A pipeline leakage monitoring method and system based on optical fiber sensing
The dichotomy eliminates the baseline jump and drift of the fiber signal, and divides the data into fewer segments for uploading, which solves the problems of data redundancy and power consumption in the fiber leakage detection method, and improves the response speed of the pipeline leakage monitoring system and Transmission efficiency.
Patent Information
- Application Number
- CN202411283416.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-13
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-09-13
AI Technical Summary
The existing fiber leak detection method has signal baseline jump and low-frequency drift problems in pipeline leakage monitoring, resulting in data overflow and demodulation failure, and data redundancy leads to excessive power consumption and storage pressure.
The dichotomy method is used to quickly find the median to eliminate signal baseline jumps and drifts, and by constructing a mapping relationship between the original data and leaked parameters, the data is divided into fewer segments for uploading, simplifying program execution and reducing transmission amount.
It effectively solves the problem of signal baseline jump and drift, reduces the computing volume and power consumption of data processing, reduces the storage pressure of cloud platforms, and improves the system response speed and transmission efficiency.
Smart Images

Figure CN118820699B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of signal processing, and in particular to a pipeline leakage monitoring method and system based on optical fiber sensing. Background Art
[0002] With the continuous development and progress of society, the supply and transportation of energy are also increasing, such as water resources, oil, gas, etc., and most energy needs to rely on pipelines of various materials for storage and transmission. In recent years, due to the waste of resources caused by pipeline leakage and even the frequent occurrence of safety accidents, huge losses and adverse effects have been brought to society. In the prior art, the technologies for pipeline leakage monitoring mainly include direct detection method, indirect detection method, negative pressure wave method, infrasound method and optical fiber leak detection method. Among them, the optical fiber leak detection method is mainly based on the balanced interferometer to convert the vibration signal into the phase change of light, and then convert the optical signal into an electrical signal through a photoelectric detector. After the hardware demodulates the vibration signal, the data is transmitted to the cloud server for leakage identification and judgment. Compared with other technical methods, the optical fiber leak detection method has the advantages of high sensitivity, strong anti-interference and low deployment cost, and has been widely used. However, the optical fiber leak detection method still has shortcomings and defects. For example, the sensor lacks an effective and reasonable preprocessing process for the picked-up vibration data, resulting in data redundancy and consuming a lot of battery power.
[0003] In practical applications, due to the influence of large external vibration interference and the integrator in demodulation, the signal baseline of the DCM demodulation output signal θs will inevitably jump and drift at low frequency, and the drift will become larger and larger over time, with a cumulative effect, and finally cause data overflow, that is, the integration starting point cannot be obtained, resulting in demodulation failure. The existing technology uses high-order filters for filtering, and high-order filters mean a huge amount of calculation. In miniaturized and low-power products, low-performance SOC chip systems cannot support real-time big data operations, and cannot eliminate the cumulative effect of the integration starting point.
[0004] In addition, the existing technology usually uploads the data filtered by the low-pass filter directly to the cloud platform for calculation of pipeline leakage judgment. However, due to the large amount of original data, it will cause a certain transmission pressure on the low-power wireless transmission module, and consume a large amount of device battery power, causing the cloud platform to store too much invalid data.
[0005] In view of the above technical problems, the present invention provides a pipeline leakage monitoring method and system based on optical fiber sensing, which eliminates the jump and drift of the signal baseline by quickly finding the median through binary search, and corrects the position of the integration starting point in real time to eliminate the cumulative effect, while well retaining the low-frequency response of the sensor; quickly finds the entry point and exclusion point of the moving window at the specific position of the sorted window data through binary search, without the need for multiple sorting, simplifies the program execution steps, and greatly speeds up the execution speed of the method and the corresponding program; by constructing a mapping relationship between the original data and the leakage parameter, a large amount of redundant and invalid original data is simplified after preprocessing to a loss parameter equal to the number of vibration data after segmentation, that is, the entire uploaded original data N is divided into N / 10 segments, and converted into N / 10 loss data for uploading, which greatly reduces and reduces the transmission amount of the original data, reduces the power consumption of the sensor itself, reduces the storage of invalid and redundant data on the cloud platform, and alleviates the transmission pressure of the low-power wireless transmission module. Summary of the invention
[0006] In view of the above problems, a pipeline leakage monitoring method and system based on optical fiber sensing are provided. The present invention uses a binary search method to quickly find the median to eliminate the jump and drift of the signal baseline, and corrects the position of the integration starting point in real time to eliminate the cumulative effect, while well retaining the low-frequency response of the sensor; the binary search method is used to quickly find the specific position of the entry point and the exclusion point of the moving window in the sorted window data, without the need for multiple sorting, which simplifies the program execution steps and greatly speeds up the execution speed of the method and the corresponding program; by constructing a mapping relationship between the original data and the leakage parameters, a large amount of redundant and invalid original data is simplified after preprocessing to the same number of missing parameters as the segmented vibration data, which greatly reduces and reduces the transmission amount of the original data, reduces the power consumption of the sensor itself, reduces the storage of invalid and redundant data on the cloud platform, and alleviates the transmission pressure of the low-power wireless transmission module.
[0007] To achieve the above object, the technical solution adopted by the present invention is.
[0008] A method for depolarizing a demodulated signal comprises the following steps:
[0009] Step 1: Set the sampling rate of the ADC module, and set two buffers buffer1 and buffer2 in the SOC module. The buffer1 can store 2640 data points, and the buffer2 can store 2401 data points. The DCM module continuously caches 10 frames of data in the buffer1.
[0010] Step 2: Arrange the 10 frames of data in buffer 1 in step 2 in ascending order according to the numerical value of the data points and copy them to buffer 2. The data of the next frame entering is sequentially cached to the last frame of buffer 1.
[0011] Step 3: Set a time window containing 2401 data points in the buffer area buffer1, and move backward one data point at a time in chronological order to find the point in data points that enter the time window and the point out data points that are excluded during the movement. One frame of data at the back end of the buffer area buffer1 needs to be moved 240 times.
[0012] Step 4: Use the binary search method to quickly find the position corresponding to the point in data point and the point out data point in step 4 in the sorted buffer area buffer2, exclude the point out data point, and then move all the data between the corresponding positions of the point in data point and the point out data point by one data point position in the direction of the point out data point position, and then input the point in data point into the sequence of the buffer area buffer2, so as to ensure that the arrangement order of the values of all data points in the buffer area buffer2 is not disrupted; at this time, the point in the middle position of the sequence of the buffer area buffer2 is the median point;
[0013] Step 5: All data points in the sixth frame of data in the buffer area buffer1 are deducted from the median point in step 5, and the output is a frame of data, which is the debiased data;
[0014] Step 6: After the loop processing of one frame of data is completed, the first frame of data in the buffer 1 is deleted, and the next 10 frames of data are moved forward by one frame, leaving the position of the 11th frame of data empty to wait for the next frame of data to enter;
[0015] Step 7: Subtract the median point of the last data point in the previous frame from all data points in the two buffers buffer1 and buffer2, and save the value of the last data point of the 10th frame of data in the buffer1 as the integration starting point for the next frame of data. Repeat the above steps and output each frame of debiased data in real time.
[0016] Preferably, the sampling rate of the ADC module is set to 24Ksps, and 240 data points are collected as one frame.
[0017] A data preprocessing method comprises the following steps:
[0018] Step S1: From the raw data output by the low-pass filter, a 10-second period of raw data is intercepted as vibration data to be framed and processed, and saved in the Flash memory of the sensor;
[0019] Step S2: Divide the vibration data in step S1 into frames and calculate the number of frames of the vibration data:
[0020]
[0021] Among them, TotalLen represents the length of vibration data; FrameLen represents the frame length; FrameInc is the frame shift, which represents the moving distance between the previous frame data and the next frame data after framing processing;
[0022] Step S3: Calculate the root mean square of each frame of the vibration data A rms , and at the same time retrieve the root mean square of the background noise data stored in the Flash memory, and calculate the signal-to-noise ratio parameter of each frame of the vibration data:
[0023]
[0024] Among them, FrameRMS represents the root mean square of each frame of vibration data; BaseRMS represents the root mean square of the background noise data stored in the internal Flash memory of the sensor;
[0025] Step S4: performing rounding operation on the signal-to-noise ratio parameter of each frame of the vibration data, and selecting the signal-to-noise ratio parameter with the largest number of equal values after rounding to an integer, i.e., the mode, as the leakage parameter of the entire vibration data;
[0026] Step S5: Return to step S1, intercept and calculate the leakage parameters of the next 10S vibration data, until a mapping relationship database between each segment of vibration data in the entire original data and the corresponding leakage parameters is constructed, and at the same time, the leakage parameters corresponding to the entire original data are output to the communication module to complete the preprocessing of the entire original data.
[0027] Preferably, in step S2, the frame length FrameLen is 55 ms, and the frame shift FrameInc is 51 ms, that is, the overlapping area between the previous frame data and the next frame data is 4 ms.
[0028] Preferably, the RMS in step S3 A rms The calculation formula is as follows:
[0029]
[0030] in, Ai represents the value of a single data point, N Indicates the length of the vibration data.
[0031] A pipeline leakage monitoring method based on optical fiber sensing comprises the following steps:
[0032] Step 1: The interferometer outputs two optical signals with a fixed phase difference, which are converted into two electrical signals with a fixed phase difference by photodetector 1 and photodetector 2 respectively and then output to the ADC module. The two electrical signals can be expressed as:
[0033]
[0034] in a 1 and a 2 Indicates the DC bias of the two signals; b 1 and b 2 is the AC amplitude; θ s is the phase to be measured, indicating the phase change caused by external vibration; β is a fixed phase difference;
[0035] Step 2: The ADC module synchronously collects two electrical signals and outputs them to the ellipse fitting module, which then demodulates the ellipse fitting module and outputs the parameters in step 1. a 1 , a 2 , b 1 , b 2 , β , and construct a phase θ s The two orthogonal signals are output to the DCM module, and the two orthogonal signals can be expressed as:
[0036]
[0037] Step 3: The two orthogonal signals are demodulated by the DCM module and then output as the phase to be measured θ s To a depolarization module, the depolarization module outputs the depolarized data to a low-pass filter for real-time filtering calculation according to the depolarization method of the demodulated signal, and outputs the filtered raw data to a data preprocessing module;
[0038] Step 4: The data preprocessing module preprocesses the original data according to the data preprocessing method, outputs the leakage parameters corresponding to the original data to the communication module, and the communication module uploads the leakage parameters to the cloud platform;
[0039] Step 5: The cloud platform determines whether there is a pipeline leakage signal based on the leakage parameters in step 4. If the leakage parameters exceed the set threshold, it is determined that there is a pipeline leakage signal. At this time, the cloud platform collects the vibration data corresponding to the leakage parameters as the vibration spectrum output by the cloud platform, and outputs a pipeline leakage alarm signal.
[0040] A pipeline leakage monitoring system comprises a sensor, wherein the sensor comprises an interferometer, a SOC module and a communication module; the SOC module comprises an ADC module, an ellipse fitting module, a DCM module, a depolarization module, a low-pass filter and a data preprocessing module; when the interferometer picks up a pipeline leakage signal, two electrical signals with a fixed phase difference are output to the ADC module in the SOC module, the ADC module synchronously collects the two electrical signals and outputs them to the ellipse fitting module, and after demodulation by the ellipse fitting module, two orthogonal signals are output to the DCM module;
[0041] Preferably, the two orthogonal signals are demodulated by the DCM module and then output as the phase to be measured. θ s To the depolarization module, the depolarization module outputs the depolarized data to the low-pass filter for real-time filtering calculation according to the depolarization method of the demodulated signal, and outputs the filtered original data to the data preprocessing module; the data preprocessing module preprocesses the original data according to the data preprocessing method, and outputs the leakage parameters corresponding to the original data to the communication module, and the communication module uploads the leakage parameters to the cloud platform; the cloud platform then determines whether there is a pipeline leakage signal according to the leakage parameters, if the leakage parameters exceed the set threshold, it is determined that there is a pipeline leakage signal, at this time, the cloud platform collects the vibration data corresponding to the leakage parameter as the vibration spectrum output by the cloud platform, and outputs a pipeline leakage alarm signal.
[0042] Preferably, the interferometer includes a laser, an elastomer, a mass block, a reference optical fiber, a sensing optical fiber, a Faraday rotation mirror, a 3x3 coupler, a photodetector 1 and a photodetector 2.
[0043] Preferably, the third output end of the 3×3 coupler is subjected to a return cancellation process.
[0044] Preferably, the sensing optical fiber is evenly wound around the surface of the elastomer; and the reference optical fiber is evenly wound around the surface of the cylindrical groove of the mass block.
[0045] Preferably, the laser output by the laser is split into sensing light and reference light after passing through a 3x3 coupler; the tiny vibration signal generated when the pipeline leaks is transmitted to the elastic body at the lower end of the interferometer through the pipeline, and drives the elastic body to produce vibrations of the same frequency and amplitude, thereby causing the sensing optical fiber wound on the elastic body to produce stretching vibration changes, the mass block is insensitive to external vibration signals, and does not produce vibrations of the same frequency and amplitude, and the corresponding reference optical fiber wound on the surface of the cylindrical groove of the mass block does not produce stretching vibration changes; at this time, the optical path difference between the sensing optical fiber and the reference optical fiber will change; the sensing light and the reference light are reflected by the Faraday rotator mirror and return to the 3x3 coupler along the original path, and two optical signals with a fixed phase difference are generated in the 3x3 coupler, and the two optical signals are converted into two optical signals with a fixed phase difference by photodetector 1 and photodetector 2 respectively, and output to the ADC module in the SOC module.
[0046] Due to the adoption of the above technical solution, the present invention has the following beneficial effects.
[0047] (1) The present invention eliminates the jump and drift of the signal baseline by quickly finding the median through the binary search method, and corrects the position of the integration starting point in real time to eliminate the cumulative effect. At the same time, it well preserves the low-frequency response of the sensor, solves the jump and drift of the signal baseline, and avoids the risk of data overflow and demodulation failure.
[0048] (2) The present invention uses a binary search method to quickly find the specific positions of the entry and exclusion points of the moving window in the sorted window data, without the need for multiple sorting, which simplifies the program execution steps and greatly speeds up the execution of the method and the corresponding program. This reduces the amount of calculation in signal processing and improves the response speed of the system. Compared with the traditional high-order filter solution, it solves the problem that small-sized and low-power devices cannot support real-time big data calculations.
[0049] (3) The present invention constructs a mapping relationship between raw data and leakage parameters, simplifies a large amount of redundant and invalid raw data into leakage parameters equal to the number of segmented vibration data after preprocessing, that is, the entire uploaded raw data N is divided into N / 10 segments, and converted into N / 10 leakage data for uploading, which greatly reduces and reduces the transmission volume of raw data, reduces the power consumption of the sensor itself, reduces the storage of invalid and redundant data on the cloud platform, and alleviates the transmission pressure of the low-power wireless transmission module.
[0050] (4) The present invention uses a balanced Michelson interferometer composed of a 3×3 coupler and a Faraday rotator mirror, combined with an ellipse fitting algorithm and differential cross multiplication (DCM) to demodulate the micro-vibration signal generated by external pipeline leakage, thereby achieving the technical effect of accurately and stably measuring the vibration signal. Compared with the traditional optical fiber leak detection technology solution, it solves the data redundancy and power consumption problems caused by the lack of effective preprocessing of vibration data by the sensor.
