CN103792932A - Fault detection method based on ECA rule - Google Patents
Fault detection method based on ECA rule Download PDFInfo
- Publication number
- CN103792932A CN103792932A CN201210430601.7A CN201210430601A CN103792932A CN 103792932 A CN103792932 A CN 103792932A CN 201210430601 A CN201210430601 A CN 201210430601A CN 103792932 A CN103792932 A CN 103792932A
- Authority
- CN
- China
- Prior art keywords
- node
- rule
- event
- eca
- action
- 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
Images
Landscapes
- Test And Diagnosis Of Digital Computers (AREA)
Abstract
The invention relates to a high-speed train fault detection method based on an ECA rule. To begin with, the ECA rule is utilized to carry out modeling on faults of a network control system of a high-speed train, so that a complex sequential relationship between fault data can be described, and the fault description ability is enhanced; then, since the ECA rule is suitable for describing active behaviors of the system, an active working mode of " event detecting-condition judging-action triggering" of a fault diagnosis system can be supported, and the defect of a conventional "request-response " mode is overcome; meanwhile, heuristic search based on rule graphs is utilized in the method, so that the occurred faults can be diagnosed as many as possible in a given time based on collected data; and since the input of the fault diagnosis based on the ECA rule is atomic events but not original collected data, data irrelevant to the diagnosis rules is filtered out, and the processing efficiency of the fault diagnosis system under the condition of mass collected data is improved.
Description
Technical field
Patent of the present invention relates to fault detection technique, is specifically related to a kind of fault detection method based on eca rule of bullet train network control system.
Background technology
ECA(event-condition-action) rule has very strong semantic meaning representation ability, can support integrality to keep, the various application such as derived data maintenance, production monitoring, market monitoring and decision support system (DSS), ECA is accepted widely as the mechanism of taking the initiative in offering a hand of active database system.
Along with developing rapidly of computer technology and automated control technology, the popularization of various application systems, control difficulty increases, association between all parts becomes increasingly complex, the generation of a fault may have an immense impact on and even cause thrashing system, and therefore enterprise is badly in need of that one sensitive, fault diagnosis system is as the assurance of safety, High-efficient Production accurately.Fault diagnosis is exactly that (comprising can digital information for the characteristic information of collecting device in operation process, mechanical parameter while operation as equipment and non-can digital information, as some is not easy to the blooming of accurately describing), determine that based on characteristic information whether equipment running status is good.Concrete failure diagnostic process may be summarized to be following several stages:
Signals collecting: in service due to vibrations at equipment, the natural processes such as heating, the parameters of equipment can constantly change.The characteristic information difference that distinct device need to be paid close attention to, therefore can carry out signal crawl at the sensor of some location arrangements particular type, such as common speed pickup, temperature sensor and optical sensor etc.;
Signal processing: the signal collecting is as raw information, may not effective especially to the description of equipment state feature, therefore need these raw informations to process, for example vibration information is carried out to the conversion of time domain to frequency domain, or comprehensive some original signals and form some composite signals, make signal after treatment can more accurately portray equipment state;
State recognition: using the relevant knowledge of Signals & Systems after treatment (as canonical parameter etc.) as input, whether investigate the described state of signal conforms to normal condition, thereby determine whether system exists fault, and the relevant information of fault, as type, character etc.
Diagnosis decision-making: this step can be considered as the extension phase of fault diagnosis; obtain after diagnostic result; according to information determining system running statuses such as fault levels; and take appropriate measures; as made shutdown maintenance decision to high-risk fault; prevent that equipment running status from worsening, and can enable alternate device to low level fault, and start corresponding overhaul procedure.
All fault diagnosis technologies all can be divided three classes at present: i.e. the method based on knowledge, the method based on analytic model and the method based on signal processing.Wherein, the method based on knowledge, with reference to the domain knowledge of long-term accumulation, does not need mathematical models, demonstrates certain intelligent characteristic simultaneously, is introduced into multiple industrial circles, for example, be applied to bullet train fault diagnosis.Wherein, rule-based diagnostic expert system is as the representative of these class methods, because diagnostic knowledge statement is visual and clear, diagnosis effect is good, and be subject to the generally attention of application, be widely used in medical diagnosis and engineering failure diagnosis, become the most active and the most ripe at present method for diagnosing faults, but such method for diagnosing faults existed following three problems:
(1) limited to the descriptive power of fault: existing fault expert system is all based on production rule, the associated employing production representation between the data that sum up according to association area knowledge and fault, and then the rule-based fault diagnosis of carrying out.But, in fault diagnosis, first need to carry out the extraction of apparatus characteristic information, under a lot of scenes for fault is carried out to accurate description, except paying close attention to the quantity, logical relation of characteristic information, sequential relationship between information also needs careful consideration, and production rule is numerous and diverse and fuzzy to the description of this type of relation.
(2) Efficiency Decreasing under large-scale data: in practical application scene, data volume is quite huge often, and often there are the data irrelevant with diagnosis, because traditional fault expert diagnostic system does not filter these data, cause these redundant datas to pour in diagnostic system, reduced the efficiency of fault diagnosis.
(3) shortcoming initiative: traditional fault diagnosis system is request-answer-mode, this makes it lack initiative.
Summary of the invention
Technical matters to be solved by this invention is: how the fault of bullet train network control system is described to how to judge efficiently occurred fault according to fault data.
In order to solve the problems of the technologies described above, the present invention proposes a kind of fault detection method based on eca rule, and its step comprises:
1) corresponding relation with eca rule according to the diagnosis rule foundation in bullet train bug list, obtains eca rule model;
2) set up the rule schema of described eca rule model, described rule schema comprises and connecting by directed edge: atomic event Event node, compound event Event node, condition C ondition node and action Action node, in the rule schema of described eca rule model, entrance is made as atom Event node, and outlet is made as Action node;
3) adopt the heuristic search of rule-based figure to start to set up directed connection from described atomic event Event node, mate in rule schema according to described Condition node, find corresponding Action node;
4) carry out the action that described Action node represents in described bullet train bug list, the information of output corresponding failure, completes detection.
Described eca rule model is set up as follows:
2-1), for every diagnosis rule in described bullet train bug list, the value condition of the each operation data being provided is expressed as an atom Event;
2-2) according to the relation between operation data value condition related in described diagnosis rule, described atom Event is combined and obtains compound Event;
2-3), according to the relation composition Boolean expression between the data attribute of related data in described diagnosis rule, obtain Condition;
2-4) extracting the Output rusults that in described bullet train bug list, the corresponding failure message of diagnosis rule obtains is Action.
In the directed edge of described eca rule model rule schema, the compound Event that it forms is pointed to by atomic event in the limit that connects atom Event node and compound Event node; This rule condition is pointed to by the compound Event of top layer in a rule in the limit that connects Event node and Condition node; Connect Condition node and Action node limit and point to rule action by rule condition.
Described compound event Event is accorded with and being connected to form by event action by some atomic events, the value condition of the each operation data providing in the corresponding described bullet train bug list of described atomic event.
Extract described atomic event model according to the diagnosis rule in bullet train bug list and obtain atomic event, and the atomic event that described extraction is generated mates with the leaf node in rule schema.
The heuristic search concrete steps of described rule-based figure are as follows:
6-1) from the leaf node having mated using the leaf node of choosing as present node;
6-2) the described expression atomic event that comprises present node or compound event content and the temporal information that produces this event are passed to all father nodes of described this atomic event or compound event;
6-3) calculate the target function value of each father node;
6-4) according to the father node of described target function value select target functional value minimum as present node, whether the event that judges this node representative occurs or whether the condition of this node representative meets, and finds action node;
6-5) the action node output of coupling will be met.
Described target function value
wherein v is node, and L is the number in the path from node v to all outlet ports node, and PEs is the set of event node on s paths in PT, PT={pt1, and pt2 ..., pt
lthe set of paths from node v to all outlet ports node, WT (v
j) by node v
jthe expection stand-by period of the event of representative.
Described step 2-3) in described diagnosis rule between the data attribute of related data by with or logical operator connect to form Boolean expression.
Duration, relative Relative, constraint sequential Tcs or Dis tetra-class compound events in the middle of described compound event comprises.
Described atomic event mates and forms data stream according to its order of occurrence input rule figure and with corresponding leaf node.
Beneficial effect of the present invention
First, the present invention utilizes eca rule to carry out modeling to the fault of bullet train network control system, can describe the complex time sequence relation between fault data, thereby has strengthened the descriptive power to fault.Secondly, because eca rule is suitable for the active behavior of descriptive system, therefore, can support the active work mode of fault diagnosis system " detection event-Rule of judgment-trigger action ", overcome the shortcoming of tradition " ask-reply " pattern.Meanwhile, adopt the heuristic search of rule-based figure as much as possible according to the data that collect within preset time, to diagnose out the fault of generation.In addition, due to the fault diagnosis based on eca rule, its input is former component, rather than acquired original data, and this has just filtered the data irrelevant with diagnostic rule, thereby has improved the treatment effeciency of fault diagnosis system under magnanimity image data.
Accompanying drawing explanation
Fig. 1 is implication and the detection method schematic diagram that the present invention is based on compound event in fault detection method one embodiment of eca rule;
Fig. 2 the present invention is based on rule schema schematic diagram in fault detection method one embodiment of eca rule;
Fig. 3 the present invention is based on the data flow diagram collecting in fault detection method one embodiment of eca rule;
Fig. 4 the present invention is based on regular triggering situation schematic diagram in fault detection method one embodiment of eca rule;
Fig. 5 the present invention is based on detection method process flow diagram in fault detection method one embodiment of eca rule.
Specific implementation method
Under regard to the present invention propose the method for diagnosing faults based on eca rule, be elaborated in conjunction with following instance.
As shown in Figure 5, for solving technical matters as above, the invention provides a kind of method for diagnosing faults based on Event-condition-action (ECA) rule, comprise the following steps:
(1) the supporting train fault table of bullet train is analyzed, each correspondence in this bug list a diagnosis rule, every diagnosis rule has been described the corresponding relation between bullet train operation data and failure code, adopt eca rule to represent each diagnosis rule, (be that a kind of active rule (can be referring to R.Adaikkalavan at this eca rule, S.Chakravarthy, " Snooplb:Interval-based event specification and detection for active databases ", IEEE Transactions on Data Knowl.Eng.59 (1), 2006, pp.139-165.), its form is On event, When condition, Do action, it is semantic in the time that regular defined event occurs, and the condition of evaluation condition part, if condition is very, performs an action) this step specifically comprises:
(1.1) for every diagnosis rule, the value condition of the each operation data being provided is expressed as an atomic event, and at this, atomic event refers to the signal that causes concern at the needs of a physics moment generation, can not decompose again.For example, to be greater than 100 degree be exactly an atomic event to train axle temperature
(1.2) relation between related operation data value condition in analysis of failure rule, select suitable compound event pattern to combine the atomic event obtaining in (1.1), form compound event, (at this, compound event has some atomic events to accord with connecting by event action to form; Conventional event action symbol comprise with, or etc.), thereby form event (Event) part based on eca rule fault model;
(1.3) relation between the data attribute of related data in analysis of failure rule, by these relations be organized as by with or the Boolean expression that is formed by connecting of logical operator, form condition (Condition) part based on eca rule fault model;
(1.4) extract the corresponding failure message of diagnosis rule, as title, grade, code etc., the action of eca rule action (Action) part corresponding with this diagnosis rule is these failure messages of output.
(2) all eca rules of describing fault being expressed as to rule schema (can be referring to Y.Qiao, X.Li, H.Wang, K.Zhong, " Real-Time Reasoning Based on Event-Condition-Action Rules ", Proceedings of International Conference on Cooperative Information Systems (Poster Session), Lecture Notes in Computer Science, Springer, Nov.2008, pp.1-2.), rule schema is a directed acyclic graph, node and directed edge are comprised.Node can be divided into event node, condition node and action node.Directed edge may connection event and condition node, condition and action node, event and event node.The orientation determination on limit a pair of set membership---directed edge points to father node from child node, thereby indicated the direction of Information Conduction---information is conducted to father node from child node.In directed edge, the compound event that it forms is pointed to by subevent in the limit of connection event and event; This rule condition is pointed to by top layer compound event in a rule in connection event and condition limit; Rule action is pointed to by rule condition in condition and action limit.Degree according to node divides, and the node that in-degree is zero is called an entrance of rule schema, and the node that out-degree is zero is called an outlet of rule schema, and in general, entrance is atomic event node, exports as action node.Regular event, condition and an action are according to first detecting event, reevaluating condition, then, then the order performing an action, but between Different Rule, there is no particular order, for example, the detection of the event to Different Rule can be concurrent.
(3) according to given atomic event model, (at this, atomic event correspondence a value condition of operation data, get by analyzing bullet train bug list), the operation data obtaining from driving recording is analyzed and filtered, according to the situation of the operation data relevant to each diagnosis rule, generate corresponding atomic event;
(4) atomic event of proposing generation is mated with the leaf node in rule schema.At this, leaf node is the node that in rule schema, in-degree is 0, is to only have father node in rule schema, there is no the node of child node.
(5) adopt the heuristic search of rule-based figure (can be referring to Y.Qiao, X.Li, H.Wang, K.Zhong, " Real-Time Reasoning Based on Event-Condition-Action Rules ", Proceedings of International Conference on Cooperative Information Systems (Poster Session), Lecture Notes in Computer Science, Springer, Nov.2008, pp.1-2.), from the leaf node having mated, in rule schema, find corresponding action node, thereby diagnose out the fault of generation, its specific practice is:
(5.1) using the leaf node of choosing as present node
(5.2) token that comprises present node information is passed to its all father nodes, at this, the information that token comprises is: the atomic event that node is represented or the content of compound event and the time that produces this event.
(5.3) calculate the target function value of each father node, at the objective function of this node v, be designated as H (v)
Wherein, PT is the set of paths from node v to all outlet ports node.PT={pt1, pt2 ..., pt
l, L is the number in the path from node v to all outlet ports node.PEs is the set of event node on s paths in PT.WT (vj) is by the expection stand-by period of the event of node vj representative.It has reacted this event and how long has occurred in addition.
(5.4) father node of select target functional value minimum is as present node, and whether the event that judges this node representative occurs or whether the condition of this node representative meets; If judged result is true, proceed to step (5.2); If judged result is not true, selects another to mate, but there is no processed leaf node, and proceed to step (5.1).
(5.5) when searching out action node, proceed to step (6)
(6) the perform an action action of node representative, exports the information of corresponding failure.
Suppose bullet train trailer system electric voltage exception fault VoltageExceptional(VE) as follows with the failure-description of the abnormal ResistivityExceptional of trailer system resistance (RE):
Bullet train trailer system electric voltage exception fault VoltageExceptional(VE): the trailer system of bullet train is carried out to voltage setting, setting voltage is V-Set, after setting, voltage produces the virtual voltage V-Actual corresponding to this setting value in 2s, in the time that the actual voltage value V-ActualEvent.ActualValue of trailer system and trailer system setting voltage value V-SetEvent.SetValue error exceed threshold epsilon, report trailer system electric voltage exception fault (VE) occurs.
1) Rule-VE:(fault VE)
On TCS(V-SetEvnet，2s,=,Duration(V-SetEvent，V-ActualEvent，Relative(V-SetEvent，2s)))
IF
V-SetEvent.MatchAttr=V-ActualEvent.MatchAttr
AND
|V-SetEvent.SetValue-V-ActualEvent.ActualValue|>0.3
Do
show VoltageExceptionalInfo
The abnormal ResistivityExceptional of trailer system resistance (RE): bullet train is in service, if the currency V-Actual.value of trailer system virtual voltage V-Actual is greater than the currency Current.value of certain limit value maxStandardVoltage or electric current Current and is less than certain limit value minStandardCurrent, report that trailer system resistance abnormal failure RE occurs.
Can adopt following eca rule to carry out established model to fault VE and fault RE:
2) Rule-RE:(fault RE)
On Dis(V-ActualEvent，CurrentEvent）
If V-ActualEvent.ActualValue>2.5OR CurrenEventt.value<1.0
Do show ResistivityExceptionalInfo
Wherein, related centre (Duration) in rule, relatively (Relative), constraint sequential (Tcs) or (Dis) this four classes compound event implication as shown in Figure 1.(the detailed detection method of each compound event can be referring to document R.Adaikkalavan, S.Chakravarthy, " SnoopIb:Interval-based event specification and detection for active databases ", IEEE Transactions on Data Knowl.Eng.59 (1), 2006, pp.139 – 165.) atomic event V-SetEvent represents voltage, the attribute that this event comprises has setting value SetValue, match attribute MatchAttr, time stamp TimeStamp etc.; Atomic event V-ActualEvent represents virtual voltage, and the attribute that this event comprises has actual value Actualvalue, match attribute MatchAttr, time stamp TimeStamp etc.; Atomic event CurrentEvent represents current electric current.VoltageExceptionalInfo is the information of trailer system electric voltage exception fault, comprises fault title, failure code, and corresponding maintenance policy etc., while confirming that fault occurs, VoltageExceptionalInfo is as action in output.ResistivityExceptionalInfo is the information of trailer system resistance abnormal failure, comprises fault title, failure code, and corresponding maintenance policy etc., while confirming that fault occurs, ResistivityExceptionalInfo is as action in output.
Above-mentioned eca rule can be expressed as to rule schema as shown in Figure 2.
Suppose that the data stream collecting is as Fig. 3, through filtration stage, the atomic event set that obtains occurring successively
1, set
2, cur
1, act
2, act
1, rel
1, set
3, act
3, rel
2, rel
3.Wherein v corresponds respectively to V-SetEvent.SetValue attribute, the V-ActualEvent.ActualValue attribute of V-ActualEvent event and the Current.value attribute of CurrentEvent of V-SetEvent, and the Set event that numeric suffix is identical has identical MatchAttr attribute with Act event.
As shown in Figure 3, along with the generation of atomic event, the information in rule schema constantly flows.According to the sequencing occurring, atomic event is transfused to rule schema according to its order of occurrence and mates with corresponding leaf node, gives father node thereby produce token.
First set1 flows into rule schema, and Set node is triggered, and to his father's node R el, Dur and TCS transmit set1, because these three father nodes sort as Rel from small to large according to heuristic function value, and Dur, TCS, therefore according to this sequential processes.In the time that set2 flows into, processing procedure and set1 are similar.In the time that cur1 flows into, Cur node is triggered, transmit cur1 to its father node Dis, thereby activate Dis node, process known this node of Dis node and be triggered, produce event instance Dis1, and pass to its father node Con2, due to cur1.value=1.5, do not meet the requirement that is less than 1.0 in Con2, therefore Dis1 is dropped.In the time that act2 flows into, Act node is triggered, and transmits act2, thereby activate Dis and Dur to its father node Dis and Dur, due to H (Dis) <H (Dur), first processes Dis.Processing known this node of Dis node is triggered, produce event instance Dis2, and send its father node Con2 to, due to act2.value=3.0, meet the requirement that is greater than 2.5 in Con2, above parameter can be defined according to the diagnosis rule in bug list, and diagnosis rule embodies in the bug list being provided as technical information by train manufacturer.
Therefore perform an action, export RE-Fault
1.Then process Dur, by act
2be stored in its buffer memory.Act
1after inflow, with act
2processing procedure is similar.Work as rel
1occur, be passed to its father node Dur, due to rel
1for the example of the right subevent of Dur, therefore take out the oldest example Set in its left subevent queue Set queue
1, utilize Set
1with all examples and rel in neutron event Act queue
1common composition Dur example dur
1, and send this example, according to the definition of Dur types of events, dur
1time of origin be rel
1time of origin.By dur
1give its father node TCS, now, in rule schema, be activated and untreated node has and only has TCS, due to dur
1for the example of the right subevent of TCS, therefore take out the oldest event instance set in the left subevent Set queue of TCS
1, retrain set because their time stamp meets the REL defining in TCS
1with dur
1information combination, form the information of TCS example, then produce TCS example tcs
1and send.Tcs
1generation, can activate Con
1node.At Con
1in node, resolve tcs
1information (derive from set
1with dur
1), at dur
1act queue in find out and set
1the identical Act type instance of MatchAttr, then relatively whether both value difference is greater than 0.3.Due to dur
1act queue in exist and set
1the example act mating by MatchAttr
1, and both value value differences are different is | 1.0-1.8|>0.3, therefore Con
1meet, perform an action, output VE-Fault
1.Subsequent treatment is similar, repeats no more.Finally, set
1, set
2, cur
1, act
2, act
1, rel
1, set
3, act
3, rel
2, rel
3flow in the process of diagnostic system, regular triggering situation as shown in Figure 4.
After diagnostic system receives above-mentioned data, diagnosable three fault diagnosis example: ResistivityExceptional1, VoltageExceptional1, the VoltageExceptional2 of obtaining.
Claims (10)
1. the fault detection method based on eca rule, its step comprises:
1) corresponding relation with eca rule according to the diagnosis rule foundation in bullet train bug list, obtains eca rule model;
2) set up the rule schema of described eca rule model, described rule schema comprises and connecting by directed edge: atomic event Event node, compound event Event node, condition C ondition node and action Action node, in the rule schema of described eca rule model, entrance is made as atom Event node, and outlet is made as Action node;
3) adopt the heuristic search of rule-based figure to start to set up directed connection from described atomic event Event node, mate in rule schema according to described Condition node, find corresponding Action node;
4) carry out the action that described Action node represents in described bullet train bug list, the information of output corresponding failure, completes detection.
2. the fault detection method based on eca rule as claimed in claim 1, is characterized in that, described eca rule model is set up as follows:
2-1), for every diagnosis rule in described bullet train bug list, the value condition of the each operation data being provided is expressed as an atom Event;
2-2) according to the relation between operation data value condition related in described diagnosis rule, described atom Event is combined and obtains compound Event;
2-3), according to the relation composition Boolean expression between the data attribute of related data in described diagnosis rule, obtain Condition;
2-4) extracting the Output rusults that in described bullet train bug list, the corresponding failure message of diagnosis rule obtains is Action.
3. the fault detection method based on eca rule as claimed in claim 1 or 2, it is characterized in that, in the directed edge of described eca rule model rule schema, the compound Event that it forms is pointed to by atomic event in the limit that connects atom Event node and compound Event node; This rule condition is pointed to by the compound Event of top layer in a rule in the limit that connects Event node and Condition node; Connect Condition node and Action node limit and point to rule action by rule condition.
4. the fault detection method based on eca rule as claimed in claim 1, it is characterized in that, described compound event Event is accorded with and being connected to form by event action by some atomic events, the value condition of the each operation data providing in the corresponding described bullet train bug list of described atomic event.
5. the fault detection method based on eca rule as claimed in claim 4, it is characterized in that, extract described atomic event model according to the diagnosis rule in bullet train bug list and obtain atomic event, and the atomic event that described extraction is generated mates with the leaf node in rule schema.
6. the fault detection method based on eca rule as described in claim 1 or 5, is characterized in that, the heuristic search concrete steps of described rule-based figure are as follows:
6-1) from the leaf node having mated using the leaf node of choosing as present node;
6-2) the described expression atomic event that comprises present node or compound event content and the temporal information that produces this event are passed to all father nodes of described this atomic event or compound event;
6-3) calculate the target function value of each father node;
6-4) according to the father node of described target function value select target functional value minimum as present node, whether the event that judges this node representative occurs or whether the condition of this node representative meets, and finds action node;
6-5) the action node output of coupling will be met.
7. the fault detection method based on eca rule as claimed in claim 6, is characterized in that, described target function value
wherein v is node, and L is the number in the path from node v to all outlet ports node,
PEs is the set of event node on s paths in PT, PT={pt1, and pt2 ..., ptL} is the set of paths from node v to all outlet ports node, WT (v
j) by node v
jthe expection stand-by period of the event of representative.
8. the fault detection method based on eca rule as claimed in claim 2, is characterized in that, described step 2-3) in described diagnosis rule between the data attribute of related data by with or logical operator connect to form Boolean expression.
9. the fault detection method based on eca rule as claimed in claim 1, is characterized in that, Duration, relative Relative, constraint sequential Tcs or Dis tetra-class compound events in the middle of described compound event comprises.
10. the fault detection method based on eca rule as claimed in claim 5, is characterized in that, described atomic event mates and forms data stream according to its order of occurrence input rule figure and with corresponding leaf node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210430601.7A CN103792932A (en) | 2012-11-01 | 2012-11-01 | Fault detection method based on ECA rule |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210430601.7A CN103792932A (en) | 2012-11-01 | 2012-11-01 | Fault detection method based on ECA rule |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103792932A true CN103792932A (en) | 2014-05-14 |
Family
ID=50668719
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210430601.7A Pending CN103792932A (en) | 2012-11-01 | 2012-11-01 | Fault detection method based on ECA rule |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103792932A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105550400A (en) * | 2015-12-04 | 2016-05-04 | 西安泛华科技开发有限公司 | Modeling method of system time sequence |
| CN107491311A (en) * | 2017-08-16 | 2017-12-19 | 广州视源电子科技股份有限公司 | Generate method, system and the computer equipment of pagefile |
| WO2021073367A1 (en) * | 2019-10-16 | 2021-04-22 | 华为技术有限公司 | Data processing method, device, and system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101814114A (en) * | 2010-01-07 | 2010-08-25 | 北京航空航天大学 | Method for performing fault diagnosis by using model conversion |
| CN101945009A (en) * | 2010-09-14 | 2011-01-12 | 国网电力科学研究院 | Positioning method and device of power communication network fault based on case and pattern matching |
-
2012
- 2012-11-01 CN CN201210430601.7A patent/CN103792932A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101814114A (en) * | 2010-01-07 | 2010-08-25 | 北京航空航天大学 | Method for performing fault diagnosis by using model conversion |
| CN101945009A (en) * | 2010-09-14 | 2011-01-12 | 国网电力科学研究院 | Positioning method and device of power communication network fault based on case and pattern matching |
Non-Patent Citations (1)
| Title |
|---|
| 李欣等: "基于ECA规则推理的故障诊断技术", 《计算机工程与设计》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105550400A (en) * | 2015-12-04 | 2016-05-04 | 西安泛华科技开发有限公司 | Modeling method of system time sequence |
| CN107491311A (en) * | 2017-08-16 | 2017-12-19 | 广州视源电子科技股份有限公司 | Generate method, system and the computer equipment of pagefile |
| WO2021073367A1 (en) * | 2019-10-16 | 2021-04-22 | 华为技术有限公司 | Data processing method, device, and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110647539B (en) | Prediction method and system for vehicle faults | |
| Cordier et al. | Alarm driven monitoring based on chronicles | |
| Vodenčarević et al. | Identifying behavior models for process plants | |
| Ren et al. | An online adaptive approach to alert correlation | |
| CN101783749B (en) | Network fault positioning method and device | |
| WO2017211593A1 (en) | System and method for the asset management of railway trains | |
| Tomola et al. | Robust disjunctive-codiagnosability of discrete-event systems against permanent loss of observations | |
| CN110535878A (en) | A kind of threat detection method based on sequence of events | |
| CN110460591A (en) | Based on the CDN Traffic anomaly detection device and method for improving separation time memory network | |
| CN107409134A (en) | Method card analysis | |
| CN110597235A (en) | Universal intelligent fault diagnosis method | |
| CN103792932A (en) | Fault detection method based on ECA rule | |
| US20040073843A1 (en) | Diagnostics using information specific to a subsystem | |
| CN115372816A (en) | Power distribution switchgear operation fault prediction system and method based on data analysis | |
| CN115800272A (en) | Power grid fault analysis method, system, terminal and medium based on topology identification | |
| Weiss | Predicting telecommunication equipment failures from sequences of network alarms | |
| CN110727669B (en) | Electric power system sensor data cleaning device and cleaning method | |
| Syfert et al. | Development and application of methods for actuator diagnosis in industrial control systems (DAMADICS): A benchmark study | |
| Cui et al. | Fault propagation reasoning and diagnosis for computer networks using cyclic temporal constraint network model | |
| Roychoudhury et al. | Comprehensive diagnosis of continuous systems using dynamic bayes nets | |
| CN114625009B (en) | Fault detection method based on system identification and optimal filtering | |
| Renders et al. | A prototype neural network to perform early warning in nuclear power plant | |
| Lo et al. | Model-based fault diagnosis in continuous dynamic systems | |
| JP3012297B2 (en) | Abnormal event identification method and device | |
| M’halla et al. | Monitoring of a milk manufacturing workshop using chronicle and fault tree approaches |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140514 |