CN112071009B - Optical fiber pipeline early warning system and method thereof - Google Patents
Optical fiber pipeline early warning system and method thereof Download PDFInfo
- Publication number
- CN112071009B CN112071009B CN201910499887.6A CN201910499887A CN112071009B CN 112071009 B CN112071009 B CN 112071009B CN 201910499887 A CN201910499887 A CN 201910499887A CN 112071009 B CN112071009 B CN 112071009B
- Authority
- CN
- China
- Prior art keywords
- signal
- intrusion
- identification information
- sub
- signals
- 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
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/22—Electrical actuation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Burglar Alarm Systems (AREA)
Abstract
The embodiment of the invention provides a signal mode acquisition method for object intrusion early warning, and belongs to the technical field of pipeline monitoring. The method comprises the following steps: acquiring signals of different types of intrusion events and determining different signal characteristic types according to the signals; decomposing the signal according to the signal characteristic class to obtain a sub-signal, and forming a sub-signal sample set corresponding to different classes of intrusion events according to the corresponding relation between the sub-signal and the signal; and obtaining a signal mode set corresponding to each sub-signal according to a preset mapping rule by utilizing the sub-signal sample set. The embodiment of the invention has low false alarm rate and can directly identify the event type according to the signal mode.
Description
Technical Field
The invention relates to the technical field of pipeline monitoring, in particular to a signal pattern acquisition method for object intrusion early warning, a construction method of an object intrusion early warning system with a signal pattern recognition function, an intrusion event recognition method of the object intrusion early warning system, equipment for object intrusion early warning and a computer-readable storage medium.
Background
The optical fiber pipeline safety early warning system is based on an optical fiber vibration sensing principle, adopts a pipeline communication optical cable as a sensing and communication medium, and achieves the safety monitoring system for early warning the destructive behaviors of illegal construction, excavation of theft oil and the like of a pipeline through detection, analysis and positioning of an environmental vibration signal of the pipeline.
The comparison document CN104565826A adopts a pipeline safety early warning system based on the coherent Rayleigh scattering principle to detect and identify multipoint vibration signals along the pipeline in real time, and due to the complex environment condition of the oil or natural gas pipeline, heavy rain, road and railway vehicles, rivers, animals and the like along the pipeline can form vibration signal sources to interfere the detection and analysis of the system, and the technology is easy to generate false alarm due to the interference of various human factors. Therefore, the early warning system needs to perform mode analysis on the detected vibration signal of the pipeline environment, find out the behavior threatening the pipeline really, and filter out the interference source, which becomes the key and difficult point of software analysis of the system.
Disclosure of Invention
The embodiment of the invention aims to provide an optical fiber pipeline early warning system and a method thereof, and the prior art has the technical problems that false alarm caused by interference signals is mistakenly reported, a real intrusion event and an along-line engineering event cannot be distinguished, and the intrusion event cannot be quickly distinguished, such as an artificial excavation intrusion event, a small-sized mechanical excavation intrusion event or a large-sized mechanical excavation intrusion event.
In order to achieve the above object, an embodiment of the present invention provides a signal pattern acquisition method for early warning of intrusion of an object, where the signal pattern acquisition method includes:
s1) obtaining signals of different types of intrusion events and determining different signal characteristic types according to the signals;
s2) decomposing the signals according to the signal feature classes to obtain sub-signals, and forming sub-signal sample sets corresponding to different classes of intrusion events according to the corresponding relation between the sub-signals and the signals;
and S3) obtaining a signal mode set corresponding to each sub-signal according to a preset mapping rule by utilizing the sub-signal sample set.
Specifically, the step S1) of obtaining signals of different types of intrusion events includes:
s101) recording signals of each intrusion event;
s102) marking identification information of each intrusion event, identifying any two intrusion events into different classes when the identification information of any two intrusion events is different, and acquiring signals of the intrusion events of different classes, wherein the identification information comprises: the intrusion detection system comprises intrusion object identification information, invaded object environment identification information, invaded object position identification information, relative position identification information of an intrusion object relative to the invaded object, intrusion occurrence time identification information and intrusion behavior mode identification information.
Specifically, after determining different signal feature classes according to the signal, the step S1) further includes:
decomposing each type of signal feature class, and obtaining a signal subclass parameter set after all signal feature classes are decomposed, wherein the signal feature classes comprise at least two types of space feature classes, time feature classes, intensity feature classes and frequency feature classes, or the signal feature classes comprise at least two types of space feature classes, time feature classes, intensity feature classes and frequency feature classes and also comprise cross feature classes, and the cross feature classes comprise feature classes formed by at least two types of the space feature classes, the time feature classes, the intensity feature classes and the frequency feature classes.
Specifically, step S2) includes:
s201) decomposing the signal according to the signal characteristic class to obtain a sub-signal and a parameter set of the sub-signal;
s202) determining parameters corresponding to each current signal subclass parameter according to the signal subclass parameter set and the parameter set, and determining the corresponding relation between the signal subclass parameters and the sub-signals to obtain a current signal subclass parameter set;
s203) determining relevant characteristics of parameters of the current signal subclass parameter set, screening signal subclasses parameters according to the relevant characteristics and the relation of preset conditions, obtaining a new signal subclass parameter set after screening, and then utilizing the new signal subclass parameter set, combining the subsignals with the corresponding relation of the signals and different types of intrusion events to form a subsignal sample set corresponding to the different types of intrusion events.
Specifically, in step S3), the signal pattern set includes a signal sub-class parameter corresponding to each sub-signal and a distribution function related to the signal sub-class parameter, or the signal pattern set includes a signal sub-class parameter corresponding to each sub-signal and a classifier related to the signal sub-class parameter.
Specifically, the signal pattern obtaining method further includes:
s4) when a new intrusion event with at least one identification information different from the identification information of each intrusion event recorded in the step S101) is acquired, returning to the step S1).
The embodiment of the invention provides a construction method of an object intrusion early warning system with a signal pattern recognition function, which comprises the following steps:
s1) marking and recording different position identification information of an object using an object intrusion early warning system and different environment identification information corresponding to the position identification information;
s2) acquiring current time identification information and configuring a signal mode identification function according to different environment identification information of each position identification information corresponding to the current time identification information.
Specifically, the step S2) of configuring the signal pattern recognition function includes:
s201) selecting a target signal mode set from all signal modes, and configuring the priority of the target signal mode set to be higher than the priority of the rest signal mode sets except the target signal mode set in all signal modes;
s202) configuring the confidence degree of each target signal mode in the target signal mode set, wherein the target signal mode set has a distribution function corresponding to target signals, and the target signals are signals of different types of intrusion events.
The embodiment of the invention provides an intrusion event identification method of an object intrusion early warning system, which comprises the following steps:
s1) acquiring a collection signal, wherein the collection signal comprises vibration detection information of an object using an object intrusion early warning system and/or vibration detection information of an environment where the object is located;
s2) calculating the current position identification information of the acquired signal, inquiring and acquiring the pre-recorded environment identification information and the pre-configured signal pattern recognition function corresponding to the current position identification information;
and S3) acquiring current time identification information, identifying and determining the signal mode of the acquired signal according to the current time identification information and the pre-configured signal mode identification function, identifying that the acquired signal is a signal belonging to different types of intrusion events when the signal mode of the acquired signal is in a corresponding relation with pre-recorded environment identification information, and determining that the intrusion event occurs.
Specifically, the identifying and determining the signal mode of the acquired signal in step S3) includes:
s301) decomposing the acquired signal to obtain parameters of parameters in all signal modes, wherein the signal modes comprise signal subclass parameters and distribution functions related to the signal subclass parameters;
s302) comparing the acquired signals with each signal mode according to a pre-configured priority order, identifying and obtaining the signal mode to be determined, and determining the signal mode of the acquired signals according to the confidence degree of the pre-configured signal mode corresponding to the signal mode to be determined.
The embodiment of the invention provides an object intrusion early warning system, which comprises:
a control system for carrying out the aforementioned method.
In another aspect, an embodiment of the present invention provides an apparatus for early warning of intrusion of an object, including:
at least one processor;
a memory coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor, the at least one processor implements the aforementioned method by executing the instructions stored by the memory.
In still another aspect, an embodiment of the present invention provides a computer-readable storage medium storing computer instructions, which, when executed on a computer, cause the computer to perform the foregoing method.
The method acquires a signal mode possibly associated with a certain type of intrusion event through collecting signal characteristic processing analysis of the intrusion event;
the invention introduces the identification information defining the type of the intrusion event, and can fully distinguish different types of conditions such as interference events, engineering events, intrusion events and the like;
according to the invention, not only are direct signal decomposition characteristics considered, but also cross classes considered, and further correlation judgment is carried out, so that during actual use, excessively correlated signal subclass parameters can be screened out;
the signal mode set is a distribution function set, and compared with other machine learning models, classification judgment is carried out on the signal mode set, and the distribution function can more quickly finish signal mode identification;
the signal mode set is continuously updated so as to prevent the signal mode set from being too old and inconsistent with the current application environment to cause the missing report and the like;
the object intrusion early warning system configured by the invention is combined with the current application environment, and the signal pattern recognition result is closely related to the environment;
the invention considers that the general pipeline objects have very long geographical span, so although the same pipeline object is configured with different signal mode priorities and confidences, the identification result is associated with the environment, and false alarm is reduced;
the invention combines the comparison process of time, environmental factors and conditions, and can quickly judge whether the position generated by the acquisition signal is an intrusion event.
Additional features and advantages of embodiments of the invention will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the embodiments of the invention without limiting the embodiments of the invention. In the drawings:
FIG. 1 is a schematic diagram of the main method of the embodiment of the present invention;
FIG. 2 is a schematic diagram of an engineering event and an intrusion event according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a signal decomposition and signal acquisition mode process according to an embodiment of the present invention;
fig. 4 is a schematic diagram of the operation of the display device according to the embodiment of the present invention.
Detailed Description
The following describes in detail embodiments of the present invention with reference to the drawings. It should be understood that the detailed description and specific examples, while indicating embodiments of the invention, are given by way of illustration and explanation only, not limitation.
Example 1
Referring to fig. 1, a signal pattern acquisition method for early warning of intrusion of an object, the signal pattern acquisition method includes:
s1) obtaining signals of different types of intrusion events and determining different signal characteristic types according to the signals;
s2) decomposing the signals according to the signal feature classes to obtain sub-signals, and forming sub-signal sample sets corresponding to different classes of intrusion events according to the corresponding relation between the sub-signals and the signals;
s3) obtaining a signal mode set corresponding to each sub-signal according to a preset mapping rule by utilizing the sub-signal sample set;
the method comprises the following steps that an object is taken as a petroleum pipeline, a signal can be obtained by collecting a laser signal which is near the landfill position of the petroleum pipeline or is directly arranged in an optical cable or an optical fiber pipeline of the petroleum pipeline, the signal has various characteristics, namely, a sub-signal (the sub-signal is further decomposed into a plurality of sub-signals) of time, space, frequency, intensity and the like can be obtained through decomposition, and a crossed sub-signal can also be formed, for example, the intensity sub-signal and the time sub-signal are crossed to form a power sub-signal; the preset mapping rule may be nonlinear mapping and iterative computation performed by the deep neural network learning model under the condition of initial condition, or statistics performed under the condition of initial condition (statistics may be a clustering algorithm, such as a C-means algorithm), or statistics performed on a sub-signal sample set, and then the statistical result is used as the input of the deep neural network learning model, and then the nonlinear mapping and iterative computation performed under the condition of initial condition; a set of signal patterns corresponding to each sub-signal is obtained, which naturally also correspond to different classes of intrusion events.
Specifically, the step S1) of obtaining signals of different types of intrusion events includes:
s101) recording signals of each intrusion event;
s102) marking identification information of each intrusion event, identifying any two intrusion events into different classes when the identification information of any two intrusion events is different, and acquiring signals of the intrusion events of different classes, wherein the identification information comprises: the method comprises the following steps of (1) identifying information of an invading object, identifying information of an invaded object, identifying information of an environment of the invaded object, identifying information of a position of the invaded object, identifying information of a relative position of the invading object relative to the invaded object, identifying information of time of invading occurrence and identifying information of invading behavior modes;
as shown in fig. 2, an optical cable Fi (i is a positive integer, and F1, F2, and F3 are three event occurrence sections of the entire optical cable) buried in underground soil together with a pipeline for transporting oil and gas resources (pipeline is a section of the entire pipeline where three events occur, and a flow direction or a signal transmission direction of fluid in the pipeline is represented by "x" and "·") is affected by vibration of the ground surface, so that signals (Ei, si) corresponding to different types of invasion events Ei are obtained in a detection device; the intrusion event can be divided into a true intrusion event (event E1 and event E2 in fig. 2) and a false intrusion event, and the false intrusion event can include an interference event and a pre-documented (in the early warning system) engineering event (event E3 in fig. 2); in some embodiments, intrusion events can be further classified as true intrusion events, suspicious intrusion events, and false intrusion events; the signal data of the intrusion event can be realized by simulating intrusion or come from real event signal records; various identification information related to the intrusion event is obtained by marking engineering environment (static state such as soil texture and dynamic state such as vehicle interference), stage and intrusion behavior characteristics (such as rhythm and behavior initiator characteristics) of field behavior personnel when triggering the vibration of the optical fiber pipeline; the identification information of the invading object can be identification information of manual excavation or identification information of mechanical excavation, the identification information of the invaded object is a liquid transportation pipeline (such as a petroleum pipeline) at the moment, the environmental identification information of the invaded object can be whether the identification information of a traffic main road or a pre-filed engineering operation area, the position identification information of the invaded object can be the geographic position identification information of relative detection equipment, the relative position identification information of the invaded object relative to the invaded object can be the identification information (right above or side part) of the relative position of an excavator relative to an optical fiber pipeline and relative distance information (obtained by calculating signals), the time identification information of invasion occurrence can be season identification information or engineering event operation time identification information, and the invasion behavior mode identification information can be the identification information of excavation mode, the identification information of small-sized excavation mechanical excavation mode or the identification information of large-sized excavation mechanical excavation mode; for example, a shovel ground work event located near a fiber optic conduit or cable line on a traffic thoroughfare, the different types of marked intrusion events may include: "winter (time) -important road (environment) -small excavator (invading object and behavior mode) -first section first number optical cable (invaded object and position) shallow burying (relative position) excavating-engineering event (environment and time)", "winter-desert-small excavator-first section second number optical cable deep burying dry soil excavating-invading event", "winter-important road-large excavator-second section first number optical cable shallow burying side excavating-engineering event" and "summer-plain-small excavator-optical cable shallow burying wet soil side excavating-invading event" and the like; for example, event E1 "summer-plaint-manual excavation-second-stage second-numbered optical cable deep-buried dry soil excavation-invasion event" and signal is (E1, S1), event E2 "summer-plaint-mini-excavator-optical cable deep-buried wet soil lateral excavation-invasion event" and signal is (E2, S2), event E3 "summer-main road-large-excavator-optical cable shallow-buried dry soil excavation-engineering event" and signal is (E3, S3); the soil dryness and humidity (the wave dotted line in fig. 2 indicates that the soil moisture content in the area is large) are introduced because the practice finds that the soil moisture content can affect the signal vibration in the optical fiber, and the influence is that the soil moisture content is small and the sensitivity of vibration detection is high; the obtained set of signal patterns also corresponds to different identification information for each type of intrusion event.
Specifically, after determining different signal feature classes according to the signal, step S1) further includes:
decomposing each type of signal feature class to obtain a signal subclass parameter set after all the signal feature classes are decomposed, wherein the signal feature classes comprise at least two types of space feature classes, time feature classes, intensity feature classes and frequency feature classes, or the signal feature classes comprise at least two types of space feature classes, time feature classes, intensity feature classes and frequency feature classes and also comprise cross feature classes, and the cross feature classes comprise feature classes formed by at least two types of the space feature classes, the time feature classes, the intensity feature classes and the frequency feature classes;
the preset mapping rule is used for carrying out nonlinear mapping and iterative calculation on the deep neural network learning model under the condition of initial conditions, meanwhile, the more abundant the shallow feature classes of the deep neural network learning model are, the more isolated the feature parameters are, the more effective the application scene adopting the deep neural network learning model is, and the more obvious feature parameters such as average power, accumulated area and the like except simple time, space, speed and central deviation are obtained through repeatedly acquiring, comparing and analyzing signals of a large number of invasion events; specifically, the characteristic parameters of the deep layer of the deep neural network learning model (in the deep neural network learning model, the ith neuron of the l layer
Wherein, a l As a function of layer i activation, W l Is the l-th layer weight value, b l For the l-th layer bias parameter, the characteristic parameter may be
) The method comprises the steps of carrying out behavior identification on different weights, continuously updating the weight values of deep parameters in a shallow layer through signals of new intrusion events, setting new characteristic parameters through signal characteristics of different types of new intrusion events, and introducing a new classifier to be used for identifying the type of the new intrusion events.
Specifically, step S2) includes:
s201) decomposing the signal according to the signal characteristic class to obtain a sub-signal and a parameter set of the sub-signal;
s202) determining parameters corresponding to each current signal subclass parameter according to the signal subclass parameter set and the parameter set, and determining the corresponding relation between the signal subclass parameters and the sub-signals to obtain a current signal subclass parameter set;
s203) determining relevant characteristics of parameters of the current signal subclass parameter set, screening signal subclass parameters according to the relevant characteristics and the relation of preset conditions, obtaining a new signal subclass parameter set after screening, and forming a sub-signal sample set corresponding to different types of intrusion events by utilizing the new signal subclass parameter set, combining the sub-signals with the corresponding relation of the signals and combining the corresponding relation of the signals and different types of intrusion events.
Specifically, in step S3), the signal pattern set includes a signal sub-class parameter corresponding to each sub-signal and a distribution function related to the signal sub-class parameter, or the signal pattern set includes a signal sub-class parameter corresponding to each sub-signal and a classifier related to the signal sub-class parameter;
as shown in fig. 3, the preset mapping rule is a statistical rule and a deep neural network model, and the physical characteristics of the vibration signals detected by the system are divided into sub-signals of a space class a, a time class B, an intensity class C, a frequency class D, a cross class E and the like, each of the classes can be divided into sub-signals of several sub-classes (e.g., < Aa, ab, ac, ba \8230;, >), each sub-class corresponds to a specific parameter into which the signal can be decomposed, and the decomposition and acquisition of the sub-class parameters of the signal can be performed by using a wavelet analysis method or a fourier transform method, where the wavelet analysis has a higher resolution compared with the fourier transform, and the fourier transform cannot analyze the signal in multiple scales; the structure and sampling rate of various filters in wavelet analysis can be selected according to the signal characteristics of the intrusion event and the final required identification precision; the target signal to be identified is detected and analyzed, and parameters (pre-learning target parameters) of signal subclass parameters < TAa, TAb, tac, \8230;) aiming at the target signal are obtained; calculating the correlation of every two pre-learning target parameters, and eliminating the parameters with overhigh correlation (the correlation coefficient is higher than a specified threshold) to obtain independent target parameters (new signal subclass parameters < TAa, TAb >); then, a great amount of cyclic training is carried out on the target signal in a deep neural network model (such as a back propagation neural network model or a feedforward neural network model), sub-signal samples < Aa, ab, ac, \8230; > are calculated, and respective distribution functions < FAa (), FAb () > (F or T are mapping function relations) or classifiers are calculated, finally, the signal mode of the target signal is formed by independent target parameters and the distribution functions or the classifiers thereof, and finally the obtained distribution functions or the classifiers can correspond to various kinds of identification information of the intrusion event through signals.
Specifically, the signal pattern obtaining method further includes:
s4) returning to the step S1) when a new intrusion event with at least one identification information different from the identification information of each intrusion event recorded in the step S101) is acquired;
when learning and sample forming are performed on target events such as new intrusion events and new engineering events, in order to adapt to complicated and changeable conditions, for the single type of target event, a large number of sample sequences for collecting signals need to be collected, for example, signal collection in various mining modes is performed on a certain area in four seasons, such as spring, summer, autumn and winter, so as to obtain feature expressions of the target events under different conditions, and complete coverage is performed on signal modes corresponding to real situations as much as possible.
Debugging personnel only need to set related initial parameters and calculation thresholds, provide recorded signals of the intrusion event and used for test training, and obtain a signal mode corresponding to the intrusion event signals after all signal acquisition, analysis, learning and mode calculation are completed;
the target signal type to be detected and pre-warned can be selected according to the actual environment and requirements, and the interference signals can be effectively filtered and filtered in the use of the system through the test and repeated learning of a large number of parameters of the sample, so that the pre-warning accuracy is improved;
on the basis of long-term learning accumulation, the system can obtain a signal pattern related to time (seasons) and a signal pattern related to environment (soil texture), and the pipeline is a characteristic parameter with relative dynamic change in different seasons or different environments.
Example 2
Based on embodiment 1, a method for constructing an object intrusion warning system with a signal pattern recognition function is provided, where a signal pattern of the object intrusion warning system used at this time is derived from the signal pattern set in embodiment 1, and the method includes:
s1) marking and recording different position identification information of an object using an object intrusion early warning system and different environment identification information corresponding to the position identification information;
s2) acquiring current time identification information and configuring a signal mode identification function according to different environment identification information of each position identification information corresponding to the current time identification information.
Specifically, the step S2) of configuring the signal pattern recognition function includes:
s201) selecting a target signal mode set from all signal modes, and configuring the priority of the target signal mode set to be higher than the priority of the rest signal mode sets except the target signal mode set in all signal modes;
s202) configuring the confidence degree of each target signal mode in the target signal mode set, wherein the target signal mode set has a distribution function corresponding to a target signal, and the target signal is a signal of different intrusion events; because the distribution function or the distribution functions correspond to the target signal patterns and each target signal pattern has a unique corresponding target signal (one target signal may have a plurality of target signal patterns according to the number of feature classes), the distribution function corresponding to the target signal exists in the target signal pattern set according to the target signal corresponding to the target signal pattern in the target signal pattern set; simultaneously configuring the confidence degrees of all distribution functions corresponding to the target signal mode, namely configuring the confidence degree of the target signal mode;
step S202) may alternatively be implemented by configuring a confidence threshold of a classifier or a plurality of classifiers corresponding to each target signal pattern in the target signal pattern set, where the confidence threshold is used to obtain a probability interval of the classifier or the plurality of classifiers for determining the event type, for example, if the probability output by the classifier or the plurality of classifiers is greater than the confidence threshold, then the classifier result appears to identify the classified feature, and the target signal pattern is a signal of a different class of intrusion event;
the object can be an optical fiber pipeline, the position identification information can be the content such as whether the object is in a required path or an area with larger soil water content, the current time identification information can be the content such as seasonal information, the environment identification information can be the content such as whether a pre-recorded engineering event exists or not and whether jumping transformation of soil water content exists or not at the position along the line of the object, the priority of each signal mode at a certain position of the object is configured corresponding to the identification information by fully researching different identification information of the same object (for example, the priority of a signal mode of a certain optical fiber pipeline without the pre-recorded engineering event in summer is higher than that of a signal mode of a certain number optical fiber cable with credibility of a certain number, namely a summer-plain-small excavator-second section optical fiber cable with buried wet soil side excavation-shallow invasion event ", is higher than that of a signal mode of a certain number, namely a winter-major path-large excavator-second section optical fiber cable with buried side excavation-engineering event"), the false alarm rate can be effectively reduced, and in the case of the selected signal mode, the confidence coefficient is obtained whether the signal collected in the signal mode in the confidence coefficient is credible or not, and the signal mode can be configured, and the relative higher than the relative high-confidence coefficient distribution function of the relative priority of the signal mode is configured.
Example 3
Based on embodiment 1 and embodiment 2, the present invention provides an intrusion event recognition method for an object intrusion early warning system, where on the basis of the object intrusion early warning system of embodiment 2, the intrusion event recognition method includes:
s1) acquiring an acquisition signal, wherein the acquisition signal comprises vibration detection information of an object using an object intrusion early warning system and/or vibration detection information of an environment where the object is located;
s2) calculating the current position identification information of the acquired signal, inquiring and acquiring the pre-recorded environment identification information and the pre-configured signal pattern recognition function corresponding to the current position identification information;
s3) acquiring current time identification information, identifying and determining a signal mode of the acquired signal according to the current time identification information and the pre-configured signal mode identification function, identifying that the acquired signal is a signal belonging to different types of intrusion events when the signal mode of the acquired signal is in a corresponding relation with pre-recorded environment identification information, and determining that the intrusion event occurs;
the environment identification information is pre-recorded, for example, no engineering event and no traffic main road exist in an object pipeline area where a source vibration signal generating position of the acquired signal is recorded.
Specifically, the identifying and determining the signal mode of the acquired signal in step S3) includes:
s301) decomposing the acquired signal to obtain parameters of parameters in all signal modes, wherein the signal modes comprise signal subclass parameters and distribution functions related to the signal subclass parameters;
s302) comparing the collected signals with each signal mode according to a pre-configured priority order, identifying and obtaining the signal mode to be determined, and determining the signal mode of the collected signals according to the confidence degree of the pre-configured signal mode corresponding to the signal mode to be determined.
The confidence degree is used for obtaining a confidence interval of the distribution function, the signal characteristic parameters in the confidence interval can represent the signal condition that the acquired signal corresponding to the signal characteristic parameters belongs to a certain type of intrusion event, or part of the signal characteristic parameters of the acquired signal are all in the confidence interval, and then the signal condition that the acquired signal belongs to the certain type of intrusion event is judged, and the range of the part of the signal characteristic parameters is at least two signal subclasses of parameters; in some implementations, suspicious events may be introduced, and on the basis of the priority and the confidence level, the suspicious events are defined as that no intrusion event is found in a signal pattern with a high priority, that an intrusion event is found in a signal pattern with a low priority, and that a confidence interval of a distribution function of the signal pattern with the low priority is satisfied, and the suspicious events may be combined with field investigation or signal acquisition performed multiple times in a short time to verify whether intrusion exists. The present embodiment has a low false alarm rate and can identify the event type directly from the signal pattern.
Example 4
As shown in fig. 4, on the basis of the system for early warning of intrusion into an object according to embodiment 3, the present invention provides a system for early warning of intrusion into an object, including: a control system for the aforementioned method; according to the actual use requirement, the system can further comprise a database with a read-write function of the controlled system, a display device for presenting the state and event information of an object monitored by the controlled system and an input/output device for the controlled system, the database can be used for storing a large number of signal sample sets and signal pattern sets, so that the controlled system can perform real-time monitoring, compare and collect signals or write new signal patterns corresponding to new intrusion events at any time by using the signal pattern sets, the display device can facilitate monitoring personnel to visually check the state of each area of the current object, the display device can be set to be a state area with red (R), yellow (Y), green (G) and three state identification lamps, an information area, an operation area and a real-time display area for collecting signals, the red identification lamp can represent the occurrence of an intrusion event, the yellow identification lamp can represent the existence of a suspicious event, and the green identification lamp can represent the normal object to be inspected, the operation area can comprise operation prompts corresponding to the intrusion event and preliminary case information of an engineering event in a certain area, the operation area can be set with an emergency information push button and a task allocation button, even the set D1 and D2 can also be set to provide an operation prompt for the suspicious event and display of an area for the unmanned aerial vehicle to display the suspicious event, and to display the area, and to display the area can be provided with a video image of the suspicious event.
Although the embodiments of the present invention have been described in detail with reference to the accompanying drawings, the embodiments of the present invention are not limited to the details of the above embodiments, and various simple modifications can be made to the technical solutions of the embodiments of the present invention within the technical idea of the embodiments of the present invention, and the simple modifications all belong to the protection scope of the embodiments of the present invention.
It should be noted that the various features described in the foregoing embodiments may be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the embodiments of the present invention do not describe every possible combination.
Those skilled in the art will understand that all or part of the steps in the method according to the above embodiments may be implemented by a program, which is stored in a storage medium and includes several instructions to enable a single chip, a chip, or a processor (processor) to execute all or part of the steps in the method according to the embodiments of the present application. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In addition, any combination of various different implementation manners of the embodiments of the present invention is also possible, and the embodiments of the present invention should be considered as disclosed in the embodiments of the present invention as long as the combination does not depart from the spirit of the embodiments of the present invention.
Claims (12)
1. A signal pattern acquisition method for early warning of object intrusion is characterized by comprising the following steps:
s1) obtaining signals of different types of intrusion events and determining different signal feature types according to the signals, wherein the signal feature types comprise at least two types of space feature types, time feature types, intensity feature types and frequency feature types and also comprise cross feature types, and the cross feature types comprise feature types formed by at least two types of the space feature types, the time feature types, the intensity feature types and the frequency feature types;
s2) decomposing the signals according to the signal feature classes to obtain sub-signals, and forming sub-signal sample sets corresponding to different classes of intrusion events according to the corresponding relation between the sub-signals and the correlation among the sub-signals;
s3) obtaining a signal mode set corresponding to each sub-signal according to a preset mapping rule by utilizing the sub-signal sample set, wherein each signal mode is configured with a priority at a position of an object;
the step S1) of obtaining signals of different types of intrusion events comprises the following steps:
s101) signals of each intrusion event are recorded,
s102) marking identification information of each intrusion event, identifying any two intrusion events into different classes when the identification information of any two intrusion events is different, and acquiring signals of the intrusion events of different classes, wherein the identification information comprises: the system comprises intrusion object identification information, invaded object environment identification information, invaded object position identification information, relative position identification information of the intrusion object relative to the invaded object, time identification information of intrusion occurrence and intrusion behavior mode identification information.
2. The signal pattern acquisition method for early warning of intrusion of an object according to claim 1, wherein the step S1) further comprises, after determining different signal feature classes according to the signal:
and decomposing each type of signal feature class to obtain a signal subclass parameter set after all the signal feature classes are decomposed, wherein the signal feature classes comprise at least two types of space feature classes, time feature classes, intensity feature classes and frequency feature classes.
3. The signal pattern acquisition method for early warning of intrusion of an object according to claim 2, wherein the step S2) comprises:
s201) decomposing the signal according to the signal characteristic class to obtain a sub-signal and a parameter set of the sub-signal;
s202) determining parameters corresponding to each current signal subclass parameter according to the signal subclass parameter set and the parameter set, and determining the corresponding relation between the signal subclass parameters and the sub-signals to obtain a current signal subclass parameter set;
s203) determining relevant characteristics of parameters of the current signal subclass parameter set, screening signal subclass parameters according to the relevant characteristics and the relation of preset conditions, obtaining a new signal subclass parameter set after screening, and forming a sub-signal sample set corresponding to different types of intrusion events by utilizing the new signal subclass parameter set, combining the sub-signals with the corresponding relation of the signals and combining the corresponding relation of the signals and different types of intrusion events.
4. The method as claimed in claim 3, wherein in step S3), the signal pattern set includes a signal sub-class parameter corresponding to each sub-signal and a distribution function related to the signal sub-class parameter, or the signal pattern set includes a signal sub-class parameter corresponding to each sub-signal and a classifier related to the signal sub-class parameter.
5. The signal pattern acquisition method for early warning of intrusion of an object as claimed in claim 1, further comprising:
s4) when a new intrusion event with at least one identification information different from the identification information of each intrusion event recorded in the step S101) is acquired, returning to the step S1).
6. A method for constructing an object intrusion alert system having a signal pattern recognition function, wherein the signal pattern is the signal pattern in the signal pattern acquisition method for object intrusion alert according to any one of claims 1 to 5, the method comprising:
s1) marking and recording different position identification information of an object using an object intrusion early warning system and different environment identification information corresponding to the position identification information;
and S2) acquiring current time identification information and configuring a signal mode identification function according to different environment identification information of each position identification information corresponding to the current time identification information.
7. The method for constructing the object intrusion warning system with the signal pattern recognition function according to claim 6, wherein the step S2) of configuring the signal pattern recognition function includes:
s201) selecting a target signal mode set from all signal modes, and configuring the priority of the target signal mode set to be higher than the priority of the rest signal mode sets except the target signal mode set in all signal modes;
s202) configuring the confidence degree of each target signal mode in the target signal mode set, wherein the target signal mode set has a distribution function corresponding to target signals, and the target signals are signals of different types of intrusion events.
8. An intrusion event recognition method of an object intrusion alert system, wherein an intrusion event is an intrusion event in the signal pattern acquisition method for object intrusion alert according to any one of claims 1 to 5, the intrusion event recognition method comprising:
s1) acquiring a collection signal, wherein the collection signal comprises vibration detection information of an object using an object intrusion early warning system and/or vibration detection information of an environment where the object is located;
s2) calculating the current position identification information of the acquired signal, inquiring and acquiring the pre-recorded environment identification information and the pre-configured signal pattern recognition function corresponding to the current position identification information;
and S3) acquiring current time identification information, identifying and determining the signal mode of the acquired signal according to the current time identification information and the pre-configured signal mode identification function, identifying that the acquired signal is a signal belonging to different types of intrusion events when the signal mode of the acquired signal is in a corresponding relation with pre-recorded environment identification information, and determining that the intrusion event occurs.
9. The intrusion event recognition method for the object intrusion warning system according to claim 8, wherein the recognizing and determining the signal pattern of the collected signal in step S3) includes:
s301) decomposing the acquired signal to obtain parameters of parameters in all signal modes, wherein the signal modes comprise signal subclass parameters and distribution functions related to the signal subclass parameters;
s302) comparing the collected signals with each signal mode according to a pre-configured priority order, identifying and obtaining the signal mode to be determined, and determining the signal mode of the collected signals according to the confidence degree of the pre-configured signal mode corresponding to the signal mode to be determined.
10. An object intrusion warning system, comprising:
a control system for performing the method of any one of claims 1 to 9.
11. An apparatus for early warning of intrusion by an object, comprising:
at least one processor;
a memory coupled to the at least one processor;
wherein the memory stores instructions executable by the at least one processor, the at least one processor implementing the method of any one of claims 1 to 9 by executing the instructions stored by the memory.
12. A computer readable storage medium storing computer instructions which, when executed on a computer, cause the computer to perform the method of any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910499887.6A CN112071009B (en) | 2019-06-11 | 2019-06-11 | Optical fiber pipeline early warning system and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910499887.6A CN112071009B (en) | 2019-06-11 | 2019-06-11 | Optical fiber pipeline early warning system and method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112071009A CN112071009A (en) | 2020-12-11 |
| CN112071009B true CN112071009B (en) | 2023-01-03 |
Family
ID=73658267
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910499887.6A Active CN112071009B (en) | 2019-06-11 | 2019-06-11 | Optical fiber pipeline early warning system and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112071009B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114882679A (en) * | 2022-04-01 | 2022-08-09 | 深圳市燃气集团股份有限公司 | Hook machine monitoring and early warning method, device, terminal and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104751580A (en) * | 2015-04-08 | 2015-07-01 | 武汉理工光科股份有限公司 | Distributed optical fiber sensing signal mode identifying method and system |
| CN106051468A (en) * | 2016-06-08 | 2016-10-26 | 无锡亚天光电科技有限公司 | Alarm model analytical algorithm of distributive type optical fiber pipe safety early-warning system |
| CN106301575A (en) * | 2016-08-29 | 2017-01-04 | 深圳艾瑞斯通技术有限公司 | The sorting technique of a kind of fiber-optic vibration signal and device and optical fiber sensing system |
| CN107995982A (en) * | 2017-09-15 | 2018-05-04 | 达闼科技（北京）有限公司 | A kind of target identification method, device and intelligent terminal |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030030557A1 (en) * | 2001-08-08 | 2003-02-13 | Trw Inc. | Apparatus and method for detecting intrusion and non-intrusion events |
-
2019
- 2019-06-11 CN CN201910499887.6A patent/CN112071009B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104751580A (en) * | 2015-04-08 | 2015-07-01 | 武汉理工光科股份有限公司 | Distributed optical fiber sensing signal mode identifying method and system |
| CN106051468A (en) * | 2016-06-08 | 2016-10-26 | 无锡亚天光电科技有限公司 | Alarm model analytical algorithm of distributive type optical fiber pipe safety early-warning system |
| CN106301575A (en) * | 2016-08-29 | 2017-01-04 | 深圳艾瑞斯通技术有限公司 | The sorting technique of a kind of fiber-optic vibration signal and device and optical fiber sensing system |
| CN107995982A (en) * | 2017-09-15 | 2018-05-04 | 达闼科技（北京）有限公司 | A kind of target identification method, device and intelligent terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112071009A (en) | 2020-12-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11562224B2 (en) | 1D-CNN-based distributed optical fiber sensing signal feature learning and classification method | |
| CN107425906B (en) | Distributing optical fiber sensing signal processing method towards underground pipe network safety monitoring | |
| CN105471882A (en) | Behavior characteristics-based network attack detection method and device | |
| CN102098180A (en) | Network security situational awareness method | |
| CN102663032B (en) | Fiber grating fence invasion event mode recognition method | |
| CN112069912B (en) | Method for identifying construction threat event of optical cable channel based on phi-OTDR | |
| CN111537056A (en) | Pipeline along-line third-party construction dynamic early warning method based on SVM and time-frequency domain characteristics | |
| CN106027529A (en) | Intrusion detection system and method based on traceability information | |
| JP7471470B2 (en) | Anomaly detection based on statistical image processing to prevent cable cuts | |
| CN112071009B (en) | Optical fiber pipeline early warning system and method thereof | |
| CN116838955A (en) | Two-stage oil and gas pipeline line interference identification method | |
| CN112032575A (en) | Pipeline safety monitoring method and system based on weak grating and storage medium | |
| Luu et al. | Flash flood and landslide susceptibility analysis for a mountainous roadway in Vietnam using spatial modeling | |
| CN109120336B (en) | False alarm prevention and false alarm prevention method based on phase sensitive optical time domain reflection sensor | |
| CN111951505B (en) | Fence vibration intrusion positioning and mode identification method based on distributed optical fiber system | |
| CN111539394B (en) | Pipeline along-line third-party construction early warning method based on time domain characteristics and space-time information | |
| CN109614926B (en) | Distributed optical fiber sensing signal mode identification method and system based on prediction model | |
| CN115187048B (en) | Rail line multi-domain limit foreign matter intrusion event condition identification method and system | |
| US20220230539A1 (en) | Road traffic extraction for unknown anomaly detection using distributed fiber optic sensing | |
| Aslangul | Detecting Tunnels for Border Security based on Fiber Optical Distributed Acoustic Sensor Data using DBSCAN. | |
| Kozińska et al. | A review of methods in the field of detecting illegal open-pit mining activities | |
| WO2022034750A1 (en) | Unconfirmed sound extraction device, unconfirmed sound extraction system, unconfirmed sound extraction method, and recording medium | |
| CN114354744A (en) | Power cable external damage prevention monitoring method, system and device | |
| Xiao et al. | Intrusion detection for high-speed railway system: a faster R-CNN approach | |
| CN112633163A (en) | Detection method for realizing illegal operation vehicle detection based on machine learning algorithm |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20221214 Address after: 100728 No. 22 North Main Street, Chaoyang District, Beijing, Chaoyangmen Applicant after: CHINA PETROLEUM & CHEMICAL Corp. Applicant after: Sinopec Safety Engineering Research Institute Co.,Ltd. Address before: 100728 No. 22 North Main Street, Chaoyang District, Beijing, Chaoyangmen Applicant before: CHINA PETROLEUM & CHEMICAL Corp. Applicant before: SINOPEC Research Institute OF SAFETY ENGINEERING |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |