CN104879295A - Large complex system fault diagnosis method based on multilevel flow model and minimal cutset of fault tree - Google Patents
Large complex system fault diagnosis method based on multilevel flow model and minimal cutset of fault tree Download PDFInfo
- Publication number
- CN104879295A CN104879295A CN201510312479.7A CN201510312479A CN104879295A CN 104879295 A CN104879295 A CN 104879295A CN 201510312479 A CN201510312479 A CN 201510312479A CN 104879295 A CN104879295 A CN 104879295A
- Authority
- CN
- China
- Prior art keywords
- fault
- tree
- multilevel flow
- subtree
- cause
- 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
- Test And Diagnosis Of Digital Computers (AREA)
Abstract
The invention discloses a large complex system fault diagnosis method based on a multilevel flow model and the minimal cutset of a fault tree. The large complex system fault diagnosis method includes the steps of establishing the multilevel flow model of a large complex system; linking cause-consequence trees of every functional status of the multilevel flow model to obtain an evidence fault tree of the large complex system; performing fault diagnosis on corresponding parts or part combinations in the large complex system according to a mixed sequence table of percentage importance degree of the minimal cutset and FV importance degree of every elementary event obtained according to the evidence fault tree. The large complex system fault diagnosis method has the advantages that the problems that a multilevel flow model is poor in accuracy and a fault tree diagnosis method is poor in readability and difficult to understand are solved, the advantages of high visualization, good readability and easiness in inspection and maintenance of the multilevel flow model and the advantages of accuracy in model establishing and quickness and accuracy in resolving of the fault tree diagnosis method are combined, economic input and time consumption are reduced, and system fault diagnosis efficiency is improved.
Description
Technical field
The present invention relates to the fault diagnosis field of large-scale complicated system, specifically a kind of large-scale complicated system method for diagnosing faults based on multilevel flow models and Minimizing Cut Sets of Fault Trees.
Background technique
Fault diagnosis refers to by the detection to system, inspection and test, and analyzes the result obtained, thus identification system fault it is taked to the process of corresponding measure.Because the content of carrying out fault diagnosis object and fault is varied, diagnostic method is also varied.Modern method for diagnosing faults has multiple classification, but may be summarized to be three classes generally: namely based on the method for analytical mathematic model, based on the method for signal transacting and Knowledge based engineering method.
Various event and accident is there will be in the processes such as the operation and maintenance of large-scale complicated system (as systems such as large-scale hadron head-on collision device, space flight and aviation system, nuclear power plants), in order to the stable operation of keeping system, needs can fault in Timeliness coverage system or potential abnormal condition and their reason, could take treatment measures timely and effectively like this, the possible loss of system is controlled minimum.Such could under the limited prerequisite of total resources, more rationally effective system to be optimized, the Economy of elevator system, stability and Security.
Multilevel flow models (Multilevel Flow Models, MFM) is a kind of KBS Knowledge Based System function modeling method proposed by Morten professor Lind of Technical University Of Denmark the phase at the beginning of the eighties in last century.Multilevel flow models is from the target of system, function and parts three levels, on the basic principle of energy conservation by by extensive for the multiple behavior of system be the processes such as the generation of matter and energy, transmission, the observation of storage and consumption and information, decision-making and execution, realize function modeling to complex system, have intuitive strong, readable good, be easy to the features such as Inspection and maintenance.
Fault Tree Analysis is a kind of deduction modeling method of system, by to the reason of thrashing may be caused from system to parts to draw out an arborizations figure launched gradually to part bed-by-bed analysis again, and then obtain the various possible component combination causing thrashing, i.e. minimal cut set.And on the basis of minimal cut set, the parameters such as the importance degree of fault tree top event failure probability or frequency and each parts or component combination can be tried to achieve by quantitatively calculating.Fault Tree Analysis has meticulous, the easy quantification of modeling and by features such as special algorithm rapid solvings.
Multilevel flow models due to be based upon systematic knowledge basis on, the complex system of reality uses, although have the advantages that intuitive is strong, readability is good, be easy to Inspection and maintenance, its modeling feature determines it and not easily calculates, and accuracy is difficult to good guarantee; Fault Tree Analysis is a kind of important method of carrying out safety analysis in large-scale complicated system, as in probabilistic safety analysis with regard to application and trouble tree analytical method be that the Security of Main Means to nuclear power station is analyzed, it has the meticulous and feature that can quick and precisely solve of modeling process, but its model extremely bulky complex, modeling process or all suitable the wasting time and energy of review process, its fault tree models readable poor, not easily understand, and later maintenance also relative complex.Therefore, the present invention will organically combine both it, propose a kind of large-scale complicated system method for diagnosing faults based on multilevel flow models and Minimizing Cut Sets of Fault Trees, solve multilevel flow models poor accuracy and readable poor, the not intelligible problem of Fault Tree Analysis.
Summary of the invention
The present invention wants technical solution problem to be: overcome the deficiencies in the prior art, a kind of large-scale complicated system method for diagnosing faults based on multilevel flow models and Minimizing Cut Sets of Fault Trees is proposed, there is intuitive simultaneously strong, readable good, be easy to Inspection and maintenance and can quantize, solve the features such as simple, overcome tradition based on threshold value in the alert analysis method for diagnosing faults of multilevel flow models method to the impact of diagnostic result, and have the advantages that to identify single part and the component combination source of trouble simultaneously, thus improve the accuracy of fault diagnosis, accelerate the speed of fault diagnosis.
The technological scheme that the present invention solves the problems of the technologies described above employing is: a kind of large-scale complicated system method for diagnosing faults based on multilevel flow models and Minimizing Cut Sets of Fault Trees, and implementation step comprises:
Step (1), set up the multilevel flow models of system;
The cause and effect subtree of each functional status of the system multilayer flow model in step (2), establishment step (1);
Step (3), to characterize according to the physical fault of system, revise and simplify the cause and effect subtree of each functional status in step (2), then setting up the evidence fault tree of system;
Step (4), evidence fault tree according to the system obtained in step (3), the minimal cut set of solving system;
Step (5), according to the minimal cut set obtaining parts of the system of trying to achieve in step (4) or the hybrid-sorting list of component combination;
Step (6), sequentially fault diagnosis is carried out to the corresponding component in system or component combination according to the hybrid-sorting list of the parts obtained in step (5) or component combination, thus reach minimizing economic input and time overhead, improve the target of system fault diagnosis efficiency.
As above based on the large-scale complicated system method for diagnosing faults of multilevel flow models and Minimizing Cut Sets of Fault Trees, it is characterized in that: the method setting up the evidence fault tree of system in described step (3) is as follows:
1) other functional statuses used in its branch are substituted into the cause and effect subtree of its correspondence and launch by the cause and effect subtree model of a certain functional status respectively in selecting system;
2) substituting into and launching in the process of cause and effect subtree, as there is repeated events, need the tree structure of blocking this, thus eliminate logic abscission ring, finally generate the complete because of fruit tree of a certain functional status;
3) the cause and effect subtree model of all need functional status to be processed in selecting system successively, repeats step 1)-2) several times, until all generate complete because of fruit tree for all functions state;
4) by AND gate combination step 3) the middle all complete causal tree model generated, form the evidence fault tree of system.
As above based on the large-scale complicated system method for diagnosing faults of multilevel flow models and Minimizing Cut Sets of Fault Trees, it is characterized in that: as follows by the method for the minimal cut set obtaining parts of system or the hybrid-sorting list of component combination in described step (5):
(A) the percentage importance degree of each cut set is calculated respectively according to the minimal cut set of the system obtained;
(B) the FV importance degree of elementary event in all cut sets is calculated respectively according to the minimal cut set of the system obtained;
(C) according to the importance degree information obtained in step (A) and step (B), the hybrid-sorting list of obtaining parts or component combination.
The present invention compared with prior art advantage is:
(1), the present invention's application multilevel flow models and Minimizing Cut Sets of Fault Trees method carry out fault diagnosis to system, solve simple by model bulky complex in the method for fault tree modeling, not easily understand, the shortcoming such as maintenance, because multilevel flow models is a kind of system functional model and obey conservation principle, have intuitive strong, readable good, be easy to the features such as Inspection and maintenance; Solve again that conventional multilayer flow model not easily quantizes, the difficult point of calculation of complex simultaneously.
(2), existing multilevel flow models method generally adopts the method for alert analysis to carry out fault diagnosis, the present invention uses the technological means such as minimal cut set and importance degree in fault tree, quantitative analysis is carried out to multilevel flow models, thus improve the specific aim of fault diagnosis, to efficiently solve in alert analysis method threshold value to the impact of diagnostic result.
(3), the present invention is by can identify the parts and/or component combination that cause fault state the method for minimal cut set and importance of basic event hybrid-sorting simultaneously, solve the problem that only can identify single failure source in existing method for diagnosing faults, also can carry out fault diagnosis to system fast when fault chain ruptures, improve the accuracy of fault diagnosis.
Accompanying drawing explanation
Fig. 1 is realization flow figure of the present invention;
Fig. 2 is a certain process flow diagram for water storage system example;
Fig. 3 is a certain multilevel flow models for water storage system;
Fig. 4 is a certain evidence fault tree models for water storage system.
Embodiment
In order to better the present invention can be understood, first the basic conception related in the present invention is briefly described:
Multilevel flow models: a kind of KBS Knowledge Based System function modeling method, describes the process state of complex system from the target of system, function and physics realization three aspects;
Fault tree: a kind of handstand tree shape model of expression system failure mode, by the model successively launching downwards to set up till the basic element of character without the need to maybe continuing expansion to thrashing reason;
Elementary event: the leaf node of fault tree, namely in fault Tree without the need to maybe cannot continue launch the basic element of character, generally corresponds to one or more failure modes of parts;
Minimal cut set: represent the minimum combination that can cause one or more failure modes of one or more basic elements of character of thrashing, correspond to the set of one or more elementary event;
Main thought of the present invention is as follows:
When carrying out fault diagnosis to large-scale complicated system, adopt multilevel flow models to have strong, readable good, the feature that is easy to Inspection and maintenance of intuitive, but its modeling feature determines it not easily calculates, accuracy is difficult to good guarantee; Fault Tree Analysis is adopted to have the meticulous and feature that can quick and precisely solve of modeling process, but its model extremely bulky complex, modeling process or all suitable the wasting time and energy of review process, and readable poor, not easily understand, later maintenance is relative complex also.Therefore, the present invention will organically combine both it, proposes a kind of large-scale complicated system method for diagnosing faults based on multilevel flow models and Minimizing Cut Sets of Fault Trees, solves multilevel flow models poor accuracy and readable poor, the not intelligible problem of Fault Tree Analysis
The present invention is described in further detail below.
Technological scheme of the present invention: a kind of large-scale complicated system method for diagnosing faults based on multilevel flow models and Minimizing Cut Sets of Fault Trees, its flow chart is as Fig. 1, and implementation step is as follows:
Step (1), set up the multilevel flow models of system;
The cause and effect subtree of each functional status of the system multilayer flow model in step (2), establishment step (1);
Step (3), to characterize according to the physical fault of system, revise and simplify the cause and effect subtree of each functional status in step (2), then setting up the evidence fault tree of system;
Step (4), evidence fault tree according to the system obtained in step (3), the minimal cut set of solving system;
Step (5), according to the minimal cut set obtaining parts of the system of trying to achieve in step (4) or the hybrid-sorting list of component combination;
Step (6), sequentially fault diagnosis is carried out to the corresponding component in system or component combination according to the hybrid-sorting list of the parts obtained in step (5) or component combination, thus reach minimizing economic input and time overhead, improve the target of system fault diagnosis efficiency.
As above based on the large-scale complicated system method for diagnosing faults of multilevel flow models and Minimizing Cut Sets of Fault Trees, it is characterized in that: the method setting up the evidence fault tree of system in described step (3) is as follows:
1) other functional statuses used in its branch are substituted into the cause and effect subtree of its correspondence and launch by the cause and effect subtree model of a certain functional status respectively in selecting system;
2) substituting into and launching in the process of cause and effect subtree, as there is repeated events, need the tree structure of blocking this, thus eliminate logical loops, finally generate the complete because of fruit tree of a certain functional status;
3) the cause and effect subtree model of all need functional status to be processed in selecting system successively, repeats step 1)-2) several times, until all generate complete because of fruit tree for all functions state;
4) by AND gate combination step 3) the middle all complete causal tree model generated, form the evidence fault tree of system.
As above based on the large-scale complicated system method for diagnosing faults of multilevel flow models and Minimizing Cut Sets of Fault Trees, it is characterized in that: as follows by the method for the minimal cut set obtaining parts of system or the hybrid-sorting list of component combination in described step (5):
(A) the percentage importance degree of each cut set is calculated respectively according to the minimal cut set of the system obtained;
(B) the FV importance degree of elementary event in all cut sets is calculated respectively according to the minimal cut set of the system obtained;
(C) according to the importance degree information obtained in step (A) and step (B), the hybrid-sorting list of obtaining parts or component combination.
The invention process example is as follows:
Fig. 2 is a certain for water storage system sketch, and this system is made up of pump, water tank and associated conduit, and pump draws water and transports to water tank.
Setting up multilevel flow models as shown in the figure, there is following fault mode in this system:
B1: motor-drive pump operation troubles;
B2: water tank is revealed;
B3: line clogging;
Suppose that sensor detects following trouble signal in certain moment:
Motor-drive pump low discharge (F2 is low);
Pipeline low discharge (F4 is low);
Set up the evidence fault tree of system thus as Fig. 4, solve this evidence fault tree, and calculate the importance sorting of all cut sets and elementary event, the item that wherein importance degree is the highest is B3P34'B1, known by analysis, show normal situation at functional unit F3, the possibility that basic fault B1 and B3 occurs is maximum.
This example can be extended in the failure diagnostic process of complex system, and by computer software, carry out rapid failure diagnosis, the decision-making for operator provides auxiliary, improves diagnosis efficiency.
Non-elaborated part of the present invention belongs to techniques well known.
The above; be only part embodiment of the present invention, but protection scope of the present invention is not limited thereto, any those skilled in the art are in the technical scope that the present invention discloses; the change that can expect easily or replacement, all should be encompassed within protection scope of the present invention.
Claims (3)
1., based on a large-scale complicated system method for diagnosing faults for multilevel flow models and Minimizing Cut Sets of Fault Trees, it is characterized in that implementation step is as follows:
Step (1), set up the multilevel flow models of system;
The cause and effect subtree of each functional status of the system multilayer flow model in step (2), establishment step (1);
Step (3), to characterize according to the physical fault of system, revise and simplify the cause and effect subtree of each functional status in step (2), then setting up the evidence fault tree of system;
Step (4), evidence fault tree according to the system obtained in step (3), the minimal cut set of solving system;
Step (5), according to the minimal cut set obtaining parts of the system of trying to achieve in step (4) or the hybrid-sorting list of component combination;
Step (6), sequentially fault diagnosis is carried out to the corresponding component in system or component combination according to the hybrid-sorting list of the parts obtained in step (5) or component combination, thus reach minimizing economic input and time overhead, improve the target of system fault diagnosis efficiency.
2., according to the large-scale complicated system method for diagnosing faults based on multilevel flow models and Minimizing Cut Sets of Fault Trees described in claim 1, it is characterized in that: the method setting up the evidence fault tree of system in described step (3) is as follows:
1) other functional statuses used in its branch are substituted into the cause and effect subtree of its correspondence and launch by the cause and effect subtree model of a certain functional status respectively in selecting system;
2) substituting into and launching in the process of cause and effect subtree, as there is repeated events, need the tree structure of blocking this, thus eliminate logical loops, finally generate the complete because of fruit tree of a certain functional status;
3) the cause and effect subtree model of all need functional status to be processed in selecting system successively, repeats step 1)-2) several times, until all generate complete because of fruit tree for all functions state;
4) by AND gate combination step 3) the middle all complete causal tree model generated, form the evidence fault tree of system.
3. according to the large-scale complicated system method for diagnosing faults based on multilevel flow models and Minimizing Cut Sets of Fault Trees described in claim 1, it is characterized in that: as follows by the method for the minimal cut set obtaining parts of system or the hybrid-sorting list of component combination in described step (5):
(A) the percentage importance degree of each cut set is calculated respectively according to the minimal cut set of the system obtained;
(B) the FV importance degree of elementary event in all cut sets is calculated respectively according to the minimal cut set of the system obtained;
(C) according to the importance degree information obtained in step (A) and step (B), the hybrid-sorting list of obtaining parts or component combination.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510312479.7A CN104879295A (en) | 2015-06-09 | 2015-06-09 | Large complex system fault diagnosis method based on multilevel flow model and minimal cutset of fault tree |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510312479.7A CN104879295A (en) | 2015-06-09 | 2015-06-09 | Large complex system fault diagnosis method based on multilevel flow model and minimal cutset of fault tree |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104879295A true CN104879295A (en) | 2015-09-02 |
Family
ID=53946928
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510312479.7A Pending CN104879295A (en) | 2015-06-09 | 2015-06-09 | Large complex system fault diagnosis method based on multilevel flow model and minimal cutset of fault tree |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104879295A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105486526A (en) * | 2015-11-30 | 2016-04-13 | 北京宇航***工程研究所 | Multi-strategy fault diagnosis system for carrier rocket test launching process |
| CN106226055A (en) * | 2016-08-04 | 2016-12-14 | 哈尔滨工程大学 | The monitoring reliability method that a kind of nuclear power plant based on fault tree valve body lost efficacy |
| CN110175359A (en) * | 2019-04-23 | 2019-08-27 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Complication system Safety Modeling Methods and device based on operation flow |
| CN114080577A (en) * | 2019-07-12 | 2022-02-22 | 西门子工业软件有限责任公司 | Ring closure and normalized representation in fault trees |
-
2015
- 2015-06-09 CN CN201510312479.7A patent/CN104879295A/en active Pending
Non-Patent Citations (2)
| Title |
|---|
| 朱大奇等: "基于故障树最小割集的故障诊断方法研究", 《数据采集与处理》 * |
| 陈强等: "基于多层流模型和故障树的靠性分析方法研究", 《原子能科学技术》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105486526A (en) * | 2015-11-30 | 2016-04-13 | 北京宇航***工程研究所 | Multi-strategy fault diagnosis system for carrier rocket test launching process |
| CN105486526B (en) * | 2015-11-30 | 2018-02-09 | 北京宇航***工程研究所 | A kind of how tactful fault diagnosis system for carrier rocket test emission process |
| CN106226055A (en) * | 2016-08-04 | 2016-12-14 | 哈尔滨工程大学 | The monitoring reliability method that a kind of nuclear power plant based on fault tree valve body lost efficacy |
| CN106226055B (en) * | 2016-08-04 | 2018-07-24 | 哈尔滨工程大学 | A kind of monitoring reliability method of nuclear power plant's valve body failure based on fault tree |
| CN110175359A (en) * | 2019-04-23 | 2019-08-27 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Complication system Safety Modeling Methods and device based on operation flow |
| CN114080577A (en) * | 2019-07-12 | 2022-02-22 | 西门子工业软件有限责任公司 | Ring closure and normalized representation in fault trees |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103399820B (en) | The breakdown judge system and method for a kind of sequence based on event tree and consequences analysis | |
| CN103116540B (en) | Dynamic symbol execution method based on global superblock domination graph | |
| CN104879295A (en) | Large complex system fault diagnosis method based on multilevel flow model and minimal cutset of fault tree | |
| CN104778370A (en) | Risk analyzing method based on Monte-Carlo simulation solution dynamic fault tree model | |
| CN104063586A (en) | Polymorphic failure tree-based bayesian network failure prediction method | |
| CN103810383A (en) | Method for identifying and analyzing single-point failure mode of launch vehicle | |
| CN105070097A (en) | Risk recognition method for operating vehicle road traffic accidents | |
| CN106155035A (en) | Method for diagnosing faults based on maintenance class data and fault diagnosis system | |
| CN104484548A (en) | Improved optimizing method for sequential fault diagnosis strategy | |
| CN104318110A (en) | Method for improving risk design and maintenance efficiency of large complex system on basis of importance and sensibility complex sequence | |
| Mhalla et al. | Estimation of failure probability of milk manufacturing unit by fuzzy fault tree analysis | |
| CN115981902A (en) | Fine-grained distributed micro-service system abnormal root cause positioning method and device | |
| Yang et al. | Bidirectional implementation of Markov/CCMT for dynamic reliability analysis with application to digital I&C systems | |
| CN117061318A (en) | Fault diagnosis method, device, equipment, medium and product | |
| CN106875015B (en) | A kind of airplane fault diagnostic method and system | |
| Han et al. | Fault diagnosis of hydraulic variable pitch for wind turbine based on qualitative and quantitative analysis | |
| CN102024084A (en) | Establishment and using method of reliability model having fuzzy polymorphism characteristic | |
| Liu et al. | Reliability analysis of a random fuzzy repairable parallel system with two non-identical components | |
| CN104050079A (en) | Real-time system testing method based on time automata | |
| CN104102562A (en) | Cellular automata-based single particle soft error system level propagation analysis method | |
| CN103197564A (en) | Digitalized human-machine interface monitoring unit number optimization method and system | |
| CN110262465B (en) | Flight control fault diagnosis method based on fault code classification | |
| CN113205215A (en) | Knowledge-based battlefield situation prediction method | |
| Rushby et al. | Evaluating the assessment of software fault-freeness | |
| CN106528401B (en) | The tripping test method and device of control system based on redundancy structure |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20150902 |
|
| RJ01 | Rejection of invention patent application after publication |