CN110285330A - A kind of water utilities net pipe burst detection method based on the local factor that peels off - Google Patents
A kind of water utilities net pipe burst detection method based on the local factor that peels off Download PDFInfo
- Publication number
- CN110285330A CN110285330A CN201910624754.7A CN201910624754A CN110285330A CN 110285330 A CN110285330 A CN 110285330A CN 201910624754 A CN201910624754 A CN 201910624754A CN 110285330 A CN110285330 A CN 110285330A
- Authority
- CN
- China
- Prior art keywords
- peels
- test point
- factor
- detection data
- booster
- 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.)
- Granted
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 83
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 title claims abstract description 47
- 238000012360 testing method Methods 0.000 claims abstract description 56
- 238000000034 method Methods 0.000 claims abstract description 22
- 238000013480 data collection Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 2
- 239000000523 sample Substances 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000008439 repair process Effects 0.000 description 2
- 240000007594 Oryza sativa Species 0.000 description 1
- 235000007164 Oryza sativa Nutrition 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- QVFWZNCVPCJQOP-UHFFFAOYSA-N chloralodol Chemical compound CC(O)(C)CC(C)OC(O)C(Cl)(Cl)Cl QVFWZNCVPCJQOP-UHFFFAOYSA-N 0.000 description 1
- 230000001427 coherent effect Effects 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000004907 flux Effects 0.000 description 1
- 239000013505 freshwater Substances 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000003331 infrared imaging Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000002045 lasting effect Effects 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012552 review Methods 0.000 description 1
- 235000009566 rice Nutrition 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 239000008400 supply water Substances 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F17—STORING OR DISTRIBUTING GASES OR LIQUIDS
- F17D—PIPE-LINE SYSTEMS; PIPE-LINES
- F17D5/00—Protection or supervision of installations
- F17D5/02—Preventing, monitoring, or locating loss
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Pipeline Systems (AREA)
Abstract
The present invention relates to field water utilities network management net detection technique fields, more particularly to a kind of water utilities net pipe burst detection method based on the local factor that peels off, include: S1: collecting the detection data at each test point current time and the history detection data of passing more synchronizations in a few days in pipe network to be detected;S2: the factor that peels off of the detection data at each test point current time is calculated according to the detection data at each test point current time and history detection data;S3: the spatial neighborhood relations of test point are obtained, and according to the factor that peels off at two adjacent test point current times, the booster probability of the pipeline section between calculating adjacent test point two-by-two;S4: judging whether the booster probability of each pipeline section is greater than given threshold, and such as larger than threshold value then determines that booster occurs for the pipeline section, on the contrary then be judged as that there is no boosters.This method requires no knowledge about the label of data, has practicability, so that pipe burst detection feasibility is higher.
Description
Technical field
The present invention relates to water utilities network management net detection technique fields, and in particular to a kind of water utilities net based on the local factor that peels off
Pipe burst detection method.
Background technique
Water resource is the indispensable resource of every country, is the indispensable component part of development of human civilization, people
Demand to water is increasing, and the pollution condition of water resource is also increasingly sharpened, and freshwater resources are originally seldom, therefore all over the world all
Very water shortage, and studied by correlation study, the water shortage problem in China is more serious, well below world average level.It is adjoint
Construction high speed, lasting development and the generally increase of population in China city, the demand of water resource become very big.Section
About water resource is the important method of the shortage of water resource to be effectively relieved, and water money can be saved by reducing water supply network leakage loss not only
Source, while water undertaking's economic benefit can also be effectively improved.Pipe leakage is a kind of very common accident.Even if pipeline is being spread
If when reached state-set standard, but due to the old of pipeline, deformation, corrosion etc., leakage can usually occur, wherein
Physics leakage loss is even more unavoidable.In recent years, the problem of China's pipe network model, is increasingly severe, especially for some water shortages
City is just able to satisfy the primary demand in its area by maximum is influenced, so that needing to supply water using long range, but due to
The influence of weather, environment etc. factor causes its cost greatly to increase;On the other hand when leakage loss reaches a certain level, ten
Partial volume is also easy to produce booster problem, that can generate sizable economic loss again.Therefore pipe network model is detected and is reduced, prevent water utilities
Pipe burst is extremely urgent.
It is divided into passive and two kinds of active leak detecting with the main method of wider judgement leakage loss at present.
Passive leak detection method it is former with than wide, this method to wait pipeline leakages for a period of time after, water is by ground end
Under after emerging gradually, then relevant staff regularly makes an inspection tour pipeline, has seen whether that water is overflowed from underground, then sentences
Disconnected whether with the presence of leak, this leakage loss point found by testing staff belongs to bright leak source, is not required to utilize other coherent detections
Tool, this detection method very simple, and also the fund spent is few, and manpower is also few, but favorably also has disadvantage.Because of this side
Method could always be found after generating a large amount of leaks, can waste many water resources, very big to the loss of pipe network.It adds
These leakage points and the optimization of pipeline section leakage point position public supply mains test point and localization of bursted pipe model research contrast are too big, past
Toward keeping excavated volume excessive, maintenance time is too long.
Active leak detecting is the method for comparing modernization, different from Passive leak detection method, does not need water and emerges just from underground
It can detected by some modern instruments.With the development of the society, various countries develop the new of many leakage survey of gas network and positioning again
Technology: probe method in the pipe based on technologies such as magnetic flux, ultrasounds, thermal infrared imaging, meteorological imaging, olfactory sensing method, radioactivity are shown
Track agent leak detecting, negative pressure wave method, pressure gradient method, neural network, optical fiber leak detecting, quality or volumetric balance method, SCADA
The outer leak detecting of the pipelines such as (supervisory control and data acqui sition system), acoustic wave detection, pressure wave detection method.
In active leak detecting, big multi-method requires a large amount of labeled data, for generating prediction model, but
It is since the acquisition of data is relatively difficult especially also to need that data are marked, this work is ten in reality
It point is difficult to realize, it is special to expend time cost and human cost.
Summary of the invention
This patent for the data for being not readily available label in existing active leak detecting, it is special expend time cost with
And human cost this disadvantage, propose it is a kind of based on part peel off the factor water utilities net pipe burst detection method come detection pipe
Net whether booster, this method requires no knowledge about the label of data, has certain practicability, so that pipe burst detects feasibility
It is higher.
Method includes the following steps:
S1: the detection data at each test point current time in pipe network to be detected and passing more in a few days same are collected
The history detection data at one moment;
S2: each test point is calculated according to the detection data at each test point current time and history detection data and is worked as
The factor that peels off of the detection data at preceding moment;
S3: obtaining the spatial neighborhood relations of test point, and according to the factor that peels off at two adjacent test point current times,
The booster probability of pipeline section between calculating adjacent test point two-by-two;
S4: judging whether the booster probability of each pipeline section is greater than given threshold, and such as larger than threshold value then determines the pipeline section
Booster occurs, it is on the contrary then be judged as that there is no boosters.
This method is using the data of the same pipeline section different time as foundation, by the data for judging a certain moment pipeline section
It peels off factor size with the historical juncture pipeline section data and calculates booster probability, so that the booster probability threshold value according to setting comes
Detect the pipeline section whether booster, do not need the label taken time and effort to data, have very high practicability so that pipe network is quick-fried
It is higher that pipe detects feasibility.
Further, in step S2, the calculating process for the factor that peels off is as follows:
If the detection data collection Υ={ X at current timeoj, history detection data integrates as X={ Xij, i is historical juncture, i
={ 1,2 ... I }, I represent the depth of historical juncture, and j is test point number, and j={ 1,2 ... J }, J represent test point sum,
The formula for then calculating the factor that peels off is as follows:
Wherein: (1) ρk(Xoj) indicate point XojK neighborhood in all the points to XojAverage reach distance, formula is as follows:
(2)Nk(Xoj) it is point XojK apart from neighborhood, meet:
Nk(Xoj)={ Xij'∈D\{Xoj}|d(Xoj,Xij')≤dk(Xoj)};
(3)dk(Xoj,Xij) it is XijPoint-to-point XojK reach distance
dk(Xoj,Xij)=max { dk(Xoj),d(Xoj,Xij)};
(4) d (X is setoj,Xij) it is two pressure spot XojAnd XijThe distance between;
(5)dk(Xoj) it is point XojKth distance, and dk(Xoj)=d (Xoj,Xij) following condition need to be met:
It a) does not at least include X in setijK point X insideoj,∈C{x≠Xij, meet d (Xij,Xoj,)≤d
(Xij,Xoj);
B) being up in set does not include XijK-1 point X including poj,∈C{x≠Xij, meet d (Xij,Xoj,)<d
(Xij,Xoj)。
Further, certain pipeline section booster probability is calculated as follows in step S3:
Test point j1 such as is obtained according to the spatial neighborhood relations between test point and test point j2 is adjacent, then the two is formed by connecting
Certain section of pipeline section booster probability P calculates as follows:
Wherein s indicates pipeline section serial number, and s={ 1,2 ... m }, m represent the maximum serial number of pipeline section;LOFk(X0j1) represent test point
The current detection data X of j10j1The factor that peels off, LOFk(X0j2) represent the current detection data X of test point j20j2Peel off because
Son;
max(LOFk(X0j)) and large (LOFk(X0j)) respectively represent in all test points it is maximum and it is second largest from
Group factor.
Further, the detection data is the pressure value in pipe.
Further, selection indicates that the K value of Size of Neighborhood takes 3 in step s 2.
Further, maximum historical depth selection 3 in step sl.
Further, the value range of the threshold value of the booster probability is 0.85~0.95.
Further, the threshold value of the booster probability takes 0.90.
Wrong report occurs on adjacent a small number of pipeline sections, that is to say, that and distance actually occurs being closer for the pipeline section of booster,
This is because be after booster occurs for actually pipeline section, certainly will influence whether the pressure detection data of adjacent tubular segments, so this
The wrong report of sample is little for the influence for finally sending staff to go to verify and repair, within the acceptable range, however such as
Fruit selects higher threshold value to be then likely to occur the case where failing to report, and from the angle of real work, a degree of wrong report is can
With receiving, and loss brought by failing to report will be serious, so the present embodiment has chosen 0.90 on this basis and is used as threshold value,
To guarantee to be not in the case where failing to report.
Detailed description of the invention
Fig. 1 is by the schematic diagram for the water supply network booster model established in the embodiment of the present invention one.
Fig. 2 is the enlarged drawing in Fig. 1 at A.
Fig. 3 is the stream of water utilities net pipe burst detection method of one of the embodiment of the present invention based on the local factor that peels off
Cheng Tu.
Specific embodiment
It is further described below by specific embodiment:
Embodiment one
The present embodiment passes through the water distribution hydraulic model of one villages and small towns rank of EPANET software building, the water supply network first
Hydraulic model includes:
One water source: gross head is set as 500, enters pipe network by water supply node;
One water pump: water pump curve can be selected arbitrarily, the equation of the curve used in this example for;
Lift=46.67-1.867 × 10-4× (flow)2, the unit rice (m) of lift, the unit of flow is liter/second (L/
s);
49 pipe network test points: for absolute altitude all in 49ft or so, basic water requirement is set as 5-10L/s;
63 sections of pipeline section numbers: caliber chooses DN100-DN400, and the coefficient of roughness is defaulted as 100;
The pipe network topological diagram of the water distribution hydraulic model built is substantially as shown in Figure 1;
It is provided with booster at 3 in pipe network, is 24,30 and No. 53 pipeline sections respectively, is circled in Fig. 2.
In order to implement and simulate water utilities net pipe burst detection method of one of the present invention based on the local factor that peels off,
After establishing water supply network booster model, established pipe network is imported by MATLAB tool, passes through relative program generation
Code reads inp file derived from EPANET, and the data in EPANET program are exported, and is stored in Excel file, normal to obtain
In the case of pressure detection data collection and booster when pressure detection data collection.
In pipe network operation, pipe network is always lasted for 24 hours, and waterpower time step is 1 hour, i.e., acquisition one in each hour
The pressure data of secondary test point, totally 49 test points, therefore the data dimension of every group of data are 1 × 49, total daily to obtain 24
Group data.
Firstly, obtain 24 hours interior conduits 24 groups of data under normal circumstances, every group of pressure detection data respectively with value
For [0.99-1.01], dimension is that 1 × 49 random vector carries out Hadamard product operation (Hadamard product), by with three
A different random vector carries out operation, obtain three groups with similar data under normal circumstances for simulation one day before, two days
Preceding and synchronization before three days pressure detection data is as training sample, along with the one under normal circumstances group number of script
According to each moment shares 4 groups of data in training sample;
Then, 24,30, No. 53 pipeline sections are set as booster respectively, simulate a pipeline section booster in three every time and obtained
In booster, the pressure detection data at 24 moment calls booster detection data in the following text as test data in one day.
Then booster detection can be carried out according to following algorithm:
The detection data collection Υ={ X at current timeoj, it is the booster detection data at certain moment, history testing number in this example
According to integrating as X={ Xij, i.e., in the training sample in this example before one day, before two days and before three days synchronization pressure detecting number
According to wherein i={ 1,2 ... I }, I represents the depth of historical juncture, the historical data of first three days has been recalled in this example, so history
The depth at moment be 3, j be test point number, j={ 1,2 ... J }, J represent test point sum, this example 49, then setup algorithm from
The formula of group factor is as follows:
Wherein: (1) ρk(Xoj) indicate point XojK neighborhood in all the points to XojAverage reach distance, calculation formula is such as
Under:
(2)Nk(Xoj) it is point XojK apart from neighborhood, meet:
Nk(Xoj)={ Xij'∈D\{Xoj}|d(Xoj,Xij')≤dk(Xoj)};
(3)dk(Xoj,Xij) it is XijPoint-to-point XojK reach distance
dk(Xoj,Xij)=max { dk(Xoj),d(Xoj,Xij)};
(4) d (X is setoj,Xij) it is two pressure detection data point XojAnd XijThe distance between;
(5)dk(Xoj) it is point XojK distance, and dk(Xoj)=d (Xoj,Xij) following condition need to be met:
It a) does not at least include X in setijK point X insideoj,∈C{x≠Xij, meet d (Xij,Xoj,)≤d
(Xij,Xoj);
B) being up in set does not include XijK-1 point X including poj,∈C{x≠Xij, meet d (Xij,Xoj,)<d
(Xij,Xoj), the value of K takes 3 in this example.
It is peeled off the factor by the part that LOF algorithm calculates each point one by one, and is arranged by ascending order.
From the spatial neighborhood relations known between test point in the network topology structure of water supply network booster model, if test point
J1 and test point j2 are adjacent, then the booster probability P for the pipeline section s that the two is formed by connecting calculates as follows:
Wherein s indicates pipeline section serial number, and it is 63 in maximum serial number this example of pipeline section that s={ 1,2 ... m }, m, which are represented,;LOFk(X0j1)
Represent the current detection data X of test point j10j1The factor that peels off, LOFk(X0j2) represent the current detection data of test point j2
X0j2The factor that peels off;
max(LOFk(X0j)) and large (LOFk(X0j)) respectively represent in all test points it is maximum and it is second largest from
Group factor.
Different booster probability threshold values is arranged in the booster probability for successively calculating each pipeline section, obtains different booster pipes
Segment number.In this example, using the booster detection data of 24,30, No. 53 pipeline section boosters respectively, in total 24 of 0 point to 24 points
Moment has made booster probability calculation, i.e. the booster probability calculation of each pipeline section of total 72 wheel, and is directed to different booster probability thresholds
The screening that value has carried out booster pipeline section calculates, the booster probability calculated when in table 1 using 12 as foundation and respectively with 0.85,
0.90 and 0.95 is booster probability threshold value, illustratively gives one group of booster testing result.
Table 1
It is in table 1 the result shows that method used by this patent, can compare when booster probability threshold value is set as 0.95
The pipeline section of booster is accurately predicted, both without leaking quick-fried or not reporting by mistake.In addition, wrong report situation occurs in the pipeline section with booster
On adjacent pipeline section.
If: rate of false alarm=the pipeline section quantity of prediction result mistake/pipeline section quantity being predicted;Rate of failing to report=booster pipeline section is not
Appear in the prediction number of number in prediction result/total;
By extending the value range of booster probability threshold value, the booster probability calculation result of all 72 wheels in this example is adopted
After carrying out finally prediction with different booster probability threshold values, rate of false alarm and rate of failing to report to prediction result are found after counting, quick-fried
The more high then false detection rate of pipe probability threshold value is lower, when booster probability threshold value up to 0.85, rate of false alarm 6.1%, and rate of failing to report 0;
When booster probability threshold value up to 0.9, rate of false alarm drops to 2.4%, rate of failing to report 0;When pipe probability threshold value up to 0.6%, but
Rate of failing to report is really significantly increased to 33.3%;On the other hand, wrong report situation occurs on the pipeline section adjacent with the pipeline section of booster, away from
It is with a distance from booster pipeline section and not far, so for influence is within the scope of can tolerate brought by manual review, as a result,
Booster probability threshold value preferably uses 0.9.
Computer simulation sets suitable threshold value the results show that passing through, and the method in the present invention can be utilized preferably
History detection data predicts that the pipeline section of booster occurs for number with the currently monitored data, and reports probability and the wrong report generation of situation by mistake
The case where within the acceptable range.
Embodiment two
In the present embodiment, the solution of the present invention is used in actual pipe network detection, and the flow chart of detection method is such as
It is specific as follows shown in Fig. 3
S1: the detection data at each test point current time in pipe network to be detected and passing more in a few days same are collected
The history detection data at one moment;
S2: each test point is calculated according to the detection data at each test point current time and history detection data and is worked as
The factor that peels off of the detection data at preceding moment;
S3: obtaining the spatial neighborhood relations of test point, and according to the factor that peels off at two adjacent test point current times,
The booster probability of pipeline section between calculating adjacent test point two-by-two;
S4: judging whether the booster probability of each pipeline section is greater than given threshold, and such as larger than threshold value then determines the pipeline section
Booster occurs, it is on the contrary then be judged as that there is no boosters.
Peeling off in this example, the factor, the calculating of booster probability are same as Example 1, and each detection node is acquisition in pipe network
The pressure sensor of hydraulic pressure, round-the-clock progress pressure data acquisition, sampling period are 5 seconds, and sampled data is aggregate to backstage
Detection service device;The depth of historical juncture equally also selects 3 days, indicates that the K value of field size also takes 3, then, in detection service
The data in passing 3 days are utilized on device, the booster probability P every five seconds in embodiment 1 is calculated only once, and booster probability threshold value is set
It is set to 0.90;When the discovery of detection service device has the booster probability of pipeline section to surpass the threshold value, then send a warning, to realize pipe
The real-time detection of net.Set by the threshold value, can accomplish not fail to report, but it is possible that wrong report the case where, in table one
As exemplified, wrong report occurs on adjacent a small number of pipeline sections, that is to say, that distance actually occurs the distance of the pipeline section of booster
It is relatively close, this is because being that certainly will influence whether the pressure detection data of adjacent tubular segments, institute after actually booster occurs for a pipeline section
It is little for the influence for finally sending staff to go to verify and repair with such wrong report, within the acceptable range.So
And the case where failing to report is likely to occur if selecting higher threshold value, and from the angle of real work, a degree of wrong report
It is acceptable, and loss brought by failing to report will be serious, so the present embodiment has chosen 0.90 conduct on this basis
Threshold value, to guarantee to be not in the case where failing to report.
Signal connection between pressure sensor and detection service device uses sensor network, utilizes LORA or NB-IoT etc.
The communication protocol of low-power consumption carries out data transmission, and corresponding pressure sensor can be clustered, in order to increase single detection service device
Coverage area, multiple communication relay stations can be set in sensor network, but no matter how communication network is arranged, and does not influence
The implementation of detection method in the present invention.
Method in the present embodiment, using the data of the same pipeline section different time as foundation, by judging a certain moment
Factor size is peeled off between the data of the pipeline section and the data of preceding several historical juncture pipeline sections to calculate booster probability, thus according to
Detected according to the booster probability threshold value of setting the pipeline section whether booster, do not need the label taken time and effort to data, have
Very high practicability, so that pipe burst detection feasibility is higher.Can be improved after practical application water factory booster occur when
The working efficiency for detecting booster, has saved water resource.
The above are merely the embodiment of the present invention, the common sense such as well known specific structure and characteristic are not made excessively herein in scheme
Description, all common of technical field that the present invention belongs to before one skilled in the art know the applying date or priority date
Technological know-how can know the prior art all in the field, and have using routine experiment means before the date
Ability, one skilled in the art can improve in conjunction with self-ability under the enlightenment that the application provides and implement we
Case, some typical known features or known method should not become the barrier that one skilled in the art implement the application
Hinder.It should be pointed out that for those skilled in the art, without departing from the structure of the invention, if can also make
Dry modification and improvement, these also should be considered as protection scope of the present invention, these all will not influence the effect that the present invention is implemented and
Patent practicability.The scope of protection required by this application should be based on the content of the claims, the specific reality in specification
Applying the records such as mode can be used for explaining the content of claim.
Claims (8)
1. a kind of water utilities net pipe burst detection method based on the local factor that peels off, it is characterised in that: the following steps are included:
S1: the detection data at each test point current time and passing more same a period of time in a few days in pipe network to be detected are collected
The history detection data at quarter;
S2: when according to the detection data at each test point current time and the current each test point of history detection data calculating
The factor that peels off of the detection data at quarter;
S3: obtaining the spatial neighborhood relations of test point, and according to the factor that peels off at two adjacent test point current times, calculates
Adjacent test point two-by-two between pipeline section booster probability;
S4: judging whether the booster probability of each pipeline section is greater than given threshold, and such as larger than threshold value then determines that the pipeline section occurs
Booster, it is on the contrary then be judged as that there is no boosters.
2. the water utilities net pipe burst detection method according to claim 1 based on the local factor that peels off, it is characterised in that:
In step S2, the calculating process for the factor that peels off is as follows:
If the detection data collection Υ={ X at current timeoj, history detection data integrates as X={ Xij, wherein i is historical juncture, i
={ 1,2 ... I }, I represent the depth capacity of historical juncture, and j is test point number, and j={ 1,2 ... J }, J represent test point sum,
The formula for then calculating the factor that peels off is as follows:
Wherein: (1) ρk(Xoj) indicate point XojK neighborhood in all the points to XojAverage reach distance, formula is as follows:
(2)Nk(Xoj) it is point XojK apart from neighborhood, meet:
Nk(Xoj)={ Xij'∈D\{Xoj}|d(Xoj,Xij')≤dk(Xoj)};
(3)dk(Xoj,Xij) it is XijPoint-to-point XojK reach distance
dk(Xoj,Xij)=max { dk(Xoj),d(Xoj,Xij)};
(4) d (X is setoj,Xij) it is two pressure spot XojAnd XijThe distance between;
(5)dk(Xoj) it is point XojKth distance, and dk(Xoj)=d (Xoj,Xij) following condition need to be met:
It a) does not at least include X in setijK point X insideoj,∈C{x≠Xij, meet d (Xij,Xoj,)≤d(Xij,
Xoj);
B) being up in set does not include XijK-1 point X including poj,∈C{x≠Xij, meet d (Xij,Xoj,)<d(Xij,
Xoj)。
3. the water utilities net pipe burst detection method according to claim 2 based on the local factor that peels off, it is characterised in that:
Certain pipeline section booster probability is calculated as follows in step S3:
It such as obtains test point j1 according to the spatial neighborhood relations between test point and test point j2 is adjacent, then certain section that the two is formed by connecting
Pipeline section booster probability P calculates as follows:
Wherein s indicates pipeline section serial number, and s={ 1,2 ... m }, m represent the maximum serial number of pipeline section;LOFk(X0j1) represent test point j1's
Current detection data X0j1The factor that peels off, LOFk(X0j2) represent the current detection data X of test point j20j2The factor that peels off;
max(LOFk(X0j)) and large (LOFk(X0j)) respectively represent in all test points it is maximum and second largest peel off because
Son.
4. the water utilities net pipe burst detection method according to claim 1 based on the local factor that peels off, it is characterised in that:
The detection data is the pressure value in pipe.
5. the water utilities net pipe burst detection method according to claim 2 based on the local factor that peels off, it is characterised in that:
It selects to indicate that the K value of Size of Neighborhood is 3 in step s 2.
6. the water utilities net pipe burst detection method according to claim 5 based on the local factor that peels off, it is characterised in that:
The historical depth selection 3 of history detection data sequence in step sl.
7. the water utilities net pipe burst detection method according to claim 1 based on the local factor that peels off, it is characterised in that:
The value range of the threshold value of the booster probability is 0.85~0.95.
8. the water utilities net pipe burst detection method according to claim 7 based on the local factor that peels off, it is characterised in that:
The threshold value of the booster probability takes 0.90.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910624754.7A CN110285330B (en) | 2019-07-11 | 2019-07-11 | Water network pipe burst detection method based on local outlier factor |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910624754.7A CN110285330B (en) | 2019-07-11 | 2019-07-11 | Water network pipe burst detection method based on local outlier factor |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110285330A true CN110285330A (en) | 2019-09-27 |
CN110285330B CN110285330B (en) | 2021-08-10 |
Family
ID=68022098
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910624754.7A Expired - Fee Related CN110285330B (en) | 2019-07-11 | 2019-07-11 | Water network pipe burst detection method based on local outlier factor |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110285330B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111367972A (en) * | 2020-02-03 | 2020-07-03 | 重庆特斯联智慧科技股份有限公司 | Community drainage pipeline exploration method and system based on artificial intelligence |
CN112097125A (en) * | 2020-09-18 | 2020-12-18 | 同济大学 | Water supply pipe network pipe burst detection and positioning method based on self-adaptive checking |
CN116380363A (en) * | 2023-03-29 | 2023-07-04 | 宁夏隆基宁光仪表股份有限公司 | Water leakage detection method and water meter |
CN117436845A (en) * | 2023-12-15 | 2024-01-23 | 山东齐芯智控科技有限公司 | Abnormal data monitoring method for intelligent district heating system |
CN117851974A (en) * | 2024-03-07 | 2024-04-09 | 艾信智能环境科技(无锡)有限公司 | Medical pneumatic type garbage bedding and clothing intelligent collection system |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102777770A (en) * | 2012-08-07 | 2012-11-14 | 河北农业大学 | Flow monitoring point optimizing arranging method for water supply pipe network leakage fault diagnosis |
US20130211762A1 (en) * | 2012-02-09 | 2013-08-15 | Microsoft Corporation | Self-tuning statistical resource leak detection |
CN104976517A (en) * | 2015-05-25 | 2015-10-14 | 安迅捷集装箱码头(深圳)有限公司 | Wharf water supply pipe network online supervision method |
CN106874532A (en) * | 2016-12-23 | 2017-06-20 | 武汉智博创享科技有限公司 | A kind of grid method for analyzing cartridge igniter |
CN108980628A (en) * | 2018-07-27 | 2018-12-11 | 上海航天动力科技工程有限公司 | A kind of water supply network booster alarm method |
CN109560984A (en) * | 2018-11-13 | 2019-04-02 | 苏宁易购集团股份有限公司 | A kind of network service response time method for detecting abnormality and device |
CN109681789A (en) * | 2019-03-01 | 2019-04-26 | 重庆邮电大学 | The relevant detection localization method of fluid line leak acoustic signal based on variation mode decomposition |
-
2019
- 2019-07-11 CN CN201910624754.7A patent/CN110285330B/en not_active Expired - Fee Related
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20130211762A1 (en) * | 2012-02-09 | 2013-08-15 | Microsoft Corporation | Self-tuning statistical resource leak detection |
CN102777770A (en) * | 2012-08-07 | 2012-11-14 | 河北农业大学 | Flow monitoring point optimizing arranging method for water supply pipe network leakage fault diagnosis |
CN104976517A (en) * | 2015-05-25 | 2015-10-14 | 安迅捷集装箱码头(深圳)有限公司 | Wharf water supply pipe network online supervision method |
CN106874532A (en) * | 2016-12-23 | 2017-06-20 | 武汉智博创享科技有限公司 | A kind of grid method for analyzing cartridge igniter |
CN108980628A (en) * | 2018-07-27 | 2018-12-11 | 上海航天动力科技工程有限公司 | A kind of water supply network booster alarm method |
CN109560984A (en) * | 2018-11-13 | 2019-04-02 | 苏宁易购集团股份有限公司 | A kind of network service response time method for detecting abnormality and device |
CN109681789A (en) * | 2019-03-01 | 2019-04-26 | 重庆邮电大学 | The relevant detection localization method of fluid line leak acoustic signal based on variation mode decomposition |
Non-Patent Citations (1)
Title |
---|
黄爱芹: "基于数据驱动的调节阀故障诊断方法研究", 《中国博士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111367972A (en) * | 2020-02-03 | 2020-07-03 | 重庆特斯联智慧科技股份有限公司 | Community drainage pipeline exploration method and system based on artificial intelligence |
CN112097125A (en) * | 2020-09-18 | 2020-12-18 | 同济大学 | Water supply pipe network pipe burst detection and positioning method based on self-adaptive checking |
CN116380363A (en) * | 2023-03-29 | 2023-07-04 | 宁夏隆基宁光仪表股份有限公司 | Water leakage detection method and water meter |
CN117436845A (en) * | 2023-12-15 | 2024-01-23 | 山东齐芯智控科技有限公司 | Abnormal data monitoring method for intelligent district heating system |
CN117436845B (en) * | 2023-12-15 | 2024-04-09 | 山东齐芯智控科技有限公司 | Abnormal data monitoring method for intelligent district heating system |
CN117851974A (en) * | 2024-03-07 | 2024-04-09 | 艾信智能环境科技(无锡)有限公司 | Medical pneumatic type garbage bedding and clothing intelligent collection system |
CN117851974B (en) * | 2024-03-07 | 2024-05-14 | 艾信智能环境科技(无锡)有限公司 | Medical pneumatic type garbage bedding and clothing intelligent collection system |
Also Published As
Publication number | Publication date |
---|---|
CN110285330B (en) | 2021-08-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110285330A (en) | A kind of water utilities net pipe burst detection method based on the local factor that peels off | |
CN109555979B (en) | Water supply pipe network leakage monitoring method | |
CN108984873B (en) | Water supply network real-time leakage detection method, device, system and storage medium | |
CN107143750B (en) | A kind of space layout method of pipe burst monitoring network | |
CN105930653B (en) | A kind of pipeline exploding early warning method based on metering subregion flow monitoring data | |
CN102235575A (en) | Data processing method and system for checking pipeline leakage | |
US20210148782A1 (en) | Efficient method for localizing leaks in water supply pipe network based on valve operations and online water metering | |
CN110245411A (en) | A kind of cities and towns central heating system and leakage point detecting method | |
CN112377817B (en) | Municipal pipe network pipe burst monitoring system and method | |
CN106874532A (en) | A kind of grid method for analyzing cartridge igniter | |
CN112036086A (en) | Dynamic risk early warning system for gas pipeline | |
CN104976517A (en) | Wharf water supply pipe network online supervision method | |
CN103927259A (en) | Fault detection and isolation synthesis method based on testability modeling data | |
CN109344708A (en) | A kind of water supply network booster abnormal signal analysis method | |
CN108732464A (en) | Submarine observation network extra large cable on-line fault diagnosis and localization method | |
CN108520267A (en) | A kind of hydrological telemetering data exception detection method based on space-time characteristic | |
Ng et al. | A design of automatic water leak detection device | |
CN107013812A (en) | A kind of THM coupling line leakage method | |
CN110119853A (en) | Water supply network leakage loss alarm threshold value choosing method based on time series Analysis on monitoring data | |
CN109405924A (en) | A kind of online remote supervision system of drainage pipeline networks liquid level and method | |
CN109886506A (en) | A kind of water supply network booster risk analysis method | |
Li et al. | Leak detection of municipal water supply network based on the cluster-analysis and fuzzy pattern recognition | |
CN113128709B (en) | Water supply network leakage management system | |
CN108615098A (en) | Water supply network pipeline burst Risk Forecast Method based on Bayesian survival analysis | |
CN102242872A (en) | Oil transportation pipeline network leakage detection method based on generalized fuzzy hyperbolic model |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210810 |