximum frequency of the on-site traffic vibration signal.

[0014] Among them, the filter is a second-order Butterworth IIR type high-pass filter.

[0015] Among them, the cut-off frequency standard of the filter is not to cause temperature drift.

[0016] Among them, the specific method of using a continuous short window to calculate the vibration amplitude of the analytic signal and statistically analyze the vibration amplitude distribution to obtain the abnormally large value on the channel:

[0017] Select a continuous short time slice and calculate the average vibration amplitude of the analytic signal;

[0018] Accumulate the average vibration amplitudes of the signals of a preset quantity and perform statistical analysis to obtain the abnormally large value on the channel.

[0019] An anti-interference method for sensing signals of drainage pipes based on distributed optical fiber sensing according to the present invention sets a suitable sampling rate, and collects the original signal data of the effective monitoring section of the drainage pipe in real time through a collection software to obtain strain rate data; based on a filter, perform real-time integration and filtering conversion on the strain rate data to obtain strain data; perform Hilbert transform on the strain data and form a sequence of analytic signals; use a continuous short window to calculate the vibration amplitude of the analytic signal and statistically analyze the vibration amplitude distribution to obtain the abnormally large value on the channel; based on the abnormally large value, the vibration waterfall Figure 2Value quantization is used to detect the traffic signal part by using the Hough transform; the data of the time period with interference signals is removed from the original data to obtain the signal of the non-interfered time period, that is, the water flow signal. This method transforms the DAS signal into image data, identifies and removes the areas significantly interfered by traffic signals, so that the remaining part can better reflect the real water flow vibration signal intensity. It identifies and removes the traffic vehicle interference signals, retains the useful signal part for monitoring, does not involve the change of the optical cable layout method, and is different from the existing method using strain compensation. It can achieve a similar anti-interference effect at a lower layout cost. It can process the collected signals to remove traffic signal interference and obtain accurate water flow signals when only one optical cable is laid at the bottom of the drainage pipe. This method only lays one optical cable inside the drainage pipe, without the combination of strain compensation optical fibers, is economically feasible and easy to promote. Through carefully designed array signal processing, the interference can be accurately located in the signal data, so a better detection effect can be obtained under the condition of using the same data. Description of the Drawings

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0021] Figure 1 It is a flowchart of an anti-interference method for drainage pipe sensing signals based on distributed optical fiber sensing provided by the present invention.

[0022] Figure 2 It is a flowchart of an anti-interference method for drainage pipe sensing signals based on distributed optical fiber sensing provided by the present invention.

[0023] Figure 3 It is a flowchart of the specific method of calculating the vibration amplitude of the analytic signal using a continuous short window, statistically analyzing the vibration amplitude distribution, and obtaining the abnormally large values on the channel.

[0024] Figure 4 It is a schematic diagram of the calculation structure of a digital integral filter system.

[0025] Figure 5 It is a box plot using quartiles and a normal distribution probability plot of the corresponding area.

[0026] Figure 6 It is a schematic diagram of the result distribution of RMS statistics. Among them, (a) shows the distribution of RMS data using a box plot, and (b) shows the distribution of abnormally large values in a waterfall plot. Detailed implementation manners

[0027] The embodiments of the present invention will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the accompanying drawings are exemplary and are intended to explain the present invention, and should not be construed as a limitation to the present invention.

[0028] Please refer to Figures 1 to 6 , the present invention provides an anti-interference method for drainage pipeline sensing signals based on distributed optical fiber sensing, including the following steps:

[0029] S1 Set an appropriate sampling rate, and collect the original signal data of the effective monitoring section of the drainage pipeline in real time through the acquisition software to obtain strain rate data;

[0030] In the embodiments of the present invention, the sampling rate is at least twice the maximum frequency of the vibration signal in the field environment (limited by the Nyquist sampling law). Among them, the setting of the acquisition rate is restricted by the Nyquist sampling theorem and the highest vibration frequency in the current sampling environment, and it is better that there is no obvious over-π data in the system after setting. The spatial resolution and the channel spacing stipulate the density of the sensors in space. In this system, in order to maximize the decoupling effect, the set spatial resolution and channel spacing are the best values that the current device can achieve (the smaller the two values, the better). The required number of sampling points is calculated according to the length of the optical cable and the channel spacing (the number of sampling points = the length of the optical cable to be measured / the channel spacing), and the number of sampling points needs to cover the entire length of the optical cable to be detected. The acquisition duration is based on the actual duration of the test, and can completely collect the entire test experiment process. After the project commissioning is completed, the data acquisition will continue uninterruptedly. After the basic parameters of the equipment are set, the entire system is tested (including whether the data network is connected, whether the file recording is normal, etc.) to ensure that the entire system starts to run. Then the DAS device demodulation module calculates the differential phase (strain rate) data and sends it to the industrial control computer. The acquisition software is used to read the data sent by the sensor module (that is, the strain rate data, corresponding to the original signal data of the effective monitoring section of the drainage pipeline). The basic parameters of the DAS device include the sampling rate, spatial resolution and channel spacing of the system, and the number of its sampling points.

[0031] S2 Based on the filter, perform real-time integration and filtering conversion on the strain rate data to obtain strain data;

[0032] In the embodiments of the present invention, the filter is a second-order Butterworth IIR high-pass filter, and the cut-off frequency standard of the filter is not to cause temperature drift.

[0033] Specifically, it is necessary to integrate the differential phase (strain rate) data to obtain strain data, and a filter is needed to filter out the interference of low-frequency noise in the strain data. The device noise existing in the DAS itself is in the low-frequency band less than 1 Hz. At the same time, temperature drift needs to be removed during data acquisition. Select an appropriate high-pass filter type according to specific application requirements, such as the Butterworth high-pass filter, to reduce the low-frequency noise component and temperature drift in the signal. Since continuous integration and filtering operations are required, in this example, a new function of fused integration filtering is used, and its system structure is as shown in Figure 4 ; where represents delaying the result once, that is, outputting the previous result; HPF is a second-order Butterworth IIR high-pass filter. The N-order difference equation of the input-output relationship of the second-order Butterworth IIR direct type II high-pass filter is:

[0034]

[0035] Then the second-order difference equation is:

[0036]

[0037] Where is the input of the system at time n, is the output of the system at time n, and are the coefficients of the equation. By mixing the input at the current moment, the input at the past moment, and the output, the output at the current moment is obtained.

[0038] The beneficial aspect of using this equation to calculate is to avoid the continuous increase of the absolute value of the numerical value after a long time in the separate integration operation, thereby causing the accumulation and overflow of the calculation error of the floating-point value used by the computer for calculation (exceeding the IEEE754 floating-point representation range).

[0039] S3 performs a Hilbert transform on the strain data and forms a sequence of analytic signals;

[0040] In the embodiment of the present invention, the Hilbert transform is performed on the strain data and it is changed to , and and are used to form a sequence of analytic signals , where j is the imaginary part, indicating that this function is a complex function (rather than an ordinary real function). Specifically, the strain data corresponding to the physical quantity can be regarded as a real-valued function, and its Hilbert transform is:

[0041]

[0042] Among them, is the signal after Hilbert transform, H is the Hilbert transform operator, the original signal, is a constant factor for normalization, is the symbol of infinite summation, representing integration over all time points, is the integral term, where is the value of the original signal at time point value, represents the current time point and the difference between the integral variable is

[0043] Its physical meaning is to delay the phase of all frequency components of the signal by 90 degrees. By , the real part and the imaginary part of the signal can be combined to obtain the analytical signal form of the signal in the complex domain. This operation improves the contrast of the vibration waterfall diagram to improve the accuracy of subsequent operations.

[0044] S4 uses a continuous short window to calculate the vibration amplitude of the analytical signal, and statistically analyzes the vibration amplitude distribution to obtain an abnormally large value on the channel;

[0045] In the embodiment of the present invention, that is, taking a few seconds as the time window, calculating the RMS (i.e., vibration intensity) within each time window, and then performing quartile statistical calculation on the RMS sequence to obtain an abnormally large value. The abnormally large value is used to distinguish non-water flow signals.

[0046] Specifically, the amplitude of the vibration signal will change significantly within a short period of time (within one cycle), so it is necessary to statistically calculate the average amplitude within a time window. In this example, 2 seconds is selected as the statistical time window length according to observation, because the duration of the vibration caused by the vehicle driving on the ordinary road can be maintained near this value. The RMS calculated statistically uses the following formula:

[0047]

[0048] Among them, is the strain amplitude of each sampling, which is the modulus of the analytical function , N is the number of samplings within the time window, and RMS, that is, the vibration intensity, represents the geometric mean of the amplitudes of all frequency signals within this time window.

[0049] Specifically, the statistical method of quartiles is used to calculate 300 RMS data within 5 minutes. Quartiles are the position values that divide a set of data into four equal parts after arranging the data in ascending order. They are the first quartile (Q1), the second quartile (Q2, i.e., the median), and the third quartile (Q3). For a stable water flow, its vibration intensity conforms to a normal distribution, that is, 99.3% of the data is between and , where . In this example, a box plot is used to show the distribution of RMS data ( Figure 6 (a) in), where values greater than are shown as outlier points (abnormally large values, corresponding to the gray-covered area in the waterfall plot ( Figure 6 (b))), and the distribution range of normal data is within the upper and lower edges of the box plot with a groove in the lower part. The part of the statistical data greater than is marked in gray in Figure 6 (b), and it can be found that it includes the traffic trajectory vibration signal shown as a slanted line and some local abnormal vibration signals.

[0050] S5 quantizes the vibration waterfall Figure 2 values based on the abnormally large values, and uses the Hough transform to detect the traffic signal part;

[0051] In the embodiment of the present invention, the progressive probabilistic Hough transform is used to detect the two endpoints of the straight line and accurately locate the straight line in the image.

[0052] S6 removes the data of the time period of the interference signal from the original data to obtain the signal of the non-interfered time period, that is, the water flow signal.

[0053] In the embodiment of the present invention, for Figure 6 (b), the part that belongs to the abnormally large value in the quartile window RMS statistical calculation and the part that is on the straight line detected by the Hough transform are removed in the vibration image (removing the outlier points in Figure 6 (a), deleting the RMS values of the corresponding time period in the RMS sequence, and recalculating the average RMS), obtaining a vibration waterfall diagram with interference removed, and recalculating the corresponding vibration intensity. After this processing, the detected water flow vibration intensity is not easily interfered by ground traffic and sudden vibration signals, and its RMS represents the vibration intensity at most times within this time window. After the RMS remains stable, the drainage alarm detection system using a threshold (alarm when RMS is greater than the threshold) reduces false alarm events.

[0054] The above-disclosed is only a preferred embodiment of a method for anti-interference of drainage pipe sensing signals based on distributed optical fiber sensing of the present invention. Of course, the scope of the rights of the present invention cannot be limited by this. Those of ordinary skill in the art can understand all or part of the processes of implementing the above embodiments, and the equivalent changes made according to the claims of the present invention still fall within the scope covered by the invention.

Claims

1. A method for anti-interference of drainage pipeline sensing signals based on distributed optical fiber sensing, characterized in that, It includes the following steps: Set an appropriate sampling rate, and collect the original signal data of the effective monitoring section of the drainage pipeline in real time through the acquisition software to obtain strain rate data; Based on the filter, perform real-time integration and filtering conversion on the strain rate data to obtain strain data; Perform Hilbert transform on the strain data and form a sequence of analytic signals; Use a continuous short window to calculate the vibration amplitude of the analytic signal, and statistically analyze the vibration amplitude distribution to obtain abnormally large values on the channel; Based on the abnormally large values, binarize the vibration waterfall diagram and use the Hough transform to detect the traffic signal part; Remove the data of the time period with interference signals from the original data to obtain the signal of the non-interfered time period, that is, the water flow signal; The specific method of using a continuous short window to calculate the vibration amplitude of the analytic signal, statistically analyze the vibration amplitude distribution, and obtain abnormally large values on the channel: Select a continuous short time slice and calculate the average vibration amplitude of the analytic signal; Accumulate the average vibration amplitudes of the signals of a preset number, and perform statistical analysis to obtain abnormally large values on the channel.

2. The anti-interference method for drainage pipeline sensing signals based on distributed optical fiber sensing according to claim 1, wherein The sampling rate is at least twice the maximum frequency of the on-site traffic vibration signal.

3. The anti-interference method for drainage pipeline sensing signals based on distributed optical fiber sensing according to claim 1, wherein The filter is a second-order Butterworth IIR high-pass filter.

4. The anti-interference method for drainage pipeline sensing signals based on distributed optical fiber sensing according to claim 1, wherein The cut-off frequency standard of the filter is not to cause temperature drift.

Images