CN101907868A - Intelligent trouble diagnosis method for tractive power supply system and system thereof - Google Patents
Intelligent trouble diagnosis method for tractive power supply system and system thereof Download PDFInfo
- Publication number
- CN101907868A CN101907868A CN 201010246219 CN201010246219A CN101907868A CN 101907868 A CN101907868 A CN 101907868A CN 201010246219 CN201010246219 CN 201010246219 CN 201010246219 A CN201010246219 A CN 201010246219A CN 101907868 A CN101907868 A CN 101907868A
- Authority
- CN
- China
- Prior art keywords
- value
- fault
- diagnosis
- power supply
- tractive power
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Abstract
The invention provides an intelligent trouble diagnosis method for a tractive power supply system, which comprises the following steps: firstly establishing a system description library for a system to be diagnosed; then collecting observed value data required by the diagnosis in real time, carrying out pretreatment, acquiring system predicted values in accordance with a system element action description library and the observed value data, and performing a fuzzy consistency check on the predicted values and the observed values to judge the difference, if not, performing a layering consistency diagnosis on the system to generate a fault candidate collection; selecting a fault action model to perform an abductive diagnosis to determine fault reasons and fault positions, thereby predicating relay protection action and breaker action; and finally comparing the consistency of a predicated action result and real action to obtain the fault reasons and the results, and carrying out alarm output. The invention also provides an intelligent trouble diagnosis system for a tractive power supply system. The invention is suitable for the tractive power supply system, can immediately and exactly find out the fault elements and the fault reasons, overcomes computational complexity and realizes real-time monitoring and fault diagnosis for the tractive power supply system.
Description
Technical field
The present invention relates to a kind of method for diagnosing faults and system thereof, particularly a kind of intelligent failure diagnosis method of electric system and system thereof.
Background technology
Fault diagnosis for electric system, method commonly used both at home and abroad at present is to adopt expert system to develop the intelligent trouble diagnosis system of electrical network, and the foundation of this type of diagnostic system fault judgement mainly comes from the action message of circuit breaker position information and protective relay; Yet the situation of tripping or malfunction can take place in protective relay and isolating switch itself, these switching value signals are usually made mistakes in transmission and gatherer process even are lost simultaneously, above-mentioned situation all can cause the expert system situations such as reasoning error, misunderstanding that break down, thereby can not carry out fault judgement and eliminating to electric system timely and accurately; In addition, because the defective that expert system itself exists, as: rely on expertise merely, can not find unknown failure etc., so the portability that causes adopting expert system development to go out the intelligent trouble diagnosis system is poor, the diagnostic result reliability is not high.
And can be good at addressing the aforementioned drawbacks based on the method for diagnosing faults of model, it has can find unknown failure and the advantages of avoiding expertise such as dependence.Based on the model method for diagnosing faults is that discrete component is set up model, by setting up system structure model the discrete component model is connected to integral body, in case changed element or add element by diagnostic system, only need the parameter of change component models or increase the model and the change component structure model that newly add element to get final product, the modification quantities is little, and cost is low.Based on the model fault diagnosis two main research methods are arranged: consistance is diagnosed and traceed back because of diagnostic method: the consistance diagnostic method is that Reiter summed up early-stage Study person in 1987 and own achievement in research proposes, this method requires the system that will diagnose is set up element normal model and system structure model, and carry out reasoning according to the description and the difference between the observed reading of system, obtain minimum conflict set, thereby produce minimum diagnosis collection by minimum conflict set, under all minimum conflict sets are positive situation, by the minimum of diagnostic system diagnosis collection is the prime implicant of minimum conflict, because the incompleteness of systematic knowledge, the solution space that is produced diagnosis by the diagnosis of consistance diagnostic method has redundancy.Trace back because of diagnostic method is to be proposed by Poole D in 1989, this method is specified the element behavior model from system failure symptom, and the appointment that selection can the interpre(ta)tive system observed reading is as diagnosis; But when incomplete or model was unreliable when system model, tracing back often can't obtain satisfactory solution because of diagnosis.
Though avoided the dependence of expertise and had advantages such as to find unknown failure based on the method for diagnosing faults of model, yet this diagnostic method is not applied to the successful case in tractive power supply system intelligent trouble diagnosis field at present as yet, its reason is: on the one hand, because tractive power supply system belongs to high pressure, large-scale power system, it has the characteristics of dynamic uncertainty, electric power system exists many disturbing factors to make the measurement data of system have a large amount of noise datas in operational process, its measurement data also has dispersiveness and inexactness simultaneously, above characteristics not only cause the complicacy of cause and effect in the diagnostic reasoning, but also can influence the result of diagnosis; Be on the other hand: tractive power supply system is not static, irregularly during this time have that more complicated system---electric locomotive passes through, during some tractive power supply systems and can drop into series compensation device, the time and don't drop into the structure dynamic that series compensation device or the like situation has all caused tractive power supply system, and based on the method for diagnosing faults of model general only at structure fix, deterministic system.Therefore, how system architecture and system element behavior are described, how to overcome the difficulty that some elements are difficult to set up accurate model, how to set up uncertain inference mechanism, how utilizing based on the model fault Diagnosis Method in time, accurately to produce fault element and failure cause, all is technical matterss that this area needs to be resolved hurrily with defective of overcoming the big and computational complexity in candidate solution space or the like.
Summary of the invention
Fundamental purpose of the present invention provide a kind of be applicable to tractive power supply system, promptly and accurately find fault element and failure cause, overcome computational complexity based on the model intelligent failure diagnosis method, to realize real-time monitoring and fault diagnosis to tractive power supply system.
For achieving the above object; tractive power supply system intelligent failure diagnosis method provided by the invention is: at first set up by the system description storehouse of diagnostic system; the real-time required observed reading of acquisition of diagnostic and data are carried out pre-service then; after the observed reading acquisition system prediction value according to system element behavior description storehouse and collection; the predicted value and the described observed reading of described system are carried out the consistance Fuzzy Test to judge difference; if when inconsistent system is carried out the diagnosis of layering consistance to produce the fault candidate collection; select the fault behavior model to trace back again because of diagnosing to determine failure cause and abort situation; thereby prediction relay protection action and isolating switch action; by the consistance of comparison prediction the result of the action and actual act, draw failure cause and result and the output of reporting to the police.
By above scheme as seen, diagnostic method of the present invention has been realized, uncertain system model representation dynamic to structure, can carry out on-line monitoring and real-time diagnosis to tractive power supply system; In case break down, can in time, accurately produce fault element and failure cause by this method, and relay protection and isolating switch action are estimated, can also find unknown failure, overcome computational complexity, reduced the solution space of diagnosis based on the model method for diagnosing faults; Owing to adopted consistency check, the diagnosis of layering consistance in this diagnostic method and traceed back because of the method for diagnosis etc., therefore can alleviate greatly the traction substation staff labour intensity, increase work efficiency, can also effectively avoid dependence, improve the robotization and the intelligent level of troubleshooting experienced staff.
Another object of the present invention provides a kind of said method that adopts tractive power supply system is detected system with fault diagnosis.
For achieving the above object, provide a kind of tractive power supply system intelligent trouble diagnosis system to comprise: storage is by the system description module in the system description storehouse of diagnostic system; Whether carry out the man-machine conversation module of real-time data acquisition, pre-service and display result and detection system normal and to the diagnostic module of fault alarm; Described system description module comprises that system element behavior description storehouse, system architecture describe storehouse and relay protection and isolating switch action description storehouse; Described man-machine conversation module comprises data in real time signals collecting and processing module and man-machine dialog interface module; Described system description module is connected with the man-machine conversation module by described diagnostic module.
Adopt native system capture system failure message timely and effectively, location fault element and explanation failure cause in polynomial time, fault diagnosis accuracy rate height; Not only can find unknown failure, and can estimate, find the situation of isolating switch and protective relaying maloperation and tripping isolating switch and relay protection action; Simultaneously can also prevent unnecessary short-circuit current rush system, damage electric power system, therefore, utilization tractive power supply system intelligent trouble diagnosis system can improve operational reliability, the security of electric power system; In addition, the native system cost is little, installs simple, portable strong; The present invention can be extended to aspects such as light rail, subway, is with a wide range of applications.
Description of drawings
Fig. 1 is the general flow chart of the present invention's tractive power supply system intelligent failure diagnosis method.
Fig. 2 is the consistency check process flow diagram of the present invention's tractive power supply system system intelligence method for diagnosing faults.
Fig. 3 is the layering consistance diagnostic flow chart of the present invention's tractive power supply system intelligent failure diagnosis method.
Fig. 4 is the present invention's the tracing back because of diagnostic flow chart of tractive power supply system intelligent failure diagnosis method.
Fig. 5 is the system model figure of the present invention's tractive power supply system intelligent trouble diagnosis system.
Fig. 6 is that the system of the present invention's tractive power supply system intelligent trouble diagnosis system forms module diagram.
Fig. 7 is the wiring diagram of the present invention's tractive power supply system intelligent failure diagnosis method embodiment.
Fig. 8 is the present invention's the predicted value inference graph of tractive power supply system intelligent failure diagnosis method embodiment when the unregulated power locomotive operation.
Fig. 9 is the present invention's the predicted value inference graph of tractive power supply system intelligent failure diagnosis method embodiment when the electric locomotive operation is arranged.
Figure 10 is the man-machine inputting interface synoptic diagram of the present invention's tractive power supply system intelligent trouble diagnosis system embodiment.
Figure 11 is the man-machine inputting interface data input synoptic diagram of the present invention's tractive power supply system intelligent trouble diagnosis system embodiment.
Figure 12 is the system dynamics demonstration and the diagnostic result figure of the present invention's tractive power supply system intelligent trouble diagnosis system embodiment.
Figure 13 is the diagnostic result synoptic diagram of the present invention's tractive power supply system intelligent trouble diagnosis system embodiment.
The invention will be further described below in conjunction with each embodiment and accompanying drawing thereof.
Embodiment
All two three-phase transformers of tractive power supply system intelligent failure diagnosis method embodiment power transformation, a is earth point mutually, power supply mode is directly power supply, a power supply mode is single side feeding, series compensation, the calculating parameter of each transformer is shown in table 1-1 and table 1-2: table 1-1 single transformer calculating parameter
Three transformer calculating parameters of table 1-2
The physical cabling figure of electric substation is referring to Fig. 7.
Referring to Fig. 1, adopt method for diagnosing faults of the present invention as follows: S1: to set up the system description storehouse; At first set up by the system description storehouse of diagnostic system, comprise that system element behavior description storehouse, system architecture describe storehouse and relay protection and isolating switch action description storehouse; Wherein: the concrete grammar of setting up the system description storehouse is: s11: set up tractive power supply system structrual description storehouse; According to the relation of input value between tractive power supply system element and the element and output valve, wiring is described to whole tractive power supply system primary equipment, adopts sparse matrix to describe the structure of total system, if the total system structure is A=[a
11...., a
Nn], a
Ij=1 expression element c
iOutput valve be element c
jInput value; S12: the behavior description storehouse of setting up each element in the tractive power supply system; Promptly be to set up qualitative or quantitative model, element to be described under certain behavior model, the relation between output valve and the input value to each element fault behavior and normal behaviour in the electric power system; S13: set up relay protection and isolating switch action description storehouse; This describes the condition of position, numbering and the action of condition that the storehouse is the relay protection type that stores tractive power supply system, position, action and isolating switch.
S2: data acquisition and pre-service; The real-time data that need of acquisition of diagnostic and data are carried out filtering, remove pre-service such as make an uproar, wherein image data comprises from supervisory control comuter and gathers busbar voltage, current transformer in real time, electric parameters such as the voltage at feeder voltage current transformer place, electric current, power; Obtain switching values such as relay protection and isolating switch action message from remote terminal unit; The concrete waveform of electric current and voltage during from fault oscillograph collection fault.
S3: obtain the system prediction value; According to system structure model, element normal behaviour model and system's input value, to system just often the measured value at observation station place predict, obtain the system prediction value, referring to Fig. 8 and Fig. 9; The step that obtains the system prediction value is: s31: choose each element normal behaviour model from element system behavior description storehouse and put into the uncertain reasoning module; S32: select an input value In_system of system, if system's input value all processed then finish; S33: judge according to system element behavior description storehouse, if system's input value is element c
iInput value In_c
i, with input value substitution element c
iIn the normal behaviour model, with producing component c in the component models that the system input value links to each other
iOutput valve Out_c
i, if the c of element
iOutput valve is the value that will measure of systematic perspective measuring point, and then this is worth and is the system prediction value; S34: describe the storehouse according to system architecture, seek and element c
iIf the element that links to each other is element c
jWith element c
iLink to each other, i.e. element c
iOutput be element c
jInput, with Out_c
iAs c
jInput, enter s35 step, if element c
iNot linking to each other with any element in the system is element c
iOutput be not any element c
jInput, then get back to s32 step; S35: with Out_c
iAs c
jInput, substitution element c
jIn the normal behaviour model, with producing component c in the component models that the system input value links to each other
jOutput valve Out_c
j, change step s34 over to and operate.
Suppose system prediction value such as following table: table 1-1 through above-mentioned steps reasoning and calculation gained
Through pretreated observed reading be so: table 1-2
S4: consistency check: referring to shown in Figure 2, to be carried out the consistance Fuzzy Test by the voltage and current mutual inductor predicted value of diagnostic system and observed reading, difference between checking system observed reading and the predicted value, system prediction value and observed reading are inconsistent, then the tractive power supply system fault; The method of consistency check is: s41: calculate the observed reading of each observation station and the degree of consistency of predicted value, if the observed reading of all observation stations is all consistent in the predicted value of this observation station with normal model, then system is normal, not the system failure; The degree of consistency computing method of the observed reading of each observation station and predicted value are: certain observation station is S, the observed reading O that obtains at observation station S
S=[a, b] is if system's operate as normal is at this point prediction value P
S=[c, d], the consistance subjection degree of S point observation value and predicted value is:
φ
cBe the subordinate function to predicted value, the subordinate function general type is as follows:
G (x) can be given by expertise or experimental data, if
Then this point observation value is consistent with predicted value, and the value of setting when σ is diagnosis is generally given by the expert, is perhaps obtained by the data training.
S42: according to the consistency check result, judge whether the observed reading of this observation station is consistent with predicted value, if systematic perspective measured value and predicted value are inconsistent, the system failure.
S5: layering consistance diagnosis; S51: as shown in Figure 3,, be System={S with system divides according to observed reading and the inconsistent observation station of predicted value
1_ sub
1, S
1_ sub
2, S
1_ sub
1=b phase bus, b phase feeder line ...., the b net-fault of joining }, S
1_ sub
2=c phase bus, c phase feeder line ...., the c net-fault of joining } two subsystems; S52: the predicted value when subsystem is normal among the step S3, and judge according to the result who among the step S4 observed reading and the predicted value of two subsystems is carried out consistency check, if all observed readings of subsystem are consistent with predicted value, then subsystem is normal, all elements are all normal in this system, finish the diagnosis of layering consistance, otherwise subsystem is the fault subsystem, wherein all elements are the fault Candidate Set; S as a result
1_ sub
2Subsystem is normal, S
1_ sub
1Subsystem fault.To S
1_ sub
1Subsystem fault is divided S
1_ sub
1={ S
2_ sub
1, S
2_ sub
2, S
2_ sub
1=b phase bus, b phase feeder line ... .}, S
2_ sub
2=...., the b net-fault of joining }, these two subsystems are carried out the diagnosis of layering consistance, the result is S
2_ sub
1Normally, S
2_ sub
2Be the fault subsystem.
S6: trace back because of diagnosis; S61: as shown in Figure 4, from the element behavior model, select the fault behavior model of fault Candidate Set, tractive power supply system is traceed back because of diagnosis; S611: in element behavior description storehouse, select the transient fault behavior model of element contact net, in observed reading, add pretreated voltage oscillogram in the fault oscillograph; S612: select a subsystem input value In_system, then finish if the subsystem input value is all processed; S613: describe the storehouse according to system architecture, if the subsystem input value is element c
iInput value In_c
i, with input value substitution element c
iIn the normal behaviour model, (in the component models that links to each other with the system input value) producing component c
iOutput valve Out_c
i, if Out_c
iBe the value that observation station will be measured, then Out_c
iDiagnosis prediction value for subsystem; S614: describe the storehouse according to system architecture, seek and element c
iIf the element that links to each other is element c
jWith element c
iLink to each other and c
j∈ S_sub
i, i.e. element c
iOutput be element c
jInput, with Out_c
iAs c
jInput, enter step S62, if element c
iNot linking to each other with any element in the system is element c
iOutput be not any element c
jInput, then get back to step S613; S615: with Out_c
iAs c
jInput, substitution element c
jIn the normal behaviour model, (in the component models that links to each other with the system input value) producing component c
jOutput valve Out_c
j, change step S613 over to and operate.
S62: with the degree check that makes an explanation of diagnosis prediction value and observed reading, the check of explanation degree adopts following method to calculate: certain observation station is S, the observed reading O that obtains at observation station S
S=[a, b] is if system's operate as normal is at this point prediction value DP
S=[c, d], then explanation degree is that the degree of consistency is
φ
dBe the subordinate function of predicted value, the subordinate function formula is as follows:
g
t(x) can be given by expertise or experimental data; S63: through check, the diagnosis prediction value of all observation stations explains that to observed reading degree has
Then this element is a fault element, changes step s61 over to, and ε is given by expertise or experimental data, is the contact net transient fault according to one of result of said method diagnosis.
S7: relatively output; Predict relay protection action and isolating switch action according to failure cause and abort situation that step S6 diagnoses out; to predict the outcome and relay protection and isolating switch actual act compare; if inconsistent then isolating switch generation tripping or malfunction; report to the police; if consistent, then export failure cause and fail result.Fail result among this embodiment is output as: the contact net permanent fault.
Traction power supply intelligent trouble diagnosis system embodiment is referring to Fig. 5 and Fig. 6, and traction power supply intelligent trouble diagnosis system comprises: storage is by the system description module in the system description storehouse of diagnostic system; Whether carry out the man-machine conversation module of real-time data acquisition, pre-service and display result and detection system normal and to the diagnostic module of fault alarm; Described system description module comprises that system element behavior description storehouse, system architecture describe storehouse and relay protection and isolating switch action description storehouse; Described man-machine conversation module comprises data in real time signals collecting and processing module and man-machine dialog interface module; Described system description module is connected with the man-machine conversation module by described diagnostic module; The consistance of described diagnostic module detection system observed reading and actual observed value: if consistent, then system is normal; If it is inconsistent, then produce the candidate diagnosis collection by layering consistance diagnosis algorithm, again by tracing back because of diagnosis generation fault element and explaining failure cause, and the relay protection and the isolating switch action prediction value of passing through diagnostic result and relay protection and the generation of isolating switch action description storehouse compare, if the actual switch amount is consistent with predicted value, then relay protection and isolating switch are all normal; Otherwise system's generation isolating switch tripping or malfunction are reported to the police and are exported.
This intelligent trouble diagnosis system can the dynamic monitoring tractive power supply system, in case system breaks down, can in time reflect in native system, the yardman inserts traction substation by man-machine interface relation information, makes diagnostic system have to deal with the ability of multiple situation.
The professional studies certain fault signature for convenience, and this system provides static input function, and its inputting interface is referring to Figure 10; Data are as shown in figure 11 inserted in space in Figure 10; that is: Umb=[27.5 30] Imb=[0.015 0.8] Umc=[27.5 30] Imc=[0.015 0.8] Ukb=[27.5 30] Ikb=[0.1 0.8] Ukc=[0 0.01] Ikc=[0.015 0.08] selection " dynamic demonstration "; and selection " relay protection is arranged " in enquirement option thereafter; its dynamic demonstration and diagnostic result are referring to Figure 12; diagnostic result is the out-phase metallic short circuit, and the diagnostic result synoptic diagram is referring to Figure 13.
Claims (9)
1. tractive power supply system intelligent failure diagnosis method is characterized in that:
At first set up by the system description storehouse of diagnostic system; the real-time required observed reading of acquisition of diagnostic and data are carried out pre-service then; after the observed reading data acquistion system predicted value according to system element behavior description storehouse and collection; the predicted value and the described observed reading of described system are carried out the consistance Fuzzy Test to judge difference; if when inconsistent system is carried out the diagnosis of layering consistance to produce the fault candidate collection; select the fault behavior model to trace back again because of diagnosing to determine failure cause and abort situation; thereby prediction relay protection action and isolating switch action; by the consistance of comparison prediction the result of the action and actual act, draw failure cause and result and the output of reporting to the police.
2. tractive power supply system intelligent failure diagnosis method according to claim 1 is characterized in that:
This method specifically comprises the steps:
Step S1: set up the system description storehouse; Foundation is by the system description storehouse of diagnostic system, comprises that system element behavior description storehouse, system architecture describe storehouse, relay protection and isolating switch action description storehouse;
Step S2: data acquisition and pre-service; The real-time required observed reading of acquisition of diagnostic and data are carried out pre-service;
Step S3: obtain the system prediction value; According to system structure model, element normal behaviour model and system's input value, to system just often the measured value at observation station place predict, to obtain the system prediction value;
Step S4: consistency check; To be carried out the consistance Fuzzy Test to judge difference by the voltage and current mutual inductor predicted value of diagnostic system and observed reading, if the systematic perspective measured value is consistent with predicted value, then tractive power supply system is normal, test ending; Otherwise tractive power supply system is undesired, enters step S5;
Step S5: layering consistance diagnosis; The inconsistent observation station of selective system observed reading and predicted value according to the observation station position, is divided into several subsystems that are not coupled with tractive power supply system, then each subsystem is carried out the diagnosis of layering consistance, produces the fault candidate collection;
Step S6: trace back because of diagnosis; From system element behavior description storehouse, select the fault behavior model of fault Candidate Set to trace back, to determine failure cause and abort situation because of diagnosis; If trace back because of diagnostic result is sky, then this fault is a unknown failure, and the element in directly consistance being diagnosed can not be explained the element fault behavior as fault element;
Step S7: relatively output; Diagnose be out of order reason and abort situation according to step S6; prediction relay protection action and isolating switch action; to predict the outcome and compare with relay protection and isolating switch actual act; if it is inconsistent; then be isolating switch generation tripping or malfunction; if system alarm consistent, is then exported failure cause and fail result.
3. according to the described tractive power supply system intelligent failure diagnosis method of claim 2, it is characterized in that:
The concrete grammar of setting up the system description storehouse among the described step S1 is:
S11: set up tractive power supply system structrual description storehouse; According to the relation of input value between tractive power supply system element and the element and output valve, wiring is described to whole tractive power supply system primary equipment, adopts sparse matrix to describe the structure of total system, if the total system structure is A=[a
11, a
Nn], a
IjExpression element c
iOutput valve be element c
jInput value;
S12: the behavior description storehouse of setting up each element in the tractive power supply system; Promptly be to set up qualitative or quantitative model, element to be described under certain behavior model, the relation between output valve and the input value to each element fault behavior and normal behaviour in the electric power system;
S13: set up relay protection and isolating switch action description storehouse; This describes the condition of position, numbering and the action of condition that the storehouse is the relay protection type that stores tractive power supply system, position, action and isolating switch.
4. according to the described tractive power supply system intelligent failure diagnosis method of claim 2, it is characterized in that:
Acquisition data acquisition and pretreated concrete grammar comprise among the described step S2:
Gather busbar voltage, current transformer in real time, the electric parameters that the voltage at feeder voltage current transformer place, electric current, power and all diagnosis need; Obtain the switching value that relay protection and isolating switch action message and all diagnosis need; The concrete waveform of electric current and voltage when collecting fault; Then the data of gathering are carried out filtering, removed the pre-service of making an uproar.
5. according to the described tractive power supply system intelligent failure diagnosis method of claim 2, it is characterized in that:
The concrete grammar that obtains the system prediction value among the described step S3 is:
S31: the normal behaviour model of choosing each element from system element behavior description storehouse is put into the uncertain reasoning module;
S32: select an input value In_system of system, if all system's input value all processed then finish;
S33: judge according to system element behavior description storehouse, if system's input value is element c
iInput value In_c
i, then with the input value substitution element c of system
iIn the normal behaviour model, with producing component c in the component models that the system input value links to each other
iOutput valve Out_c
i, if the c of element
iOutput valve is the value that the systematic perspective measuring point will be measured, then the c of element
iOutput valve is the system prediction value;
S34: describe the storehouse according to system architecture, seek and element c
iIf the element that links to each other is element c
jWith element c
iLink to each other, i.e. element c
iOutput be element c
jInput, with Out_c
iAs c
jInput, enter step s35, if element c
iDo not link to each other, i.e. element c with any element in the system
iOutput be not any element c
jInput, then get back to step s32;
S35: with Out_c
iAs c
jInput, substitution element c
jIn the normal behaviour model, with producing component c in the component models that the system input value links to each other
jOutput valve Out_c
j, change step s34 then over to and operate.
6. according to the described tractive power supply system intelligent failure diagnosis method of claim 2, it is characterized in that:
The concrete grammar of consistency check is among the described S4:
S41: calculate the consistance subjection degree of the observed reading and the predicted value of each observation station, computing method are: certain observation station is S, the observed reading O that obtains at observation station S
S=[a, b] is if system's operate as normal is at this point prediction value P
S=[c, d], the consistance subjection degree of S point observation value and predicted value is
φ
cBe the subordinate function to predicted value, general type is as follows:
G (x) is given by expertise or experimental data, and the value that σ set for when diagnosis is given or obtained by the data training by the expert;
S42: whether the observed reading of judging this observation station is consistent with predicted value, if
Then this observed reading is consistent with predicted value, otherwise observed reading and predicted value are inconsistent, and system breaks down;
S43:, then obtain the observed reading of next observation station and change step s41 over to if do not obtained the observed reading of all observation stations;
S44: judge that if the observed reading of all observation stations is all consistent in the predicted value of this observation station with normal model, then system is normal, otherwise the system failure.
7. according to the described tractive power supply system intelligent failure diagnosis method of claim 2, it is characterized in that:
The concrete grammar of layering consistance diagnosis is among the described step S5:
S51: selecting observed reading and the inconsistent and described observed reading of predicted value is the observation station of one of them element input of system, according to the observation station position, will be divided into two subsystem System={S that are not coupled by diagnostic system
1_ sub
1, S
1_ sub
2;
S52: according to producing subsystem S among the step S3
1_ sub
iPredicted value just often, and according among the step S4 to the S of system
1_ sub
iObserved reading and the predicted value result that carries out consistency check judge, if all observed readings of subsystem are consistent with predicted value, subsystem S then
1_ sub
iNormally, all elements are all normal in this system, finish the diagnosis of layering consistance, otherwise subsystem are the fault subsystem, and wherein all elements are the fault Candidate Set;
S53: if S
1_ sub
iBe judged as the fault subsystem, and also have observation station at S
1_ sub
iIn, need be to S
1_ sub
iContinue to divide, get back to step s51.
8. according to the described tractive power supply system intelligent failure diagnosis method of claim 2, it is characterized in that:
Trace back among the described step S6 because of the concrete grammar of judging and be:
S61: S_sub in the fault subsystem
iThe middle element c that selects
k∈ S_sub
i, calculate the diagnosis prediction value, it is as follows to calculate diagnosis prediction value step:
S611: in system element behavior description storehouse, select element c
kThe fault behavior model and the normal behaviour model of all the other elements;
S612: select a subsystem input value In_system, then finish if the subsystem input value is all processed;
S613: describe the storehouse according to system architecture, if the subsystem input value is element c
iInput value In_c
i, with input value substitution element c
iIn the normal behaviour model, producing component c in the component models that links to each other with the system input value
iOutput valve Out_c
i, if Out_c
iBe the value that observation station will be measured, then Out_c
iDiagnosis prediction value for subsystem;
S614: describe the storehouse according to system architecture, seek and element c
iIf the element that links to each other is element c
jWith element c
iLink to each other and c
j∈ S_sub
i, i.e. element c
iOutput be element c
jInput, with Out_c
iAs c
jInput, enter step s62, if element c
iNot linking to each other with any element in the system is element c
iOutput be not any element c
jInput, then get back to step s613;
S615: with Out_c
iAs c
jInput, substitution element c
jIn the normal behaviour model, producing component c in the component models that links to each other with the system input value
jOutput valve Out_c
j, change step s613 over to and operate;
S62: with the degree check that makes an explanation of diagnosis prediction value and observed reading, the check of explanation degree adopts following method to calculate: the observed reading O that obtains at observation station S
S=[a, b] is if system's operate as normal is at this point prediction value DP
S=[c, d], then explanation degree is that the degree of consistency is
φ
dBe the subordinate function of predicted value, the subordinate function formula is as follows:
g
t(x) given by expertise or experimental data, S is an observation station;
S63: if the diagnosis prediction value of all observation stations explains that to observed reading degree has
Then this element is a fault element, changes over to
Step s61, ε is given by expertise or experimental data.
9. traction power supply intelligent trouble diagnosis system is characterized in that:
Comprise:
Storage is by the system description module in the system description storehouse of diagnostic system;
The man-machine conversation module of carrying out real-time data acquisition, pre-service and display result reaches
Whether detection system is normal and to the diagnostic module of fault alarm;
Described system description module comprises that system element behavior description storehouse, system architecture describe storehouse and relay protection and isolating switch action description storehouse; Described man-machine conversation module comprises data in real time signals collecting and processing module and man-machine dialog interface module; Described system description module is connected with the man-machine conversation module by described diagnostic module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010102462191A CN101907868B (en) | 2010-08-05 | 2010-08-05 | Intelligent trouble diagnosis method for tractive power supply system and system thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010102462191A CN101907868B (en) | 2010-08-05 | 2010-08-05 | Intelligent trouble diagnosis method for tractive power supply system and system thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101907868A true CN101907868A (en) | 2010-12-08 |
| CN101907868B CN101907868B (en) | 2012-01-25 |
Family
ID=43263349
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010102462191A Expired - Fee Related CN101907868B (en) | 2010-08-05 | 2010-08-05 | Intelligent trouble diagnosis method for tractive power supply system and system thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101907868B (en) |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102811143A (en) * | 2012-07-25 | 2012-12-05 | 北京星网锐捷网络技术有限公司 | Method and device for monitoring failure of data link layer, and network equipment |
| CN102830626A (en) * | 2012-09-13 | 2012-12-19 | 武汉瑞莱保能源技术有限公司 | Multilevel power system fault diagnosis system based on fault tree |
| CN103065195A (en) * | 2011-10-24 | 2013-04-24 | 索尼公司 | Power demand forecast device, method and system and power failure detection system |
| CN103197168A (en) * | 2013-02-21 | 2013-07-10 | 华东电网有限公司 | Fault diagnosis control method in power system based on event set causal chain |
| CN104915777A (en) * | 2015-06-12 | 2015-09-16 | 北京交通大学 | Relay protection judgment method based on dynamic tendency analysis |
| CN105469156A (en) * | 2014-09-11 | 2016-04-06 | 国网四川省电力公司电力科学研究院 | MOA condition management and fault prediction method and MOA condition management and fault prediction system |
| CN108363379A (en) * | 2018-01-23 | 2018-08-03 | 北京航空航天大学 | A kind of method for diagnosing faults based on satellite hot control system |
| CN108508359A (en) * | 2017-02-24 | 2018-09-07 | 通用电器技术有限公司 | Machine self monitor and diagnostic program based on electric characteristic analysis |
| CN109978345A (en) * | 2019-03-05 | 2019-07-05 | 清华大学 | A kind of bullet train trailer system combined failure dynamic risk analysis method based on characteristic quantity |
| CN110674193A (en) * | 2019-10-25 | 2020-01-10 | 国网四川省电力公司技能培训中心 | Intelligent substation relay protection fault information modeling method |
| CN110997447A (en) * | 2017-07-31 | 2020-04-10 | 西门子交通有限公司 | Method and device for monitoring a drive system of a rail vehicle |
| CN111061148A (en) * | 2018-10-17 | 2020-04-24 | 帆宣系统科技股份有限公司 | Intelligent pre-diagnosis and health management system and method |
| CN111707906A (en) * | 2020-07-14 | 2020-09-25 | 广州白云电器设备股份有限公司 | Method for realizing fault location of subway direct-current traction power supply system |
| CN114910763A (en) * | 2022-07-18 | 2022-08-16 | 广东电网有限责任公司佛山供电局 | GIS partial discharge monitoring method and device |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020121902A1 (en) * | 2001-01-11 | 2002-09-05 | Nissan Motor Co., Ltd. | Ground detection apparatus for electric vehicle |
| CN2581235Y (en) * | 2002-11-21 | 2003-10-22 | 成都铁华电子科技实业有限公司 | Detector for operating fault of locomotive tracting motor |
| KR20050093566A (en) * | 2004-03-19 | 2005-09-23 | 한국철도기술연구원 | A directiona and differential protective relay system for the traction power supply system to the dc operated electric rail car and the method thereof |
| CN101306653A (en) * | 2008-04-08 | 2008-11-19 | 北京交通大学 | Traction power supply equipment based on PWM rectifier and control method |
-
2010
- 2010-08-05 CN CN2010102462191A patent/CN101907868B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020121902A1 (en) * | 2001-01-11 | 2002-09-05 | Nissan Motor Co., Ltd. | Ground detection apparatus for electric vehicle |
| CN2581235Y (en) * | 2002-11-21 | 2003-10-22 | 成都铁华电子科技实业有限公司 | Detector for operating fault of locomotive tracting motor |
| KR20050093566A (en) * | 2004-03-19 | 2005-09-23 | 한국철도기술연구원 | A directiona and differential protective relay system for the traction power supply system to the dc operated electric rail car and the method thereof |
| CN101306653A (en) * | 2008-04-08 | 2008-11-19 | 北京交通大学 | Traction power supply equipment based on PWM rectifier and control method |
Non-Patent Citations (3)
| Title |
|---|
| 《中国测试技术》 20050828 张孟训等 牵引变压器模糊故障诊断技术 , 第04期 2 * |
| 《城市轨道交通研究》 20070330 叶俊等 磁浮列车系统故障模型的建立与分析 , 第03期 2 * |
| 《电气化铁道》 20061020 陈明忠等 基于Web的牵引变压器在线监测诊断系统 , 第05期 2 * |
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103065195A (en) * | 2011-10-24 | 2013-04-24 | 索尼公司 | Power demand forecast device, method and system and power failure detection system |
| US9880578B2 (en) | 2011-10-24 | 2018-01-30 | Sony Corporation | Power demand forecast device, method and system and power failure detection system |
| CN102811143A (en) * | 2012-07-25 | 2012-12-05 | 北京星网锐捷网络技术有限公司 | Method and device for monitoring failure of data link layer, and network equipment |
| CN102830626A (en) * | 2012-09-13 | 2012-12-19 | 武汉瑞莱保能源技术有限公司 | Multilevel power system fault diagnosis system based on fault tree |
| CN103197168A (en) * | 2013-02-21 | 2013-07-10 | 华东电网有限公司 | Fault diagnosis control method in power system based on event set causal chain |
| CN103197168B (en) * | 2013-02-21 | 2015-08-12 | 华东电网有限公司 | Realize the method for fault diagnosis control based on the event set chain of causation in electric system |
| CN105469156A (en) * | 2014-09-11 | 2016-04-06 | 国网四川省电力公司电力科学研究院 | MOA condition management and fault prediction method and MOA condition management and fault prediction system |
| CN104915777A (en) * | 2015-06-12 | 2015-09-16 | 北京交通大学 | Relay protection judgment method based on dynamic tendency analysis |
| CN104915777B (en) * | 2015-06-12 | 2018-11-02 | 北京交通大学 | Relay protection criterion method based on the analysis of dynamic trend degree |
| CN108508359A (en) * | 2017-02-24 | 2018-09-07 | 通用电器技术有限公司 | Machine self monitor and diagnostic program based on electric characteristic analysis |
| CN110997447A (en) * | 2017-07-31 | 2020-04-10 | 西门子交通有限公司 | Method and device for monitoring a drive system of a rail vehicle |
| CN108363379A (en) * | 2018-01-23 | 2018-08-03 | 北京航空航天大学 | A kind of method for diagnosing faults based on satellite hot control system |
| CN111061148A (en) * | 2018-10-17 | 2020-04-24 | 帆宣系统科技股份有限公司 | Intelligent pre-diagnosis and health management system and method |
| CN109978345A (en) * | 2019-03-05 | 2019-07-05 | 清华大学 | A kind of bullet train trailer system combined failure dynamic risk analysis method based on characteristic quantity |
| CN110674193A (en) * | 2019-10-25 | 2020-01-10 | 国网四川省电力公司技能培训中心 | Intelligent substation relay protection fault information modeling method |
| CN111707906A (en) * | 2020-07-14 | 2020-09-25 | 广州白云电器设备股份有限公司 | Method for realizing fault location of subway direct-current traction power supply system |
| CN111707906B (en) * | 2020-07-14 | 2021-08-03 | 广州地铁集团有限公司 | Method for realizing fault location of subway direct-current traction power supply system |
| CN114910763A (en) * | 2022-07-18 | 2022-08-16 | 广东电网有限责任公司佛山供电局 | GIS partial discharge monitoring method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101907868B (en) | 2012-01-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101907868B (en) | Intelligent trouble diagnosis method for tractive power supply system and system thereof | |
| CN108255649B (en) | Diagnosis strategy design method based on modeling simulation cooperative analysis | |
| Bastidas-Rodríguez et al. | Model-based degradation analysis of photovoltaic modules through series resistance estimation | |
| CN103454516B (en) | Intelligent transformer substation secondary equipment health state diagnostic method | |
| Cai et al. | A multiphase dynamic Bayesian networks methodology for the determination of safety integrity levels | |
| EP2845286B1 (en) | Fault detection in energy supply networks | |
| CN104504248A (en) | Failure diagnosis modeling method based on designing data analysis | |
| CN106226621A (en) | A kind of secondary device fault diagnosis based on grey correlation analysis and method for early warning | |
| CN105445591B (en) | Current transformer method for diagnosing faults and current transformer trouble-shooter | |
| CN104091622A (en) | Method and system for evaluating availability ratio of nuclear power plant digital control system | |
| CN104076813A (en) | TCAS system fault comprehensive diagnosis method and system based on Bayesian decision tree | |
| CN105445585A (en) | Fault diagnosis method and system for primary loop of power grid | |
| CN102306244B (en) | Fault eliminating method based on evaluation of detecting points | |
| CN106104282A (en) | A kind of electric measurement method based on reference energy | |
| CN201449436U (en) | Failure diagnosing and positioning system for power system earthing net | |
| CN109030976B (en) | Power electronic equipment monitoring and diagnosing system and method | |
| CN101571566A (en) | System for diagnosing and positioning ground grid faults of electric system | |
| Hsiao et al. | Special protection system reliability assessment | |
| CN110412417B (en) | Micro-grid data fault diagnosis method based on intelligent power monitoring instrument | |
| CN105676044A (en) | Online test method and system for lightning arrester and monitor thereof | |
| Luo et al. | Automated analysis of digital relay data based on expert system | |
| Manyari-Rivera et al. | Integrated fault diagnosis based on Petri net models | |
| CN110398666B (en) | Power distribution network fault diagnosis method based on relay protection time sequence information characteristics | |
| Zhang et al. | Distributed Architecture of Power Grid Asset Management and Future Research Directions | |
| CN115906350A (en) | Micro-grid fault positioning and diagnosing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120125 Termination date: 20120805 |