CN106227907B - A kind of liquid hydrogen loading system security assessment method based on evolution clustering learning - Google Patents
A kind of liquid hydrogen loading system security assessment method based on evolution clustering learning Download PDFInfo
- Publication number
- CN106227907B CN106227907B CN201610366767.5A CN201610366767A CN106227907B CN 106227907 B CN106227907 B CN 106227907B CN 201610366767 A CN201610366767 A CN 201610366767A CN 106227907 B CN106227907 B CN 106227907B
- Authority
- CN
- China
- Prior art keywords
- liquid hydrogen
- loading system
- hydrogen loading
- dictionary
- safety
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/36—Circuit design at the analogue level
- G06F30/367—Design verification, e.g. using simulation, simulation program with integrated circuit emphasis [SPICE], direct methods or relaxation methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/18—Manufacturability analysis or optimisation for manufacturability
Abstract
The invention discloses a kind of liquid hydrogen loading system security assessment methods based on evolution clustering learning, belong to space launch security evaluation field, comprising the following steps: S1: choosing training sample from history perception data；S2: piecemeal processing is carried out to training sample；S3: pass through l2Norm distance measure carries out similarity measurement；S4: pass through block coordinate relaxed algorithm solution group sparse coefficient；S5: Training Support Vector Machines, real-time update evolution cluster block dictionary；S6: in conjunction with system and state motion rule, quantitative analysis liquid hydrogen loading system security measure standard；S7: construction liquid hydrogen loading system safety evaluation index, the safety grades of describing system；S8: when on-line operation, above-mentioned S2 is repeated to S7 step, realizes assessment in real time.Evolution clustering learning is introduced into safety evaluation theoretical frame by the present invention, portrays dictionary learning and update mechanism under data-driven, constructs safety evaluation index, realizes liquid hydrogen loading system safety evaluation.
Description
Technical field
The present invention implements to be related to security evaluation more particularly to a kind of liquid hydrogen filling system based on evolution clustering learning
System security assessment method.
Background technique
The 21 century mankind will move towards space comprehensively, explore space, development space and using space, space launch, as commenting
The important symbol for sentencing a national overall national strength, has important strategic importance.Space launch be one be typically related to more people,
Multimachine, multi-environment large-scale complex engineering system, are made of spacecraft, carrier rocket and launch facility etc., in emission process
In, any security risk, personnel misoperation and the system failure all may cause interruption of service even calamity, exploration, examination
The property tested, risk and the social safety for determining space launch have very important status.
With the continuous development of science and technology, space launch system and equipment composition tend to system networking, so that abnormal
Behavior is presented multi-to-multi with system unit and is associated with, and affects the safe operation of system.In the tasks such as test, commander, operation, control
It is related to more people, multioperation and remote operating etc., failure and human error constitute the major safety risks of system operation.Emission system
Severe natural environment (such as high temperature, with high salt, high humidity or severe cold, dust storm etc.) is run on, by complexity during launch mission
Physical-chemical-mechanics comprehensive function (physical influence of such as cryogenic propellant to pipe-line system, sound of the rocket launching to system
Wave, mechanics impact and high temperature spray chemical attack of fuel etc.).Interim intermittent duty repeatedly is to system equipment, device
Performance generate strong bad effect, cause each subsystem hidden trouble of equipment occur, show different system or equipment failures.
Therefore, system moving law of the space launch system under external disturbance, stress variation and operating mistake and dynamic how is analyzed
State behavior is studied the evolution of " risk factor-safety accident " under the intermittent duty of space launch system stage, is set in exposing system
Standby performance degradation trend, the system failure is propagated and accident genesis mechanism, is that space launch system operational safety is assessed in real time
Critical issue.
By taking liquid hydrogen fills as an example, the fuel of liquid hydrogen propellant and oxidant are delivered to the combustion of carrier rocket from ground storage tank
In hopper, as shown in Figure 1.Liquid hydrogen filling is cryogenic liquid propellant combination, does fuel with liquid hydrogen, liquid oxygen makees oxidant, this two
Substance boiling point is extremely low, readily volatilized, needs to be perfused at elevated pressures, can just be allowed to keep liquid, add to cryogenic propellant
It is very high to infuse time precision requirement.The every hair task of "Long March" series of carrier rockets fills the hundreds of tons of conventional fuel, has very strong corruption
Corrosion and toxicity.And cryogenic propellant liquid hydrogen liquid oxygen has a volatile and evaporation characteristic, the explosion limit of liquid hydrogen is 4.0%~
75.6% (volumetric concentration) when that is, aerial volumetric concentration is between 4%~75.6%, encounters fire source and just causes explosion.
Therefore, guarantee that high security is primary during space launch.
For safeguards system safe and reliable operation, need to carry out operating parameter acquisition, more to space launch liquid hydrogen loading system
It is secondary to penetrate front simulation flight and test general inspection etc., so that the data volume generated in emission system operational process constantly accumulates, in conjunction with
Data analysis is excavated and is handled, and overcomes the non-linear and various uncertain noises of process to procedures system safe and reliable operation
It influences, has great significance.However, liquid hydrogen loading system failure mode is considerably complicated, the inducement of influence safety is more, is difficult to
Prediction；So that there is a large amount of unknown rule in entire liquid hydrogen filling process so that the process have it is very big not really
It is qualitative.Therefore, it analyzes removal system by the subsequent interpretation of test data and is operating abnormally and guarantee emission security with failure, need to study
The real-time Assessment theory of safety in operation and method during task.
The real-time evaluation problem of space launch liquid hydrogen loading system operational safety is the base for guaranteeing space launch safe operation
Plinth.Currently, the security evaluation theory and method of all kinds of complex engineering system operations specifically includes that
(1) the operational safety appraisal procedure based on qualitative analysis, including Fault Tree Analysis, Risk Assessment Code method,
The methods of safety checklist, preparatory hazard analysis, fault modes and effect analysis, dangerous operability analysis.
(2) the operational safety appraisal procedure based on quantitative analysis, including event tree method, Markov Process as Applied, Event Sequence Diagram
Method, analytics, analog simulation method etc..
(3) comprehensive estimation method, including risk coordinate evaluation and probability distributive function method etc..
As it can be seen that existing security assessment method, transfinite in the technological parameter of system/device, failure and maloperation etc.
Under the influence of risk factor, do not fully consider that the extensive close coupling between subsystem of system makes the hair of system interruption of service
Sequence shows complexity and uncertainty to life reason at any time with Evolution.Therefore, in different time sequence and different works
Under condition, how the changing features of Efficient Characterization procedure parameter and state behavior, and establish risk factor effect under " process ginseng
Evolutionary model between number-safety accident " is key scientific problems urgently to be resolved.
Summary of the invention
In view of this, technical problem to be solved by the invention is to provide a kind of, the liquid hydrogen based on evolution clustering learning is filled
Evolution clustering learning is introduced into safety evaluation theoretical frame by security of system appraisal procedure, is portrayed and is developed under data-driven
The study and update mechanism of dictionary are clustered, safety evaluation index is constructed, realizes the safety of space launch liquid hydrogen loading system
Assessment in real time.
The substantially process of liquid hydrogen loading system safety evaluation based on evolution clustering learning of the invention as shown in Fig. 2,
The specific implementation steps are as follows for it:
S1: training sample is chosen.Training sample is chosen from the history perception data of liquid hydrogen loading system；
S2: piecemeal processing.Known according to the priori such as operating condition, system motion rule under liquid hydrogen loading system time series
Know, piecemeal processing is carried out to training sample；
S3: block similarity measurement.Pass through non local operation operator, joint operation algorithm and l2Norm distance measure block
Similarity measurement；
S4: solution group structure sparse coefficient.Pass through block coordinate relaxed algorithm solution group sparse coefficient；
S5: determining the parameter of supporting vector machine model (Support Vector Machines, SVM), updates the cluster that develops
Block dictionary.It determines the parameter of support vector machine classifier, and is timely updated dictionary according to history tab；
S6: the building of liquid hydrogen loading system degree of safety and its module；In conjunction with the characteristics of motion and state behavior of system
The characteristics of motion, the degree of safety and its module of quantitative analysis liquid hydrogen loading system, probability distributive function；
S7: the safety grades of describing system.Construct the safety evaluation index of liquid hydrogen loading system, adaptive adjustment danger
Dangerous factor weight provides the safety grades of system, verifies security of system grade under history data；
S8: online real time security assessment.
Further, the block similarity measurement in the step S3, as shown in Figure 3, the specific steps are as follows:
S31: using the prior information of the non local similitude of block, by non local operation operator and joint operation operator,
Obtain the rarefaction representation of block.
S32 by reference to block and candidate blocks l2Norm distance similarity measurement, obtains the cluster of similar block.
Further, the parameter of the determination supporting vector machine model in the step S5 updates evolution cluster block dictionary, specifically
Steps are as follows:
S51: according to the prior information of the corresponding system safe condition behavior label of procedure parameter, to l2SVM classifier
Model is trained, and obtains its optimized parameter, as shown in Figure 5.
S52: it for test information or online awareness information, is compared, is constructed with legitimate reading by classification prediction result
Develop and clusters the update mechanism of dictionary.
Further, the building of the liquid hydrogen loading system degree of safety and its module in the step S6, specific steps are such as
Under:
S61: by failure tree analysis (FTA), the architecture logic block diagram of liquid hydrogen loading system under possible accident is obtained, minimum is utilized
The cut set upper limit and minimal path Lower-bound Method carry out the state behavior of approximation system, and divide and be independent elementary event, such as Fig. 6
It is shown.
S62: it by the classification accuracy rate of failure and the probability of happening of elementary event, defines liquid hydrogen loading system safety and comments
The degree of safety estimated.
Further, the online real time security assessment in the step S8, the specific steps are as follows:
S81: above-mentioned S2 is repeated to S7 step, realizes that liquid hydrogen loading system safety is assessed in real time.
S82: specifically, when repeating step S4 and S5, the update parallel processing of failure predication classification and dictionary: firstly,
Dictionary updating principle -- by compared with real system safe condition, and then renewal learning dictionary；Secondly, when on-line operation,
Feature extraction, classification and the prediction of failure do not consider the update of dictionary, only consider the dictionary of last moment.
The present invention has the advantages that evolution clustering learning is introduced into safety evaluation theoretical frame, data-driven is portrayed
Lower dictionary learning and update mechanism construct safety evaluation index, realize liquid hydrogen loading system safety evaluation.
Detailed description of the invention
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing to the present invention do into
The detailed description of one step, wherein
Fig. 1 is certain liquid rocket low temperature propellant liquid hydrogen loading system；
Fig. 2 is the safety evaluation theoretical frame based on evolution clustering learning；
Fig. 3 is similar block cluster；
Fig. 4 is that signal is clustered and is grouped
Fig. 5 is support vector machines basic schematic diagram；
Fig. 6 is fault tree synthesis schematic diagram.
In Fig. 1,1- deflation valve；2- three-stage rocket helium heat exchanger；3- gaseous helium；4- three-level hydrogen case；5- pylon is put
Air pipe；6- fills valve；7- filter；The main filling valve of 8-；9- gas helium mouth；10- fuel burns pond；11- second level helium heat is handed over
Parallel operation；12- second level hydrogen case；13- deflation valve；14- filler pipe is deflated；15- gas helium mouth；16- evaporator；The storage of 17- liquid hydrogen is held
Device；18- filter pool.
In Fig. 3, (a)-original signal；(b)-block cluster；(c)-array dimension.
In Fig. 4, (a)-actual signal and its region of search；(b)-reference block and its similar block；(c)-signal group.
Specific embodiment
Below with reference to attached drawing, a preferred embodiment of the present invention will be described in detail；It should be appreciated that preferred embodiment
Only for illustrating the present invention, rather than limiting the scope of protection of the present invention.
The present invention, substantially process such as Fig. 2 institute of the liquid hydrogen loading system security assessment method based on evolution clustering learning
Show, the specific steps are as follows:
S1: training sample is chosen.From the history run perception data of liquid hydrogen loading system, training as much as possible is chosen
Data reduce the interference of noise during data acquisition, transmission, storage etc..
S2: piecemeal processing.Known according to the priori such as operating condition, system motion rule under liquid hydrogen loading system time series
Know, piecemeal processing is carried out to training sample；
S3: the similarity measurement of block.Pass through non local operation operator, joint operation algorithm and l2Norm distance measure
Block similarity measurement.
S31: effectively characterizing signal with behaviour's operator is combined by block-based non local operation operator, piecemeal processing
Process is as shown in Figure 3.For given signal X ∈ RN, using the block splitting signal of fixed size (L × 1), non local operation is calculated
Sub-definite are as follows:
In formula, vjFor the index of block grouping；For vjThe block of group；D2-dimensionFor two-dimensional transform.In order to
The sparse representation of block is effectively obtained, the group of optimization is sparse to be indicated are as follows:
zj=Ajxi (2)
Wherein, joint operation operatorAnd meetTo really
Protecting every piece can be assigned in a group.
S32: l is utilized2Norm distance carrys out measured similarity.For effectively packet signal block, as shown in figure 4, given search
The measure definitions of the similitude of rope region Ω and reference block T, reference block and candidate blocks are as follows:
S4: solution group structure sparse coefficient.Cluster dictionary is obtained using KSVD training, and is asked by block coordinate relaxed algorithm
System of solutions structure sparse coefficient, detailed process are described as follows:
Consider that perception information obtains model:
X=DZ+ ε (4)
It is typical to solve objective function are as follows:
Wherein, X ∈ RNFor discrete signal；Z∈RCFor the k- feature space of acquisition；D∈RM×NFor dictionary；ε be noise to
Amount；λ is regularization parameter.In formula (5), first item is reconstructed error, for the validity of the feature space of measurement & characterization signal；
Section 2 is characterized the sparsity in space.
In practical applications, Setting signal forms X={ x by block1,…,xi,…}.Wherein block xiIt can be characterized as xi=
zidi, then the objective function for organizing structure rarefaction representation can be expressed as
It is approached by KSVD dictionary learning and the iteration optimization of block coordinate relaxed algorithm, obtains the optimal sparse word of group structure
Allusion quotation, and approach the sparse coefficient of original signal.
S5: determining the parameter of SVM, updates evolution cluster block dictionary.Its detailed process is described as follows:
S51: according to the prior information of the corresponding system safe condition behavior label of procedure parameter, to l2SVM classifier
Model is trained, and obtains its optimized parameter, as shown in Figure 5.
S52: it for test information or online awareness information, is compared, is constructed with legitimate reading by classification prediction result
Develop and clusters the update mechanism of dictionary.
S6: the building of liquid hydrogen loading system degree of safety and its module.In conjunction with the characteristics of motion and state behavior of system
The characteristics of motion, the degree of safety and its module of quantitative analysis liquid hydrogen loading system, probability distributive function.Its detailed process is such as
Under:
S61: by failure tree analysis (FTA), the architecture logic block diagram of liquid hydrogen loading system under possible accident is obtained, minimum is utilized
The cut set upper limit and minimal path Lower-bound Method carry out the state behavior of approximation system, and divide and be independent elementary event, such as Fig. 6
It is shown.
S62: pass through the classification accuracy rate of failureWith the probability of happening of elementary eventDefine liquid hydrogen loading system peace
The degree of safety of full property assessment.The degree of safety of liquid hydrogen loading system interruption of service (Top-event) may be defined as
S7: the safety grades of liquid hydrogen loading system are portrayed.In conjunction with key factor shadow under different time sequence, different operating conditions
Loud difference, the incidence relation of establishment process parameter and safety accident, the weight of reasonable distribution key factor；Pass through expertise
With key factor weight, the safety evaluation index of liquid hydrogen loading system is constructed, provides the safety grades of system, verifies history
Security of system grade under operation data.
S8: online real time security assessment.Detailed process is as follows for it:
S81: above-mentioned S2 is repeated to S7 step, it is " crucial under different time sequence, different operating conditions to establish critical risk factor
The incidence relation of factor (perception data) -- safety accident " realizes that liquid hydrogen loading system safety is assessed in real time.
S82: specifically, when repeating step S4 and S5, the update parallel processing of failure predication classification and dictionary: firstly,
Dictionary updating principle -- by compared with real system safe condition, and then renewal learning dictionary；Secondly, when on-line operation,
Feature extraction, classification and the prediction of failure do not consider the update of dictionary, only consider the dictionary of last moment.
The above description is only a preferred embodiment of the present invention, is not intended to restrict the invention, it is clear that those skilled in the art
Various changes and modifications can be made to the invention by member without departing from the spirit and scope of the present invention.If in this way, of the invention
Within the scope of the claims of the present invention and its equivalent technology, then the present invention is also intended to encompass these to these modifications and variations
Including modification and variation.
Claims (7)
1. a kind of liquid hydrogen loading system security assessment method based on evolution clustering learning, it is characterised in that: such as including step
Under:
S1: training sample is chosen from the history perception data of space launch liquid hydrogen loading system；
S2: according to the priori knowledge of liquid hydrogen loading system, piecemeal processing is carried out to training sample, the priori knowledge includes the time
Operating condition and system motion rule under sequence；
S3: pass through non local operation operator, joint operation algorithm and l2Norm distance measure block similarity measurement；
S4: pass through KSVD and block coordinate relaxed algorithm solution group sparse coefficient；
S5: determining the parameter of supporting vector machine model SVM, updates evolution cluster block dictionary；
S6: in conjunction with the characteristics of motion of system and the characteristics of motion of state behavior, space launch liquid hydrogen loading system degree of safety and its
The building of module；
S7: the safety evaluation index of construction space launch liquid hydrogen loading system, it is adaptive to adjust risk factor weight, it portrays and is
The safety grades of system；
S8: online real time security assessment.
2. the liquid hydrogen loading system security assessment method according to claim 1 based on evolution clustering learning, feature
It is: in the step S3, block is realized by non local operation operator, joint operation algorithm and l2 norm distance measure
Similarity measurement, specific steps are described as follows:
S31: it is obtained using the prior information of the non local similitude of block by non local operation operator and joint operation operator
The rarefaction representation of block；
S32: it is measured by reference to the norm distance similarity of block and candidate blocks, obtains the cluster of similar block.
3. the liquid hydrogen loading system security assessment method according to claim 1 based on evolution clustering learning, feature
Be: in the step S4, by KSVD and block coordinate relaxed algorithm solution group sparse coefficient, detailed process is as follows:
Consider that perception information obtains model:
X=DZ+ ε (1)
It is typical to solve objective function are as follows:
Wherein, X ∈ RNFor discrete signal；Z∈RCFor the k- feature space of acquisition；D∈RM×NFor dictionary；RNIndicate the discrete reality of N-dimensional
Number；RCIndicate the complex number space of C dimension；RM×NIndicate the real number matrix of M × N；ε is noise vector；λ is regularization parameter；Formula (2)
In, first item is reconstructed error, for the validity of the feature space of measurement & characterization signal；Section 2 is characterized the sparse of space
Property；
In practical applications, Setting signal forms X={ x by block1,…,xi,…}；Wherein block xiIt can be characterized as xi=zidi,
Middle ziIndicate i-th of element in k- feature space, diIt indicates i-th of atom in dictionary, then organizes the target of structure rarefaction representation
Function can be expressed as
It is approached by KSVD dictionary learning and the iteration optimization of block coordinate relaxed algorithm, obtains optimal group structure sparse dictionary,
And approach the sparse coefficient of original signal.
4. the liquid hydrogen loading system security assessment method according to claim 1 based on evolution clustering learning, feature
It is: in the step S5, determines the parameter of SVM, update evolution cluster block dictionary；Its detailed process is described as follows:
S51: according to the prior information of the corresponding system safe condition behavior label of procedure parameter, to l2SVM classifier model
It is trained, obtains its optimized parameter；
S52: it for test information or online awareness information, is compared by classification prediction result with legitimate reading, building is developed
It clusters the update mechanism of dictionary: passing through KSVD dictionary learning and l2Svm classifier is not involved in dictionary when realizing security level assessment
Update step；By comparing prediction security level and practical security level, it is determined whether update dictionary.
5. the liquid hydrogen loading system security assessment method according to claim 1 based on evolution clustering learning, feature
It is: in the step S6, in conjunction with the characteristics of motion of system and the characteristics of motion of state behavior, quantitative analysis liquid hydrogen loading system
Degree of safety and its module, probability distributive function；Detailed process is as follows for it:
S61: by failure tree analysis (FTA), the architecture logic block diagram of loading system under possible accident is obtained, the minimal cut set upper limit is utilized
With minimal path Lower-bound Method, carry out the state behavior of approximation system, and divides and be independent elementary event；
S62: pass through the classification accuracy rate of failureWith the probability of happening of elementary eventDefine the safety of security of system assessment
Degree；The degree of safety of liquid hydrogen loading system interruption of service is defined as
6. the liquid hydrogen loading system security assessment method according to claim 1 based on evolution clustering learning, feature
It is: the safety grades of system in the step S7, in conjunction with crucial hazards under different time sequence, different operating conditions
Difference, the incidence relation of establishment process parameter and safety accident, the weight of reasonable distribution key factor；By expertise with
Key factor weight constructs the safety evaluation index of liquid hydrogen loading system, realizes liquid hydrogen loading system safety under data-driven
Property is assessed in real time.
7. the liquid hydrogen loading system security assessment method according to claim 1 based on evolution clustering learning, feature
Be: online assessment in real time in the step S8, detailed process is as follows:
S81: repeating above-mentioned S2 to S7 step, establish critical risk factor under different time sequence, different operating conditions " it is crucial because
The incidence relation of element -- safety accident ", the key factor includes perception data, realizes that liquid hydrogen loading system safety is commented in real time
Estimate；
S82: specifically, when repeating step S4 and S5, the update parallel processing of failure predication classification and dictionary: firstly, dictionary
Update principle -- by compared with real system safe condition, and then renewal learning dictionary；Secondly, when on-line operation, failure
Feature extraction, classification and prediction do not consider the update of dictionary, only consider the dictionary of last moment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610366767.5A CN106227907B (en) | 2016-05-30 | 2016-05-30 | A kind of liquid hydrogen loading system security assessment method based on evolution clustering learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610366767.5A CN106227907B (en) | 2016-05-30 | 2016-05-30 | A kind of liquid hydrogen loading system security assessment method based on evolution clustering learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106227907A CN106227907A (en) | 2016-12-14 |
| CN106227907B true CN106227907B (en) | 2019-06-07 |
Family
ID=57520271
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610366767.5A Active CN106227907B (en) | 2016-05-30 | 2016-05-30 | A kind of liquid hydrogen loading system security assessment method based on evolution clustering learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106227907B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109120451B (en) * | 2018-08-31 | 2021-08-06 | 深圳市麦斯杰网络有限公司 | Equipment evaluation method and equipment based on Internet of things and computer-readable storage medium |
| CN116705184B (en) * | 2023-05-29 | 2024-04-05 | 上海海德利森科技有限公司 | Liquid hydrogen evaporation loss prediction method, device, equipment and medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101315545A (en) * | 2008-06-27 | 2008-12-03 | 浙江大学 | Three-level charging-up optimizing control method and system for hydrogenation station high-efficiency hydrogenation |
| JP2013019453A (en) * | 2011-07-11 | 2013-01-31 | Nippon Sharyo Seizo Kaisha Ltd | Piping structure for low-temperature liquid container |
| CN104156418A (en) * | 2014-08-01 | 2014-11-19 | 北京系统工程研究所 | Knowledge reuse based evolutionary clustering method |
| CN104898714A (en) * | 2015-04-02 | 2015-09-09 | 北京航天发射技术研究所 | High-reliability redundancy liquid hydrogen filling system and method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20100129464A1 (en) * | 2008-11-21 | 2010-05-27 | Isao Suzuki | Manufacturing method and apparatus for producing substances that include negative hydrogen ion |
-
2016
- 2016-05-30 CN CN201610366767.5A patent/CN106227907B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101315545A (en) * | 2008-06-27 | 2008-12-03 | 浙江大学 | Three-level charging-up optimizing control method and system for hydrogenation station high-efficiency hydrogenation |
| JP2013019453A (en) * | 2011-07-11 | 2013-01-31 | Nippon Sharyo Seizo Kaisha Ltd | Piping structure for low-temperature liquid container |
| CN104156418A (en) * | 2014-08-01 | 2014-11-19 | 北京系统工程研究所 | Knowledge reuse based evolutionary clustering method |
| CN104898714A (en) * | 2015-04-02 | 2015-09-09 | 北京航天发射技术研究所 | High-reliability redundancy liquid hydrogen filling system and method |
Non-Patent Citations (4)
| Title |
|---|
| 《Adaptive sliding-mode control for fractional-order uncertain linear systems with nonlinear disturbances》;L Chen等;《Nonlinear Dynamics》;20151231;第80卷(第1期);第51-58页 |
| 《Hilbert marginal spectrum analysis for automatic seizure detection in EEG signals》;K Fu等;《Biomedical Signal Processing & Control》;20151231(第18期);第179-185页 |
| 《协同演化算法在聚类中的应用》;董红斌 等;《模式识别与人工智能》;20121231;第25卷(第4期);第676-683页 |
| 《智能化航天发射系统及其关键技术研究》;柴毅;《国防科技》;20160301;第37卷(第1期);第7-13页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106227907A (en) | 2016-12-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kadri et al. | The impact of natural disasters on critical infrastructures: A domino effect-based study | |
| Guo et al. | Disruption prediction on EAST tokamak using a deep learning algorithm | |
| Shin et al. | Itad: integrative tensor-based anomaly detection system for reducing false positives of satellite systems | |
| CN112070215B (en) | Processing method and processing device for dangerous situation analysis based on BP neural network | |
| Seo et al. | A methodology for determining efficient gas detector locations on offshore installations | |
| Caputo et al. | Problems and perspectives in seismic quantitative risk analysis of chemical process plants | |
| Hassan et al. | A data base oriented dynamic methodology for the failure analysis of closed loop control systems in process plant | |
| CN106227907B (en) | A kind of liquid hydrogen loading system security assessment method based on evolution clustering learning | |
| Lee et al. | An online operator support tool for severe accident management in nuclear power plants using dynamic event trees and deep learning | |
| Li et al. | Prediction of BLEVE loads on structures using machine learning and CFD | |
| Zhen et al. | An interpretable and augmented machine-learning approach for causation analysis of major accident risk indicators in the offshore petroleum industry | |
| Chavoshi et al. | Data‐driven prediction of the probability of creep–fatigue crack initiation in 316H stainless steel | |
| Nanyonga et al. | Natural Language Processing and Deep Learning Models to Classify Phase of Flight in Aviation Safety Occurrences | |
| CN113962164A (en) | Organic gas leakage diffusion real-time intelligent early warning method considering uncertainty reasoning | |
| Sun et al. | Towards limiting potential domino effects from single flammable substance release in chemical complexes by risk-based shut down of critical nearby process units | |
| Andrade et al. | Wildfire Emergency Response Hazard Extraction and Analysis of Trends (HEAT) through Natural Language Processing and Time Series | |
| Chen et al. | Risk analysis of oilfield gathering station | |
| Harhara et al. | Process safety consequence modeling using artificial neural networks for approximating heat exchanger overpressure severity | |
| Park et al. | Development of fire consequence prediction model in fuel gas supply system room with changes in operating conditions during liquefied natural gas bunkering | |
| Heitzer et al. | Reliability analysis of elasto-plastic structures under variable loads | |
| Musgrave et al. | FDIA Detection Methods on a Navy Smart Grid AMI Data Set using Autoenocoder Neural Networks: A Case Study | |
| Andrade et al. | Machine learning framework for Hazard Extraction and Analysis of Trends (HEAT) in wildfire response | |
| Yang et al. | Machine Learning in Process Safety and Asset Integrity Management | |
| Cai et al. | Risk identification of civil aviation engine control system based on particle swarm optimization-mean impact value-support vector machine | |
| Nikula et al. | Capturing deviations from design intent in building simulation models for risk assessment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 |