CN109357167A - A kind of gas pipeline device for detecting leak point and detection method - Google Patents

A kind of gas pipeline device for detecting leak point and detection method Download PDF

Info

Publication number
CN109357167A
CN109357167A CN201811276735.1A CN201811276735A CN109357167A CN 109357167 A CN109357167 A CN 109357167A CN 201811276735 A CN201811276735 A CN 201811276735A CN 109357167 A CN109357167 A CN 109357167A
Authority
CN
China
Prior art keywords
leakage
gas pipeline
module
gas
data
Prior art date
Application number
CN201811276735.1A
Other languages
Chinese (zh)
Other versions
CN109357167B (en
Inventor
贺玉珍
曲晓威
宋延卫
李英韬
杨玉东
赵新峰
周琼
乔树远
Original Assignee
长春市万易科技有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 长春市万易科技有限公司 filed Critical 长春市万易科技有限公司
Priority to CN201811276735.1A priority Critical patent/CN109357167B/en
Publication of CN109357167A publication Critical patent/CN109357167A/en
Application granted granted Critical
Publication of CN109357167B publication Critical patent/CN109357167B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • FMECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
    • F17STORING OR DISTRIBUTING GASES OR LIQUIDS
    • F17DPIPE-LINE SYSTEMS; PIPE-LINES
    • F17D5/00Protection or supervision of installations
    • F17D5/02Preventing, monitoring, or locating loss

Abstract

A kind of gas pipeline device for detecting leak point and detection method, are related to gas pipeline detection field, gas pipeline device for detecting leak point of the invention includes sensor module, acquires flow, the pressure, infrasound data of gas pipeline;The data transmission module being connected with sensor module;The fixation IP route being connected with data transmission module;The server being connected with fixed IP route, flow, pressure, infrasound data are uploaded in server by data transmission module and fixed IP route;The fuel gas pipeline leakage alarm management system of installation in the server receives flow, pressure, infrasound data and is calculated it and analyzed, combustion gas operating status is shown and is alarmed.The management computer being connected with server, for being managed to fuel gas pipeline leakage alarm management system.The invention proposes the combustion gas characteristic spectrums of different sections of highway, reduce the influence factors such as environment, reduce the rate of false alarm of gas alarm, and detection accuracy and alarm are high-efficient.

Description

A kind of gas pipeline device for detecting leak point and detection method
Technical field
The present invention relates to Detection Technique of Gas Pipeline fields, and in particular to a kind of gas pipeline device for detecting leak point and detection Method.
Background technique
Natural gas has many advantages, such as that heat is high, energy efficiency is high, is considered as mesh as the energy most clean in fossil fuel The emerging high efficient energy sources of preceding most future.Pipeline transmission because have cost is low, security closed, freight volume greatly, quality can protect It demonstrate,proves, be easy to control the advantages such as management, it has also become the preferred manner of natural gas transportation.But natural gas has inflammable, explosive spy Point, once pipeline leakage accident occurs, it is possible to a series of disasters such as explosion, fire, poisoning, environmental pollution are caused, if Occur in resident accumulation regions, it will cause even more serious harm.The reason of natural gas line leaks mainly includes external force Damage, material installation corrosion and ageing, violation operation and natural calamity etc..
Natural gas is leaked during production, processing, transport and storage etc. and is difficult to avoid that, the long pipeline of tens of thousands of kilometers On have thousands of a source of leaks, pipeline maintenance is an arduous process.The detection method of gas pipeline leakage has at present Many kinds, acoustic wave detection, suction wave detection method, pressure gradient method etc. are at present using more several method.
Acoustic wave detection technical principle:
When pipeline generates leak source, gas erupts the sound wave source that formed by crackle or corruption aperture, and then passes through and pipeline Reaction force, sound wave source releases energy to outside pipeline constitutes transmitted wave.And it is concentrated mainly on low-frequency range, with specific sensor (sonic sensor, microphone etc.) can be captured.Since the moment for only having line pressure to generate fluctuation just has sound wave Transmitting, therefore only need to be that is, discernable whether to there is potential leakage to threaten by the fluctuation of detecting sound wave curve.Once natural gas tube Road leaks, and the high-pressure natural gas gas in pipeline will discharge suddenly, to generate high-pressure shocking wave at leak hole, in turn Generate sound wave, that is to say, that the sound wave of leakage leaks excitation by gas and generates, and is continuous acoustic signals.
The formula of impulsive model is as follows:
Wherein, t is time value, and unit is the second;t0It is time constant, it corresponds to the e that acoustic attenuation amplitude is initial magnitude-1 At the time of;t1At the time of leakage for pipeline, the unit second;p0For t1Sound pressure amplitude corresponding to moment, unit pa;When p is t Sound pressure amplitude corresponding to quarter value, unit pa.
Fourier transformation is applied to formula (1), obtained frequency spectrum function is as follows:
Wherein, p0For t1Sound pressure amplitude corresponding to moment, unit pa;P (w) is single sine wave after being fourier transformed, Unit pa;t0It is time constant, it corresponds to the e that acoustic attenuation amplitude is initial magnitude-1At the time of.
After leakage is formed, persistently there can be gas to discharge through leakage hole, form the micro-vibration of pipeline, thus can generate lasting Stablize sound source, it is a kind of continuous acoustic signals.This is also one of the reason of sonic method is superior to other methods.At leakage The frequency range of sound wave is very wide, and wherein the overwhelming majority is in 175KHz~750KHz.In the communication process of sound wave, in signal Radio-frequency component can gradually decay in pipeline medium, and low-frequency component can long-distance communications, be based on this feature, acoustic measurement Method can apply to gas pipeline leakage detection.
Gas leakage point location formula are as follows:
Wherein, v is infrasonic sound velocity of wave propagation, unit meter per second;U is speed of the combustion gas in pipeline, unit meter per second;△t For the time difference of two infrasonic sensors, unit second;L is distance between two sensors, unit rice;X is leakage point position, single Position rice.
Pressure distribution situation formula in pipeline are as follows:
Wherein, PQ、PZFor the pressure of the monitoring point Q and the monitoring point Z, unit pa;L is the distance of the monitoring point Q and the monitoring point Z, single Position rice;X is the distance of the monitoring point leakage point distance Q, unit rice;PxFor X point pressure, unit pa.
The environment as locating for urban underground gas pipeline is complicated, and live disturbed condition is numerous, different pipeline material, no With diameter, pressure situations such as it is generated alarm be all different, if depend merely on given threshold carry out leakage judgement will be significantly Increase leakage False Rate and misdetection rate.
Accident happens occasionally caused by leaking in recent years because of urban pipe network, and existing Detection Techniques are difficult to meet growing The demand of Developing Natural Gas Industry, urgent need study a kind of scheme that gas leakage quickly detects, and the accident of reduction occur have weight Big social effect.
Summary of the invention
To solve the problems, such as gas leakage, the present invention provides a kind of gas pipeline device for detecting leak point and detection method.
Used technical solution is as follows in order to solve the technical problem by the present invention:
A kind of gas pipeline device for detecting leak point of the invention, comprising:
Sensor module acquires flow, the pressure, infrasound data of gas pipeline;
The data transmission module being connected with sensor module;
The fixation IP route being connected with data transmission module;
The server being connected with fixed IP route, flow, pressure, infrasound data pass through data transmission module and fixed IP Route is uploaded in server;
Installation fuel gas pipeline leakage alarm management system in the server receives flow, pressure, infrasound data and right It is calculated and is analyzed, and combustion gas operating status is shown and is alarmed.
The management computer being connected with server, for being managed to fuel gas pipeline leakage alarm management system.
Further, the sensor module includes the flow sensor being connected with data transmission module, multiple pressure Sensor and multiple infrasonic sensors;The flow sensor is mounted at the entrance of gas pipeline, in gas pipeline 1 pressure sensor and 1 infrasonic sensor are installed at interval of 1Km, 1 infrasonic sound is installed at interval of 1Km outside gas pipeline Wave sensor, the infrasonic sensor being mounted in gas pipeline and the infrasonic sensor one being mounted on outside gas pipeline are a pair of It should be arranged, the vertical range of the infrasonic sensor and gas pipeline outer wall that are mounted on outside gas pipeline is 10-15cm.
Further, the data transmission module includes shell and installation input module inside the shell, multichannel AD conversion Module, memory module and transmission module;The input module is connected with multichannel AD conversion module, multichannel AD conversion module and storage Module is connected, and memory module is connected with transmission module, and transmission module is connected with fixed IP route;The input module is that four roads are defeated Enter module;Transmission module is NB-IoT mode;
It is converted into digital signal through multichannel AD conversion module by the analog signal that input module receiving sensor module transmits, It is saved in memory module, combustion was passed to by transmission module and fixed IP route in the detection data after the completion of storing every 30 seconds Feed channel leakage alarm management system.
Further, the fuel gas pipeline leakage alarm management system includes: the data receiver being connected with fixed IP route Module, the combustion gas operating status display module being connected with data reception module and pattern recognition module and pattern recognition module phase Alarm module even and gas leakage intelligence learning module, the combustion gas operating status display module and pattern recognition module phase Even;The data reception module receives the detection passed over by sensor module by data transmission module and fixed IP route Data, while will test data by way of message transmission and passing to combustion gas operating status display module and pattern-recognition mould Block;The detection data of gas pipeline is shown that the pattern recognition module is to inspection by the combustion gas operating status display module Measured data is calculated, if detection data is more than alarming value, the side that alarm signal is passed through message transmission by pattern recognition module Formula is sent to alarm module and combustion gas operating status display module, and alarm module is alarmed, combustion gas operating status display module Warning message is shown, the gas leakage intelligence learning module completes different combustion gas by learning various detection datas The alarm model of duct section is arranged, and alarm model is cured in pattern recognition module.
Further, CLIENT PROGRAM management system is installed, the management computer passes through net in the management computer Network browser and CLIENT PROGRAM management system access fuel gas pipeline leakage alarm management system use it after authentication In data reception module, combustion gas operating status display module, pattern recognition module, alarm module, gas leakage intelligence learning Module completes fuel gas pipeline leakage alarming and managing function.
A kind of detection method of gas pipeline device for detecting leak point of the invention, comprising the following steps:
Step 1: dynamic fingerprint acquires
Flow, the pressure, infrasound data of gas pipeline are acquired by sensor module, and by NB-IoT network and admittedly Determine IP route and is uploaded to fuel gas pipeline leakage alarm management system;
Step 2: leakage analog data acquisition
In gas pipeline, 100 gas leakage points are simulated, each gas leakage point simulated leakage diameter range is 1-20 Centimetre, leakage is greater than 10 minutes every time;Preceding 1 hour of acquisition leakage is to 1 hour flow of gas pipeline after leakage, pressure, secondary Sonic data, and fuel gas pipeline leakage alarm management system is uploaded to by NB-IoT network and fixed IP route;
Step 3: feature monochrome audible spectrum identifies
S301: single-frequency is split
Infrasound data are subjected to unifrequency fractionation, and single audio frequency amplitude is identified;
S302: frequency spectrum association analysis
Each infrasonic sound wave frequency rate in gas pipeline and outside gas pipeline is calculated separately to occur in non-leakage point and leakage point Probability;
S303: spectral noise is rejected
To in gas pipeline and before the external leakage of gas pipeline, in leakage, occur each time of three periods after leakage Frequency of sound wave compares and analyzes, and rejects the infrasonic sound wave frequency rate outside gas pipeline according to the infrasonic sound wave frequency rate in gas pipeline;
S304: frequency spectrum classification
Infrasonic sound wave frequency rate in gas pipeline is arranged using small wave converting method, calculates the infrasonic sound wave frequency rate in gas pipeline Before leakage, leakage in, leakage after three periods occur probability;
S305: temporal is rejected
Reject gas pipeline in infrasonic sound wave frequency rate before leakage, leakage in, leakage after three periods occur repetition Frequency spectrum;
S306: characteristic frequency spectrum extracts
It rejects after repeating frequency spectrum, if the residual spectrum quantity of period is less than 2 in leakage, infrasonic sound intensity of wave will be filtered It is adjusted to 5%, i.e. the frequency spectrum below of average audio intensity 5% weeds out first, step S303 to S305 is repeated, at this point, if letting out Less than 2 then 5%, i.e. average audio intensity is turned up in filtering infrasonic sound intensity of wave by the residual spectrum quantity of period again in leakage 10% frequency spectrum below weeds out first, repeats step S303 to S305, until the residual spectrum number in leakage in the period Until amount is greater than 2, these residual spectrums are defined as gas leakage feature monochrome frequency spectrum;
Step 4: amplitude characteristics identify
Infrasound strength change laws corresponding to the gas leakage feature monochrome frequency spectrum obtained according to step 3, are made with this To identify that the verifying map of gas leakage identifies amplitude characteristics;
S401: the sound intensity is cut out
The infrasonic sound intensity of wave P that will be acquired in gas pipeline0It is compared with the infrasonic sound intensity of wave acquired outside gas pipeline, to combustion Infrasonic sound intensity of wave P in feed channel0Make a proportional cutting, forms P1, the value range of a is between 1-10;
S402: singular point is found
By P1Second dervative is sought according to the time, and calculates singular point;
S403: feature wave amplitude extracts
Using singular point as origin, by P1First 30 seconds, characteristic wave of 30 seconds latter, the rear strong curve of 60 seconds waves as gas leakage Width spectral line;
Step 5: gas pipe pressure perturbation wave identifies
Each test point in gas pipeline follows pressure balance formula, makees the change curve of a pressure and time, meter Pressure time change curve is calculated to the second dervative of time, then calculates extreme value, the time point of extreme value be leak when Between point be the pressure disturbance time, take this as a foundation to determine whether gas leakage event and nearest leakage point occurs;
Step 6: leakage point position calculates
S601: the position of leakage point X is calculated using infrasound method;
S602: pressure disturbance method calculates leakage point
The time point of most similar two pressure sensor senses, root are calculated according to the pressure disturbance time that step 5 obtains Distance and time difference according to the two time points calculate the distance X that point occurs for its gas leakage1, calculation formula is as follows:
Wherein, PQ、PZThe respectively pressure of the monitoring point Q and the monitoring point Z, unit pa;L be the monitoring point Q and the monitoring point Z away from From unit rice;X1At a distance from the monitoring point leakage point X and Q, unit rice;
S603: position correction parameter calculates
Resulting X and X will be calculated1With actual location of leak X0It is verified and is rectified a deviation, A value is calculated using formula (6) With B value:
X0=A × X+B × X1 (6)
Wherein, A, B are position correction parameter;
Step 7: gas pipeline leak source detection parameters solidify
Step 3 to the resulting various parameters of step 6 is saved in pattern recognition module.
Further, further include step 8 after step 7: gas pipeline operating status is shown
S801: gas pipeline operating status simulation
Combustion gas operating status display module shows actual gas pipeline operating status and expection using the twin method of number Virtual gas pipeline operating status, core drive engine be prognostic and health management engine PHM, i.e., digital twin number It is analogue data according to the data calculated in the PHM of driving, passes through pressure sensor, flow sensor, infrasonic sensor The actual motion state of real-time perception gas pipeline;The expected virtual gas pipeline operation of combustion gas operating status display module simulation State, virtual gas pipeline data transmissions are enough same with the transmission of practical gas pipeline data under the driving of the twin data of number Step operation;
S802: gas leakage closes valve processing
In gas pipeline any place, it is assumed that incident of leakage occur, in the PHM of the twin data-driven of number, use is outstanding Color shows the user for needing the valve closed and influence;
S803: parameter adjustment
It is pressure data in extraction gas pipeline, data on flows, secondary from gas leakage to closing after valve in 2-4 hour Sonic data, and compared with the analogue data calculated in the PHM of the twin data-driven of number, and adjust PHM ginseng Number is allowed to consistent with the practical gas pipeline operating status of detection.
Further, further comprising the steps of between step 3 and step 4: feature composite audio frequency spectrum discerning;Specifically It operates as follows:
Frequency spectrum after step S303 spectral noise is rejected is as data source, respectively 10 seconds before gas leakage, after leakage Assemble core for data within 10 seconds and cluster a possibility that based on core is calculated using fuzzy core aggregation algorithms;It is clustered by possibility Afterwards, each sample correspondence obtains C degree of membership coefficient, and representative sample is under the jurisdiction of the degree of every class, in [0,1] interior value, sample Be divided into two major classes: first kind sample is close to certain a kind of center and far from other classifications, is non-supporting vector;Second class sample This is in different classes of boundary position, can be under the jurisdiction of multiple classes, is supporting vector;By first kind sample be directly divided into most close to Class without consider and other classes relationship, without being added in the training set of support vector machines;Second class sample is used to prop up The training for holding vector machine obtains gas leakage character symbol sum of fundamental frequencies spectrum after meeting given threshold.
The beneficial effects of the present invention are:
The present invention uses gas leakage intelligence learning algorithm, piecewise acquisition is carried out to actual gas pipeline, to various days The characteristic wave spectrum of right gas leakage carries out speced learning, forms natural gas operational network and leaks map, to solve to lean on given threshold Problem is failed to judge and judged by accident to bring, improves alarm accuracy.
The invention proposes the combustion gas characteristic spectrums of different sections of highway, reduce the influence factors such as environment, reduce gas alarm Rate of false alarm improves detection accuracy, improves alarm efficiency and gas safe management level, promotes maintenance social public security ability Level reduces accident and the frequency occurs, casualties and property loss caused by reducing after the accident, the safety management to industry Demonstration effect is formed, enterprise core competence is enhanced, improves the access barrier of industry, promotes Business Economic Benefit.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of gas pipeline device for detecting leak point of the invention.
Fig. 2 is the structural schematic diagram of data transmission module in gas pipeline device for detecting leak point of the invention.
Fig. 3 is the structural representation of fuel gas pipeline leakage alarm management system in gas pipeline device for detecting leak point of the invention Figure.
Fig. 4 is the flow chart of gas pipeline leak source detection method of the invention.
Specific embodiment
Invention is further described in detail with reference to the accompanying drawings and examples.
As shown in Figure 1, a kind of gas pipeline device for detecting leak point of the invention by sensor module, data transmission module, Fixed IP route, server, management computer and fuel gas pipeline leakage alarm management system composition.
Sensor module is connected with data transmission module, and data transmission module is connected with fixed IP route, fixed IP route It is connected with server, management computer is connected with server by network, and fuel gas pipeline leakage alarm management system is mounted on clothes It is engaged in device, manages in computer and CLIENT PROGRAM management system is installed.
The signal of sensor module acquisition is wired to data transmission module, and data transmission module passes through NB-IoT The signal of acquisition is passed to the fuel gas pipeline leakage alarm management system of installation in the server, pipe by network, fixed IP route Reason computer is managed fuel gas pipeline leakage alarm management system by CLIENT PROGRAM management system.
Sensor module includes: 1 flow sensor L, multiple pressure sensor Y, multiple infrasonic sensor C.Flow Sensor L, pressure sensor Y, infrasonic sensor C connect with data transmission module T-phase.Flow sensor L is mounted on combustion gas At the entrance of pipeline.1 pressure sensor Y and 1 infrasonic sensor C is installed at interval of 1Km in gas pipeline.Fuel gas conduit 1 infrasonic sensor C is installed at interval of 1Km outside road, the infrasonic sensor C being mounted in gas pipeline and is mounted on combustion gas Infrasonic sensor C outside pipeline is arranged in a one-to-one correspondence, that is, the infrasonic sensor C being mounted in gas pipeline and is mounted on combustion Line segment is vertical with gas pipeline after infrasonic sensor C line outside feed channel.The infrasound sensing being mounted on outside gas pipeline The vertical range of device C and gas pipeline outer wall is 10-15cm.
As shown in Fig. 2, data transmission module is by shell, input module, multichannel AD conversion module, memory module, transmission mould Block composition.Shell is that the gas pipeline of diameter 15cm buries the plastic part of mark, is placed on outside gas pipeline graveyard, has 50 meters of waterproof abilities.Input module, multichannel AD conversion module, memory module, transmission module are installed in interior of shell.Input mould Block is connected with multichannel AD conversion module, and multichannel AD conversion module is connected with memory module, and memory module is connected with transmission module, passes Defeated module is connected with fixed IP route.Input module is four road input modules;Transmission module is NB-IoT mode;By input module The analog signal of receiving sensor module transmission is converted into digital signal through multichannel AD conversion module, is saved in memory module. Memory module sample frequency is 100HZ, and data storage file format is time, sensor number, detection numerical value.After the completion of storage Fuel gas pipeline leakage alarm management system was passed to by transmission module and fixed IP route in detection data every 30 seconds.
As shown in figure 3, fuel gas pipeline leakage alarm management system shows mould by data reception module, combustion gas operating status Block, pattern recognition module, alarm module, gas leakage intelligence learning module composition.Fixed IP route and data reception module phase Even, data reception module is connected with combustion gas operating status display module and pattern recognition module, combustion gas operating status display module It is connected with pattern recognition module, pattern recognition module is connected with alarm module and gas leakage intelligence learning module.Data receiver Module receives the detection data passed over by sensor module by data transmission module and fixed IP route, while by disappearing The mode of breath transmitting will test data and pass to combustion gas operating status display module and pattern recognition module.Combustion gas operating status exhibition Show that module shows the detection data of gas pipeline on combustion gas operation figure, pattern recognition module uses pre-set meter Calculation method calculates detection data, if detection data is more than alarming value, pattern recognition module passes through alarm signal The mode of message transmission is sent to alarm module and combustion gas operating status display module, and alarm module is alarmed, combustion gas operation State display module schemes upper display alarm information in combustion gas operation.Gas leakage intelligence learning module is by learning various testing numbers According to the alarm model for completing different gas pipeline sections is arranged, and alarm module is cured in pattern recognition module.
It manages computer and fuel gas pipeline leakage alarming and managing system is accessed by web browser and CLIENT PROGRAM management system System, after authentication, uses data reception module therein, combustion gas operating status display module, pattern recognition module, report Alert module, gas leakage intelligence learning module complete every fuel gas pipeline leakage alarming and managing function.
As shown in figure 4, a kind of gas pipeline leak source detection method of the invention, steps are as follows for concrete implementation:
Step 1: dynamic fingerprint acquires
Under gas pipeline normal operation, sensor module acquire 7*24 hours the flow of gas pipeline, pressure, The flow of acquisition, pressure, infrasound data were passed through NB-IoT network, fixed IP route at interval of 30 seconds by infrasound data Pass to fuel gas pipeline leakage alarm management system.
Step 2: leakage analog data acquisition
In gas pipeline, 100 gas leakage points are simulated, each gas leakage point simulated leakage diameter range is 1-20 Centimetre, leakage is greater than 10 minutes every time;Preceding 1 hour of acquisition leakage is to 1 hour flow of gas pipeline after leakage, pressure, secondary Sonic data passed through the flow of acquisition, pressure, infrasound data on NB-IoT network, fixed IP route at interval of 30 seconds It passes in server.
Step 3: feature monochrome audible spectrum identifies
S301: single-frequency is split
By the infrasound data in server, (infrasound data are in gas pipeline, the outer complicated audio letter of gas pipeline Number) 1 minute or 2 minutes is pressed as time interval, unifrequency fractionation is carried out, and be identified to single audio frequency amplitude.
S302: frequency spectrum association analysis
Each infrasonic sound wave frequency rate in gas pipeline and outside gas pipeline is calculated separately in non-leakage point (normal point) and is let out The probability that leak source occurs.
S303: spectral noise is rejected
To in gas pipeline and before the external leakage of gas pipeline, in leakage, occur each time of three periods after leakage Frequency of sound wave compares and analyzes, and rejects the infrasonic sound wave frequency rate outside gas pipeline according to the infrasonic sound wave frequency rate in gas pipeline, with This rejects influence of the environmental audio to gas pipeline audio.
S304: frequency spectrum classification
Using the method for wavelet transformation, the infrasonic sound wave frequency rate in gas pipeline is arranged, calculates the infrasound in gas pipeline Frequency before leakage, leakage in, leakage after three periods occur probability.
S305: temporal is rejected
Reject gas pipeline in infrasonic sound wave frequency rate before leakage, leakage in, leakage after three periods occur repetition Frequency spectrum.
S306: characteristic frequency spectrum extracts
It rejects after repeating frequency spectrum, if the residual spectrum quantity of period is less than 2 in leakage, infrasonic sound intensity of wave will be filtered It is adjusted to 5%, i.e. the frequency spectrum below of average audio intensity 5% weeds out first, S303 to S305 is repeated, at this point, if in leakage Filtering infrasonic sound intensity of wave is then turned up 5% less than 2 by the residual spectrum quantity of period again, i.e., average audio intensity 10% with Under frequency spectrum weed out first, S303 to S305 is repeated, until the residual spectrum quantity in leakage in the period is greater than 2 is Only, these residual spectrums are then defined as gas leakage feature monochrome frequency spectrum.
Step 4: feature composite audio frequency spectrum discerning
If gas leakage feature monochrome frequency spectrum can not be found after step 3 or gas leakage feature monochrome frequency spectrum exists Effect is undesirable in Late Stage Verification, then carries out feature composite audio frequency spectrum discerning using this step.
Frequency spectrum after step S303 spectral noise is rejected is as data source, using fuzzy core aggregation algorithms, by low-dimensional sky Between in be not easy the data clustered and be mapped to high-dimensional feature space come the difference between extension mode class, then in height using kernel function It is clustered in dimension space, increases the optimization to sample characteristics, by will linearly can not in observation space using kernel function Point sample Nonlinear Mapping to higher-dimension feature space and become linear separability, such sample characteristics are differentiated well, are mentioned After taking and amplifying, more accurate cluster may be implemented, to reach better Clustering Effect.Using the thought of kernel learning method, Respectively with 10 seconds before gas leakage, leakage after 10 seconds for data assemble core, can obtain being based on core in conjunction with fuzzy core aggregation algorithms A possibility that cluster.
After cluster, each sample can obtain C (C is cluster number) degree of membership coefficients, and representative sample is under the jurisdiction of often The degree of class, in [0,1] interior value, sample can be divided into two major classes: a kind of situation is that sample is under the jurisdiction of certain a kind of degree very Greatly, to the subjection degree very little of other classes, this kind of sample is usually quite close to certain a kind of center and far from other classifications, and one As will not be supporting vector;Another situation is that sample is not much different to all kinds of distances, in different classes of boundary position, Multiple classes may be under the jurisdiction of, this kind of sample is possible to as supporting vector.For the former, it can be directly divided into most close to class And do not have to consider the relationship with other classes, without being added in the training set of support vector machines;For the latter, need with support to Amount machine is trained, and after meeting given threshold, obtains gas leakage character symbol sum of fundamental frequencies spectrum.
Step 5: amplitude characteristics identify
The compound frequency of gas leakage feature that the gas leakage feature monochrome frequency spectrum or step 4 obtained according to step 3 obtains The corresponding infrasound strength change laws of spectrum, in this, as the verifying map of identification gas leakage.
S501: the sound intensity is cut out
The infrasonic sound intensity of wave P that will be acquired in gas pipeline first0It is compared with the infrasonic sound intensity of wave acquired outside gas pipeline, To the infrasonic sound intensity of wave P in gas pipeline0Make a proportional cutting, forms P1, the value range of a, can be according to reality between 1-10 Border detection on-the-ground test determines the specific value of a.
S502: singular point is found
By the infrasonic sound intensity of wave P in gas pipeline0Make the P formed after a proportional cutting1Second dervative is sought according to the time, and is counted Calculate singular point.
S503: feature wave amplitude extracts
Using singular point as origin, by P1First 30 seconds, characteristic wave of 30 seconds latter, the rear strong curve of 60 seconds waves as gas leakage Width spectral line.
Step 6: gas pipe pressure perturbation wave identifies
Since under gas pipeline normal operation, gas pipe pressure is a dynamic platform, each test point is abided by Pressure balance formula is followed, the change curve of a pressure and time is made.In case of incident of leakage, it is dynamic necessarily to break this pressure State equation of equilibrium calculates pressure time curve to the second dervative of time, then calculates its extreme value, and the time point of extreme value is can Can the time point i.e. pressure disturbance time that leak, take this as a foundation to determine whether occurs gas leakage event with recently Leakage point.
Step 7: leakage point position calculates
S701: infrasound method calculates leakage point
Leakage point position is calculated according to infrasound data, that is, audio signal time difference of infrasonic sensor acquisition, specifically The position of leakage point X is calculated using gas leakage point location formula (3).
S702: pressure disturbance method calculates leakage point
When due to gas leakage, the pressure of leakage point is close to 0.Thus, when according to the pressure disturbance of step 6 calculating Between, the time point of most similar two pressure sensor senses is calculated, according to the distance at the two time points and time difference, is calculated The distance X of point occurs for its gas leakage1.Specific formula is as follows:
Wherein, PQ、PZThe respectively pressure of the monitoring point Q and the monitoring point Z, unit pa;L be the monitoring point Q and the monitoring point Z away from From unit rice;X1At a distance from the monitoring point leakage point X and Q, unit rice.
S703: position correction parameter calculates
Resulting X and X will be calculated1With actual location of leak X0It is verified and is rectified a deviation, A value is calculated using formula (6) With B value:
X0=A × X+B × X1 (6)
Wherein, A, B are position correction parameter i.e. correction constant, and the correction constant of different zones is different.
Step 8: gas pipeline leak source detection parameters solidify
Step 3 to the resulting various parameters of step 7 is saved in pattern recognition module.
Step 9: gas pipeline operating status is shown
S901: gas pipeline operating status simulation
Combustion gas operating status display module shows actual gas pipeline operating status and expection using the twin method of number Virtual gas pipeline operating status, core drive engine be prognostic and health management engine (PHM), i.e., it is digital twin The data calculated in the PHM of data-driven are analogue data.Pass through pressure sensor, flow sensor, infrasound sensing The actual motion state of device real-time perception gas pipeline;The expected virtual gas pipeline fortune of combustion gas operating status display module simulation Row state, the transmission of virtual gas pipeline data can be transmitted with practical gas pipeline data under the driving of the twin data of number Synchronous operation.
S902: gas leakage closes valve processing
In system operation, a gas leakage situation is at least simulated every year.In gas pipeline any place, it is assumed that let out Leakage event shows the user for needing the valve closed and influence with color outstanding in the PHM of the twin data-driven of number.
S903: parameter adjustment
It is pressure data in extraction gas pipeline, data on flows, secondary from gas leakage to closing after valve in 2-4 hour Sonic data, and compared with the analogue data calculated in the PHM of the twin data-driven of number, and adjust PHM ginseng Number is allowed to consistent with the practical gas pipeline operating status of detection.
Although the embodiments of the present invention have been disclosed as above, but its is not only in the description and the implementation listed With.It can be adapted for various suitable the field of the invention completely.It for those skilled in the art, can be easily real Now other modification.Therefore without departing from the general concept defined in the claims and the equivalent scope, the present invention is not limited to Specific details and legend shown and described herein.

Claims (8)

1. a kind of gas pipeline device for detecting leak point characterized by comprising
Sensor module acquires flow, the pressure, infrasound data of gas pipeline;
The data transmission module being connected with sensor module;
The fixation IP route being connected with data transmission module;
The server being connected with fixed IP route, flow, pressure, infrasound data pass through data transmission module and fixed IP route It is uploaded in server;
Installation fuel gas pipeline leakage alarm management system in the server, receive flow, pressure, infrasound data and to its into Row calculates and analysis, and combustion gas operating status is shown and is alarmed.
The management computer being connected with server, for being managed to fuel gas pipeline leakage alarm management system.
2. a kind of gas pipeline device for detecting leak point according to claim 1, which is characterized in that the sensor module packet Include the flow sensor being connected with data transmission module, multiple pressure sensors and multiple infrasonic sensors;The flow Sensor is mounted at the entrance of gas pipeline, installs 1 pressure sensor and 1 time at interval of 1Km in gas pipeline Sonic sensor installs 1 infrasonic sensor, the infrasound being mounted in gas pipeline at interval of 1Km outside gas pipeline Sensor is arranged in a one-to-one correspondence with the infrasonic sensor being mounted on outside gas pipeline, and the infrasound being mounted on outside gas pipeline passes The vertical range of sensor and gas pipeline outer wall is 10-15cm.
3. a kind of gas pipeline device for detecting leak point according to claim 1, which is characterized in that the data transmission module Input module, multichannel AD conversion module, memory module and transmission module including shell and installation inside the shell;The input mould Block is connected with multichannel AD conversion module, and multichannel AD conversion module is connected with memory module, and memory module is connected with transmission module, passes Defeated module is connected with fixed IP route;The input module is four road input modules;Transmission module is NB-IoT mode;
It is converted into digital signal through multichannel AD conversion module by the analog signal that input module receiving sensor module transmits, is saved Into memory module, fuel gas conduit was passed to by transmission module and fixed IP route in detection data every 30 seconds after the completion of storing Road leakage alarm management system.
4. a kind of gas pipeline device for detecting leak point according to claim 1, which is characterized in that the fuel gas pipeline leakage Alarm management system includes: the data reception module being connected with fixed IP route, the combustion gas operation being connected with data reception module State display module and pattern recognition module, the alarm module and gas leakage intelligence learning mould being connected with pattern recognition module Block, the combustion gas operating status display module are connected with pattern recognition module;The data reception module is received by sensor die The detection data that block is passed over by data transmission module and fixed IP route, while will test by way of message transmission Data pass to combustion gas operating status display module and pattern recognition module;The combustion gas operating status display module is by fuel gas conduit The detection data in road is shown that the pattern recognition module calculates detection data, if detection data is more than alarming value, Then alarm signal is sent to alarm module by pattern recognition module by way of message transmission and combustion gas operating status shows mould Block, alarm module are alarmed, and combustion gas operating status display module shows warning message, and the gas leakage is intelligently learned Module is practised by learning various detection datas, completes the alarm model setting of different gas pipeline sections, and alarm model is solidified Into pattern recognition module.
5. a kind of gas pipeline device for detecting leak point according to claim 1, which is characterized in that in the management computer CLIENT PROGRAM management system is installed, the management computer accesses combustion gas by web browser and CLIENT PROGRAM management system Pipe leakage alarm management system shows mould using data reception module therein, combustion gas operating status after authentication Block, pattern recognition module, alarm module, gas leakage intelligence learning module complete fuel gas pipeline leakage alarming and managing function.
6. a kind of detection method of gas pipeline device for detecting leak point as described in any one of claim 1 to 5, feature It is, comprising the following steps:
Step 1: dynamic fingerprint acquires
Flow, the pressure, infrasound data of gas pipeline are acquired by sensor module, and pass through NB-IoT network and fixed IP Route is uploaded to fuel gas pipeline leakage alarm management system;
Step 2: leakage analog data acquisition
In gas pipeline, 100 gas leakage points are simulated, each gas leakage point simulated leakage diameter range is 1-20 lis Rice, leakage is greater than 10 minutes every time;Flow, pressure, the infrasonic sound of first 1 hour 1 hour gas pipeline to after leaking of acquisition leakage Wave number evidence, and fuel gas pipeline leakage alarm management system is uploaded to by NB-IoT network and fixed IP route;
Step 3: feature monochrome audible spectrum identifies
S301: single-frequency is split
Infrasound data are subjected to unifrequency fractionation, and single audio frequency amplitude is identified;
S302: frequency spectrum association analysis
Calculate separately each infrasonic sound wave frequency rate in gas pipeline and outside gas pipeline occur in non-leakage point and leakage point it is general Rate;
S303: spectral noise is rejected
To in gas pipeline and before the external leakage of gas pipeline, in leakage, each infrasound that three periods occur after leakage Frequency compares and analyzes, and rejects the infrasonic sound wave frequency rate outside gas pipeline according to the infrasonic sound wave frequency rate in gas pipeline;
S304: frequency spectrum classification
Infrasonic sound wave frequency rate in gas pipeline is arranged using small wave converting method, the infrasonic sound wave frequency rate calculated in gas pipeline is being let out Before leakage, leakage in, leakage after three periods occur probability;
S305: temporal is rejected
Reject gas pipeline in infrasonic sound wave frequency rate before leakage, leakage in, leakage after three periods occur repetition frequency spectrum;
S306: characteristic frequency spectrum extracts
It rejects after repeating frequency spectrum, if the residual spectrum quantity of period is less than 2 in leakage, filtering infrasonic sound intensity of wave is adjusted to 5%, i.e. the frequency spectrum below of average audio intensity 5% weeds out first, repeats step S303 to S305, at this point, if in leakage Filtering infrasonic sound intensity of wave is then turned up 5% less than 2 by the residual spectrum quantity of period again, i.e., average audio intensity 10% with Under frequency spectrum weed out first, step S303 to S305 is repeated, until the residual spectrum quantity in leakage in the period is greater than 2 Until a, these residual spectrums are defined as gas leakage feature monochrome frequency spectrum;
Step 4: amplitude characteristics identify
Infrasound strength change laws corresponding to the gas leakage feature monochrome frequency spectrum obtained according to step 3, in this, as knowledge The verifying map of other gas leakage identifies amplitude characteristics;
S401: the sound intensity is cut out
The infrasonic sound intensity of wave P that will be acquired in gas pipeline0It is compared with the infrasonic sound intensity of wave acquired outside gas pipeline, to fuel gas conduit Infrasonic sound intensity of wave P in road0Make a proportional cutting, forms P1, the value range of a is between 1-10;
S402: singular point is found
By P1Second dervative is sought according to the time, and calculates singular point;
S403: feature wave amplitude extracts
Using singular point as origin, by P1First 30 seconds, 30 seconds latter, the rear strong curve of 60 seconds waves composes as the feature wave amplitude of gas leakage Line;
Step 5: gas pipe pressure perturbation wave identifies
Each test point in gas pipeline follows pressure balance formula, makees the change curve of a pressure and time, calculates pressure Power time changing curve is to the second dervative of time, then calculates extreme value, and the time point of extreme value is the time point leaked That is the pressure disturbance time takes this as a foundation to determine whether gas leakage event and nearest leakage point occurs;
Step 6: leakage point position calculates
S601: the position of leakage point X is calculated using infrasound method;
S602: pressure disturbance method calculates leakage point
The time point that most similar two pressure sensor senses are calculated according to the pressure disturbance time that step 5 obtains, according to this The distance at two time points and time difference calculate the distance X that point occurs for its gas leakage1, calculation formula is as follows:
Wherein, PQ、PZThe respectively pressure of the monitoring point Q and the monitoring point Z, unit pa;L is the distance of the monitoring point Q and the monitoring point Z, single Position rice;X1At a distance from the monitoring point leakage point X and Q, unit rice;
S603: position correction parameter calculates
Resulting X and X will be calculated1With actual location of leak X0It is verified and is rectified a deviation, A value and B are calculated using formula (6) Value:
X0=A × X+B × X1 (6)
Wherein, A, B are position correction parameter;
Step 7: gas pipeline leak source detection parameters solidify
Step 3 to the resulting various parameters of step 6 is saved in pattern recognition module.
7. detection method according to claim 6, which is characterized in that further include after step 7
Step 8: gas pipeline operating status is shown
S801: gas pipeline operating status simulation
Combustion gas operating status display module shows actual gas pipeline operating status and expected void using the twin method of number Quasi- gas pipeline operating status, it is prognostic and health management engine PHM that core, which drives engine, i.e., digital twin data are driven The data calculated in dynamic PHM are analogue data, real-time by pressure sensor, flow sensor, infrasonic sensor Perceive the actual motion state of gas pipeline;The expected virtual gas pipeline of combustion gas operating status display module simulation runs shape State, virtual gas pipeline data transmissions reach and practical gas pipeline data transmission synchronization under the driving of the twin data of number Operation;
S802: gas leakage closes valve processing
In gas pipeline any place, it is assumed that incident of leakage occur, in the PHM of the twin data-driven of number, with color outstanding Show the user for needing the valve closed and influence;
S803: parameter adjustment
Pressure data, data on flows, infrasound after from gas leakage to closing valve in 2-4 hours, in extraction gas pipeline Data, and compared with the analogue data calculated in the PHM of the twin data-driven of number, and adjust PHM parameter, make It is consistent with the practical gas pipeline operating status of detection.
8. detection method according to claim 6, which is characterized in that further include following step between step 3 and step 4 It is rapid: feature composite audio frequency spectrum discerning;Concrete operations are as follows:
Frequency spectrum after step S303 spectral noise is rejected is as data source, respectively 10 seconds before gas leakage, 10 seconds after leakage Assemble a possibility that core is calculated using fuzzy core aggregation algorithms based on core cluster for data;After being clustered by possibility, often A sample correspondence obtains C degree of membership coefficient, and representative sample is under the jurisdiction of the degree of every class, and in [0,1] interior value, sample is divided into Two major classes: first kind sample is close to certain a kind of center and far from other classifications, is non-supporting vector;Second class sample is in Different classes of boundary position, can be under the jurisdiction of multiple classes, be supporting vector;By first kind sample be directly divided into most close to class and The relationship with other classes is not considered, without being added in the training set of support vector machines;Second class sample is used for supporting vector The training of machine obtains gas leakage character symbol sum of fundamental frequencies spectrum after meeting given threshold.
CN201811276735.1A 2018-10-30 2018-10-30 Gas pipeline leakage point detection device and detection method Active CN109357167B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811276735.1A CN109357167B (en) 2018-10-30 2018-10-30 Gas pipeline leakage point detection device and detection method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811276735.1A CN109357167B (en) 2018-10-30 2018-10-30 Gas pipeline leakage point detection device and detection method

Publications (2)

Publication Number Publication Date
CN109357167A true CN109357167A (en) 2019-02-19
CN109357167B CN109357167B (en) 2020-08-25

Family

ID=65347116

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811276735.1A Active CN109357167B (en) 2018-10-30 2018-10-30 Gas pipeline leakage point detection device and detection method

Country Status (1)

Country Link
CN (1) CN109357167B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110646181A (en) * 2019-08-30 2020-01-03 广州文冲船厂有限责任公司 Test method and test device for testing strength of LNG pipeline system
CN110792928A (en) * 2019-09-24 2020-02-14 中国石油化工股份有限公司 Pipeline leakage diagnosis combined algorithm based on big data

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5866802A (en) * 1995-10-17 1999-02-02 Matsushita Electric Industrial Co., Ltd. Piping leakage detecting apparatus
CN101319955A (en) * 2007-06-07 2008-12-10 北京昊科航科技有限责任公司 Method for extracting leakage of pipe monitored by infrasonic wave
CN202082629U (en) * 2011-04-06 2011-12-21 黄定军 Natural gas pipeline leakage monitoring system
CN203147291U (en) * 2013-03-27 2013-08-21 黄鹏 System capable of monitoring pipeline leakage by means of infrasonic waves, flow balance and negative pressure waves
CN103438359A (en) * 2013-08-06 2013-12-11 毛振刚 Oil pipeline leakage detection and positioning system
CN105156905A (en) * 2015-07-09 2015-12-16 南京声宏毅霆网络科技有限公司 Leakage monitoring system, method and device for pipeline and server

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5866802A (en) * 1995-10-17 1999-02-02 Matsushita Electric Industrial Co., Ltd. Piping leakage detecting apparatus
CN101319955A (en) * 2007-06-07 2008-12-10 北京昊科航科技有限责任公司 Method for extracting leakage of pipe monitored by infrasonic wave
CN202082629U (en) * 2011-04-06 2011-12-21 黄定军 Natural gas pipeline leakage monitoring system
CN203147291U (en) * 2013-03-27 2013-08-21 黄鹏 System capable of monitoring pipeline leakage by means of infrasonic waves, flow balance and negative pressure waves
CN103438359A (en) * 2013-08-06 2013-12-11 毛振刚 Oil pipeline leakage detection and positioning system
CN105156905A (en) * 2015-07-09 2015-12-16 南京声宏毅霆网络科技有限公司 Leakage monitoring system, method and device for pipeline and server

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110646181A (en) * 2019-08-30 2020-01-03 广州文冲船厂有限责任公司 Test method and test device for testing strength of LNG pipeline system
CN110792928A (en) * 2019-09-24 2020-02-14 中国石油化工股份有限公司 Pipeline leakage diagnosis combined algorithm based on big data

Also Published As

Publication number Publication date
CN109357167B (en) 2020-08-25

Similar Documents

Publication Publication Date Title
CN106442720B (en) A kind of acoustic shock type track bridge health monitoring device, system and method
EP2912416B1 (en) Detecting leaks in a fluid distribution system
Arattano et al. Systems and sensors for debris-flow monitoring and warning
Erdik et al. Istanbul earthquake rapid response and the early warning system
Tejedor et al. A novel fiber optic based surveillance system for prevention of pipeline integrity threats
André et al. Listening to the deep: live monitoring of ocean noise and cetacean acoustic signals
Satriano et al. Earthquake early warning: Concepts, methods and physical grounds
Gasparini et al. Earthquake early warning systems
CN100552668C (en) Leakage detecting and locating method based on pressure and sound wave information fusion
CA2786303C (en) Method and system for automated location dependent natural disaster forecast
CN103236127B (en) A kind of fiber fence system for monitoring intrusion and mode identification method thereof
CN103144937B (en) System and method for intelligently monitoring belt-type conveyer for coal mine steel wire rope core
Song et al. Distribution characteristics of vehicle-specific power on urban restricted-access roadways
Allen et al. Earthquake early warning: Advances, scientific challenges, and societal needs
CN105182450B (en) A kind of strong convective weather nowcasting warning system
US7202797B2 (en) System and a method for detecting, locating and discerning an approach towards a linear installation
Kibblewhite et al. Wave–wave interactions, microseisms, and infrasonic ambient noise in the ocean
Iannaccone et al. A prototype system for earthquake early-warning and alert management in southern Italy
Alamdari et al. A spectral-based clustering for structural health monitoring of the Sydney Harbour Bridge
JPWO2013145493A1 (en) Pipe management support device and pipe management support system
CN104412087B (en) The ultrasonic gas leakage detector distinguished with false alarm
US20130338920A1 (en) Apparatus and method for providing environmental predictive indicators
US10401254B2 (en) Method and system for continuous remote monitoring of the integrity of pressurized pipelines and properties of the fluids transported
CN106197910B (en) Bridge detection method and detection system based on vehicle-bridge coupling vibration analysis
CN105042339B (en) One kind is based on nondimensional leakage of finished oil pipeline amount estimating system and method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant