US 12,748,011 B2Granted

Pipeline leakage monitoring method and system based on optical fiber sensing

AssigneeAnhui Zhibo Photoelectric Technology Co., Ltd.

InventorsPengfei Li, Heng Luo, Jiatong Luo, Rui Wang, Yusheng Wang

Life of this patent

Estimating the term…

Loading patent…

Abstract

A pipeline leakage monitoring method and system based on optical fiber sensing are provided, and belongs to the technical field of pipeline leakage monitoring. The disclosure eliminates the jump and drift of the signal baseline by quickly finding the median by dichotomy, and corrects the position of the integration starting point in real time to eliminate the cumulative effect, while retaining the low-frequency response of the sensor. The specific positions of the entry point and the exclusion point of the moving window in the sorted window data are quickly found by dichotomy, the execution speed of the method is greatly accelerated. By constructing the mapping relationship between original data and leakage parameter, a large number of redundant and invalid original data are simplified into leakage parameters equal to the number of segmented vibration data after preprocessing.

Claims7

Dependent claims

A de-biasing method for a demodulated signal, comprising:

S1:
setting a sampling rate of ADC module, and setting two buffer areas, buffer1 and buffer2, in SOC module;
wherein buffer area buffer1 is capable of storing 2640 data points, buffer area buffer2 is capable of storing 2401 data points, and DCM module continuously caches 10 frames data in buffer area buffer1;
S2:
arranging the 10 frames data in buffer area buffer1 in S2 according to numerical order of data points from small to large, and then copying to buffer area buffer2, and sequentially caching next frame entering data to last frame of buffer area buffer1 in order;
S3:
setting a time window with 2401 data points in buffer area buffer1, and moving backward data points one by one according to time sequence, so as to find out point in data points entering time window and point out data points being excluded during moving process, wherein a frame data at a back end of buffer area buffer1 needs to be moved 240 times;
S4:
quickly finding positions corresponding to point in data points and point out data points in S4 in sorted buffer area buffer2 by using dichotomy, and after excluding the point out data points, moving all data in a middle of corresponding positions of the point in data points and the point out data points to a point out data point position direction by a data point position;
then inputting point in data points into a sequence of buffer area buffer2, so as to ensure arrangement order of all data point values in buffer area buffer2 being not disturbed;
now, a point in a middle position of a sequence of buffer area buffer2 is a median point;
S5:
sequentially subtracting all data points in a 6th frame data in a middle of buffer area buffer1 from the median point in S5, wherein an output frame data is de-biased data;
S6:
after a frame data cycle processing is completed, deleting a first frame data in buffer area buffer1, and simultaneously moving last 10 frames data forward by one frame position, and vacanting a position of 11th frame data to wait for a next frame data to enter;
S7:
subtracting a median point after a last data point of previous frame data being processed from all data points in the two buffer areas buffer1 and buffer2, and saving a value of a last data point of 10th frame data in buffer area buffer1 as an integration starting point of a next frame data, and executing circularly according to above steps to output de-biased data of each frame in real time.

14 more elements

A pipeline leakage monitoring method based on optical fiber sensing, wherein the pipeline leakage monitoring method applies a de-biasing method for a demodulated signal and a data preprocessing method,

wherein the de-biasing method for a demodulated signal comprises:
S1:
setting a sampling rate of ADC module, and setting two buffer areas, buffer1 and buffer2, in SOC module;
wherein buffer area buffer1 is capable of storing 2640 data points, buffer area buffer2 is capable of storing 2401 data points, and DCM module continuously caches 10 frames data in buffer area buffer1;
S2:
arranging the 10 frames data in buffer area buffer1 in S2 according to numerical order of data points from small to large, and then copying to buffer area buffer2, and sequentially caching next frame entering data to last frame of buffer area buffer1 in order;
S3:
setting a time window with 2401 data points in buffer area buffer1, and moving backward data points one by one according to time sequence, so as to find out point in data points entering time window and point out data points being excluded during moving process, wherein a frame data at a back end of buffer area buffer1 needs to be moved 240 times;
S4:
quickly finding positions corresponding to point in data points and point out data points in S4 in sorted buffer area buffer2 by using dichotomy, and after excluding the point out data points, moving all data in a middle of corresponding positions of the point in data points and the point out data points to a point out data point position direction by a data point position;
then inputting point in data points into a sequence of buffer area buffer2, so as to ensure arrangement order of all data point values in buffer area buffer2 being not disturbed;
now, a point in a middle position of a sequence of buffer area buffer2 is a median point;
S5:
sequentially subtracting all data points in a 6th frame data in a middle of buffer area buffer1 from the median point in S5, wherein an output frame data is de-biased data;
S6:
after a frame data cycle processing is completed, deleting a first frame data in buffer area buffer1, and simultaneously moving last 10 frames data forward by one frame position, and vacanting a position of 11th frame data to wait for a next frame data to enter;
S7:
subtracting a median point after a last data point of previous frame data being processed from all data points in the two buffer areas buffer1 and buffer2, and saving a value of a last data point of 10th frame data in buffer area buffer1 as an integration starting point of a next frame data, and executing circularly according to above steps to output de-biased data of each frame in real time;
wherein the data preprocessing method comprises:
S1:
intercepting a segment original data with a 10 s duration from original data output by a low-pass filter as vibration data to be frame processed, and storing in Flash memory of a sensor;
S2:
performing frame processing on the vibration data in S1 to calculate frame number of the vibration data;
FrameNum
=
(
TotalLen
-
FrameLen
+
FrameInc
)
/
FrameInc
;
wherein, TotalLen represents a length of the vibration data;
FrameLen represents a frame length;
FrameInc is frame shift, indicating a moving distance between a previous frame data and a next frame data after framing processing;
S3:
calculating root mean square Arms of each frame data in the vibration data, simultaneously retrieving root mean square of bottom noise data stored in the Flash memory, and calculating a signal-to-noise ratio parameter of each frame data in the vibration data;
SNR
=
20
*
ln
⁡
(
FrameRMS
/
BaseRMS
)
;
wherein, FrameRMS represents the root mean square of each frame data in the vibration data;
BaseRMS represents the root mean square of the bottom noise data stored in the Flash memory inside the sensor;
S4:
performing rounding operation of selecting integer on the signal-to-noise ratio parameter of each frame data in the vibration data, and selecting a signal-to-noise ratio parameter with a largest number of equal values after selecting integer, that is, mode is used as a leakage parameter of whole vibration data;
S5:
returning to S1, intercepting and calculating a leakage parameter of a next segment 10 s duration vibration data until a mapping relationship database between each segment vibration data and corresponding a leakage parameter in whole original data is constructed, and simultaneously outputting leakage parameters corresponding to the whole original data to a communication module, thus completing preprocessing of the whole original data;
wherein the pipeline leakage monitoring method comprises:
S1:
outputting two optical signals with fixed phase difference by an interferometer, converting into two electrical signals with fixed phase difference by photodetector I and photodetector II respectively, and then outputting to ADC module;
wherein the two electrical signals are expressed as:
V
1
=
a
1
+
b
1
⁢
cos
⁡
(
θ
s
)
;
V
2
=
a
2
+
b
2
⁢
cos
⁡
(
θ
s
-
β
)
;
wherein a1 and a2 represent direct current bias of two signals;
b1 and b2 are alternating current amplitudes;
θs is phase to be measured, indicating phase change caused by external vibration;
β is a fixed phase difference;
S2:
synchronously collecting two electrical signals by the ADC module, then outputting to an ellipse fitting module, and outputting parameters a1, a2, b1, b2 and β in S1 after demodulation by the ellipse fitting module, and constructing two orthogonal signals containing phase θs to be measured and outputting to a DCM module, wherein the two orthogonal signals are expressed as:
V
1
′
=
cos
⁢
θ
s
=
(
V
1
-
a
1
)
/
b
1
;
V
2
′
=
sin
⁢
θ
s
=
[
V
2
-
a
2
-
b
2
⁢
cos
⁢
θ
s
⁢
cos
⁢
β
]
/
b
2
⁢
sin
⁢
β
;
S3:
after the two orthogonal signals are demodulated by the DCM module, outputting phase θs to be measured to a de-biasing module, and outputting de-biased data to a low-pass filter for real-time filtering calculation by the de-biasing module according to the de-biasing method for a demodulated signal, and outputting filtered original data to a data preprocessing module;
S4:
performing preprocessing on original data by the data preprocessing module according to the data preprocessing method, outputting a leakage parameter corresponding to the original data to a communication module, and uploading the leakage parameter to a cloud platform by the communication module;
S5:
judging whether there is a pipeline leakage signal according to the leakage parameter in S4 by the cloud platform, wherein if the leakage parameter exceeds set thresholds, there is the pipeline leakage signal is judged, and now, vibration data corresponding to leakage parameter are copied by the cloud platform is as a vibration spectrum output by the cloud platform, and a pipeline leakage alarm signal is output.

161 more elements

Dependencies are inferred from the claim text.