CN108615098A - Water supply network pipeline burst Risk Forecast Method based on Bayesian survival analysis - Google Patents
Water supply network pipeline burst Risk Forecast Method based on Bayesian survival analysis Download PDFInfo
- Publication number
- CN108615098A CN108615098A CN201810448471.7A CN201810448471A CN108615098A CN 108615098 A CN108615098 A CN 108615098A CN 201810448471 A CN201810448471 A CN 201810448471A CN 108615098 A CN108615098 A CN 108615098A
- Authority
- CN
- China
- Prior art keywords
- booster
- pipeline
- risk
- covariant
- survival analysis
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Educational Administration (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Pipeline Systems (AREA)
Abstract
The present invention relates to a kind of water supply network pipeline burst Risk Forecast Methods based on Bayesian survival analysis, and this method comprises the following steps：(1) booster database is established according to the booster historical data of collection, extraction key message is as covariant；(2) space cluster analysis is carried out to booster point, the space distribution information of booster point, which is quantified the covariant new as one, to be supplemented in booster database；(3) it is based on booster database and booster risk forecast model is built using Bayesian survival analysis method；(4) the booster risk of booster risk forecast model prediction pipeline is used.Compared with prior art, prediction result of the present invention is more accurate reasonable.
Description
Technical field
The present invention relates to a kind of water supply network pipeline burst Risk Forecast Methods, are given birth to based on Bayes more particularly, to one kind
Deposit the water supply network pipeline burst Risk Forecast Method of analysis.
Background technology
The water supply network public infrastructure one of important as city, is the main artery in entire city, carry to
The important task of family conveying water.In recent ten years, Chinese Urbanization development speed is constantly accelerated, and Urban Water Demand surges, and city supplies
The scale of grid increasingly increases.However there is also section of tubing to be laid with overlong time for China's water supply network, aging phenomenon is tight
A series of problems, such as weight, tubing (such as gray cast iron tube) inferior occupy larger proportion in pipe network.China is in pipe network operation simultaneously
Safety guarantee and administrative skill aspect, for the prevention and control of pipe burst accident, in information system management, pipeline material matter
Amount, network security detection, accident reaction and treatment technology etc. lack scientific research and the technology application of system, pipe network operation
Safety and administrative skill level are relatively low, and the thing followed is exactly that China's public supply mains pipe explosion accident takes place frequently, and is not only wasted
Valuable water resource, has also seriously affected the drinking water safety of people, the economic loss thereby resulted in and social negative effect etc.
A series of chain reactions are very important.Therefore, the operating status that water supply network is assessed using the method for science, is found out quick-fried in pipe network
Manage-style nearly higher pipeline, targetedly reinforces maintenance work, is necessary to reduce pipe explosion accident.
There are many domestic and international research in relation to this respect, and booster risk forecast model can simply be divided into physical model, statistics mould
Three kinds of type and data mining model are below some representative researchs：
1) physical model
Such as document：
[1]:Moglia M,Davis P,Burn S.Strong exploration of a cast iron pipe
failure model.Reliability Engineering&System Safety,2008,93(6):885-896.
The technical measures that such method uses：Based on Theory of Fracture Mechanics, assume initially that the remaining surrender of pipeline is strong
Degree meets Weibull distributions, is then simulated by Monte-Carlo Simulation and has obtained ash using historical data progress regression analysis
The booster risk forecast model of mouth cast-iron pipe.
Advantage and disadvantage：Such methods advantage is that the model established is to be based on some specific booster physical mechanisms, prediction knot
Fruit is more accurate.But disadvantage is that：(1) physical mechanism of pipeline burst is all more complicated and booster risk factor is numerous
More, physical prediction model can not often include all factors, larger using difficulty；(2) data needed for model are established to be difficult to
It obtains or obtains and cost dearly.
2) statistical model
Such as document：
[2]:Park S,Jun H,Agbenowosi N,et al.The proportional hazards modeling
of water main failure data incorporating the time-dependent effects of
covariates.Water resources management,2011,25(1):1-19.
[3]:Kimutai E,Betrie G,Brander R,et al.Comparison of statistical
models for predicting pipe failures:Illustrative example with the City of
Calgary water main failure.Journal of Pipeline Systems Engineering and
Practice,2015,6(4):04015005.
The technical measures that such method uses：It, will by statistical analysis based on a large amount of pipeline burst historical data
The factor and the object functions such as pipeline burst risk or booster time for influencing pipeline burst establish certain relationship, to establish mould
Type.
Advantage and disadvantage：Such methods advantage is that model form is fairly simple, and operability is strong, smaller using difficulty.But
Such method has certain requirement to the quality and quantity of booster historical data, less or incomplete in data
The result arrived is not ideal enough.
3) data mining model
Such as document：
[4]:Ahmad A,Mcbean E,Bahram G,et al.Forecasting watermain failure
using artificial neural network modelling.Canadian Water Resources Journal,
2013,38(1):24-33.
[5]:Harvey R,Mcbean EA,Gharabaghi B.Predicting the Timing of Water
Main Failure Using Artificial Neural Networks.Journal of Water Resources
Planning&Management,2013,140(4):425-434.
The technical measures that such method uses：Based on a large amount of pipeline burst historical data, by applying various numbers
Historical data is analyzed according to mining algorithm, certain principle is finally based on and pipeline burst risk is predicted.
Advantage and disadvantage：Such methods advantage is to be not limited to specific functional form.But disadvantage is that：(1) it needs
A large amount of booster historical data is wanted to be trained；(2) prediction result that pure data-driven model obtains may be with reality
The result observed is inconsistent.
Invention content
It is an object of the present invention to overcome the above-mentioned drawbacks of the prior art and provide one kind to be given birth to based on Bayes
Deposit the water supply network pipeline burst Risk Forecast Method of analysis.
The purpose of the present invention can be achieved through the following technical solutions：
A kind of water supply network pipeline burst Risk Forecast Method based on Bayesian survival analysis, this method include following step
Suddenly：
(1) booster database is established according to the booster historical data of collection, extraction key message is as covariant；
(2) space cluster analysis is carried out to booster point, the space distribution information quantization of booster point is new as one
Covariant be supplemented in booster database；
(3) it is based on booster database and booster risk forecast model is built using Bayesian survival analysis method；
(4) the booster risk of booster risk forecast model prediction pipeline is used.
Step (2) is using whether there is or not other booster quantized results covariants new as one in booster point setting space length
Amount, specifically：When having other boosters in booster point setting space length, then quantized result is 1, when booster point sets space
Apart from interior without other boosters, then quantized result is 0.
Step (3) is specially：
(31) the booster historical data standardization in booster database：Every booster pipeline historical record is divided into three portions
Point,
Wherein, DiIndicate i-th booster pipeline historical record, tiIndicate i-th booster pipeline life span, ZiIndicate i-th
Covariant vector corresponding to booster pipeline, δiIndicate the data type indicator variable of i-th booster pipeline, δi=1 indicates the
I booster pipeline historical record is complete data, δi=0 indicates that i-th booster pipeline historical record is Random censorship, i=1,
2 ... ..., N, N indicate booster pipeline sum；
(32) baseline risk function is established：Baseline risk function is quick-fried with pipeline using booster pipeline life span as independent variable
Manage-style is nearly dependent variable, and the pipeline burst risk is specially the booster number of annual unit pipe range；
(33) choose the principal element for influencing pipeline burst risk as covariant, establish Bayesian survival analysis model into
Row parameter Estimation determines covariant estimates of parameters；
(34) booster risk forecast model is determined according to baseline risk function and covariant estimates of parameters.
Step (33) is specially：
(331) life span of all booster pipelines is divided into 0 ＜ s of section1＜ s2＜ sJ＜ ∞ remember i-th
The life span of booster pipeline is ti, and to all tiBoth less than sJ, obtain J time interval (0, s1],(s1,
s2],···,(sJ-1,sJ]；
(332) assume that there are one fixed benchmark dangerous function h for each time interval0(ti)=λj, ti∈(sj-1,
sj-1), obtain Bayesian survival analysis model function：
Wherein, h (ti,Zi) be i-th pipeline booster risk, tiIndicate i-th pipeline life span, ZiIndicate i-th
Covariant vector corresponding to pipeline, βTFor the corresponding regression coefficient vector of covariant, λjIt is dangerous for the benchmark of each time interval
Function, j=1,2 ... ..., J, J are the time interval sum divided, i=1,2 ... ..., N, N expression booster pipeline sums.
Booster risk model is in step (34)：
H (t, Z)=(at2+bt+c)exp(βTZ)；
Wherein, h (t, Z) is pipeline burst risk, and t indicates pipeline life span, at2Risk function on the basis of+bt+c, a,
The fitting coefficient of risk function on the basis of b and c, Z indicate that the covariant for influencing the principal element composition of pipeline burst risk is vectorial,
βTFor the corresponding regression coefficient vector of covariant.
Space cluster analysis is carried out to booster point using DBSCAN clustering algorithms.
Compared with prior art, the invention has the advantages that：
(1) when calculating covariant regression coefficient, the achievement in research for introducing forefathers makes the present invention as prior distribution
The covariant regression coefficient acquired more meets general booster rule, and the prediction result of booster risk forecast model is more accurately closed
Reason；
(2) present invention incorporates the advantages of bayes method and traditional survival analysis method, more adapt to booster history note
Record less and incomplete situation.
Description of the drawings
Fig. 1 is that the present invention is based on the pipeline burst risk profile general flow charts that Bayesian survival is analyzed；
Fig. 2 is the flow diagram of DBSCAN clustering algorithms of the present invention.
Specific implementation mode
The present invention is described in detail with specific embodiment below in conjunction with the accompanying drawings.Note that the following embodiments and the accompanying drawings is said
Bright is substantial illustration, and the present invention is not intended to be applicable in it object or its purposes is defined, and the present invention does not limit
In the following embodiments and the accompanying drawings.
Embodiment
As shown in Figure 1, a kind of water supply network pipeline burst Risk Forecast Method based on Bayesian survival analysis, this method
Include the following steps：
(1) booster database is established according to the booster historical data of collection, extraction key message is as covariant；
(2) space cluster analysis is carried out to booster point, the space distribution information quantization of booster point is new as one
Covariant be supplemented in booster database；
(3) it is based on booster database and booster risk forecast model is built using Bayesian survival analysis method；
(4) the booster risk of booster risk forecast model prediction pipeline is used.
Step (1) is directed to booster event, records relevant booster information in detail, is to analyze the reason of booster occurs, rule,
Or even the data basis prevented with emergency policy is formulated, the data collection carried out for burst analysis should include following a few class numbers
According to：
1) pipeline burst safeguards record：It is primarily referred to as water undertaking and letter is recorded to the maintenance that the pipeline of pipe explosion accident occurs
Breath, including a system such as pipeline ID, pipe range, tubing, booster time, pipeline laying time, booster geographical location and maintenance measure
The data of row related data, collection are more detailed, and booster Producing reason and relevant booster factor are also more clear；
2) pipeline network GIS data：It is primarily referred to as the static data of public supply mains, including the ID of each pipeline, tubing and pipe
The data such as self structures data and the position of each water factory or booster station such as long；
3) SCADA monitoring data：It is primarily referred to as the Real-time Monitoring Data of each pressure monitoring point in pipe network, may therefrom be obtained
Real-time pressure data when to pipeline burst；
4) other data：It is primarily referred to as tube circumference environmental information, such as soil corrosion index, the pipeline around pipeline laying
Ground load or data, these data such as economic development level all have certain relationship with pipeline burst, collect these data
More accurate booster risk forecast model can be established.
After being collected into booster historical data, the data that wrong false information and record repeat in data are removed, by certain
Pretreatment after establish corresponding booster database.
To a certain extent, the booster risk in the intensive place of booster point, pipeline is also higher, therefore step (2) is adopted
Space cluster analysis is carried out to booster point with DBSCAN clustering algorithms, booster point sets in space length to that whether there is or not other is quick-fried
The pipe quantized result covariant new as one, specifically：When there are other boosters in booster point setting space length, then quantify
As a result it is 1, when without other boosters, then quantized result is 0 in booster point setting space length.Specifically, the algorithm such as Fig. 2 institutes
Show, including following steps：
1) one point k of random selection is used as and clusters starting point from data set, from k in the neighborhood of searching data centrostigma k
The reachable object of density；
If 2) k is kernel object, all the points in its neighborhood are divided into a cluster, and will be under these point conducts
Then the candidate point of one wheel is constantly searched and extends the cluster where them from the reachable point of candidate dot density, until that can not look for
The point reachable to density, these points found are a complete cluster；
It is if 3) k is not kernel object, i.e., reachable from k density without, then k is temporarily labeled as noise spot.Then, right
Next point in entire data set repeats the above process, after all objects in data set were investigated, a cluster
Extension just complete；
If 4) there is also not processed point in data set, the extension of another cluster is carried out；Otherwise, by data set
In be not belonging to the point of any cluster and be determined as noise spot.
It can be quantified the input variable as booster risk model after the completion of cluster according to cluster result.
Step (3) is specially：
(31) the booster historical data standardization in booster database：Every booster pipeline historical record is divided into three portions
Point,
Wherein, DiIndicate i-th booster pipeline historical record, tiIndicate i-th booster pipeline life span, ZiIndicate i-th
Covariant vector corresponding to booster pipeline, δiIndicate the data type indicator variable of i-th booster pipeline, δi=1 indicates the
I booster pipeline historical record is complete data, δi=0 indicates that i-th booster pipeline historical record is Random censorship, i=1,
2 ... ..., N, N indicate booster pipeline sum；
(32) baseline risk function is established：Baseline risk function is quick-fried with pipeline using booster pipeline life span as independent variable
Manage-style is nearly dependent variable, and the pipeline burst risk is specially the booster number of annual unit pipe range, pipeline burst risk list
Position is：Secondary/(akm)；
(33) choose the principal element for influencing pipeline burst risk as covariant, establish Bayesian survival analysis model into
Row parameter Estimation determines covariant estimates of parameters；
(34) booster risk forecast model is determined according to baseline risk function and covariant estimates of parameters.
Step (33) is specially：
(331) life span of all booster pipelines is divided into 0 ＜ s of section1＜ s2＜ sJ＜ ∞ remember i-th
The life span of booster pipeline is ti, and to all tiBoth less than sJ, obtain J time interval (0, s1],(s1,
s2],···,(sJ-1,sJ]；
(332) assume that there are one fixed benchmark dangerous function h for each time interval0(ti)=λj, ti∈(sj-1,
sj-1), obtain Bayesian survival analysis model function：
Wherein, h (ti,Zi) be i-th pipeline booster risk, tiIndicate i-th pipeline life span, ZiIndicate i-th
Covariant vector corresponding to pipeline, βTFor the corresponding regression coefficient vector of covariant, λjIt is dangerous for the benchmark of each time interval
Function, j=1,2 ... ..., J, J are the time interval sum divided, i=1,2 ... ..., N, N expression booster pipeline sums.
Booster risk model is in step (34)：
H (t, Z)=(at2+bt+c)exp(βTZ)；
Wherein, h (t, Z) is pipeline burst risk, and t indicates pipeline life span, at2Risk function on the basis of+bt+c, a,
The fitting coefficient of risk function on the basis of b and c, Z indicate that the covariant for influencing the principal element composition of pipeline burst risk is vectorial,
βTFor the corresponding regression coefficient vector of covariant.
After the completion of booster risk forecast model structure, in conjunction with the GIS pipe network informations of practical water supply network, you can calculate and supply water
The booster risk of each pipeline in pipe network, and pipeline burst risk distribution figure is drawn according to result of calculation.It can be according to pipeline burst wind
Pipeline burst danger classes is divided into high-risk, danger, low dangerous and four grades of safety by the result of calculation of danger successively.Meanwhile
In order to verify the reasonability of prediction result, the present invention carries out spatial analysis contrast verification using prediction result and practical booster point
The forecasting accuracy of model.
It is directed to the present embodiment, specifically：
(1) collection and pretreatment of booster historical data
For the example pipe network, it is collected into 449 original booster historical records and the GIS pipes of the water supply network in total
Net establishes corresponding booster database according to these booster historical datas, main caliber, tubing, the pipe for including booster pipeline
Long, 23 fields such as lay time, the time of being informed of a case.Remove that field record is imperfect in original booster record or artificial damage causes
The record of booster obtains altogether 275 booster historical records that can be used for modeling.
(2) space cluster analysis of booster point
According to the characteristic of spatial distribution of booster point, the present invention takes the distance threshold Eps=100m of DBSCAN clustering algorithms
With density threshold MinPts=2, clustering is carried out to booster point and obtains cluster result.In the cluster knot of quantization booster point
When fruit, according to booster point 100m, whether there is or not other booster points to carry out assignment, and using assigned result as booster risk profile mould
The covariant of type.The cluster assignment of every booster historical record indicates that then assignment formula such as following formula is indicated with variable Cluster：
(3) booster risk forecast model is built
Booster historical record after having pre-processed is organized into reference format, and establishes corresponding baseline risk function.Root
The characteristics of according to booster historical record, chooses caliber D and this two principal element of cluster assignment Cluster as influence pipeline burst wind
The covariant of danger, meanwhile, the prior distribution being uniformly distributed as caliber D being chosen in (- 1,0) section.And cluster is assigned
Value Cluster, it is assumed that the covariant regression coefficient of cluster assignment Cluster meets normal distribution.Each covariant regression coefficient
Prior distribution is as shown in following formula：
f(β1)~U (- 1,0),
In formula, f (β1) and f (β2) be respectively covariant D and Cluster regression coefficient prior distribution, β1And β2Respectively
For the corresponding regression coefficient of covariant D and Cluster, Bayesian survival can be used after determining the prior distribution of each covariant
Analysis method acquires the coefficient of each covariant.In addition, in order to verify the convergence of Bayesian model, using two MCMC chains, respectively
The initial value of covariant regression coefficient is respectively (β in chainCluster=0, βD=0) and (βCluster=0.5, βD=-0.5) it, carries out
After 2000 pre- iteration, then the Posterior distrbutionp that 10000 iteration can be obtained each covariant regression coefficient is carried out, and posteriority is equal
Value finally obtains booster risk forecast model as final parameter estimation result.
(4) the booster risk of booster risk forecast model prediction pipeline is applied
In conjunction with GIS pipe networks, the booster risk of each pipeline in water supply network is calculated using the booster risk forecast model of structure.
In this example, booster risk be more than 0.04 time/(akm) be high-risk pipeline, booster risk is more than and 0.02 and is less than or equal to
0.04 time/(akm) be dangerous pipeline, booster risk be more than 0.01 and less than or equal to 0.02 time/(akm) be low dangerous manage
Road, booster risk be less than or equal to 0.01 time/(akm) then be safety corridor.
The above embodiment is only to enumerate, and does not indicate that limiting the scope of the invention.These embodiments can also be with other
Various modes are implemented, and can make in the range of not departing from technical thought of the invention it is various omit, displacement, change.
Claims (6)
1. a kind of water supply network pipeline burst Risk Forecast Method based on Bayesian survival analysis, which is characterized in that this method
Include the following steps：
(1) booster database is established according to the booster historical data of collection, extraction key message is as covariant；
(2) space cluster analysis is carried out to booster point, the space distribution information of booster point is quantified to the association new as one
Variable is supplemented in booster database；
(3) it is based on booster database and booster risk forecast model is built using Bayesian survival analysis method；
(4) the booster risk of booster risk forecast model prediction pipeline is used.
2. a kind of water supply network pipeline burst risk profile side based on Bayesian survival analysis according to claim 1
Method, which is characterized in that step (2) sets booster point new as one whether there is or not other booster quantized results in space length
Covariant, specifically：When having other boosters in booster point setting space length, then quantized result is 1, when booster point is set
Without other boosters in space length, then quantized result is 0.
3. a kind of water supply network pipeline burst risk profile side based on Bayesian survival analysis according to claim 1
Method, which is characterized in that step (3) is specially：
(31) the booster historical data standardization in booster database：Every booster pipeline historical record is divided into three parts,
Wherein, DiIndicate i-th booster pipeline historical record, tiIndicate i-th booster pipeline life span, ZiExpression i-th is quick-fried
Covariant vector corresponding to pipeline, δiIndicate the data type indicator variable of i-th booster pipeline, δi=1 indicates i-th
Booster pipeline historical record is complete data, δi=0 indicates that i-th booster pipeline historical record is Random censorship, i=1,
2 ... ..., N, N indicate booster pipeline sum；
(32) baseline risk function is established：Baseline risk function is using booster pipeline life span as independent variable, with pipeline burst wind
Danger is dependent variable, and the pipeline burst risk is specially the booster number of annual unit pipe range；
(33) principal element for influencing pipeline burst risk is chosen as covariant, is established Bayesian survival analysis model and is joined
Number estimation determines covariant estimates of parameters；
(34) booster risk forecast model is determined according to baseline risk function and covariant estimates of parameters.
4. a kind of water supply network pipeline burst risk profile side based on Bayesian survival analysis according to claim 3
Method, which is characterized in that step (33) is specially：
(331) life span of all booster pipelines is divided into 0 ＜ s of section1＜ s2＜ sJ＜ ∞ remember i-th of booster
The life span of pipeline is ti, and to all tiBoth less than sJ, obtain J time interval (0, s1],(s1,s2],···,
(sJ-1,sJ]；
(332) assume that there are one fixed benchmark dangerous function h for each time interval0(ti)=λj, ti∈(sj-1,sj-1), it obtains
To Bayesian survival analysis model function：
Wherein, h (ti,Zi) be i-th pipeline booster risk, tiIndicate i-th pipeline life span, ZiIndicate i-th pipeline
Corresponding covariant vector, βTFor the corresponding regression coefficient vector of covariant, λjFor the benchmark danger letter of each time interval
Number, j=1,2 ... ..., J, J are the time interval sum divided, i=1,2 ... ..., N, N expression booster pipeline sums.
5. a kind of water supply network pipeline burst risk profile side based on Bayesian survival analysis according to claim 1
Method, which is characterized in that booster risk model is in step (34)：
H (t, Z)=(at2+bt+c)exp(βTZ)；
Wherein, h (t, Z) is pipeline burst risk, and t indicates pipeline life span, at2Risk function on the basis of+bt+c, a, b and c
On the basis of risk function fitting coefficient, Z indicate influence pipeline burst risk principal element composition covariant vector, βTFor
The corresponding regression coefficient vector of covariant.
6. a kind of water supply network pipeline burst risk profile based on Bayesian survival analysis according to claim 1 or 2
Method, which is characterized in that space cluster analysis is carried out to booster point using DBSCAN clustering algorithms.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810448471.7A CN108615098B (en) | 2018-05-11 | 2018-05-11 | Bayesian survival analysis-based water supply network pipeline pipe burst risk prediction method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810448471.7A CN108615098B (en) | 2018-05-11 | 2018-05-11 | Bayesian survival analysis-based water supply network pipeline pipe burst risk prediction method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108615098A true CN108615098A (en) | 2018-10-02 |
| CN108615098B CN108615098B (en) | 2021-11-09 |
Family
ID=63663052
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810448471.7A Active CN108615098B (en) | 2018-05-11 | 2018-05-11 | Bayesian survival analysis-based water supply network pipeline pipe burst risk prediction method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108615098B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886506A (en) * | 2019-03-14 | 2019-06-14 | 重庆大学 | A kind of water supply network booster risk analysis method |
| CN110705018A (en) * | 2019-08-28 | 2020-01-17 | 泰华智慧产业集团股份有限公司 | Water supply pipeline pipe burst positioning method based on hot line work order and pipeline health assessment |
| CN115293656A (en) * | 2022-10-08 | 2022-11-04 | 西南石油大学 | Parallel oil and gas pipeline domino effect risk analysis method based on Bayesian network |
| CN116090605A (en) * | 2022-12-07 | 2023-05-09 | 南栖仙策(南京)科技有限公司 | Pipe network early warning method and device, electronic equipment and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060002412A1 (en) * | 2004-06-30 | 2006-01-05 | Bijoy Bose | Methods and apparatus for supporting programmable burst management schemes on pipelined buses |
| CN102222169A (en) * | 2011-06-21 | 2011-10-19 | 天津大学 | Method for predicting and analyzing pipe burst of urban water supply network |
| CN103226741A (en) * | 2013-05-10 | 2013-07-31 | 天津大学 | Urban water supply network tube explosion prediction method |
| CN107145691A (en) * | 2017-06-23 | 2017-09-08 | 广东青藤环境科技有限公司 | A kind of public supply mains booster prediction analysis method |
| US20180080812A1 (en) * | 2017-07-25 | 2018-03-22 | University Of Electronic Science And Technology Of China | Distributed optical fiber sensing signal processing method for safety monitoring of underground pipe network |
-
2018
- 2018-05-11 CN CN201810448471.7A patent/CN108615098B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060002412A1 (en) * | 2004-06-30 | 2006-01-05 | Bijoy Bose | Methods and apparatus for supporting programmable burst management schemes on pipelined buses |
| CN102222169A (en) * | 2011-06-21 | 2011-10-19 | 天津大学 | Method for predicting and analyzing pipe burst of urban water supply network |
| CN103226741A (en) * | 2013-05-10 | 2013-07-31 | 天津大学 | Urban water supply network tube explosion prediction method |
| CN107145691A (en) * | 2017-06-23 | 2017-09-08 | 广东青藤环境科技有限公司 | A kind of public supply mains booster prediction analysis method |
| US20180080812A1 (en) * | 2017-07-25 | 2018-03-22 | University Of Electronic Science And Technology Of China | Distributed optical fiber sensing signal processing method for safety monitoring of underground pipe network |
Non-Patent Citations (4)
| Title |
|---|
| TAO TAO ET AL.: ""a pipe network skeleton method based on GIS network analysis technologis"", 《ASCE LIBRARY》 * |
| 张忠贵 等: ""一种通用的生命线工程网络事件空间聚类分析算法"", 《灾害学》 * |
| 徐哲等: "数据驱动的城市供水管网异常事件侦测方法", 《浙江大学学报(工学版)》 * |
| 王祎 等: ""供水管网系统爆管预测"", 《天津大学学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109886506A (en) * | 2019-03-14 | 2019-06-14 | 重庆大学 | A kind of water supply network booster risk analysis method |
| CN109886506B (en) * | 2019-03-14 | 2023-05-23 | 重庆大学 | Water supply network pipe explosion risk analysis method |
| CN110705018A (en) * | 2019-08-28 | 2020-01-17 | 泰华智慧产业集团股份有限公司 | Water supply pipeline pipe burst positioning method based on hot line work order and pipeline health assessment |
| CN110705018B (en) * | 2019-08-28 | 2023-03-10 | 泰华智慧产业集团股份有限公司 | Water supply pipeline pipe burst positioning method based on hot line work order and pipeline health assessment |
| CN115293656A (en) * | 2022-10-08 | 2022-11-04 | 西南石油大学 | Parallel oil and gas pipeline domino effect risk analysis method based on Bayesian network |
| CN116090605A (en) * | 2022-12-07 | 2023-05-09 | 南栖仙策(南京)科技有限公司 | Pipe network early warning method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108615098B (en) | 2021-11-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108894282B (en) | Municipal pipe network operation safety dynamic early warning method | |
| CN108615098A (en) | Water supply network pipeline burst Risk Forecast Method based on Bayesian survival analysis | |
| CN109583520B (en) | State evaluation method of cloud model and genetic algorithm optimization support vector machine | |
| CN117236673B (en) | Urban water network multi-scale flood control and drainage combined optimization scheduling method and system | |
| CN113971463A (en) | Heat supply pipeline risk distribution analysis method and routing inspection path planning system | |
| CN110929359B (en) | Pipe network siltation risk prediction modeling method based on PNN neural network and SWMM technology | |
| CN105678481A (en) | Pipeline health state assessment method based on random forest model | |
| CN107133400A (en) | A kind of aircraft structure fatigue reliability Bayes's combination forecasting method | |
| CN103793854A (en) | Multiple combination optimization overhead transmission line operation risk informatization assessment method | |
| CN108536926A (en) | Based on the corrosion oil-gas pipeline Residual Life method for improving adaptive GEV distributions | |
| CN110516883B (en) | Water supply pipe network region leakage prediction method based on space metering model | |
| CN103530818A (en) | Water supply pipe network modeling method based on BRB (belief-rule-base) system | |
| CN109307159A (en) | A kind of pipe network model alarm method based on water consumption optimal prediction model | |
| CN111859779B (en) | Method and device for early warning of third party construction damage risk of gas pipe network | |
| CN109886506B (en) | Water supply network pipe explosion risk analysis method | |
| CN106759546B (en) | Based on the Deep Foundation Distortion Forecast method and device for improving multivariable grey forecasting model | |
| CN112101811A (en) | Water supply network pipe explosion risk prediction method and system | |
| CN112330187A (en) | Underground facility flood risk assessment method, system, terminal and storage medium | |
| CN115654381A (en) | Water supply pipeline leakage detection method based on graph neural network | |
| Han et al. | An Online safety monitoring system of hydropower station based on expert system | |
| Wu et al. | Local and global Bayesian network based model for flood prediction | |
| CN109858732A (en) | A kind of urban water supply pipe network health status evaluation method | |
| CN116989277A (en) | Dynamic water supply network leakage risk assessment method based on Adaboost and BP neural network | |
| CN112434887B (en) | Water supply network risk prediction method combining network kernel density estimation and SVM | |
| CN114462688A (en) | Tube explosion detection method based on LSTM model and dynamic threshold determination algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |