CN112883802A - Method for identifying destructive event of pipeline optical fiber vibration safety early warning system - Google Patents
Method for identifying destructive event of pipeline optical fiber vibration safety early warning system Download PDFInfo
- Publication number
- CN112883802A CN112883802A CN202110075378.8A CN202110075378A CN112883802A CN 112883802 A CN112883802 A CN 112883802A CN 202110075378 A CN202110075378 A CN 202110075378A CN 112883802 A CN112883802 A CN 112883802A
- Authority
- CN
- China
- Prior art keywords
- pipeline
- identifying
- optical fiber
- image
- early warning
- 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.)
- Pending
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
- G06F2218/12—Classification; Matching
-
- 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
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/02—Mechanical actuation
- G08B13/12—Mechanical actuation by the breaking or disturbance of stretched cords or wires
- G08B13/122—Mechanical actuation by the breaking or disturbance of stretched cords or wires for a perimeter fence
- G08B13/124—Mechanical actuation by the breaking or disturbance of stretched cords or wires for a perimeter fence with the breaking or disturbance being optically detected, e.g. optical fibers in the perimeter fence
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Signal Processing (AREA)
- Probability & Statistics with Applications (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a method for identifying specific event monitoring, in particular to a method for identifying vibration signals of a buried pipeline vibration safety early warning system. The invention discloses a method for identifying destructive events of a pipeline optical fiber vibration safety early warning system, which comprises the steps of utilizing an optical fiber to obtain real-time signals of soil vibration around a pipeline, analyzing and processing the signals, training by using a classification algorithm according to the characteristics and corresponding events to establish a model for identifying the pipeline destructive events, and carrying out classification identification on the actually measured signals on the basis. The invention has the advantages of high identification accuracy, capability of overcoming distortion and high identification speed.
Description
Technical Field
The invention relates to a method for identifying specific event monitoring, in particular to a method for identifying vibration signals of a buried pipeline vibration safety early warning system.
Background
The long-distance oil pipeline is a large-scale pipeline system with a large caliber and a long distance, is a key project of national energy aorta and national economy, bears the important responsibility of providing energy guarantee for the rapid development of the national economy health, and has important social and economic significance in the safe operation. Traditional pipeline safety protection is watched through patrolling line workman's lasting pipeline, and this kind of mode not only takes the people hard, and when the action that takes place to destroy the pipeline takes place, patrolling line workman probably can not in time discover moreover, probably causes the serious destruction of pipeline, so said pipeline safety early warning system and have very important effect to the transportation energy of pipeline.
Pipeline optic fibre vibration safety early warning system utilizes pipe laying optic fibre as the sensor, and the soil vibration signal along the real-time induction pipeline is along the line, through intelligent identification analysis, carries out early warning and location to destructive events such as mechanical construction, artifical excavation that threaten pipeline safety, notifies the patrolling line personnel to rush to the scene and look over, prevents that the destruction event from further worsening.
Generally, an interference area caused by factors such as an image removal environment is realized through filtering, irrelevant areas can be removed to a certain extent through filtering, but normal and effective areas can be weakened at the same time, and the image is easy to generate a distortion phenomenon. According to the method, the interference image area in the signal waterfall graph is removed by using a clustering abnormal value detection method, meanwhile, the image area information caused by real damage is not lost, the difficulty of the classifier in identifying the image information is reduced, and the classification accuracy is improved.
Chinese patent 2013106132601 discloses a device and method for expanding the monitoring area of an optical cable vibration intrusion monitoring system. The device comprises a passive infrared detector, a vibration signal transmission medium and a vibration signal decoder, wherein the passive infrared detector and the vibration signal decoder are connected through the vibration signal transmission medium. And (4) utilizing a passive infrared detector to sense intrusion behavior. This patent is assisted optic fibre vibration intrusion monitoring system and is carried out the invasion and take precautions against, has expanded optical cable vibration intrusion monitoring system's monitoring area, reduces the invasion monitoring blind spot, reduces the requirement of sensing optical cable to the rail, but can't distinguish and cause the optical cable vibration classification, consequently its mistake reports to the policeThe rate is large. Similar patents such as ZL 2016486278, ZL2016104767007, etc. suffer from the same disadvantages. To solve the defects of the prior art, the chinese invention patent 2010105235522 discloses a vibration signal identification method of an optical fiber perimeter system, which comprises the following steps: signal acquisition, windowing, band-pass filtering, wavelet denoising, vibration event detection, characteristic parameter extraction, mode matching and classification, and the beneficial effects are that: more characteristic parameters such as short-time energy E, short-time average amplitude M, short-time average zero-crossing rate Z and detail signal energy E of each scale of wavelet decomposition are introducedwAnd a vibration signal power spectrum P, so that the category of the external vibration signal can be more accurately judged. A similar patent is also 2015105567662. However, due to the factors causing the fiber vibration, the complexity of the signal and the signal characteristics makes the characteristic parameters introduced at the early stage rather difficult, and at the same time, the exact type of the intrusion signal is difficult to judge, so the prior art makes the fiber damage protection work difficult to be practically applied
The prior art has the following defects:
1. the patent utilizes the waterfall picture of signal to go as the medium to discern the type of destroying pipeline safety, and prior art is mostly gone through the direct processing and the analysis to the signal and is judged, and signal processing and analysis degree of difficulty are great, and this patent utilizes the waterfall picture of signal discernment to turn into visual image with the signal, reduces the analysis and judges the degree of difficulty.
2. The interference area caused by factors such as waterfall image environment is removed through image filtering, irrelevant areas can be removed to a certain extent through the filtering, but normal and effective areas can be weakened at the same time, and the image is easy to generate distortion.
3. In the prior art, image features are extracted by a plurality of methods, and various features have different influences on classification, namely good or bad.
4. The prior art generally makes use of random forest classification methods alone.
Disclosure of Invention
The invention provides a method for identifying destructive events of a buried pipeline vibration safety early warning system by using optical fibers, which can overcome the defects of the prior art.
The invention discloses a method for identifying destructive events of a pipeline optical fiber vibration safety early warning system, which comprises the following steps of acquiring real-time signals of soil vibration around a pipeline by using an optical fiber, analyzing and processing the signals, identifying the real-time signals, and judging whether the destructive events occur around the pipeline, wherein the specific method comprises the following steps: the method comprises the steps of obtaining a real-time signal of optical fiber refractive index change caused by vibration of soil around a pipeline, converting the real-time signal into a corresponding waterfall graph, carrying out an abnormal value detection method on the waterfall graph to remove interference areas caused by factors such as environment in an image, extracting image characteristics, training by using the characteristics and corresponding events through a classification algorithm to establish a model of pipeline damage events, and carrying out classification recognition on the actually measured signal on the basis.
Preferably, the method for identifying the destructive event of the pipeline optical fiber vibration safety early warning system extracts a gray level co-occurrence matrix of an image from a signal waterfall diagram after an interference area caused by factors such as environment in the image is removed by using an abnormal value detection method based on a clustering algorithm, obtains five image characteristics including energy, entropy, contrast, autocorrelation and an inverse matrix from the gray level co-occurrence matrix, performs characteristic selection by using an Embedded method according to the characteristics, and finally trains and identifies a model of the destructive pipeline event through a machine learning classification algorithm Adaboost-SVM so as to identify the destructive and non-destructive type events.
Preferably, the identification method of the destructive event of the pipeline optical fiber vibration safety early warning system detects the abnormal value of the waterfall graph by a clustering method and removes the abnormal value point.
Preferably, the clustering method in the method for identifying the destructive event of the pipeline optical fiber vibration safety early warning system is any one of a DBSCAN clustering algorithm, a statistical method detection outlier, an outlier detection based on proximity, an outlier detection based on density, or an outlier detection based on clustering.
Preferably, the clustering method specifically used by the identification method of the destructive event of the pipeline optical fiber vibration safety early warning system is an abnormal value detection method based on clustering. The cluster-based abnormal value detection method is more suitable for being used on images as detection objects, the utilized clustering algorithm DBSCAN method does not need to artificially specify the number of clusters, and the clustering speed and the clustering effect are better represented on waterfall image. The advantages are that: in the prior art, the removal of the image interference area is mainly realized by filtering the image, and the filtering can remove irrelevant areas to a certain extent, but can weaken normal and effective areas, and the image is easy to generate distortion.
Preferably, the method for identifying the destructive event of the pipeline optical fiber vibration safety early warning system provided by the invention trains a classification model after the characteristic selection is performed on the image, and judges whether the destructive event occurs around the pipeline.
More preferably, the algorithm of the training model in the method for identifying the destructive event of the pipeline optical fiber vibration safety early warning system is realized by adopting an Adaboost-SVM algorithm.
The invention has the following advantages:
1. the invention utilizes the waterfall graph of the signal as an identification object to judge the type of damaging the pipeline safety, most of the prior art judges through direct processing and analysis of the signal, the signal processing and analysis difficulty is larger, the waterfall graph identification of the signal is utilized to convert the signal into a visual image, the analysis difficulty is reduced, and the identification accuracy is increased. The analysis and identification of the destructive behavior along the pipeline are realized through a more intuitive signal processing medium frequency spectrum waterfall diagram.
2. The interference area caused by factors such as waterfall image environment is removed through image filtering, irrelevant areas can be removed to a certain extent through the filtering, but normal and effective areas can be weakened at the same time, and the image is easy to generate distortion. The invention realizes the purpose of removing the interference area caused by factors such as waterfall graph environment and the like by a method based on abnormal value detection.
3. The waterfall image recognition obtained by the SVM algorithm in the pipeline safety monitoring system is a weak classifier, the accuracy is not high, and the Adaboost-SVM method is utilized by the patented method to solve the problem of low accuracy.
4. The method comprises the steps of extracting a gray level co-occurrence matrix of the image, obtaining five characteristic items with the largest information amount through a principal component analysis method, and finally obtaining the optimal image characteristic through characteristic selection.
At present, in the field of pipeline safety monitoring, a proper method does not exist in a feature extraction and identification algorithm of an analysis object, and a patent utilization method is a scheme with high identification accuracy and high identification speed.
Drawings
FIG. 1 is a schematic diagram of a working process of a pipeline optical fiber vibration monitoring system.
Fig. 2 is a spectrum waterfall diagram of manual work.
Fig. 3 is a spectral waterfall plot of a mechanical excavation operation.
Fig. 4 is a gray level co-occurrence matrix calculated from the waterfall graph.
Detailed Description
The invention will now be described with reference to one embodiment, which is the preferred embodiment of the invention.
The first step is as follows: optical fiber vibration safety early warning system
The method comprises the steps of utilizing a pipeline optical fiber vibration safety early warning system to obtain vibration signals of soil around a pipeline in real time, placing an optical cable about 0.5 meter below a detected pipeline, and utilizing optical fibers to sense the excavation condition of the soil when a destructive behavior occurs, and referring to attached drawing 1.
The second step is that: and converting the vibration signal in actual measurement into a corresponding frequency spectrum waterfall graph. The transformation method comprises the following steps:
sampling is carried out every 10ms on the length of an optical fiber of a monitoring pipeline, wherein the sampling carried out every 10ms is that laser sends a pulse every 10ms, an analog signal on the optical fiber takes a first value, and a signal variance calculation formula is as follows:
wherein the mean value
xiIs the sampled signal amplitude of the signal, and N is the number of analog signal samples.
In the whole monitored pipeline, values are continuously taken on the analog signals of the optical fibers according to a sampling period, and according to the statistical result of the signal intensity difference of each value, a two-dimensional graph which takes the longitudinal axis as a time axis, the transverse axis as a distance and the initial point of the distance as the starting point of the pipeline is obtained till the whole pipeline is finished, namely the length of the vibrating optical fiber is obtained.
At this time, the signal monitored in real time is changed into two-dimensional data, a waterfall graph obtained by manual operation is formed, the waterfall graph is formed, the Y axis is used as a time axis (sampling period), the X axis is used as a distance, the starting point is used as a pipeline starting point, and the whole pipeline is ended, namely the two-dimensional graph of the distance of the vibrating optical fiber is obtained, the attached figure 2 is a waterfall graph, the image area of the waterfall graph is distributed in intermittent areas at intervals, and the state distribution is consistent with the action of the manual operation. With continuous sampling, the behavior of the event finally forms a waterfall image through the image.
The third step: an abnormal value detection method based on a clustering method is utilized for the frequency spectrum waterfall graph to remove interference areas caused by factors such as environment in the image, and various existing clustering algorithms can be specifically adopted, such as: any one of a DBSCAN clustering algorithm, an outlier detection method based on a statistical method, an outlier detection method based on proximity, an outlier detection method based on density, or an outlier detection method based on clustering. The method for detecting the abnormal value of the cluster is adopted in the best embodiment of the invention, and the method is more suitable for being used on an image as a detection object. The clustering method utilized by the invention is a DBSCAN algorithm. The method of the invention firstly uses DBSCAN clustering algorithm to cluster the image, clusters the image into several areas, then calculates Euclidean distance P, point (x) from each point in the clustering area to the center of the clustering area1，y1) To the central point (x)2，y2) The Euclidean distance calculation formula is as follows:
and obtaining the average distance L from all points in the clustering area to the center, setting a spec value according to multiple tests, wherein the test process is from 0.1 to 10, and taking the value as a test value every 0.5. And calculating the Euclidean distance S from each point in the clustering region to the center of the clustering region, and if S-L is greater than spec, judging the point as an abnormal value point. Abnormal values are represented in pixels or areas of the waterfall graph, when damage behaviors are happening, for example, surrounding vehicles passing through can vibrate at the same time, and vibration signals of the two events can be converted on the waterfall graph at the same time. And removing the points which are judged to be abnormal values from the waterfall image to remove the image area caused by the non-recognition event.
The fourth step: extracting a gray level co-occurrence matrix of the image from the obtained signal waterfall graph, wherein the gray level co-occurrence matrix can be used for calculating to obtain 12 image characteristics, and analyzing the degree of image information content contained in each characteristic by utilizing a principal component analysis method for the 12 characteristics. The gray level co-occurrence matrix calculates 5 image characteristics as follows: firstly, obtaining a gray level co-occurrence matrix of an image, wherein the calculation process comprises the following steps: the gray scale values of any point (x, y) and another point (x + a, y + b) deviated from the point in the image N × N are set as (g1, g 2). When the dot (x, y) is moved over the entire screen, various values (g1, g2) are obtained, and when the number of gradation values is k and the value range of k is 0 to 255, the combination of (g1, g2) shares the square of k. For the whole picture, the number of times each (g1, g2) value appears is counted, then arranged into a square matrix, and the (g1, g2) total number of times are used for normalizing the values into probability P of appearance (g1, g2), and the square matrix is called a gray level co-occurrence matrix. The distance difference values (a, b) take different numerical value combinations, and joint probability matrixes under different conditions can be obtained. The values of (a and b) are selected according to the characteristics of the periodic distribution of the textures, and for the finer textures, small difference values such as (1, 0), (1, 1), 2, 0) and the like are selected.
When a is 1 and b is 0, the pixel pair is horizontal, i.e. 0 degree scan; when a is 0 and b is 1, the pixel pair is vertical, i.e. 90 degree scan; when a is 1 and b is 1, the pixel pair is right diagonal, i.e. 45 degree scan;
when a is-1 and b is 1, the pixel pair is the left diagonal, i.e. 135 degree scan. Thus, the probability of two pixel gray levels occurring simultaneously converts the spatial coordinates of (x, y) into a description of "gray pairs" (G1, G2), forming a gray co-occurrence matrix G (i, j).
The energy is the sum of squares of the gray level co-occurrence matrix element values, reflects the image gray level distribution uniformity degree and the texture thickness degree, and has the calculation formula as follows:
entropy is a measure of the amount of information an image has, and is calculated by the formula:
the contrast reflects the contrast of the brightness of a certain pixel value and the pixel value in the field, and the calculation formula is as follows:
the autocorrelation reflects the consistency of the image texture, and the calculation formula is as follows:
wherein the gray level co-occurrence matrix transverse mean
Variance (variance)
Longitudinal mean value of moment gray level co-occurrence array
Variance (variance)
The calculation formula of the inverse matrix is as follows:
five feature vectors ASM, ENT, CON, COR, and IDM are obtained by calculating the above five features in the entire image. Obtaining an image gray pixel point matrix shown in fig. 4 according to the feature vector, wherein the left matrix in fig. 4 is a gray pixel point of the image, the right matrix is a gray co-occurrence matrix obtained by calculation, and the feature vector of five images including energy, entropy, contrast, autocorrelation and inverse matrix of the image is obtained through the gray co-occurrence matrix.
The fifth step: for the five different gray level co-occurrence matrix image characteristics calculated above, an Embedded method, a Filter and wrapper method and the like can be adopted for characteristic selection.
And a sixth step: training a classification model for identifying destructive events of vibration signals of a pipeline optical fiber vibration safety early warning system, respectively carrying out artificial labeling on an artificial damage operation, a mechanical damage operation, a waterfall image caused by vehicle passing and an environment which are acquired in advance, inputting the images into an Adaboost-SVM alternative classification training method through the first step to the fifth step, carrying out iterative training on an SVM through an Adaboost method, wherein the Adaboost and a plurality of methods can be combined. Key parameters of the model: the method in Adaboost utilizes SAMME with a learning rate of 0.4. The kernel function used by the SVM selects a Radial Basis Function (RBF), the kernel parameter is 100, the penalty factor is 10, and the optimal parameters of the SVM are used by the classification algorithm, so that the effect is better. And when the classification model identifies a manual operation event or a mechanical operation event, the monitoring system gives an alarm prompt.
According to the invention, as an abnormal value detection method based on clustering is used for removing irrelevant areas of the image, and simultaneously, a principal component analysis method is used for obtaining the gray level co-occurrence matrix image characteristics with the largest information content, through a better Adaboost-SVM combined algorithm, the recognition accuracy rate is greatly improved, and the event of damaging the pipeline safety is accurately recognized.
In order to verify the accuracy of model identification, data are derived from Distributed Vibration Sensing (DVS) software of a pipeline optical fiber Vibration safety monitoring system of an applicant, the field environment is a project of a channel safety monitoring system of Yuan dam town of Sichuan province, the length of a channel is 22.41Km in total, the length of an optical fiber is 26.51Km in total, the geographic environment of a pipeline is complex, mountainous areas, rivers, farmlands, roads and densely populated areas exist, and great difficulty is added to identification of the type of a Vibration event.
The alarm position can happen at any point of 22.41Km total length of the pipeline, the damage position is random, and several positions can be simultaneously alarmed.
Under normal conditions, the signals are relatively stable and cannot generate a waterfall pattern, when severe vibration occurs and reaches a signal capture threshold, the system can capture abnormal vibration signals at the moment, and the captured signals are converted into the waterfall pattern. Waterfall charts such as those of fig. 2 and 3 may generate alarms when the type of manual work and mechanical work events are identified. When the vibration amplitude of signals sensed by optical fibers around the pipeline exceeds a capture threshold value, the signals are captured and converted into a frequency spectrum waterfall graph, the converted waterfall graph is identified by the method, and an alarm is given when the event that the pipeline safety is damaged by manual operation and mechanical operation is identified. For example, waterfall plots for manual work and mechanical work are shown in fig. 2 and 3. The image distribution of the two events is consistent with the occurrence rule of the behavior of the destruction event. The abnormal point removal is mainly to remove an interference image area caused by a non-recognition event in the waterfall image, for example, when a destructive behavior occurs, the surrounding vehicles passing through can also generate vibration, and the vibration of the two events can be simultaneously shown on the waterfall graph. In fig. 1 and 2, the surrounding white discontinuous light-like spots are possible areas of other vibration generation, interfering to some extent with the identification of destructive events.
Claims (7)
1. The utility model provides a pipeline optic fibre vibration safety precaution system destructive event's identification method, utilizes optic fibre to acquire the real-time signal of soil vibration around the pipeline, through to signal analysis and processing back, discerns real-time signal, judges whether there is destructive event's emergence around the pipeline, its characterized in that: the method comprises the steps of obtaining a real-time signal of optical fiber refractive index change caused by vibration of soil around a pipeline, converting the real-time signal into a corresponding waterfall graph, carrying out an abnormal value detection method on the waterfall graph to remove interference areas caused by factors such as environment in an image, extracting image characteristics, training by using the characteristics and corresponding events through a classification algorithm to establish a model for identifying pipeline damage events, and carrying out classification identification on the actually measured signal on the basis.
2. The method for identifying the destructive event of the pipeline optical fiber vibration safety early warning system according to claim 1, is characterized in that: extracting a gray level co-occurrence matrix of an image from a signal waterfall pattern after an interference area caused by factors such as environment in the image is removed by using an abnormal value detection method based on a clustering algorithm, obtaining five image characteristics of energy, entropy, contrast, autocorrelation and an inverse matrix from the gray level co-occurrence matrix, performing characteristic selection by using an Embedded method according to the characteristics, and finally training and identifying a model of a damaged pipeline event through a machine learning classification algorithm Adaboost-SVM to identify a damaged and non-damaged type event.
3. The method for identifying the destructive event of the pipeline optical fiber vibration safety early warning system according to claim 2, is characterized in that: and carrying out abnormal value detection on the waterfall graph by using a clustering method, and removing abnormal value points.
4. The method for identifying the destructive event of the pipeline optical fiber vibration safety early warning system according to claim 3, wherein the clustering method is any one of a DBSCAN clustering algorithm, a statistical method detection outlier, a proximity-based outlier detection, a density-based outlier detection or a clustering-based outlier detection method.
5. The method for identifying the destructive events of the pipeline optical fiber vibration safety early warning system according to claim 4, wherein the clustering method is a clustering-based abnormal value detection method.
6. The method for identifying the destructive event of the pipeline optical fiber vibration safety early warning system according to any one of claims 1 to 5, wherein after the image is subjected to feature selection, a classification model is trained to judge whether the destructive event occurs around the pipeline.
7. The method for identifying the destructive events of the pipeline optical fiber vibration safety early warning system according to claim 6, wherein an algorithm of a training model is realized by adopting an Adaboost-SVM algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110075378.8A CN112883802A (en) | 2021-01-20 | 2021-01-20 | Method for identifying destructive event of pipeline optical fiber vibration safety early warning system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110075378.8A CN112883802A (en) | 2021-01-20 | 2021-01-20 | Method for identifying destructive event of pipeline optical fiber vibration safety early warning system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112883802A true CN112883802A (en) | 2021-06-01 |
Family
ID=76050740
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110075378.8A Pending CN112883802A (en) | 2021-01-20 | 2021-01-20 | Method for identifying destructive event of pipeline optical fiber vibration safety early warning system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112883802A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| CN115691018A (en) * | 2022-10-26 | 2023-02-03 | 哈尔滨市科佳通用机电股份有限公司 | Railway perimeter intrusion monitoring and early warning method and system based on multi-sensor fusion |
| CN116757534A (en) * | 2023-06-15 | 2023-09-15 | 中国标准化研究院 | Intelligent refrigerator reliability analysis method based on neural training network |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103235954A (en) * | 2013-04-23 | 2013-08-07 | 南京信息工程大学 | Improved AdaBoost algorithm-based foundation cloud picture identification method |
| CN105931402A (en) * | 2016-06-27 | 2016-09-07 | 上海波汇科技股份有限公司 | Optical fiber perimeter intrusion monitoring method based on image recognition |
| CN107591002A (en) * | 2017-09-21 | 2018-01-16 | 电子科技大学 | A kind of freeway traffic parameter real-time estimation method based on distribution type fiber-optic |
| US20200005036A1 (en) * | 2018-06-28 | 2020-01-02 | Nec Laboratories America, Inc | Time-space de-noising for distributed sensors |
| CN110660184A (en) * | 2019-09-18 | 2020-01-07 | 西安科技大学 | Adaboost-based railway perimeter early warning method of fiber laser radar |
| CN111222461A (en) * | 2020-01-06 | 2020-06-02 | 兰州奥普信息技术有限公司 | Method for identifying invasion signal of optical fiber vibration detection system |
| US20200401784A1 (en) * | 2019-06-19 | 2020-12-24 | Nec Laboratories America, Inc | Distributed intelligent traffic informatics using fiber sensing |
-
2021
- 2021-01-20 CN CN202110075378.8A patent/CN112883802A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103235954A (en) * | 2013-04-23 | 2013-08-07 | 南京信息工程大学 | Improved AdaBoost algorithm-based foundation cloud picture identification method |
| CN105931402A (en) * | 2016-06-27 | 2016-09-07 | 上海波汇科技股份有限公司 | Optical fiber perimeter intrusion monitoring method based on image recognition |
| CN107591002A (en) * | 2017-09-21 | 2018-01-16 | 电子科技大学 | A kind of freeway traffic parameter real-time estimation method based on distribution type fiber-optic |
| US20200005036A1 (en) * | 2018-06-28 | 2020-01-02 | Nec Laboratories America, Inc | Time-space de-noising for distributed sensors |
| US20200401784A1 (en) * | 2019-06-19 | 2020-12-24 | Nec Laboratories America, Inc | Distributed intelligent traffic informatics using fiber sensing |
| CN110660184A (en) * | 2019-09-18 | 2020-01-07 | 西安科技大学 | Adaboost-based railway perimeter early warning method of fiber laser radar |
| CN111222461A (en) * | 2020-01-06 | 2020-06-02 | 兰州奥普信息技术有限公司 | Method for identifying invasion signal of optical fiber vibration detection system |
Non-Patent Citations (1)
| Title |
|---|
| ASLANGUL, S.: "Detecting Tunnels for Border Security based on Fiber Optical Distributed Acoustic Sensor Data using DBSCAN", IN PROCEEDINGS OF THE 9TH INTERNATIONAL CONFERENCE ON SENSOR NETWORKS (SENSORNETS 2020), pages 78 - 84 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| 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 |
| CN115691018A (en) * | 2022-10-26 | 2023-02-03 | 哈尔滨市科佳通用机电股份有限公司 | Railway perimeter intrusion monitoring and early warning method and system based on multi-sensor fusion |
| CN115691018B (en) * | 2022-10-26 | 2023-04-21 | 哈尔滨市科佳通用机电股份有限公司 | Railway perimeter intrusion monitoring and early warning method and system based on multi-sensor fusion |
| CN116757534A (en) * | 2023-06-15 | 2023-09-15 | 中国标准化研究院 | Intelligent refrigerator reliability analysis method based on neural training network |
| CN116757534B (en) * | 2023-06-15 | 2024-03-15 | 中国标准化研究院 | Intelligent refrigerator reliability analysis method based on neural training network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112883802A (en) | Method for identifying destructive event of pipeline optical fiber vibration safety early warning system | |
| US11562224B2 (en) | 1D-CNN-based distributed optical fiber sensing signal feature learning and classification method | |
| CN105931402B (en) | Optical fiber perimeter intrusion detection method based on image identification | |
| CN110570613A (en) | Fence vibration intrusion positioning and mode identification method based on distributed optical fiber system | |
| CN103442209B (en) | Video monitoring method of electric transmission line | |
| CN102496232B (en) | Transmission facility monitoring method and system | |
| CN104766099B (en) | Distribution type fiber-optic circumference vibration signal processing and recognition methods based on image | |
| CN107392885A (en) | A kind of method for detecting infrared puniness target of view-based access control model contrast mechanism | |
| CN102928435A (en) | Aircraft skin damage identification method and device based on image and ultrasound information fusion | |
| CN111222461A (en) | Method for identifying invasion signal of optical fiber vibration detection system | |
| CN115620212B (en) | Behavior identification method and system based on monitoring video | |
| CN111537056A (en) | Pipeline along-line third-party construction dynamic early warning method based on SVM and time-frequency domain characteristics | |
| CN109034038B (en) | Fire identification device based on multi-feature fusion | |
| CN114510960A (en) | Method for recognizing distributed optical fiber sensor system mode | |
| CN114093106B (en) | Intrusion signal alarm method and system based on EfficientNET classification network | |
| CN109635717A (en) | A kind of mining pedestrian detection method based on deep learning | |
| CN111539394B (en) | Pipeline along-line third-party construction early warning method based on time domain characteristics and space-time information | |
| CN111325708B (en) | Transmission line detection method and server | |
| CN114354744B (en) | Power cable external damage prevention monitoring method, system and device | |
| CN116052035A (en) | Power plant personnel perimeter intrusion detection method based on convolutional neural network | |
| CN114925722A (en) | Perimeter security intrusion signal detection method based on generalized S transformation and transfer learning | |
| Yuanbin et al. | Fire detection based on image processing in coal mine | |
| CN109034125A (en) | Pedestrian detection method and system based on scene complexity | |
| CN117312828B (en) | Public facility monitoring method and system | |
| CN115761460B (en) | Greenhouse risk identification method and device, electronic equipment and storage medium |
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 |