CN112069912B - Method for identifying construction threat event of optical cable channel based on phi-OTDR - Google Patents
Method for identifying construction threat event of optical cable channel based on phi-OTDR Download PDFInfo
- Publication number
- CN112069912B CN112069912B CN202010811484.3A CN202010811484A CN112069912B CN 112069912 B CN112069912 B CN 112069912B CN 202010811484 A CN202010811484 A CN 202010811484A CN 112069912 B CN112069912 B CN 112069912B
- Authority
- CN
- China
- Prior art keywords
- signal
- construction
- threat event
- optical cable
- otdr
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
- G06F2218/04—Denoising
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Signal Processing (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Geophysics And Detection Of Objects (AREA)
- Optical Communication System (AREA)
Abstract
The invention provides an optical cable channel construction threat event identification method based on phi-OTDR, which comprises the following steps: connecting the phi-OTDR with a core of redundant optical fiber of an optical cable to be monitored, and collecting optical fiber signals to obtain vibration signals; arranging the vibration signals into a two-dimensional array X according to time-space; carrying out average value processing on the two-dimensional array X to obtain a signal Y; denoising and normalizing the signal Y to obtain an initial signal G (i, j); extracting signal characteristics; and building a neural network training model, taking the signal characteristic Z (i, j) of the construction threat event as an input signal, and identifying the type of the construction threat event near the optical cable channel after passing through the neural network. The invention can analyze all vibration information on the optical cable in real time, can accurately analyze the mechanical construction type of each threat event, provides real-time early warning service for the safe operation of the optical cable, and improves the reliability and safety of the operation of the optical cable.
Description
Technical Field
The invention relates to the technical field of optical cable monitoring, in particular to an identification method of an optical cable channel construction threat event based on phi-OTDR.
Background
With the rapid development of modern cities at present, the coverage area of a communication network is continuously enlarged, and each metropolitan area forms a complicated underground optical cable distribution network. The wider the distribution of these underground fiber optic cable lines, the greater the line inspection pressure for the service personnel. On the other hand, municipal construction, road maintenance and the like lead to the increase of pavement construction sites, and accordingly a large number of mechanical operations such as an excavator, a crusher, a pneumatic pick and the like damage the ground, once an underground optical cable is damaged to cause fiber breakage, daily product supply such as power transmission and the like can be seriously influenced, and the safety of optical cable communication is jeopardized.
At present, line patrol personnel can be arranged for daily line patrol by referring to an optical cable map aiming at the situation by power supply companies in various places, so that line safety is ensured, hidden dangers along the optical cable are checked, but the manual monitoring has a plurality of limitations: the related affairs units often have no report in advance, are not reported in time after the related affairs units, and are hidden. The cable operation and maintenance unit lacks an effective technical monitoring means for preventing the power cable from being broken outwards, so that urban cable accidents frequently occur. The traditional monitoring means mainly depend on a video monitoring technology, can only monitor fixed points, and cannot monitor external broken events with strong randomness and high sporadic performance in real time.
At present, although some optical cable state monitoring systems exist for the problems, because of diversity of monitoring principles, construction signals and noise sources, a large number of false triggers and false alarms increase the workload of line inspection personnel. Considering that among various construction influences, the influence of mechanical construction operation on an underground electric power pipe gallery is most direct and serious, and once an optical cable is split, serious economic loss is caused by optical fiber breakage, so that a method capable of identifying a construction threat event in an optical cable channel is needed.
Disclosure of Invention
The invention provides an identification method of an optical cable channel construction threat event based on phi-OTDR, which adopts the phi-OTDR (phase sensitive optical time domain reflectometer) to be connected with an optical cable to be tested to acquire vibration data distribution along the optical cable, further processes data characteristics through a scheme of signal processing and mode identification, and further identifies various mechanical operation signals.
The technical scheme adopted by the invention is that the method for identifying the construction threat event of the optical cable channel based on the phi-OTDR comprises the following steps:
step 4, denoising and normalizing the signal Y to obtain an initial signal G (i, j);
step 5, extracting signal characteristics: obtaining a typical mechanical construction signal through experimental simulation and field monitoring, marking the signal as a sample library initial signal G (i, j), dividing the sample library initial signal G (i, j) in space according to an image segmentation process based on the projection of the signal on a space domain, and cutting out signal characteristics Z (i, j) of each possible construction threat event under the same frequency domain-space domain scale, wherein the three types of signal characteristics are marked as initial signal characteristics of the mechanical construction signal;
step 7, building a neural network training model, respectively simulating the mechanical construction signals, taking the vertical distance from mechanical vibration to an optical cable as a variable, continuously changing the distance, sequentially adopting different signal characteristics of each mechanical construction signal according to the steps, taking the signal characteristics of the mechanical construction signals as a characteristic training set of the neural network, taking the signal characteristics Z (i, j) of a construction threat event as an input signal, outputting which mechanical type the signal belongs to after passing through the neural network, and taking the characteristic type as a temporary output result;
and 8, calculating the matching confidence coefficient R of the signal characteristic Z (i, j) and the initial signal characteristic obtained in the step 5 after the neural network is judged, and determining that the output result of the step 7 is the finally recognized threat event mechanical type when the matching confidence coefficient R is greater than a threshold value T.
Further, when the matching confidence R in step 8 is determined not to exceed the threshold T, the output result obtained in step 7 is determined to be invalid, and then step 6 is executed:
step 6, expanding the initial signal characteristic type based on experiments: and (3) manually determining the new threat event mechanical type again, adding the initial signal of the sample library, extracting the signal characteristics, judging again, and continuously expanding the mechanical type sample library by analogy to obtain more initial signals of the sample library and corresponding initial signal characteristics, and then executing the step (7).
Further, typical machine construction signals include at least three types: g1 (i, j), G2 (i, j), G3 (i, j) are optical fiber vibration signals caused by a crusher, an excavator and a small air pick during line side operation.
Further, in step 3, the two-dimensional array X is subjected to mean processing according to Y (i, j) = (X (i, j) -X (i, j-1))/Xmean (i) to obtain a signal Y, where |xmean (i) = (|x (i, j) |/(i X j)) represents the result of averaging all elements of the i-th row in the X array, the L1 norm of the matrix is found.
Further, in step 4, the signal Y is denoised and normalized to obtain an initial signal G (i, j), G (i, j) =nor (Y (i, j) ×gauss (i, j)), where NOR () represents the normalization process, gauss (i, j) is a two-dimensional gaussian filter function, and x represents the convolution operation.
Further, in step 8, the threshold T is set to 0.8.
The invention can analyze all vibration information on the optical cable in real time, can accurately analyze the mechanical construction type of each threat event, provides real-time early warning service for the safe operation of the optical cable, and improves the reliability and safety of the operation of the optical cable.
Drawings
FIG. 1 is a flow chart of one embodiment of a method for identifying a threat event for cable channel construction based on phi-OTDR in accordance with the present invention;
FIG. 2 is a signal characteristic template of a crusher, an excavator, a small air pick in an embodiment of the present invention;
fig. 3 is a schematic diagram of a neural network identification process according to the present invention.
Fig. 4 is a schematic diagram showing the effect of the verification and identification method according to the embodiment of the invention.
Detailed Description
The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings.
Referring to fig. 1 to 3, the method for identifying the construction threat event of the optical cable channel based on the phi-OTDR provided by the invention can accurately analyze the construction signal type of each optical cable channel threat event based on the vibration signal along the optical cable monitored by the phi-OTDR. The identification method specifically comprises the following steps:
step 1: connecting the phi-OTDR with a core of redundant optical fiber of the optical cable to be monitored, and collecting optical fiber signals in the redundant optical fiber to obtain vibration signals;
step 2: the obtained vibration signals are arranged into a two-dimensional array X according to time-space, and the element X (i, j) refers to the vibration signal intensity of the space point i at the moment j;
step 3: the two-dimensional array X is subjected to mean processing according to Y (i, j) = (X (i, j) -X (i, j-1))/Xmean (i), to obtain a signal Y, wherein |xmean (i) = (|x (i, j) |/(i×j)) represents the result of averaging all elements of the ith row in the X array, and is the L1 norm of the matrix;
step 4: the signal Y is subjected to denoising, normalization and the like to obtain an initial signal G (i, j), G (i, j) =nor (Y (i, j) ×gauss (i, j)), wherein NOR () represents the normalization process, gauss (i, j) is a two-dimensional gaussian filter function, and is a convolution operation, and the step can denoise and trending the data. The invention uses the two-dimensional Gaussian convolution template to denoise the signal, normalizes the signal and the like, and can facilitate the extraction of more prominent signal characteristics in the subsequent steps.
Step 5: extracting signal characteristics: through experimental simulation and on-site monitoring, typical mechanical construction signals are obtained according to the processing steps of the step 3 and the step 4, and the typical mechanical construction signals in the embodiment are divided into three types: g1 (i, j), G2 (i, j), G3 (i, j) are optical fiber vibration signals caused by a crusher, an excavator and a small air pick during line side operation, and the three optical fiber vibration signals are marked as sample library initial signals; the sample library initial signal G (i, j) is spatially segmented according to the image segmentation flow based on the projection of the signal on the spatial domain, and the signal characteristic Z (i, j) of each possible construction threat event is segmented under the same frequency domain-spatial domain scale. In this embodiment, signal features Z1 (i, j), Z2 (i, j), and Z3 (i, j) of the crusher, the excavator, and the small air pick are respectively shown in fig. 2, and these three types of signal features are marked as initial signal features of the mechanical construction signal.
Step 7: building a neural network training model, respectively simulating the three mechanical construction signals, continuously performing experiments, taking the vertical distance from mechanical vibration to an optical cable as a variable, continuously changing the distance, and sequentially collecting different signal characteristics of a crusher, an excavator and a small air pick according to the steps, wherein the different signal characteristics are respectively marked as Z11, Z12. Z21, Z22..z 2n; z31, Z32. The signal characteristics of these mechanical construction signals are used as a characteristic training set of the neural network. When a vibration event acts on the optical fiber, the respective characteristics are reflected in the backward Rayleigh scattering signal, the signal characteristic Z (i, j) of the construction threat event is obtained after the processing of the steps 2-6 and is used as an input signal, the mechanical type (shown in fig. 3) of the signal is output after the signal passes through a neural network, and the characteristic type is used as a temporary output result A. The neural network judgment process is shown in fig. 1.
Step 8: and (3) after the neural network is judged, calculating the confidence coefficient R1, R2 and R3 of the matching between the signal characteristic Z (i, j) and the initial signal characteristic Z1 (i, j) obtained in the step (5), namely the crusher, Z2 (i, j), namely the excavator and Z3 (i, j), and setting a mechanical construction threshold T to be 0.8 according to field experiments and experience, and carrying out the step (9) when the R1, R2 and R3 are smaller than the threshold T, and carrying out the step (10) when a certain matching confidence coefficient R is larger than the threshold T. According to the invention, after the mechanical type of the input signal is obtained through the neural network model, the matching degree between the input signal and various mechanical templates is judged at the same time, and if the matching degree meets the threshold requirement, the mechanical type is finally determined, so that the recognition accuracy can be improved.
Step 9: if R1, R2 and R3 are all smaller than T, the output result obtained in the step 7 is invalid, the signal is judged to be not of the three mechanical types, the signal is marked as other construction signal characteristics Z4 (i, j), and the step 6 is executed.
Step 6: based on the experiment, the initial signal feature type is extended. And (3) manually determining new threat event mechanical types again, adding initial signals G4 (i, j) of the sample library, extracting signal characteristics Z4 (i, j), judging again, and continuously expanding the mechanical type sample library to obtain more initial signals G4 (i, j), G5 (i, j) of the sample library, gn (i, j) and corresponding initial signal characteristics Z4 (i, j), Z5 (i, j) of Zn (i, j), and then executing step 7.
Step 10: if a certain matching confidence level R is greater than T, the output result of the step 7 is the finally identified threat event mechanical category.
As fig. 4 shows two examples of the identification breaker and the small air pick of the invention, the identification results are consistent with the experimental machinery, and it can be seen that the identification method can accurately identify the specific mechanical operation type of the threat event near the optical cable.
The foregoing is merely illustrative embodiments of the present invention, and the present invention is not limited thereto, and any changes or substitutions that may be easily contemplated by those skilled in the art within the scope of the present invention should be included in the scope of the present invention.