[0051] (5) The present invention adopts a data preprocessing method to upload the processed leakage parameters to the cloud platform, thereby achieving the technical effects of reducing the transmission pressure of the wireless transmission module, reducing the battery power consumption of the device, and reducing the platform's storage of invalid data. Compared with the traditional solution of directly transmitting the original data, it solves the problem of low transmission and storage efficiency caused by the large amount of original data. BRIEF DESCRIPTION OF THE DRAWINGS
[0052] The following is a detailed discussion of the making and application of the preferred embodiments of the present invention. However, it should be understood that the present invention provides many applicable inventive concepts that can be embodied in various specific environments. The specific embodiments discussed are only for the purpose of illustrating the specific ways of making and using the present invention, and do not limit the scope of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative work.
[0053] Figure 1 Flow chart of the depolarization method of the present invention.
[0054] Figure 2 The figure is a flow chart of the data preprocessing method of the present invention.
[0055] Figure 3 It is a schematic diagram of the structure of the present invention.
[0056] Figure 4 This is a comparison chart of the data processing effects after adopting the monitoring method of the present invention.
[0057] Among them, 1-mass block; 2-elastic body; 3-Faraday rotation mirror; 4-reference optical fiber; 5-reflection elimination processing; 6-sensing optical fiber; 7-3x3 coupler; 8-photodetector 1; 9-laser; 10-photodetector 2; 11-SOC module; 12-ADC module; 13-ellipse fitting module; 14-DCM module; 15-depolarization module; 16-low-pass filter; 17-data preprocessing module; 18-communication module. DETAILED DESCRIPTION
[0058] The following is a detailed discussion of the making and application of the preferred embodiments of the present invention. However, it should be understood that the present invention provides many applicable inventive concepts that can be embodied in various specific environments. The specific embodiments discussed are only intended to illustrate specific ways to make and use the present invention and do not limit the scope of the present invention.
[0059] The present invention uses a binary search method to quickly find the median to eliminate the jump and drift of the signal baseline, and corrects the position of the integration starting point in real time to eliminate the cumulative effect. At the same time, it well retains the low-frequency response of the sensor, solves the jump and drift phenomenon of the signal baseline, and avoids the risk of data overflow and demodulation failure. In addition, the present invention uses a binary search method to quickly find the specific position of the entry point and exclusion point of the moving window in the sorted window data, without the need for multiple sorting, which simplifies the program execution steps and greatly speeds up the execution speed of the method and the corresponding program. This reduces the amount of calculation in signal processing, improves the response speed of the system, and solves the problem of being unable to support real-time big data calculations on miniaturized and low-power devices compared to traditional high-order filter solutions.
[0060] The following is combined with Figure 1-4 Further elaboration. Example 1
[0061] like Figure 1 The depolarization method of the demodulated signal comprises the following steps:
[0062] Step 1: Set the sampling rate of the ADC module 12 to 24Ksps, and collect 240 data points as one frame of data. The data collected by the ADC module 12 is demodulated by the ellipse fitting module 13 and the DCM module 14 in turn and then output to the depolarization module 15.
[0063] Step 2: Two buffer areas buffer1 and buffer2 are set in the SOC module 11. The buffer area buffer1 can store 2640 data points, and the buffer area buffer2 can store 2401 data points. The DCM module 14 continuously caches 10 frames of data in the buffer area buffer1.
[0064] Step 3: Arrange the 10 frames of data in buffer 1 in step 2 from small to large according to the numerical order of the data points and copy them to buffer 2. The data of the next frame entering is cached in sequence to the last frame of buffer 1, that is, buffer 1 has a total of 11 frames of data.
[0065] Step 4: Set a time window containing 2401 data points in the buffer area buffer1, and move backward one data point at a time in chronological order to find the point in data points that enter the time window and the point out data points that are excluded during the movement. One frame of data at the back end of the buffer area buffer1 needs to be moved 240 times.
[0066] Step 5: Use the binary search method to quickly find the positions corresponding to the point in data point and the point out data point in step 4 in the sorted buffer area buffer2, exclude the point out data point, and then move all the data between the corresponding positions of the point in data point and the point out data point by one data point position in the direction of the point out data point position, and then input the point in data point into the sequence of the buffer area buffer2 to ensure that the arrangement order of the values of all data points in the buffer area buffer2 is not disrupted; at this time, the point in the middle position of the sequence of the buffer area buffer2 is the median point.
[0067] The jump of the signal baseline is eliminated by quickly finding the median by the binary division method in step 5. After debiasing, the time domain waveform of the signal baseline jump is eliminated as follows: Figure 4 As shown, the low-frequency response of the sensor is well preserved, the jump phenomenon of the signal baseline is solved, and the risk of data overflow and demodulation failure is avoided. In addition, the entry point and exclusion point of the moving window are quickly found in the specific position of the sorted window data through the binary search method, without the need for multiple sorting, which simplifies the program execution steps and greatly speeds up the execution speed of this method and the corresponding program. This reduces the amount of calculation in signal processing and improves the response speed of the system. Compared with the traditional high-order filter solution, it solves the problem of being unable to support real-time big data calculations on miniaturized and low-power devices.
[0068] Step 6: A frame of data is outputted after all data points in the sixth frame of data in the buffer area buffer1 are successively subtracted from the median point in step 5, which is the debiased data.
[0069] The drift of the signal baseline is eliminated by quickly finding the median through the binary division method in step 5. After depolarization processing, the time domain waveform of the signal baseline drift elimination is as follows: Figure 4 As shown in the figure, the drift of the signal baseline is solved, and the risk of data overflow and demodulation failure is avoided.
[0070] Step 7: When the loop processing of one frame of data is completed, the first frame of data in the buffer area buffer1 is deleted, and the next 10 frames of data are moved forward by one frame, and the position of the 11th frame of data is vacated to wait for the entry of the next frame of data.
[0071] Step 8: Subtract the median point after processing the last data point of the previous frame of data from all data points in the two buffers buffer1 and buffer2, and save the value of the last data point of the 10th frame of data in the buffer1 as the integration starting point for the next frame of data.
[0072] In step 8, the value of the last data point of the 10th frame of data in buffer 1 is saved as the integration starting point of the next frame of data, and the position of the integration starting point is corrected in real time to eliminate the cumulative effect. After debiasing, the time domain waveform that eliminates the cumulative effect is as follows: Figure 4 shown.
[0073] Step 9: Execute the above steps in a loop, output each frame of de-biased data to the low-pass filter 16 in real time, and complete the de-biasing processing of the demodulated data of the DCM module 14. Example 2
[0074] like Figure 2 The data preprocessing method shown includes the following steps: Step S1: from the raw data output by the low-pass filter 16, a 10s-long raw data is intercepted as vibration data to be framed and processed, and saved in the Flash memory of the sensor.
[0075] Step S2: Divide the vibration data in step S1 into frames and calculate the number of frames of the vibration data:
[0076]
[0077] Wherein, TotalLen represents the length of vibration data; FrameLen represents the frame length; FrameInc represents the frame shift, which represents the moving distance between the previous frame data and the next frame data after the frame processing. The frame length FrameLen is 55ms, and the frame shift FrameInc is 51ms, that is, the overlapping area of the previous frame data and the next frame data is 4ms.
[0078] Step S3: Calculate the root mean square of each frame of the vibration data A rms , and at the same time retrieve the root mean square of the background noise data stored in the Flash memory, and calculate the signal-to-noise ratio parameter of each frame of the vibration data:
[0079]
[0080] Among them, FrameRMS represents the root mean square of each frame of vibration data; BaseRMS represents the root mean square of the background noise data stored in the internal Flash memory of the sensor.
[0081] The RMS A rms The calculation formula is as follows:
[0082]
[0083] in, A i represents the value of a single data point, N Indicates the length of the vibration data.
[0084] Step S4: rounding the signal-to-noise ratio parameter of each frame of the vibration data to an integer, and selecting the signal-to-noise ratio parameter with the most equal values after rounding to an integer, that is, the mode, as the leakage parameter of the entire vibration data.
[0085] Step S5: Return to step S1, intercept and calculate the leakage parameters of the next 10S vibration data, until a mapping relationship database between each segment of vibration data in the entire original data and the corresponding leakage parameters is constructed, and at the same time, the leakage parameters corresponding to the entire original data are output to the communication module 18 to complete the preprocessing of the entire original data.
[0086] The data preprocessing method in this embodiment constructs a mapping relationship between the original data and the leakage parameters, and simplifies a large amount of redundant and invalid original data into leakage parameters equal to the number of segmented vibration data after preprocessing. That is, the entire uploaded original data N is divided into N / 10 segments and converted into N / 10 leakage data for uploading, which greatly reduces and reduces the transmission volume of original data, reduces the power consumption of the sensor itself, reduces the storage of invalid and redundant data on the cloud platform, and alleviates the transmission pressure of the low-power wireless transmission module.
[0087] In addition, the data preprocessing method in this embodiment uploads the processed leakage parameters to the cloud platform by adopting the data preprocessing method, thereby achieving the technical effects of reducing the transmission pressure of the wireless transmission module, reducing the battery power consumption of the device, and reducing the platform's storage of invalid data. Compared with the traditional direct transmission of raw data, it solves the problem of low transmission and storage efficiency caused by the large amount of raw data. Example 3
[0088] A pipeline leakage monitoring method based on optical fiber sensing includes the following steps: Step 1: The interferometer outputs two optical signals with a fixed phase difference, which are converted into two electrical signals with a fixed phase difference by a photodetector 1 8 and a photodetector 2 10 respectively and then output to an ADC module 12. The two electrical signals can be expressed as:
[0089]
[0090] in a1 and a 2 Indicates the DC bias of the two signals; b 1 and b 2 is the AC amplitude; θ s is the phase to be measured, indicating the phase change caused by external vibration; β is a fixed phase difference.
[0091] Step 2: The ADC module 12 synchronously collects two electrical signals and outputs them to the ellipse fitting module 13. The ellipse fitting module 13 demodulates and outputs the parameters in step 1. a 1 , a 2 , b 1 , b 2 , β , and construct a phase θ s The two orthogonal signals are output to the DCM module 14, and the two orthogonal signals can be expressed as:
[0092] .
[0093] Step 3: The two orthogonal signals are demodulated by the DCM module 14 and then output as the phase to be measured θ s To the depolarization module 15, the depolarization module 15 outputs the depolarized data to the low-pass filter 16 for real-time filtering calculation according to the depolarization method of the demodulated signal, and outputs the filtered original data to the data preprocessing module 17.
[0094] Step 4: The data preprocessing module 17 preprocesses the original data according to the data preprocessing method, and outputs the leakage parameters corresponding to the original data to the communication module 18, and the communication module 18 uploads the leakage parameters to the cloud platform.
[0095] Step 5: The cloud platform determines whether there is a pipeline leakage signal based on the leakage parameters in step 4. If the leakage parameters exceed the set threshold, it is determined that there is a pipeline leakage signal. At this time, the cloud platform collects the vibration data corresponding to the leakage parameters as the vibration spectrum output by the cloud platform, and outputs a pipeline leakage alarm signal. Example 4
[0096] like Figure 3The pipeline leakage monitoring system includes a sensor, wherein the sensor includes an interferometer, a SOC module 11 and a communication module 18; the SOC module 11 includes an ADC module 12, an ellipse fitting module 13, a DCM module 14, a depolarization module 15, a low-pass filter 16 and a data preprocessing module 17; when the interferometer picks up a pipeline leakage signal, two electrical signals with a fixed phase difference are output to the ADC module 12 in the SOC module 11, and the ADC module 12 synchronously collects the two electrical signals and outputs them to the ellipse fitting module 13, and after demodulation by the ellipse fitting module 13, two orthogonal signals are output to the DCM module 14;
[0097] The two orthogonal signals are demodulated by the DCM module 14 and then output as the phase to be measured. θ s To the depolarization module 15, the depolarization module 15 outputs the depolarized data to the low-pass filter 16 for real-time filtering calculation according to the depolarization method of the demodulated signal, and outputs the filtered original data to the data preprocessing module 17; the data preprocessing module 17 preprocesses the original data according to the data preprocessing method, and outputs the leakage parameters corresponding to the original data to the communication module 18, and the communication module 18 uploads the leakage parameters to the cloud platform; the cloud platform then determines whether there is a pipeline leakage signal according to the leakage parameters. If the leakage parameters exceed the set threshold, it is determined that there is a pipeline leakage signal. At this time, the cloud platform collects the vibration data corresponding to the leakage parameters as the vibration spectrum output by the cloud platform, and outputs a pipeline leakage alarm signal.
[0098] The interferometer includes a laser 9, an elastic body 2, a mass block 1, a reference optical fiber 4, a sensing optical fiber 6, a Faraday rotation mirror 3, a 3x3 coupler 7, a photodetector 1 8 and a photodetector 2 10. The third output end of the 3×3 coupler is subjected to a return elimination process 4. The sensing optical fiber 6 is evenly wound on the surface of the elastic body 2; the reference optical fiber 4 is evenly wound on the surface of the cylindrical groove of the mass block 1.
[0099] The principle of the interferometer picking up the tiny vibration signal leaked from the external pipeline is as follows: the laser output by the laser 9 is split into sensing light and reference light after passing through the 3x3 coupler 7; the tiny vibration signal generated when the pipeline leaks is transmitted to the elastic body 2 at the lower end of the interferometer through the pipeline, and drives the elastic body 2 to produce vibrations of the same frequency and amplitude, thereby causing the sensing optical fiber 6 wound on the elastic body 2 to produce stretching vibration changes, the mass block 1 is insensitive to the external vibration signal, does not produce vibrations of the same frequency and amplitude, and the corresponding reference optical fiber 4 wound on the surface of the cylindrical groove of the mass block 1 does not produce stretching vibration changes; at this time, the optical path difference between the sensing optical fiber 6 and the reference optical fiber 4 will change; the sensing light and the reference light are reflected by the Faraday rotator 3 and return to the 3x3 coupler 7 along the original path, and two optical signals with a fixed phase difference are generated in the 3x3 coupler 7, and the two optical signals are converted into two optical signals with a fixed phase difference by the photodetector 1 8 and the photodetector 2 10 respectively, and output to the ADC module 12 in the SOC module 11. The two electrical signals can be expressed as:
[0100]
[0101] in a 1 and a 2 Indicates the DC bias of the two signals; b 1 and b 2 is the AC amplitude; θ s is the phase to be measured, indicating the phase change caused by external vibration; β is a fixed phase difference.
[0102] The ADC module 12 synchronously collects two electrical signals and outputs them to the ellipse fitting module 13. The ellipse fitting module 13 demodulates and outputs the parameters in step 1. a 1 , a 2 , b 1 , b 2 , β , and construct a phase θ s The two orthogonal signals are output to the DCM module 14, and the two orthogonal signals can be expressed as:
[0103]
[0104] The two orthogonal signals are demodulated by the DCM module 14 and then output as the phase to be measured. θs To the depolarization module 15, the depolarization module 15 outputs the depolarized data to the low-pass filter 16 for real-time filtering calculation according to the depolarization method of the demodulated signal, and outputs the filtered original data to the data preprocessing module 17; the data preprocessing module 17 preprocesses the original data according to the data preprocessing method, and outputs the leakage parameters corresponding to the original data to the communication module 18, and the communication module 18 uploads the leakage parameters to the cloud platform; the cloud platform determines whether there is a pipeline leakage signal according to the leakage parameters, if the leakage parameters exceed the set threshold, it is determined that there is a pipeline leakage signal, at this time, the cloud platform collects the vibration data corresponding to the leakage parameters as the vibration spectrum output by the cloud platform, and outputs a pipeline leakage alarm signal.
[0105] The present invention uses a balanced Michelson interferometer composed of a 3×3 coupler and a Faraday rotator mirror 3, combined with an ellipse fitting algorithm and differential cross multiplication (DCM) to demodulate the micro-vibration signal generated by external pipeline leakage, thereby achieving the technical effect of accurately and stably measuring the vibration signal. Compared with the traditional optical fiber leak detection technology solution, it solves the data redundancy and power consumption problems caused by the lack of effective preprocessing of vibration data by the sensor.
[0106] Although the specification has been described in detail, it should be understood that various changes, substitutions and modifications may be made without departing from the spirit and scope of the invention as defined by the appended claims. In addition, the specific embodiments described are not intended to limit the scope of the invention, and those of ordinary skill in the art can easily understand based on the present invention that currently existing or later to be developed processes, machines, manufactures, material compositions, means, methods, or steps can perform substantially the same functions as the embodiments of the present invention or obtain substantially the same results. Therefore, the appended claims are intended to include such processes, machines, manufactures, material compositions, means, methods or steps within their scope.
Claims
1. A method for depolarizing a demodulated signal, characterized in that: The following steps are involved: Step 1: Set the sampling rate of the ADC module, and set two buffers buffer1 and buffer2 in the SOC module. The buffer1 can store 2640 data points, and the buffer2 can store 2401 data points. The DCM module continuously caches 10 frames of data in the buffer1. Step 2: Arrange the 10 frames of data in buffer 1 in step 2 in ascending order according to the numerical value of the data points and copy them to buffer 2. The data of the next frame entering is sequentially cached to the last frame of buffer 1. Step 3: Set a time window containing 2401 data points in the buffer area buffer1, and move backward one data point at a time in chronological order to find the point in data points that enter the time window and the point out data points that are excluded during the movement. One frame of data at the back end of the buffer area buffer1 needs to be moved 240 times. Step 4: Use the binary search method to quickly find the position corresponding to the point in data point and the point out data point in step 4 in the sorted buffer area buffer2, exclude the point out data point, and then move all the data between the corresponding positions of the point in data point and the point out data point by one data point position in the direction of the point out data point position, and then input the point in data point into the sequence of the buffer area buffer2, so as to ensure that the arrangement order of the values of all data points in the buffer area buffer2 is not disrupted; at this time, the point in the middle position of the sequence of the buffer area buffer2 is the median point; Step 5: All data points in the sixth frame of data in the buffer area buffer1 are deducted from the median point in step 5, and the output is a frame of data, which is the debiased data; Step 6: After the loop processing of one frame of data is completed, the first frame of data in the buffer 1 is deleted, and the next 10 frames of data are moved forward by one frame, leaving the position of the 11th frame of data empty to wait for the next frame of data to enter; Step 7: Subtract the median point of the last data point in the previous frame from all data points in the two buffers buffer1 and buffer2, and save the value of the last data point of the 10th frame of data in the buffer1 as the integration starting point for the next frame of data. Repeat the above steps and output each frame of debiased data in real time.
2. A method for depolarizing a demodulated signal as claimed in claim 1, characterized in that: The sampling rate of the ADC module is set to 24Ksps, and 240 data points are collected as one frame.
3. A data preprocessing method, characterized in that: The following steps are involved: Step S1: From the raw data output by the low-pass filter, a 10-second period of raw data is intercepted as vibration data to be framed and processed, and saved in the Flash memory of the sensor; Step S2: Divide the vibration data in step S1 into frames and calculate the number of frames of the vibration data: FrameNum=(TotalLen-FrameLen+FrameInc) / FrameInc Among them, TotalLen represents the length of vibration data; FrameLen represents the frame length; FrameInc is the frame shift, which represents the moving distance between the previous frame data and the next frame data after framing processing; Step S3: Calculate the root mean square A of each frame of the vibration data rms , and at the same time retrieve the root mean square of the background noise data stored in the Flash memory, and calculate the signal-to-noise ratio parameter of each frame of the vibration data: SNR = 20 * ln (FrameRMS / BaseRMS) Among them, FrameRMS represents the root mean square of each frame of vibration data; BaseRMS represents the root mean square of the background noise data stored in the internal Flash memory of the sensor; Step S4: performing rounding operation on the signal-to-noise ratio parameter of each frame of the vibration data, and selecting the signal-to-noise ratio parameter with the largest number of equal values after rounding to an integer, i.e., the mode, as the leakage parameter of the entire vibration data; Step S5: Return to step S1, intercept and calculate the leakage parameters of the next 10S vibration data, until a mapping relationship database between each segment of vibration data in the entire original data and the corresponding leakage parameters is constructed, and at the same time, the leakage parameters corresponding to the entire original data are output to the communication module to complete the preprocessing of the entire original data.
4. A data preprocessing method as claimed in claim 3, characterized in that: In step S2, the frame length FrameLen is 55 ms, and the frame shift FrameInc is 51 ms, that is, the overlapping area between the previous frame data and the next frame data is 4 ms.
5. A data preprocessing method as claimed in claim 3, characterized in that: The root mean square A in step S3 rms The calculation formula is as follows: Among them, A i Represents the value of a single data point, and N is the length of the vibration data.
6. A pipeline leakage monitoring method based on optical fiber sensing, using the demodulated signal debiasing method as claimed in claim 1 and the data preprocessing method as claimed in claim 3, characterized in that: The following steps are involved: Step 1: The interferometer outputs two optical signals with a fixed phase difference, which are converted into two electrical signals with a fixed phase difference by photodetector 1 and photodetector 2 respectively and then output to the ADC module. The two electrical signals can be expressed as: V1=a1+b1cos(θ s ) V2=a2+b2cos(θ s -β) Where a1 and a2 represent the DC bias of the two signals; b1 and b2 are the AC amplitudes; θ s is the phase to be measured, indicating the phase change caused by external vibration; β is the fixed phase difference; Step 2: After synchronously collecting two electrical signals, the ADC module outputs the signals to the ellipse fitting module. After demodulation by the ellipse fitting module, the parameters a1, a2, b1, b2, β in step 1 are output, and a phase θ to be measured is constructed. s The two orthogonal signals are output to the DCM module, and the two orthogonal signals can be expressed as: V1′=cosθ s =(V1-a1) / b1 V2′=sinθ s =[V2-a2-b2cosθ s cosβ] / b2sinβ Step 3: The two orthogonal signals are demodulated by the DCM module and then the phase θ to be measured is output s To a depolarization module, the depolarization module outputs the depolarized data to a low-pass filter for real-time filtering calculation according to the depolarization method of the demodulated signal, and outputs the filtered raw data to a data preprocessing module; Step 4: The data preprocessing module preprocesses the original data according to the data preprocessing method, outputs the leakage parameters corresponding to the original data to the communication module, and the communication module uploads the leakage parameters to the cloud platform; Step 5: The cloud platform determines whether there is a pipeline leakage signal based on the leakage parameters in step 4. If the leakage parameters exceed the set threshold, it is determined that there is a pipeline leakage signal. At this time, the cloud platform collects the vibration data corresponding to the leakage parameters as the vibration spectrum output by the cloud platform, and outputs a pipeline leakage alarm signal.
7. A pipeline leakage monitoring system according to the pipeline leakage monitoring method of claim 6, characterized in that: The invention comprises a sensor, wherein the sensor comprises an interferometer, a SOC module and a communication module; the SOC module comprises an ADC module, an ellipse fitting module, a DCM module, a depolarization module, a low-pass filter and a data preprocessing module; when the interferometer picks up a pipeline leakage signal, two electrical signals with a fixed phase difference are output to the ADC module in the SOC module, and the ADC module synchronously collects the two electrical signals and outputs them to the ellipse fitting module, and after demodulation by the ellipse fitting module, two orthogonal signals are output to the DCM module; The two orthogonal signals are demodulated by the DCM module and then output as the phase θ to be measured. s To the depolarization module, the depolarization module outputs the depolarized data to the low-pass filter for real-time filtering calculation according to the depolarization method of the demodulated signal, and outputs the filtered original data to the data preprocessing module; the data preprocessing module preprocesses the original data according to the data preprocessing method, and outputs the leakage parameters corresponding to the original data to the communication module, and the communication module uploads the leakage parameters to the cloud platform; The cloud platform then determines whether there is a pipeline leakage signal based on the leakage parameters. If the leakage parameters exceed the set threshold, it is determined that there is a pipeline leakage signal. At this time, the cloud platform collects the vibration data corresponding to the leakage parameters as the vibration spectrum output by the cloud platform, and outputs a pipeline leakage alarm signal.
8. The pipeline leakage monitoring system based on optical fiber sensing according to claim 7, characterized in that: The interferometer comprises a laser, an elastomer, a mass block, a reference optical fiber, a sensing optical fiber, a Faraday rotation mirror, a 3x3 coupler, a photodetector 1 and a photodetector 2.
9. A pipeline leakage monitoring system as claimed in claim 8, characterized in that: The third output end of the 3×3 coupler is subjected to a return elimination process; the sensing optical fiber is evenly wound around the surface of the elastic body; and the reference optical fiber is evenly wound around the surface of the cylindrical groove of the mass block.
10. A pipeline leakage monitoring system according to claim 8, characterized in that: The laser output by the laser is split into sensing light and reference light after passing through a 3x3 coupler; the tiny vibration signal generated when the pipeline leaks is transmitted to the elastic body at the lower end of the interferometer through the pipeline, and drives the elastic body to generate vibrations of the same frequency and amplitude, thereby causing the sensing optical fiber wound on the elastic body to generate telescopic vibration changes, the mass block is insensitive to external vibration signals, and does not generate vibrations of the same frequency and amplitude, and the corresponding reference optical fiber wound on the surface of the cylindrical groove of the mass block does not generate telescopic vibration changes; at this time, the optical path difference between the sensing optical fiber and the reference optical fiber will change; the sensing light and the reference light are reflected by the Faraday rotator mirror and return to the 3x3 coupler along the original path, and generate two optical signals with a fixed phase difference in the 3x3 coupler, and the two optical signals are converted into two optical signals with a fixed phase difference by photodetector 1 and photodetector 2 respectively, and output to the ADC module in the SOC module.
