CN109800525A - Liquefied natural gas warehouse cause of accident inversion method and device - Google Patents
Liquefied natural gas warehouse cause of accident inversion method and device Download PDFInfo
- Publication number
- CN109800525A CN109800525A CN201910086465.6A CN201910086465A CN109800525A CN 109800525 A CN109800525 A CN 109800525A CN 201910086465 A CN201910086465 A CN 201910086465A CN 109800525 A CN109800525 A CN 109800525A
- Authority
- CN
- China
- Prior art keywords
- library
- confidence level
- accident
- transition
- threshold value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of liquefied natural gas warehouse cause of accident inversion method and devices.The described method includes: the reason of being occurred according to liquefied natural gas warehouse heat insulation layer failure accidents chain, determine that accident obscures variable weight model, it includes the multiple transition in one's power of multiple libraries in variable weight model that accident, which obscures, and according to multiple library institutes, determines and close up to set of library, closed up to set of library conjunction and adjacent set of library immediately；According to expert estimation determine library initial confidence level, and according to liquefied natural gas warehouse heat insulation layer failure accidents historical data determine transition confidence level and transition ignition threshold value；According to library initial confidence level, transition confidence level and transition ignition threshold value, determine library final confidence level；Screening immediately up to set of library close in final confidence level be greater than the accident library of transition ignition threshold value in one's power accident library sequence, it determines the final maximum accident library institute of confidence level in the institute of accident library, ignition threshold value is changed according to the maximum accident library of final confidence level in one's power and determines accident path.
Description
Technical field
The present invention relates to liquefied natural gas warehouse cause of accident inversion technique field, espespecially a kind of liquefied natural gas deposits
Library cause of accident inversion method and device.
Background technique
As natural gas is increasingly becoming new century irreplaceable one of the energy, usage amount and amount of storage increasingly increase,
Safety is paid more and more attention.Liquefied natural gas warehouse stores a large amount of liquefied natural gas, is that factory safety requirement is highest
One of place.The low temperature storage of liquefied natural gas is most important storage mode, and the quality of storage tank cold insulation performance can be direct
Influence the generation that reserves even result in superpressure explosion accident.Therefore, LNG (liquefied natural gas) warehouse is insulated
Layer failure accidents have become very important one of accident.As heavy construction facility, LNG storage tank has the characteristics that ultralow temperature,
Once heat insulation layer failure accidents occur, LNG heat absorption evaporation may will lead to superpressure explosion in storage tank.Therefore to storage tank heat insulation layer
The inverting of the reason of failure accidents and positioning are anticipated for ensureing that liquefied natural gas is safely and effectively stored with extremely important reality
Justice.
The accident inversion research for carrying out liquefied natural gas warehouse at present is less, and general common inversion method has finite element
Emulation, GST-MQ Radial Basis Function Method, neural network model etc., finite element simulation method are only applicable to for accident process
Reappear the intuitive displaying with consequence；Basis function method is suitable for convection-diffusion effect accident problem；And neural network method can not be comprehensive
Analyze the dynamic reason factor including people because including.Although common these types method has analysis effect well in certain field
Fruit, but can not obtain satisfactorily analyzing result in terms of dynamic modeling and reasoning by analogy.
Summary of the invention
In order to solve the problems, such as to analyze result inaccuracy in existing accident inversion technology, the embodiment of the present invention provides a kind of liquid
Change gas storage library cause of accident inversion method, which comprises
The reason of being occurred according to liquefied natural gas warehouse heat insulation layer failure accidents chain, determines that accident obscures variable weight model,
It includes the multiple transition in one's power of multiple libraries in variable weight model that the accident, which obscures, and according to multiple library institutes, determination is reachable immediately
Set of library is closed, is closed up to set of library conjunction and adjacent set of library；
According to expert estimation determine the library initial confidence level, and according to the liquefied natural gas warehouse heat insulation layer
Failure accidents historical data determines transition confidence level and transition ignition threshold value；
According to the library initial confidence level, the transition confidence level and the transition ignition threshold value, determine the library
Final confidence level；
The screening accident library for being greater than the transition ignition threshold value up to final confidence level described in set of library conjunction immediately
In one's power accident library sequence, determine the final maximum accident library institute of confidence level in the institute of the accident library, finally set according to described
The transition ignition threshold value determines accident path in one's power in reliability maximum accident library.
Optionally, in an embodiment of the present invention, it is described according to the library initial confidence level, the transition confidence level
And the transition ignition threshold value, determine the library final confidence level include: the initialization library initial confidence level, institute
State transition confidence level and the corresponding input matrix of the transition ignition threshold value and input vector；Utilize the input matrix and input
Vector determines the equivalent inpnt confidence level of each transition；The equivalent inpnt removed less than the transition ignition threshold value is set
Reliability, according to the equivalent inpnt confidence level after removing, determine the library output confidence level；According to the output confidence
Degree updates the library institute, with the determination library final confidence level.
Optionally, in an embodiment of the present invention, described that the library institute is updated according to the output confidence level, to determine
State library final confidence level include: that the library institute is updated according to the output confidence level, determine changeable weight input matrix；Root
According to changeable weight input matrix, reuses the output confidence level and update the library institute, until the output confidence level is not sent out
Changing, output confidence level at this time for the library final confidence level.
Optionally, in an embodiment of the present invention, described described in one's power according to the final maximum accident library of confidence level
Transition ignition threshold value determines that accident path includes: the institute's confidence level of library described in comparison and the transition ignition threshold value size, selects institute
The accident path that library institute's confidence level is greater than the transition ignition threshold value is stated, and when the accident path is at least two
When, judge whether the logical relation in the accident path meets preset requirement；If so, judging that the accident path is corresponding most
Whether the minimum value of whole confidence level is greater than the transition ignition threshold value, if so, determining that the accident path is that accident can occur
Path.
The embodiment of the present invention also provides a kind of liquefied natural gas warehouse cause of accident inverting device, and described device includes:
Model building module, for determining according to chain the reason of the generation of liquefied natural gas warehouse heat insulation layer failure accidents
Accident obscures variable weight model, and it includes the multiple transition in one's power of multiple libraries that the accident, which obscures in variable weight model, and according to multiple described
Library institute determines and closes up to set of library, closes immediately up to set of library conjunction and adjacent set of library；
Threshold determination module, for according to expert estimation determine the library initial confidence level, and according to the liquefaction
Gas storage library heat insulation layer failure accidents historical data determines transition confidence level and transition ignition threshold value；
Confidence determination module, for according to the library initial confidence level, the transition confidence level and the transition
Ignition threshold value, determine the library final confidence level；
Cause of accident determining module, it is described immediately up to final confidence level described in set of library conjunction greater than described for screening
The sequence for changing the accident library of ignition threshold value accident library institute in one's power determines in the institute of the accident library the finally maximum accident of confidence level
Library institute, according to the final maximum accident library of confidence level, the transition ignition threshold value determines accident path in one's power.
Optionally, in an embodiment of the present invention, the confidence determination module includes: initialization unit, for initial
Change the library initial confidence level, the transition confidence level and the corresponding input matrix of the transition ignition threshold value and input to
Amount；Equivalent inpnt confidence level unit determines the equivalent inpnt of each transition for utilizing the input matrix and input vector
Confidence level；Clearing cell, for removing the equivalent inpnt confidence level for being less than the transition ignition threshold value, after removing
The equivalent inpnt confidence level, determine the library output confidence level；Final confidence level unit, for being set according to the output
Reliability updates the library institute, with the determination library final confidence level.
Optionally, in an embodiment of the present invention, the final confidence level unit includes: changeable weight subelement, is used for
The library institute is updated according to the output confidence level, determines changeable weight input matrix；Final confidence level subelement, is used for basis
Changeable weight input matrix reuses the output confidence level and updates the library institute, until the output confidence level does not occur
Variation, output confidence level at this time for the library final confidence level.
Optionally, in an embodiment of the present invention, the cause of accident determining module includes: path selection unit, is used for
Compare library institute's confidence level and the transition ignition threshold value size, library institute's confidence level is selected to be greater than transition igniting threshold
The accident path of value, and when the accident path is at least two, judge that the logical relation in the accident path is
It is no to meet preset requirement；If so, judging whether the minimum value of the corresponding final confidence level in the accident path is greater than the change
Ignition threshold value is moved, if so, determining that the accident path is that accident path can occur.
The embodiment of the present invention also provides a kind of computer equipment, including memory, processor and storage are on a memory simultaneously
The computer program that can be run on a processor, the processor perform the steps of when executing the computer program
The reason of being occurred according to liquefied natural gas warehouse heat insulation layer failure accidents chain, determines that accident obscures variable weight model,
It includes the multiple transition in one's power of multiple libraries in variable weight model that the accident, which obscures, and according to multiple library institutes, determination is reachable immediately
Set of library is closed, is closed up to set of library conjunction and adjacent set of library；
According to expert estimation determine the library initial confidence level, and according to the liquefied natural gas warehouse heat insulation layer
Failure accidents historical data determines transition confidence level and transition ignition threshold value；
According to the library initial confidence level, the transition confidence level and the transition ignition threshold value, determine the library
Final confidence level；
The screening accident library for being greater than the transition ignition threshold value up to final confidence level described in set of library conjunction immediately
In one's power accident library sequence, determine the final maximum accident library institute of confidence level in the institute of the accident library, finally set according to described
The transition ignition threshold value determines accident path in one's power in reliability maximum accident library.
The embodiment of the present invention also provides a kind of computer readable storage medium, is stored thereon with computer program, the meter
Calculation machine program performs the steps of when being executed by processor
The reason of being occurred according to liquefied natural gas warehouse heat insulation layer failure accidents chain, determines that accident obscures variable weight model,
It includes the multiple transition in one's power of multiple libraries in variable weight model that the accident, which obscures, and according to multiple library institutes, determination is reachable immediately
Set of library is closed, is closed up to set of library conjunction and adjacent set of library；
According to expert estimation determine the library initial confidence level, and according to the liquefied natural gas warehouse heat insulation layer
Failure accidents historical data determines transition confidence level and transition ignition threshold value；
According to the library initial confidence level, the transition confidence level and the transition ignition threshold value, determine the library
Final confidence level；
The screening accident library for being greater than the transition ignition threshold value up to final confidence level described in set of library conjunction immediately
In one's power accident library sequence, determine the final maximum accident library institute of confidence level in the institute of the accident library, finally set according to described
The transition ignition threshold value determines accident path in one's power in reliability maximum accident library.
The present invention is by fuzzy message quantification, to achieve the purpose that accurate inverting.According to specific crash analysis and accident phase
The installations and facilities of pass establish the inverse model for being more bonded accident scene, thus more accurate carry out inverting positioning.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, embodiment will be described below
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is a kind of flow chart of liquefied natural gas warehouse cause of accident inversion method of the embodiment of the present invention；
Fig. 2 is that variable weight Petri net model schematic diagram is obscured in the embodiment of the present invention；
Fig. 3 is a kind of structural schematic diagram of liquefied natural gas warehouse cause of accident inverting device of the embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
There are three clear advantages for fuzzy variable weight Petri net model tool: 1) it is a kind of patterned modeling and analysis tool,
Accident-causing factor can intuitively be shown；2) it can be by Fuzzy Data Fusion into Petri network, so that the accident of analysis
Reason factor is more comprehensively；3) it can dynamically assign weight in same inference rule, solve condition importance and independence,
It is also suitable under partial condition anything definite.However, liquefied natural gas deposit library facilities are complicated, the generations of different accidents may be because
For different equipment or human factor, therefore also need according to specific crash analysis and accident relevant device reason factor, specifically
It is comprehensive to carry out Inverse modeling.
It is as shown in Figure 1 a kind of flow chart of liquefied natural gas warehouse cause of accident inversion method of the embodiment of the present invention,
The reason of method as shown in the figure includes: step S1, occurred according to liquefied natural gas warehouse heat insulation layer failure accidents chain, determines thing
Therefore fuzzy variable weight model, it includes the multiple transition in one's power of multiple libraries that the accident, which obscures in variable weight model, and according to multiple libraries
Institute determines and closes up to set of library, closes immediately up to set of library conjunction and adjacent set of library；
Step S2, according to expert estimation determine the library initial confidence level, and according to the liquefied natural gas deposit
Library heat insulation layer failure accidents historical data determines transition confidence level and transition ignition threshold value；
Step S3, according to the library initial confidence level, the transition confidence level and the transition ignition threshold value, determine
The library final confidence level；
Step S4, screening is described to be greater than the transition ignition threshold value up to final confidence level described in set of library conjunction immediately
Accident library in one's power accident library sequence, the final maximum accident library institute of confidence level in the institute of the accident library is determined, according to described
The transition ignition threshold value determines accident path in one's power in the final maximum accident library of confidence level.
As an embodiment of the present invention, it is described according to the library initial confidence level, the transition confidence level and
The transition ignition threshold value, determine the library final confidence level include: the initialization library initial confidence level, described
Change confidence level and the corresponding input matrix of the transition ignition threshold value and input vector；Using the input matrix and input to
Amount, determines the equivalent inpnt confidence level of each transition；Remove the equivalent inpnt confidence for being less than the transition ignition threshold value
Degree, according to the equivalent inpnt confidence level after removing, determine the library output confidence level；According to the output confidence level
Update the library institute, with the determination library final confidence level.
In the present embodiment, described that the library institute is updated according to the output confidence level, with the determination library it is final
Confidence level includes: to update the library institute according to the output confidence level, determines changeable weight input matrix；It is defeated according to changeable weight
Enter matrix, reuse the output confidence level update library institute, until the output confidence level does not change, at this time
Export confidence level for the library final confidence level.
As an embodiment of the present invention, described according to the final maximum accident library of the confidence level transition in one's power
Ignition threshold value determines that accident path includes: the institute's confidence level of library described in comparison and the transition ignition threshold value size, selects the library
Institute's confidence level is greater than the accident path of the transition ignition threshold value, and when the accident path is at least two, sentences
Whether the logical relation in the accident path of breaking meets preset requirement；It is finally set if so, judging that the accident path is corresponding
Whether the minimum value of reliability is greater than the transition ignition threshold value, if so, determining that the accident path is that accident path can occur.
In a specific embodiment of the invention, in order to overcome the problems, such as that existing research exists, the present invention analyzes first to be caused
All reason factors that liquefied natural gas warehouse heat insulation layer failure accidents occur, and then reason factor is combined into fuzzy variable weight
Petri network establishes heat insulation layer failure accidents inverting qualitative model；Then each library institute is analyzed according to the inverse model established
Immediately up to set of library close, up to set of library close and adjacent set of library close, this be carry out analysis of causes reasoning basis；It connects
Combine factory's status to carry out expert estimation to each reason factor (library institute) and determine transition confidence level and ignition threshold value；Further according to
Matrix inference method makes inferences to obtain the final confidence level of library institute；Finally combine the qualitative inverse model of Petri network and library institute confidence
The reason of degree and ignition threshold value reasoning lead to heat insulation layer failure accidents and path.This method combines fuzzy variable weight Petri net model
With matrix reasoning algorithm, cause of accident inverse model is established.The model for accident have the complete analysis of causes in more detail and
Reasoning process, thus more accurately inverting localized accident reason.
Specifically, a kind of liquefied natural gas warehouse cause of accident inverting inference method provided in an embodiment of the present invention, institute
Stating method includes 3 modules:
(1) LNG warehouse accident inversion qualitative model is constructed, the specific steps are 1) clearly analyze liquefied natural gas warehouse
The reason of heat insulation layer failure accidents occur establishes liquefied natural gas warehouse heat insulation layer in conjunction with fuzzy variable weight Petri net model and loses
Imitate accident inversion qualitative model.Constantly analysis improves refinement heat insulation layer failure accidents reason factor in establishment process；2) it establishes
All libraries immediately up to set of library close, up to set of library close and adjacent set of library close, as carry out qualitative reasoning it is specifically former
The basis of cause.
(2) LNG warehouse accident inversion model confidence reasoning.Library institute is determined according to expert estimation the specific steps are 1) bright
Confidence level determines transition confidence level according to field condition, determines transition ignition threshold value according to risk severity；2) matrix is initialized
Each input matrix and vector of reasoning algorithm；3) the equivalent inpnt confidence level of each transition is calculated；4) by equivalent inpnt confidence level
With transition threshold value comparison, and remove confidence level be less than transition threshold value equivalent inpnt；5) calculate activation transition output library
Confidence level；6) update all libraries Tokken value；7) changeable weight input matrix is obtained according to M_ (k+1).
(3) reasoning cause of accident and generation path, 1) specifically including leads to liquefied natural gas warehouse heat insulation layer failure thing
Therefore reason；2) lead to the specific path of liquefied natural gas warehouse heat insulation layer failure accidents.
Specifically, the specific implementation step of 3 modules of liquefied natural gas warehouse cause of accident inversion method and device
As follows.
Stage 1: building LNG warehouse accident inversion qualitative model
Step 1: analysis cause of accident simultaneously establishes dynamic variable weight Fuzzy Petri Nets Model
The reason of analysis and summary causes LNG warehouse heat insulation layer failure accidents to occur chain establishes the failure of LNG storage tank heat insulation layer
Accident Fuzzy Petri Net qualitative model, and definition storehouse institute pi, transition tn.This is to carry out LNG warehouse heat insulation layer failure accidents original
Because of the basis of positioning analysis.
Step 2: establishing three kinds of set of library and close
Establish all libraries immediately up to set of library close IRS (pi), up to set of library close RS (pi), adjacent set of library close
AP(pij)。
There is following library to be defined in fuzzy variable weight Petri net model, this is helpful to last reasoning localized reason.
Define 1: if tnIt is an arbitrary transition, p in Petri net modeli, pkFor two library institutes for connecting the transition, if
pi∈I(tn), pk∈O(tn), then claim pkIt is piImmediately up to library institute, i.e. expression piBy tnThe i.e. reachable p of primary transitionkLibrary
Institute, and piThrough any one all library that can reach of transition set be just piImmediately up to set of library close, be expressed as
IRS(pi).In this method, in the reasoning reason stage, in conjunction with the possible thing that can analyze can be quickly positioned up to library immediately
Therefore path.
Define 2: by piSet composed by a series of all libraries institute that can be reached after transition connected to it occur
Referred to as piReachable set of library close, be expressed as RS (pi).In this method, the identification closed up to set of library can be accurately positioned and divide
Analysis is by piA series of generation of caused accident factors and the generation of final damage sequence.
Define 3: if pi∈I(tn), pk∈O(tn), i.e. piWith pjEqually through same transition tnSame output library can be reached
Institute, then claim piWith piFor about transition tnAdjacent library institute；All piAdjacent library composed by collection be collectively referred to as piAdjacent library institute
Set, is expressed as AP (pii).In this method, if all adjacent library institutes in some library, judge that it whether can up to library immediately
The premise of generation is that its adjacent library institute and this library can result in it and occurred immediately up to library.
Define 4: if for library institute pi, there is no transition tnIt is set to meet condition pi∈O(tn), i.e. piOnly directed arc is directed toward
When changing and being directed toward it without transition, then claim piFor the initial storehouse institute of the Fuzzy Petri Net.
Define 5: if for library institute pi, there is no transition tnIt is set to meet condition pi∈I(tn), i.e., only directed arc is from change
It moves and is directed toward piAnd the p that has no way ofiWhen the directed arc of extraction, then claim piFor the object library institute of the Fuzzy Petri Net.LNG in the method
Tank field heat insulation layer fails in dynamic fuzzy Petri net model, and initial storehouse institute represents various causes of accident, it only exists output transition
Input transition may be not present；Object library then indicates final heat insulation layer failure accidents, it only exists input transition and may be not present
Output transition.The purpose of LNG tank area heat insulation layer failure dynamic fuzzy Petri net is exactly the accident occurred according to object library
(heat insulation layer failure) come search cause object library institute's accident occur initial storehouse institute process.
The warehouse heat insulation layer failure accidents inverse model confidence level reasoning of stage 2:LNG
Step 3: confidence data and ignition threshold value determine
The reason of causing heat insulation layer to fail represented by library of the present invention element, confidence alpha (p corresponding to libraryi) indicate element
The corresponding assessed value of state.Change tjThe development of expression event changes tjFuzzy factor μjA possibility that indicating development.Pass through
Expert estimation determines initial confidence level of place according to the marking rule of table 1, determines μ according to factory site situation etc.jSize.
1 expert estimation rule list of table
The threshold value that can change occur is exactly the ignition threshold value changed, and this method determines transition threshold value according to the size of risk
It is 0.6.
Step 4: carrying out matrix reasoning
1. defining Θ: C=A Θ B of comparison operator, then work as aijBijWhen, cij=aij；aij< bijWhen, cij=0；Wherein A,
B, C is m * n matrix；Multiplication operatorThenWherein A, B, C are respectively m × 1,1 × n, m
× n matrix.The two operators are the easy representation methods that matrix calculating is carried out in matrix reasoning process, for matrix operation meeting
It is more convenient.
2. each input matrix and vector of initialization algorithm.Initial changeable weight input matrix is W0, element ωijValue
See formula (1)；Confidence level output matrix is Z, element zij=μj, j=1,2 ..., m；Transition threshold vector be λ, initial marking to
Measure M0=(α0(p1), α0(p2), α0(p3) ... α0(pn))T.If k is reasoning number, k=0 is enabled.
3. calculating the equivalent inpnt confidence level of each transition.
E in formulak+1=(e1, e2..., em)T, i.e., by multiple Indistinct Inputs of same transition according to they confidence level and
Weight coefficient is equivalent to the Indistinct Input that weight is 1；MkFor n dimensional vector.
4. by equivalent inpnt confidence level and transition threshold value comparison, and removing the equivalent inpnt that confidence level is less than transition threshold value.
Gk+1=Ek+1Θλ (3)
G in formulak+1For m dimensional vector, at this point, Gk+1The only equivalent inpnt confidence level comprising transition can be made to activate.
5. calculate activation transition output library confidence level.
M in formulak+1For n dimensional vector.
6. update all libraries Tokken value.
Mk+1=M0+Mk+1 (5)
M in formulak+1Indicate the confidence level of all propositions obtained after k+1 reasoning.
7. according to Mk+1Obtain changeable weight input matrixEnable k=k+1, repeat step 3.~6., until Mk=Mk+1,
I.e. all proposition confidence levels are no longer changed, and obtain final library institute confidence level.
Stage 3: reasoning cause of accident and generation path
According to the ignition threshold value of library institute's confidence level and transition is obtained, the LNG warehouse heat insulation layer failure accidents of foundation are compareed
Inverting dynamic fuzzy Petri net determines the path and final cause that heat insulation layer failure accidents occur.
Step 5:LNG warehouse heat insulation layer failure accidents inverting reason reasoning
1. searching all library institutes immediately up to heat insulation layer failure accidents carried out by library, screening library institute's confidence level is greater than 0.6
Then all library institutes compare its library institute confidence level size, select maximum confidence library institute.
2. search immediately up to library thus maximum confidence library library institute, 1. with step, screening confidence level be greater than 0.6
Library institute and sort, first analysis confidence level maximum library institute path, recycle until initial storehouse institute.
3. if reach certain library path have two, if path is "AND" relationship, compare min confidence on path
Library whether reach ignition threshold value, this path is if it is selected, if not then indicating that this path can not occur.
In a specific embodiment of the invention, superpressure explosion occurs for certain LNG warehouse.Because personnel issue has filled difference
Density LNG causes the storage tank that layering rolling occurs to same storage tank, and pressure inside the tank rises rapidly, and adjusts each valve of pressure
Cause to fail not in time because usually overhauling, PTIA presignal delay leads to the manual pressure release delay of personnel again, and final storage tank superpressure occurs
Explosion.
According to above-mentioned steps as to this thus establish inverse model carry out reason inverting.
Stage 1: building LNG warehouse accident inversion qualitative model
Step 1: analysis heat insulation layer failure accidents reason chain simultaneously establishes dynamic fuzzy Petri net model
Heat insulation layer failure accidents reason chain is analyzed according to heat insulation layer failure accidents forward model, and establishes qualitative fuzzy variable weight
Petri net model is as shown in Figure 2.
2 LNG storage tank heat insulation layer of table failure Fuzzy Petri Nets Model subordinate list
By Fig. 2 and table 2 it can be seen that LNG storage tank heat insulation layer failure (p1) most immediate cause may be because that tank skin is torn, absolutely
Thermosphere sedimentation, dampness, tank body corrosion failure, fire and explosion cause.
And the reason of causing tank skin to tear, may be because that designing material is unqualified；Heat insulation layer sedimentation, dampness, tank body are corroded through
Hole, which may be because, does not inspect periodically replacement storage tank insulation layer material.Fire and explosion cause is caused to have very much, immediate cause has
LNG is largely leaked, related valve fails, fire source.Other remote causes such as Fig. 2 shows.
Step 2: establishing three kinds of set of library and close
It can be established according to heat insulation layer failure accidents inverting Petri network qualitative model and close IRS (p up to set of library immediatelyi), can
RS (p is closed up to set of libraryi), adjacent set of library close AP (pij) such as table 3.
3 library of table immediately up to set of library close, up to set of library close, adjacent set of library close
The warehouse heat insulation layer failure accidents inverse model confidence level reasoning of stage 2:LNG
Step 3: confidence data and ignition threshold value determine
Live expert estimation is carried out according to the status of the factory, obtained initial confidence level of place is as shown in table 4.
4 initial confidence level of place α (p of tablei)
Historical data is summarized through consulting literatures and the factory site situation, determining transition confidence level such as table 5 show.
Table 5 changes confidence level μj
Step 4: carrying out matrix reasoning
In conjunction with the data and Petri network of preceding step, library institute confidence level is respectively obtained after MATLAB program calculation is
Shown in table 6- table 9.
6 initial confidence level of place M of table1
7 initial confidence level of place M of table2
8 initial confidence level of place M of table3
9 initial confidence level of place M of table4
To M4When confidence level no longer change, therefore table 9 is the final confidence level in library.
Stage 3: reasoning cause of accident and generation path
Step 5:LNG warehouse heat insulation layer failure accidents inverting reason reasoning
For heat insulation layer failure, the path that can directly cause heat insulation layer failure has 5, as shown in Fig. 2, being respectively
P2-p1, p20-p1, (p23, p22, p21)-p1, p24-p1, p17-p1.The strategy preferentially selected according to confidence level maximum is changed,
Only the confidence level of p2 is greater than ignition threshold value 0.6, other paths cannot cause p1.Continue to check immediately up to p2 carried out by library
Library all p3, p4, p5 and p10, the library institute confidence level p23 for path (p23, p22, p21)-p1 that wherein confidence level ranked second,
P22, p21 are below t23 ignition threshold value 0.6, it is thus impossible to cause the generation of p1.Similarly p20-p1 can not in residual paths
Cause p1.Successively Induction matrix p24-p1.Path (p3, p4)-p2 is first calculated, discovery p3 and p4 confidence level is both greater than lighted a fire
Threshold value 0.6, therefore will lead to p1；The second paths are checked, wherein only p16, p13, p11, p14, p10, p9, p7, p5
Confidence level is greater than 0.6, and therefore, it is that p4 and p3 AND gate leads to p2, p16-p14 and p13-p11 that path, which occurs, for accident
AND gate leads to p10.P10 and p9-p7-p5 AND gate also leads to p2, and then p1 occurs.
Therefore the path that accident occurs can be regarded as: because personnel safety operation consciousness weakness is filled, by different densities LNG
It resets into a storage tank and causes layering rolling, and then pressure of storage tank is caused to increase, and due to storage tank pressure relief valve and alarm for high voltage
Device lacks maintenance and causes to fail, can not timely pressure release, eventually lead to superpressure explosion accident generation.The reason of generation, is: (1) people
Member's safety fills poor consciousness；(2) layering rolling occurs for storage tank；(3) because pressure relief valve failure causes pressure release to fail；(4) factory
Inspection and repair system is unqualified, lacks too long to the maintenance of each valve of warehouse or maintenance interval.
Common accident inversion method tends not to establish more intuitive dynamic inverse model, for complication system and difference
System factor ambiguity can not accurate modeling and accurate inverting positioning, and under certain conditions deletion conditions can not to because
Plain quantification, thus cannot accurate inverting reason reason.Therefore, it is necessary to by accident inversion and fuzzy variable weight Petri net model
Combine, with intuitive patterned displaying, by fuzzy message quantification, and in reasoning dynamic change weight to reach accurate
The purpose of inverting.Also, the complexity of reason factor and the uncertain precision that will lead to inverting reduce in different accidents, because
This, needs to establish the inverting Petri network mould for being more bonded accident scene according to specific crash analysis installations and facilities relevant to accident
Type, thus more accurate carry out inverting positioning.
The present invention analyzes each factor and reason chain of liquefied natural gas accident generation first；Next establishes LNG warehouse
Heat insulation layer failure accidents inverting obscures variable weight Petri net model, then establishes each set of library and closes, with matrix reasoning and calculation library institute
Confidence level；The reason of finally combining Petri net model reasoning that heat insulation layer failure accidents is caused to occur and path.This method combination mould
Variable weight Petri net model is pasted, Inverse modeling is carried out to LNG warehouse heat insulation layer failure accidents, and combine heat insulation layer failure accidents
Feature and relevant device information make inferences inverting.Fuzzy message in heat insulation layer failure accidents is embodied as changeable weight, and
Dynamic Inference calculation is carried out, it is final precisely to determine accident occurrence cause and path occurs.
Inversion method of the invention is by fuzzy message quantification, to achieve the purpose that accurate inverting.According to specific accident point
Installations and facilities relevant to accident are analysed, the inverse model for being more bonded accident scene is established, so that more accurate carry out inverting is fixed
Position.
It is illustrated in figure 3 a kind of structural representation of liquefied natural gas warehouse cause of accident inverting device of the embodiment of the present invention
Figure, device as shown in the figure includes: model building module 10, for being occurred according to liquefied natural gas warehouse heat insulation layer failure accidents
The reason of chain, the accident that determines obscures variable weight model, and it includes the multiple transition in one's power of multiple libraries in variable weight model that the accident, which obscures, and
According to multiple library institutes, determines and close up to set of library, closed up to set of library conjunction and adjacent set of library immediately；
Threshold determination module 20, for according to expert estimation determine the library initial confidence level, and according to the liquid
Change gas storage library heat insulation layer failure accidents historical data and determines transition confidence level and transition ignition threshold value；
Confidence determination module 30, for according to the library initial confidence level, the transition confidence level and the change
Move ignition threshold value, determine the library final confidence level；
Cause of accident determining module 40, it is described immediately up to final confidence level described in set of library conjunction greater than institute for screening
The sequence for stating the accident library of transition ignition threshold value accident library institute in one's power determines in the institute of the accident library the finally maximum thing of confidence level
Therefore library institute, according to the final maximum accident library of confidence level, the transition ignition threshold value determines accident path in one's power.
As an embodiment of the present invention, the confidence determination module includes: initialization unit, for initializing
State library initial confidence level, the transition confidence level and the corresponding input matrix of the transition ignition threshold value and input vector；
Equivalent inpnt confidence level unit determines that the equivalent inpnt of each transition is set for utilizing the input matrix and input vector
Reliability；Clearing cell, for removing the equivalent inpnt confidence level for being less than the transition ignition threshold value, according to the institute after removing
State equivalent inpnt confidence level, determine the library output confidence level；Final confidence level unit, for according to the output confidence
Degree updates the library institute, with the determination library final confidence level.
In the present embodiment, the final confidence level unit includes: changeable weight subelement, for being set according to the output
Reliability updates the library institute, determines changeable weight input matrix；Final confidence level subelement, for inputting square according to changeable weight
Battle array reuses the output confidence level update library institute, until the output confidence level does not change, output at this time
Confidence level for the library final confidence level.
As an embodiment of the present invention, the cause of accident determining module includes: path selection unit, for comparing
Library institute's confidence level and the transition ignition threshold value size select library institute's confidence level to be greater than the transition ignition threshold value
The accident path, and when the accident path is at least two, judge whether the logical relation in the accident path is full
Sufficient preset requirement；If so, judging whether the minimum value of the corresponding final confidence level in the accident path is greater than the transition point
Fiery threshold value, if so, determining that the accident path is that accident path can occur.
In one particular embodiment of the present invention, a kind of liquefied natural gas warehouse cause of accident inverting of the invention dress
It sets, may also include 3 modules:
(1) LNG warehouse accident inversion model construction device.The device can be according to LNG warehouse and with heat insulation layer phase
The installations and facilities of pass, the factors such as each factor, including personnel, environment that identification causes heat insulation layer failure accidents to occur.And according to
The reason of content that analysis obtains, establishes LNG warehouse accident inversion model, and analysis and summary causes the accident path, and energy
Each set of library in inverse model is analyzed to close；
(2) confidence level matrix reasoning device.The device can be determined in conjunction with expert graded and the analysis of matrix reasoning algorithm
Library institute confidence level in model tentatively recognizes the direct library institute caused the accident；
(3) the accurate inverting reasoning device of cause of accident, the device can be searched according to library institute's confidence level and transition ignition threshold value
Rope inquires each paths, determines the reason of causing the accident, so that obtaining accident occurs concrete reason and path.
Conceived based on application identical with a kind of above-mentioned liquefied natural gas warehouse cause of accident inversion method, the present invention is also
Provide a kind of above-mentioned liquefied natural gas warehouse cause of accident inverting device.Due to a kind of liquefied natural gas warehouse accident
The principle that reason inverting device solves the problems, such as is similar to a kind of liquefied natural gas warehouse cause of accident inversion method, thus this one
The implementation of kind liquefied natural gas warehouse cause of accident inverting device may refer to a kind of liquefied natural gas warehouse cause of accident
The implementation of inversion method, overlaps will not be repeated.
Inverting device of the invention is by fuzzy message quantification, to achieve the purpose that accurate inverting.According to specific accident point
Installations and facilities relevant to accident are analysed, the inverse model for being more bonded accident scene is established, so that more accurate carry out inverting is fixed
Position.
The embodiment of the present invention also provides a kind of computer equipment, including memory, processor and storage are on a memory simultaneously
The computer program that can be run on a processor, the processor perform the steps of when executing the computer program
The reason of being occurred according to liquefied natural gas warehouse heat insulation layer failure accidents chain, determines that accident obscures variable weight model,
It includes the multiple transition in one's power of multiple libraries in variable weight model that the accident, which obscures, and according to multiple library institutes, determination is reachable immediately
Set of library is closed, is closed up to set of library conjunction and adjacent set of library；
According to expert estimation determine the library initial confidence level, and according to the liquefied natural gas warehouse heat insulation layer
Failure accidents historical data determines transition confidence level and transition ignition threshold value；
According to the library initial confidence level, the transition confidence level and the transition ignition threshold value, determine the library
Final confidence level；
The screening accident library for being greater than the transition ignition threshold value up to final confidence level described in set of library conjunction immediately
In one's power accident library sequence, determine the final maximum accident library institute of confidence level in the institute of the accident library, finally set according to described
The transition ignition threshold value determines accident path in one's power in reliability maximum accident library.
The embodiment of the present invention also provides a kind of computer readable storage medium, is stored thereon with computer program, the meter
Calculation machine program performs the steps of when being executed by processor
The reason of being occurred according to liquefied natural gas warehouse heat insulation layer failure accidents chain, determines that accident obscures variable weight model,
It includes the multiple transition in one's power of multiple libraries in variable weight model that the accident, which obscures, and according to multiple library institutes, determination is reachable immediately
Set of library is closed, is closed up to set of library conjunction and adjacent set of library；
According to expert estimation determine the library initial confidence level, and according to the liquefied natural gas warehouse heat insulation layer
Failure accidents historical data determines transition confidence level and transition ignition threshold value；
According to the library initial confidence level, the transition confidence level and the transition ignition threshold value, determine the library
Final confidence level；
The screening accident library for being greater than the transition ignition threshold value up to final confidence level described in set of library conjunction immediately
In one's power accident library sequence, determine the final maximum accident library institute of confidence level in the institute of the accident library, finally set according to described
The transition ignition threshold value determines accident path in one's power in reliability maximum accident library.
Conceived based on application identical with a kind of above-mentioned liquefied natural gas warehouse cause of accident inversion method, the present invention is also
Provide a kind of above-mentioned computer equipment and a kind of computer readable storage medium.Due to a kind of computer equipment and a kind of meter
The principle that calculation machine readable storage medium storing program for executing solves the problems, such as is similar to a kind of liquefied natural gas warehouse cause of accident inversion method, therefore
A kind of computer equipment and a kind of implementation of computer readable storage medium may refer to a kind of liquefied natural gas warehouse thing
Therefore the implementation of reason inversion method, overlaps will not be repeated.
A kind of computer equipment of the invention and a kind of computer readable storage medium are by fuzzy message quantification, to reach
The purpose of accurate inverting.According to specific crash analysis installations and facilities relevant to accident, the inverting for being more bonded accident scene is established
Model, thus more accurate carry out inverting positioning.
Those of ordinary skill in the art will appreciate that implementing the method for the above embodiments can lead to
Program is crossed to instruct relevant hardware and complete, which can be stored in a computer readable storage medium, such as
ROM/RAM, magnetic disk, CD etc..
Particular embodiments described above has carried out further in detail the purpose of the present invention, technical scheme and beneficial effects
Describe in detail it is bright, it should be understood that the above is only a specific embodiment of the present invention, the guarantor being not intended to limit the present invention
Range is protected, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should be included in this
Within the protection scope of invention.
Claims (10)
1. a kind of liquefied natural gas warehouse cause of accident inversion method, which is characterized in that the described method includes:
The reason of being occurred according to liquefied natural gas warehouse heat insulation layer failure accidents chain, the accident that determines obscures variable weight model, described
It includes the multiple transition in one's power of multiple libraries in variable weight model that accident, which obscures, and according to multiple library institutes, is determined immediately up to library institute
Set is closed up to set of library conjunction and adjacent set of library；
According to expert estimation determine the library initial confidence level, and according to the liquefied natural gas warehouse heat insulation layer fail
Accident histories data determine transition confidence level and transition ignition threshold value；
According to the library initial confidence level, the transition confidence level and the transition ignition threshold value, determine the library
Final confidence level；
Screening is described to be greater than the accident library of the transition ignition threshold value in one's power up to final confidence level described in set of library conjunction immediately
Accident library sequence, the final maximum accident library institute of confidence level in the institute of the accident library is determined, according to the final confidence level
The transition ignition threshold value determines accident path in one's power in maximum accident library.
2. the method according to claim 1, wherein it is described according to the library initial confidence level, the change
The reliability that resites and the transition ignition threshold value, determine the library final confidence level include:
Initialize the library initial confidence level, the transition confidence level and the corresponding input matrix of the transition ignition threshold value
And input vector；
Using the input matrix and input vector, the equivalent inpnt confidence level of each transition is determined；
The equivalent inpnt confidence level for being less than the transition ignition threshold value is removed, according to the equivalent inpnt confidence after removing
Degree, determine the library output confidence level；
Update the library institute according to the output confidence level, with the determination library final confidence level.
3. according to the method described in claim 2, it is characterized in that, described update the library institute according to the output confidence level,
With the determination library final confidence level include:
The library institute is updated according to the output confidence level, determines changeable weight input matrix；
According to changeable weight input matrix, reuses the output confidence level and update the library institute, until the output confidence
Degree does not change, output confidence level at this time for the library final confidence level.
4. the method according to claim 1, wherein described according to the final maximum accident library institute of confidence level
And the transition ignition threshold value determines that accident path includes: the institute's confidence level of library described in comparison and the transition ignition threshold value size,
Library institute's confidence level is selected to be greater than the accident path of the transition ignition threshold value, and when the accident path is at least
At two, judge whether the logical relation in the accident path meets preset requirement；If so, judging that the accident path is corresponding
The minimum value of final confidence level whether be greater than the transition ignition threshold value, if so, determining that the accident path is that can occur
Accident path.
5. a kind of liquefied natural gas warehouse cause of accident inverting device, which is characterized in that described device includes:
Model building module, for determining accident according to chain the reason of the generation of liquefied natural gas warehouse heat insulation layer failure accidents
Fuzzy variable weight model, it includes the multiple transition in one's power of multiple libraries in variable weight model that the accident, which obscures, and according to multiple library institutes,
It determines and closes up to set of library, closed up to set of library conjunction and adjacent set of library immediately；
Threshold determination module, for according to expert estimation determine the library initial confidence level, and it is natural according to the liquefaction
Gas warehouse heat insulation layer failure accidents historical data determines transition confidence level and transition ignition threshold value；
Confidence determination module, for according to the library initial confidence level, the transition confidence level and the transition light a fire
Threshold value, determine the library final confidence level；
Cause of accident determining module, it is described immediately up to final confidence level described in set of library conjunction greater than the transition for screening
The accident library of ignition threshold value in one's power accident library sequence, determine the final maximum accident library of confidence level in the institute of the accident library
Institute, according to the final maximum accident library of confidence level, the transition ignition threshold value determines accident path in one's power.
6. device according to claim 5, which is characterized in that the confidence determination module includes:
Initialization unit, for initialize the library initial confidence level, the transition confidence level and the transition igniting threshold
It is worth corresponding input matrix and input vector；
Equivalent inpnt confidence level unit determines the equivalent defeated of each transition for utilizing the input matrix and input vector
Enter confidence level；
Clearing cell, for removing the equivalent inpnt confidence level for being less than the transition ignition threshold value, according to the institute after removing
State equivalent inpnt confidence level, determine the library output confidence level；
Final confidence level unit, for updating the library institute according to the output confidence level, with the determination library final set
Reliability.
7. device according to claim 6, which is characterized in that the final confidence level unit includes:
Changeable weight subelement determines changeable weight input matrix for updating the library institute according to the output confidence level；
Final confidence level subelement, for reusing described in the output confidence level update according to changeable weight input matrix
Library institute, until the output confidence level does not change, output confidence level at this time for the library final confidence level.
8. device according to claim 5, which is characterized in that the cause of accident determining module includes: Path selection list
Member selects library institute's confidence level to be greater than the change for library institute's confidence level and the transition ignition threshold value size
The accident path of ignition threshold value is moved, and when the accident path is at least two, judges patrolling for the accident path
Whether the relationship of collecting meets preset requirement；If so, judging whether the minimum value of the corresponding final confidence level in the accident path is big
In the transition ignition threshold value, if so, determining that the accident path is that accident path can occur.
9. a kind of computer equipment including memory, processor and stores the meter that can be run on a memory and on a processor
Calculation machine program, which is characterized in that the processor performs the steps of when executing the computer program
The reason of being occurred according to liquefied natural gas warehouse heat insulation layer failure accidents chain, the accident that determines obscures variable weight model, described
It includes the multiple transition in one's power of multiple libraries in variable weight model that accident, which obscures, and according to multiple library institutes, is determined immediately up to library institute
Set is closed up to set of library conjunction and adjacent set of library；
According to expert estimation determine the library initial confidence level, and according to the liquefied natural gas warehouse heat insulation layer fail
Accident histories data determine transition confidence level and transition ignition threshold value；
According to the library initial confidence level, the transition confidence level and the transition ignition threshold value, determine the library
Final confidence level；
Screening is described to be greater than the accident library of the transition ignition threshold value in one's power up to final confidence level described in set of library conjunction immediately
Accident library sequence, the final maximum accident library institute of confidence level in the institute of the accident library is determined, according to the final confidence level
The transition ignition threshold value determines accident path in one's power in maximum accident library.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
It is performed the steps of when being executed by processor
The reason of being occurred according to liquefied natural gas warehouse heat insulation layer failure accidents chain, the accident that determines obscures variable weight model, described
It includes the multiple transition in one's power of multiple libraries in variable weight model that accident, which obscures, and according to multiple library institutes, is determined immediately up to library institute
Set is closed up to set of library conjunction and adjacent set of library；
According to expert estimation determine the library initial confidence level, and according to the liquefied natural gas warehouse heat insulation layer fail
Accident histories data determine transition confidence level and transition ignition threshold value；
According to the library initial confidence level, the transition confidence level and the transition ignition threshold value, determine the library
Final confidence level；
Screening is described to be greater than the accident library of the transition ignition threshold value in one's power up to final confidence level described in set of library conjunction immediately
Accident library sequence, the final maximum accident library institute of confidence level in the institute of the accident library is determined, according to the final confidence level
The transition ignition threshold value determines accident path in one's power in maximum accident library.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910086465.6A CN109800525A (en) | 2019-01-29 | 2019-01-29 | Liquefied natural gas warehouse cause of accident inversion method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910086465.6A CN109800525A (en) | 2019-01-29 | 2019-01-29 | Liquefied natural gas warehouse cause of accident inversion method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109800525A true CN109800525A (en) | 2019-05-24 |
Family
ID=66559210
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910086465.6A Pending CN109800525A (en) | 2019-01-29 | 2019-01-29 | Liquefied natural gas warehouse cause of accident inversion method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109800525A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020188870A1 (en) * | 2001-06-11 | 2002-12-12 | Mcnc | Intrusion tolerant server system |
| CN102175460A (en) * | 2011-01-28 | 2011-09-07 | 中国人民解放军海军工程大学 | Gas path fault diagnostic method for marine gas turbine based on fuzzy Petri net (FPN) |
| CN103278328A (en) * | 2013-05-16 | 2013-09-04 | 华北电力大学 | Method for diagnosing failure of hydraulic variable-pitch system of wind turbine generator based on fuzzy Petri net |
| CN104765965A (en) * | 2015-04-15 | 2015-07-08 | 国家电网公司 | GIS fault diagnosis and reliability analysis method based on fuzzy Petri |
| CN109039739A (en) * | 2018-08-01 | 2018-12-18 | 南瑞集团有限公司 | A kind of power communication field failure diagnostic method improving fuzzy Petri |
-
2019
- 2019-01-29 CN CN201910086465.6A patent/CN109800525A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020188870A1 (en) * | 2001-06-11 | 2002-12-12 | Mcnc | Intrusion tolerant server system |
| CN102175460A (en) * | 2011-01-28 | 2011-09-07 | 中国人民解放军海军工程大学 | Gas path fault diagnostic method for marine gas turbine based on fuzzy Petri net (FPN) |
| CN103278328A (en) * | 2013-05-16 | 2013-09-04 | 华北电力大学 | Method for diagnosing failure of hydraulic variable-pitch system of wind turbine generator based on fuzzy Petri net |
| CN104765965A (en) * | 2015-04-15 | 2015-07-08 | 国家电网公司 | GIS fault diagnosis and reliability analysis method based on fuzzy Petri |
| CN109039739A (en) * | 2018-08-01 | 2018-12-18 | 南瑞集团有限公司 | A kind of power communication field failure diagnostic method improving fuzzy Petri |
Non-Patent Citations (1)
| Title |
|---|
| 包玉剑: "基于加权模糊Petri网的输电网故障诊断方法研究", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109766583B (en) | Aircraft engine life prediction method based on unlabeled, unbalanced and initial value uncertain data | |
| Bulut et al. | A new approach to determine maintenance periods of the most critical hydroelectric power plant equipment | |
| Zio et al. | Industrial disasters: Extreme events, extremely rare. Some reflections on the treatment of uncertainties in the assessment of the associated risks | |
| Rathnayaka et al. | Accident modeling approach for safety assessment in an LNG processing facility | |
| CN104715163B (en) | A kind of buried oil-gas pipeline methods of risk assessment | |
| Dormido-Canto et al. | Development of an efficient real-time disruption predictor from scratch on JET and implications for ITER | |
| Liang et al. | Risk assessment for long-distance gas pipelines in coal mine gobs based on structure entropy weight method and multi-step backward cloud transformation algorithm based on sampling with replacement | |
| CN111178758A (en) | Concrete dam monitoring data intelligent management and real-time evaluation system based on BIM | |
| Zemenkova et al. | Intelligent monitoring of the condition of hydrocarbon pipeline transport facilities using neural network technologies | |
| CN109492790A (en) | Wind turbines health control method based on neural network and data mining | |
| Naser | Autonomous fire resistance evaluation | |
| Mata et al. | Construction of decision rules for early detection of a developing concrete arch dam failure scenario. A discriminant approach | |
| CN109344490A (en) | A kind of ocean platform based on BRANN model fires risk analysis method | |
| Wen et al. | Data‐driven transient frequency stability assessment: A deep learning method with combined estimation‐correction framework | |
| Omidvar et al. | Development of a framework for resilience measurement: Suggestion of fuzzy Resilience Grade (RG) and fuzzy Resilience Early Warning Grade (REWG) | |
| Xu et al. | Dynamic risk assessment for underground gas storage facilities based on Bayesian network | |
| CN114169711A (en) | Accident emergency disposal scheme generation method, device and system for distillation device | |
| CN109800525A (en) | Liquefied natural gas warehouse cause of accident inversion method and device | |
| SHAHROKHI | Developing an Approach to Calculate Fuzzy Reliability Based on Fuzzy Failure Rate | |
| Santana et al. | Fire and Explosion Economic Losses (FEEL) Index: A new approach for quantifying economic damages due to accidents in hydrocarbon storage sites | |
| Ibrahim et al. | Bayesian network-based failure analysis of fire safety barriers in floating LNG facility | |
| Yan | Construction of digital twin ecosystem for coal-fired generating units | |
| Feijoo-Martinez et al. | Event analysis on power communication networks with big data for maintenance forms | |
| Li et al. | Research status and development trends of mine floor water inrush grade prediction | |
| Wang et al. | A multicriteria decision model based on analytic hierarchy process for managing safety in coal mines |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190524 |