Claims (6)
1. A method for identifying construction threat event of optical cable channel based on phi-OTDR is characterized by comprising the following steps: comprises the following steps of;
step 1, connecting phi-OTDR with a core of redundant optical fiber of an optical cable to be monitored, and collecting optical fiber signals in the redundant optical fiber to obtain vibration signals;
step 2, arranging the obtained vibration signals into a two-dimensional array X according to time-space, wherein an element X (i, j) refers to the vibration signal intensity of a space point i at a moment j;
step 3, carrying out average value processing on the two-dimensional array X to obtain a signal Y;
step 4, denoising and normalizing the signal Y to obtain an initial signal G (i, j);
step 5, extracting signal characteristics: obtaining a typical mechanical construction signal through experimental simulation and field monitoring, marking the signal as a sample library initial signal G (i, j), dividing the sample library initial signal G (i, j) in space according to an image segmentation process based on the projection of the signal on a space domain, and cutting out signal characteristics Z (i, j) of each possible construction threat event under the same frequency domain-space domain scale, wherein the three types of signal characteristics are marked as initial signal characteristics of the mechanical construction signal;
step 7, building a neural network training model, respectively simulating the mechanical construction signals, taking the vertical distance from mechanical vibration to an optical cable as a variable, continuously changing the distance, sequentially adopting different signal characteristics of each mechanical construction signal according to the steps, taking the signal characteristics of the mechanical construction signals as a characteristic training set of the neural network, taking the signal characteristics Z (i, j) of a construction threat event as an input signal, outputting which mechanical type the signal belongs to after passing through the neural network, and taking the characteristic type as a temporary output result;
and 8, calculating the matching confidence coefficient R of the signal characteristic Z (i, j) and the initial signal characteristic obtained in the step 5 after the neural network is judged, and determining that the output result of the step 7 is the finally recognized threat event mechanical type when the matching confidence coefficient R is greater than a threshold value T.
2. The method for identifying a cable channel construction threat event based on Φ -OTDR according to claim 1, wherein: and when the matching confidence coefficient R is judged to not exceed the threshold value T in the step 8, judging that the output result obtained in the step 7 is invalid, and then executing the step 6:
step 6, expanding the initial signal characteristic type based on experiments: and (3) manually determining the new threat event mechanical type again, adding the initial signal of the sample library, extracting the signal characteristics, judging again, and continuously expanding the mechanical type sample library by analogy to obtain more initial signals of the sample library and corresponding initial signal characteristics, and then executing the step (7).
3. The method for identifying a cable channel construction threat event based on Φ -OTDR according to claim 1, wherein: typical machine construction signals include at least three types: g1 (i, j), G2 (i, j), G3 (i, j) are optical fiber vibration signals caused by a crusher, an excavator and a small air pick during line side operation.
4. The method for identifying a cable channel construction threat event based on Φ -OTDR according to claim 1, wherein: in step 3, the two-dimensional array X is subjected to mean processing according to Y (i, j) = (X (i, j) -X (i, j-1))/Xmean (i) to obtain a signal Y, where |xmean (i) = (|x (i, j) |/(i X j)) represents the result of averaging all elements of the i-th row in the X array, the L1 norm of the matrix is found.
5. The method for identifying a cable channel construction threat event based on Φ -OTDR according to claim 1, wherein: in step 4, denoising and normalizing the signal Y to obtain an initial signal G (i, j), G (i, j) =nor (Y (i, j) ×gauss (i, j)), where NOR () represents normalization, gauss (i, j) is a two-dimensional gaussian filter function, and x represents convolution operation.
6. The method for identifying a cable channel construction threat event based on Φ -OTDR according to claim 1, wherein: in step 8, the threshold T is set to 0.8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010811484.3A CN112069912B (en) | 2020-08-13 | 2020-08-13 | Method for identifying construction threat event of optical cable channel based on phi-OTDR |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010811484.3A CN112069912B (en) | 2020-08-13 | 2020-08-13 | Method for identifying construction threat event of optical cable channel based on phi-OTDR |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112069912A CN112069912A (en) | 2020-12-11 |
| CN112069912B true CN112069912B (en) | 2023-05-12 |
Family
ID=73661587
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010811484.3A Active CN112069912B (en) | 2020-08-13 | 2020-08-13 | Method for identifying construction threat event of optical cable channel based on phi-OTDR |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112069912B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113188461B (en) * | 2021-05-06 | 2022-05-17 | 山东大学 | OFDR large strain measurement method under high spatial resolution |
| CN113258996B (en) * | 2021-07-05 | 2021-09-17 | 南京华脉科技股份有限公司 | Optical cable monitoring method in submarine cable production and laying process based on artificial intelligence |
| CN114323244A (en) * | 2021-11-30 | 2022-04-12 | 贵州电网有限责任公司 | Cable pipeline collapse signal monitoring method based on phi-OTDR |
| CN114414029A (en) * | 2021-11-30 | 2022-04-29 | 贵州电网有限责任公司 | Pier collision online monitoring method of communication optical cable based on data algorithm |
| CN114166331A (en) * | 2021-12-03 | 2022-03-11 | 武汉理工光科股份有限公司 | Non-threat signal identification method and device, electronic equipment and storage medium |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101183899B (en) * | 2007-12-19 | 2010-09-01 | 天津大学 | BP network based pipeline security identifying method for optical fiber pipeline leakage monitoring device |
| CN111222743B (en) * | 2019-11-15 | 2022-07-29 | 电子科技大学 | Method for judging vertical offset distance and threat level of optical fiber sensing event |
| CN111507310B (en) * | 2020-05-21 | 2023-05-23 | 国网湖北省电力有限公司武汉供电公司 | Method for identifying artificial cable touching operation signals in optical cable channel based on phi-OTDR |
-
2020
- 2020-08-13 CN CN202010811484.3A patent/CN112069912B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112069912A (en) | 2020-12-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112069912B (en) | Method for identifying construction threat event of optical cable channel based on phi-OTDR | |
| US11562224B2 (en) | 1D-CNN-based distributed optical fiber sensing signal feature learning and classification method | |
| CN110647539B (en) | Prediction method and system for vehicle faults | |
| CN108509850B (en) | Intrusion signal identification method based on distributed optical fiber system | |
| CN107256635B (en) | Vehicle identification method based on distributed optical fiber sensing in intelligent traffic | |
| CN111507310B (en) | Method for identifying artificial cable touching operation signals in optical cable channel based on phi-OTDR | |
| CN112149522A (en) | Intelligent visual external-damage-prevention monitoring system and method for cable channel | |
| CN115643159B (en) | Equipment abnormity early warning method and system based on edge calculation | |
| CN112362304B (en) | Method for identifying target optical cable in multiple optical cables and corresponding system | |
| CN111912513B (en) | Identification method of excavator construction event along optical cable based on phi-OTDR | |
| CN112883802A (en) | Method for identifying destructive event of pipeline optical fiber vibration safety early warning system | |
| CN115481940A (en) | Oil and gas pipeline area risk monitoring system based on big data | |
| CN114741255A (en) | Fault self-healing technology based on automatic execution of service scene | |
| CN113780237A (en) | External damage prevention early warning method, device and system for underground pipeline | |
| CN108765814A (en) | Boundary line periphery intrusion preventing alarm system based on Fibre Optical Sensor and method | |
| CN115690659A (en) | Transformer substation safety detection method and device under shielding condition | |
| CN111951505B (en) | Fence vibration intrusion positioning and mode identification method based on distributed optical fiber system | |
| CN109682457A (en) | Incident optical fiber recognition methods based on angle detection | |
| CN208834447U (en) | Boundary line periphery intrusion preventing alarm system based on Fibre Optical Sensor | |
| CN114139583A (en) | Method and system for detecting abnormal events of expressway | |
| CN112071009B (en) | Optical fiber pipeline early warning system and method thereof | |
| CN117668669B (en) | Pipeline safety monitoring method and system based on improvement YOLOv (YOLOv) | |
| CN117034093B (en) | Intrusion signal identification method based on optical fiber system | |
| CN115082865B (en) | Bridge crane intrusion dangerous behavior early warning method and system based on visual image recognition | |
| CN111639711B (en) | Oil pipeline leakage monitoring method based on pressure monitoring time sequence data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |