CN101661070A - Method for conducting power system fault diagnosis by combining information theory with expert system - Google Patents
Method for conducting power system fault diagnosis by combining information theory with expert system Download PDFInfo
- Publication number
- CN101661070A CN101661070A CN200910093007A CN200910093007A CN101661070A CN 101661070 A CN101661070 A CN 101661070A CN 200910093007 A CN200910093007 A CN 200910093007A CN 200910093007 A CN200910093007 A CN 200910093007A CN 101661070 A CN101661070 A CN 101661070A
- Authority
- CN
- China
- Prior art keywords
- protection
- fault
- information
- failure
- equipment
- 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
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a method for conducting power system fault diagnosis by combining information theory with expert system, belonging to the technical field of power system safety treatment. Themethod includes the following steps: firstly establishing a fault diagnosis information transfer model of an actual channel, determining all possible pieces of equipment with fault through forward reasoning of the expert system according to a power cut area where the power system fails, then respectively assuming suspicious equipment, and reasoning out the corresponding action state of switching and protection reversely to obtain all possible information sources in actual communication. Finally, a control center obtains a group of data sequences consisting of the action state of switching andprotection, establishes the fault diagnosis information transfer model based on the actual channel and implements examination, repair and removal of faults according to the fault diagnosis informationtransfer model.
Description
Technical field
The invention belongs to the power system security processing technology field, particularly a kind of information theory combines with expert system and carries out the method for power system failure diagnostic.
Background technology
The electric power system fault process itself exists uncertainty, and this uncertainty mainly is that factors such as telecontrol information garble cause by protection or isolating switch malfunction, tripping.In the time only passing through SCADA (data acquisition and monitoring) system acquisition failure symptom signal, the uncertainty in the electric network failure diagnosis is just more obvious, influences the real-time diagnosis of electric network fault.Bayesian network and information theory can realize the uncertainty of failure message is quantized all based on theory of probability.
Wu Xin, Guo the people such as innovates at " Power System and its Automation journal " 2005,8,17 (4): deliver on the 11-15 in " based on the power system failure diagnostic method of Bayesian network " literary composition. troubleshooting issue is expressed as decision problem under uncertain and the incomplete information, set up distributed treatment model based on Bayesian network, uncertainty to information quantizes, and has realized the power system failure diagnostic under uncertain and incomplete information.But the method for diagnosing faults based on Bayesian network needs a large amount of complete physical fault sample training to improve diagnostic accuracy, and, existing method for diagnosing faults based on Bayesian network is when handling failure information uncertain, all supposed the remote signalling amount of having known disappearance in advance, utilize all the other failure messages that disappearance information is estimated again, carry out fault diagnosis then.But such hypothesis is difficult to set up in actual applications, and no matter the SCADA system in the practical application at present is to adopt former CDT stipulations promulgated by the ministries or commissions of the Central Government or 101 stipulations of adopting international standards, and the remote signalling amount all adopts and send mode in the displacement, to save volume of transmitted data.When breaking down,, also be difficult to determine specifically to be which remote signalling measures and showed disappearance at dispatching terminal even the situation that remote signalling is lost occurred.Therefore, still do not possess practicality in present stage based on the uncertainty that the method for diagnosing faults of Bayesian network brings for remote signalling information transmission mistake.Same problem also is present in the different fault diagnosis method.
Information theory has scientifically solved the problem of measure of probabilistic information from theory of probability, can measure accurately uncertain information.At present, under the guidance of information science, existing scholar is applied to information movement and reconstruction theory in the electrical network on-line fault diagnosis, and the failure diagnostic process of electric system is regarded as information exchanging process, calculates the probability distribution of fault diagnosis solution space by the Combinatorial Optimization of asking for information loss.Tang Lei, Sun Hongbin, Zhang Baiming etc. are at " Proceedings of the CSEE ", and 2003,7,23 (7): deliver " based on the electric system on-line fault diagnosis of information theory " literary composition among the 5-11, the diagnostic method in the literary composition has promoted further developing of fault diagnosis technology.But when the process of descriptor motion, adopt the protection action that model of communication system describes electric power system fault and cause, the process of switch trip fully, will " protection action, switch trip " link is abstract is channel model.Actual reception to sign regard definite information as, ignored actual channel, promptly ignored the uncertainty in the information transmission, like this can not the complete reaction failure symptom and fault between uncertainty.And in fact, the information that SCADA provides under situation about breaking down may be inconsistent with the situation at scene, mistake has appearred in the information that is to say in transmission, directly influence the safety of electrical network and yardman's correct judgement, if ignore the uncertainty in the transmission of SCADA signal, will influence the accuracy of fault handling.
In above-mentioned abstract Information Transmission Model, all devices in the fault outage zone and combination thereof all are regarded as information source.M equipment is arranged in the electric system of supposing to be diagnosed, and this M equipment can produce N kind sign signal altogether in failure process, and then the input of this channel is the random vector of a M dimension, and output is the random vector of a N dimension.The combination of the method equipment possible breakdown has 2
MKind, the combination of possible sign has 2
NKind.Be one group of huge data, though adopted the modern optimization algorithm to overcome the excessive problem of solution space, during practical application, still can have influence on diagnosis speed, especially during complex fault, the information source solution space is very big, is unfavorable for on-line fault diagnosis.
Guo's innovation, Zhu Chuanbai, Cao Yijia, people such as Wu Xin are in " Power System and its Automation " 2006,4,30 (8): introduced expert system in " present Research of power system failure diagnostic and development trend " literary composition of delivering on the 98-103 and in fault diagnosis, obtained successful application, but because the ambiguity and the uncertainty of SCADA information are completely exposed the shortcoming of expert system in failure diagnostic process.A kind of situation is the corresponding multiple suspected fault of failure symptom that the dispatching center obtains, another kind of situation is because the erroneous transmissions of information, the failure symptom that causes receiving can't be complementary with expert system rule, and can't handle this uncertainty based on the diagnostic method of expert system this moment.Therefore, under the condition of having only a data source of SCADA, how the uncertainty between handling failure and the sign is a major issue that needs research, expert system must be combined with other technologies, could improve the accuracy and the practicality of fault diagnosis.
Summary of the invention
The purpose of this invention is to provide a kind of information theory and combine with expert system and carry out the method for power system failure diagnostic, it is characterized in that, set up failure diagnosis information TRANSFER MODEL based on actual channel.The power supply interrupted district that breaks down according to electric system at first; utilize expert system, the process forward reasoning obtains the equipment of all possible breakdowns, supposes respectively that then suspect device breaks down; backward reasoning goes out the corresponding switch and the operating state of protection, obtains all possible information source in the practical communication.The prior probability that obtains the tripping probability of join protection behind the source symbol, switch and equipment failure calculates the self-information or the prior probability of information source, and the uncertainty in the failure process just is included in the information source like this.In the SCADA of reality system; one group of data sequence that the sign information that the dispatching center is seen is made up of the operating state of protection and switch is the stay of two nights; corresponding relation according to the information source and the stay of two nights; calculate the transition probability of channel in the communication, realize quantitative description information uncertainty in the actual channel.Utilize the prior probability of the transition probability of channel and information source to calculate the condition self-information amount of information source at last, the condition self-information amount according to information source sorts to suspected fault again, and the dispatching center implements fault according to ranking results and gets rid of.
The concrete calculation procedure of the failure diagnostic process of described actual channel is as follows:
1. at first according to the step switch record that SCADA collected, obtain all power supply interrupted districts after the fault by topological analysis;
2. by remote measurement, remote signalling and protection type and actuation time information judge the character of action protection in the dead electricity zone, if main protection action, with its object of protection as suspected fault equipment, if back-up protection is moved or can't be judged protective nature, with all electrical equipments in its reserve scope as suspected fault equipment, dwindle the fault zone, the dead electricity regional record of not protecting information is the fault zone;
3. by expert system the suspect device in the fault zone is made fundamental analysis, if have only an equipment to be confirmed as suspected fault equipment this moment, then at this moment phylogenetic is simple fault, when if the corresponding various faults of the failure symptom that obtains may or can't utilize expert system rule to make a strategic decision, the analysis (backward reasoning) that enters complex fault;
4. backward reasoning: suppose that respectively suspected fault equipment concentrates all single electrical equipment malfunctions, consider the situation of combined fault simultaneously.Set up information source according to the action protection character of 2. middle judgement then.If judge it is the main protection action, when supposing equipment failure, only consider it to be isolated by the main protection of suspect device.If actual is back-up protection action, when supposing equipment failure, backward reasoning is isolated it to back-up protection thus.If protective nature is complicated or can't judge protective nature, fault isolation is carried out the fault hypothesis respectively according to main protection or back-up protection action.Infer protection, the switch motion sequence in the zone that causes power failure, set up one group of failure symptom combination of sets (being information source);
5. the tripping probability calculation of the prior probability of bonding apparatus fault, protection and switch goes out the prior probability of each information source, then according to the failure symptom information that receives, utilize information transmission Theoretical Calculation channel transition probability, calculate the part self-information amount (or posterior probability bar) of information source at last;
6. with the ascending arrangement of condition self-information amount of information source; the posterior probability that is about to information source is according to descending arrangement; come most possible generation of fault hypothesis of top information source correspondence, whether correct operation and information exist in transmission course and lose or the situation of error code can to judge protection and switch simultaneously.
Described protection type is main protection action and back-up protection action;
Main protection: satisfy system stability and safety equipment requirement, can excise the protection of protected equipment and line fault selectively with prestissimo.The i.e. protection of snap action when device fails;
Back-up protection: other link damages, isolating switch tripping in the loop of protective device tripping, protection, working power is undesired and even disappearance etc. happens occasionally, and causes main protection can not excise fault fast, at this moment needs back-up protection to excise fault.I.e. protection of moving again when main protection is failure to actuate generally having time-delay to judge that whether main protection move.
The invention has the beneficial effects as follows and can the uncertainty in the fault be quantitatively described, when the physical fault sign may be corresponding during the various faults supposed situation, perhaps because the remote signalling mistake, when the strictly all rules of being stored in the failure symptom that the dispatching center obtains and the rule base of expert system all can't mate, carry out fault diagnosis according to institute's extracting method, can play the effect of aid decision making to the yardman according to the possibility size that the fault hypothesis takes place.
Owing to adopted forward and reverse reasoning of expert system to determine possible information source; the information source combination of sets is very limited; and the appointment of all information sources all is assumed to be the basis with fault; basically be not subjected to the influence of the failure symptom that remote signalling receives; uncertain conditions such as the protection tripping in the time of not only can taking into account fault and take place and the tripping of switch, malfunction, various uncertain problems such as can also handle that the remote signalling that takes place in the actual channel (telecontrol channel) lacks, makes mistakes.This algorithm synthesis has utilized the advantage of expert system and information theory, has reached the requirement of practicability substantially.
The method that adopts information theory to combine with expert system, the uncertain problem of remote signalling when emphasis solution electric system is broken down, remote measurement (message transmitting procedure) information makes the method for diagnosing faults can practicability.Utilize the strong reasoning from logic function of expert system to simplify information source as much as possible, solve the excessive drawback of solution space of information source in the abstracted information mode.Utilize the tight reasoning and calculation function of information theory, uncertainty in the failure process and the uncertainty in the information transmission are united, quantize with quantity of information, calculate the possibility that the fault hypothesis takes place, finally the possibility size that takes place according to the fault hypothesis is arranged disconnected to all possible failure condition, instruct the yardman to go to search the trouble spot, really play the effect of aid decision making according to the possibility size that physical device breaks down.Simultaneously can diagnose out the protection of suspected fault correspondence and the action situation of switch, also can reflect the transmission situation of information in actual channel.
Description of drawings
Fig. 1 is the failure diagnosis information TRANSFER MODEL based on actual channel.
Fig. 2 is for considering the Troubleshooting Flowchart of SCADA information uncertainty.
Fig. 3 is the complex fault legend.
Fig. 4 is Shenzhen Ping'an's station failure synoptic diagram.
Embodiment
At two kinds of drawbacks that expert system exists, the diagnostic procedure of this paper algorithm is described respectively with two examples below in fault diagnosis.
Calculate row one certain distribution network failure diagnosis example
In Fig. 3, transformer station 1 is a mains side, and transformer station 1 powers to transformer station 2 by 1111 circuits.
Suppose that the failure symptom that the SCADA system obtains is: 1111 switch trips of transformer station's 1 bus C side; The action of 1111 route protections.In order to verify the validity of this algorithm, not anti-hypothesis 1111 circuits by analysis is with its fault isolation by the back-up protection action.
By the SCADA information that receives to this diagnosing malfunction.Go out the dead electricity zone according to the trip switch topological analysis.In the dead electricity zone; the action of 1111 line backup protections; expert system is according to the back-up protection scope forward reasoning zone of being out of order, and counts that suspected fault equipment is in the fault zone: 1111 circuits, bus A, 1112 circuits, main transformer, bus B enter the complex fault analysis.
Suppose that equipment failure is incident ai, because the protection information of border action is 1111 line backup protections, the main protection of suspect device or switch tripping cause fault coverage to enlarge, and are moved fault isolation by back-up protection at last.Corresponding protection and the on off state combination of sets of each fault hypothesis is mij.Suppose bus B fault, as shown in table 1 by the information source that the expert system reasoning obtains.In like manner also can infer information source m21, the m22 of main transformer fault a2 correspondence; The information source m31 of 1111 line fault a3 correspondences, information source m41, the m42 of bus-bar fault a4 correspondence; Information source m51, the m52 of 1112 line fault a5 correspondences; Owing to have only a protection action in this example, the probability of combined fault hypothesis is much smaller than individual equipment fault hypothesis, thus ignore the combined fault situation in this example, but can not influence diagnostic result.
Annotate: ' 1 '--action, ' 0 '--not action, ' √ '--correct operation, ' * '--malfunction
Comprehensive national electric reliability statistical study over the years and the situation analysis of nationwide integrated power grid relay protected operation, the prior probability that calculates element fault sees Table 2.Calculate the prior probability of each information source in conjunction with the tripping probability (seeing Table 3) of relay protection and switch.
Table 1 bus B fault information source sign combination sequential analysis table
The prior probability of table 2 part primary equipment fault
Table 3 probabilistic information statistical form
Because whether correct operation is independently for each protection and switch, has condition independence between all assumed conditions.The prior probability of information source mij equals the amassing of probability of assumed condition, if the back-up protection action, with the prior probability correction of weights λ to information source.During circuit 1111 faults, the protection domain of back-up protection is 15% of this line end, gets λ=0.15 herein in this example.And information source must be a perfect set, and the prior probability of actual information source sees Table 4 after the normalization.
Calculate the condition self-information amount I (mij|m) of information source then by information theory, i.e. the remaining uncertainty of information source after the communication.
I(mij|m)=-logp(mij|m)(1)
Because each intersymbol of mij of transmission is separate, the dawn is (0,1) sequence sets, so can regard discrete memoryless channel(DMC) as, the symbol lengths of establishing mij and m is n, note m=s
1s
2S
n, mij=l
1l
2L
n, s
i, l
i∈ 0,1}
When channel one timing of transmission information, the bit error rate of channel can be obtained by statistical property, and the bit error rate that might as well establish channel is ps
i, then the correct probability that transmits of signal is 1-ps
iP (s in each binary channel
i| l
i), i ∈ N is decided by corresponding symbol among mij and the m.
Get ps in this example
i=0.0001, then when receiving failure symptom m, according to the corresponding relation of each symbol among mij in the table 1 and the m, by the transition probability p (m|mij) of formula (3) and (4) calculating channel, the posterior probability of being calculated information source mij by formula (2) sees Table 5.Calculate fault by formula (1) and suppose remaining quantity of information.To sorting, diagnostic result sees Table 6 according to information source remaining information amount.The quantity of information of information source is more little, illustrates that this information source uncertainty of diagnosis back is lower, and the actual possibility that takes place is bigger.The sign that receives has been eliminated the uncertainty of information source hypothesis, more helps proving actual situation about breaking down.
When the failure symptom information that receives is: 1111 switch trips of transformer station's 1 bus C side; The action of 1111 route protections; The action of 1112 route protections.The just actual failure symptom m that obtains is { 1100100000}.Have only this moment a kind of fault hypothesis identical with the actual failure symptom that obtains; the diagnostic result that utilizes this algorithm to obtain is: 1112 line faults; the action of 1112 route protections; 1112 switch trippings; the action of 1111 route protections; 1111 switch motions of transformer station's 1 bus C side, and all correctly transmission of information.
Example two Shenzhen Ping'an station bus-bar fault diagnosis example.
Be example with safety station, Shenzhen safety station on the 3rd 220kV bus loss of pressure January in 2008 below, Fig. 4 is the synoptic diagram of this fault.Utilize the inventive method to this diagnosing malfunction.
The sign information that the SCADA system obtains is as follows: safety station 220kV bus differential protection, station, Xiangshan 220kV resemble flat first, second line protection, the Xixiang station 220kV all actions of first, second line protection of being setting; Safety station # 3 main transformer uprises switch 2203,2012 switches, Xiang Pingjia, second line 2807,2808 switches, preceding flat A-wire safety station 2890 switches, preceding flat second line 2891 switches, first, the second of being setting line 2944,2945 switch trips.Because information mistake occurred in the process of transmission, safety station # 1 main transformer uprises that 2201 switches, #2 main transformer uprise 2202 switches, station, the preceding gulf of preceding flat A-wire 2890 switch S CADA are shown as closing position.
At first go out the dead electricity zone according to the step switch information topological analysis that SCADA collected.Because the protection of action is main protection, can judge that the fault zone is all circuit and equipment that link to each other with safety station 220kv bus.Counting suspected fault equipment is: bus I, bus II, mother join the dead band, resemble and put down first and second lines, preceding flat first and second lines, first and second lines that are setting, three main-transformers; Utilize expert system that the failure symptom that receives is made a strategic decision then, do not have corresponding rule to match, enter the analysis of complex fault.
Expert system is carried out backward reasoning, all suspect device are carried out the fault hypothesis according to main protection with its isolation, the corresponding sign collection of each fault, setting up one group at last causes the set of symptom intersection of regional dead electricity as shown in table 7 as information source, m1~m12 is single electrical equipment malfunction, and m13~m24 is a combined fault; Calculate the prior probability of each information source at last.Because female connection dead band is the part of bus, the prior probability that might as well get female connection dead band fault is one of percentage of bus-bar fault.The posterior probability of calculating information source mij with the method for example one has: p (m3|m)>p (m13|m)>>p (m2|m)>p (m1|m).The possibility of all the other faults is less, can get rid of the possibility that breaks down substantially.Diagnostic result such as table 8.
The prior probability of actual information source after table 4 normalization (various sign)
??p m11=5.7776×10 -5 | ??p m12=1.1442×10 -6 |
??p m13=5.7204×10 -7 | ??p m14=1.1329×10 -8 |
??p m21=2.1612×10 -2 | ??p m22=4.2793×10 -4 |
??p m31=0.8777 | ??p m41=5.7206×10 -3 |
??p m42=5.6632×10 -5 | ??p m51=5.9114×10 -2 |
??p m52=3.5326×10 -2 |
The posterior probability of table 5 information source
??p(m11|m)=6.3258×10 -13 | ??p(m12|m)=1.2526×10 -10 |
??p(m13|m)=6.2626×10 -11 | ??p(m14|m)=1.2401×10 -8 |
??p(m21|m)=2.3656×10 -6 | ??p(m22|m)=4.6844×10 -4 |
??p(m31|m)=0.9608 | ??p(m41|m)=6.2618×10 -7 |
??p(m42|m)=6.1993×10 -5 | ??p(m51|m)=6.4707×10 -6 |
??p(m52|m)=0.0387 |
The diagnostic result of table 6 example one
Sequence number | Diagnostic result (information is correctly transmitted) | Quantity of information/bit |
??1 | 1111 line faults, 1111 switch trips of the action of 1111 route protections, bus C side | ??0.0577 |
??2 | 1112 line faults, 1112 route protection trippings, the action of 1111 route protections, 1111 switch trips of bus C side | ??4.6915 |
??3 | The main transformer fault, 1111 switch trips of main transformer protection tripping, bus C side | ??11.0586 |
??4 | Bus A fault, 1111 switch trips of bus A protection tripping, the action of 1111 route protections, bus C side | ??13.9775 |
The all possible information source combination of table 7
Incident mi | ??m1 | ??m2 | ??m3 | ??m4 | ??m5 | ??m6 | ??m7 | ??m8 | ??m9 | ??m10 | ??m11 | ??m12 |
The assumed fault place | Mother I | Mother II | Female connection dead band | Resemble flat A-wire | Resemble flat second line | Preceding flat A-wire | Preceding flat second line | A-wire is setting | The second of being setting line | The 1# main transformer | The 2# main transformer | The 3# main transformer |
Incident mi | ??m13 | ??m14 | ??m15 | ??m16 | ??m17 | ??m18 | ??m19 | ??m20 | ??m21 | ??m22 | ??m23 | ??m24 |
The assumed fault place | Bus I, II | Resemble flat first and second | Preceding flat first and second | Be setting first and second | ??1、2 ??# | ??1、3 ??# | ??2、3 ??# | ??1、2、 ??3 ??# | Before, resemble line | The west, resemble line | Preceding, western line | Before, the west, resemble line |
Table 8 diagnostic result
Sequence number | Diagnostic result (field condition) | The information of erroneous transmissions | Quantity of information/bit |
? ??1 | Female connection dead band fault, the action of safety station 220kV bus differential protection, tripping mother I, the female switch that connects of II | #12201 switch, #22202 switch, station, the preceding gulf of preceding flat A-wire 2890 switches close a floodgate opposite with field condition | ? ??0.9259 |
? ??2 | Bus I, II fault, the action of safety station 220kV bus differential protection, tripping mother I, the female switch that connects of II | #12201 switch, #22202 switch, station, the preceding gulf of preceding flat A-wire 2890 switches close a floodgate opposite with field condition | ? ??1.0780 |
Claims (3)
1. an information theory combines with expert system and carries out the method for power system failure diagnostic, it is characterized in that, at first set up the failure diagnosis information TRANSFER MODEL of actual channel, the power supply interrupted district that breaks down according to electric system, utilize expert system, the process forward reasoning obtains the equipment of all possible breakdowns, supposes respectively that then suspect device breaks down, backward reasoning goes out the corresponding switch and the operating state of protection, obtains all possible information source in the practical communication; Last join protection; the tripping probability of switch and the prior probability of equipment failure calculate the self-information or the prior probability of information source; uncertainty in the failure process just is included in the information source like this; in the SCADA of reality system; one group of data sequence that the sign information that the dispatching center is seen is made up of the operating state of protection and switch; calculate the condition self-information amount of information source like this according to the transition probability of channel in the communication; realized quantitative description to information uncertainty in the actual channel; foundation is based on the failure diagnosis information TRANSFER MODEL of actual channel, and the dispatching center implements fault according to this failure diagnosis information TRANSFER MODEL and gets rid of.
2. combine with expert system according to the described information theory of claim 1 and carry out the method for power system failure diagnostic, it is characterized in that the concrete calculation procedure of the failure diagnostic process of described actual channel is as follows:
1. at first according to the step switch record that SCADA collected, obtain all dead electricity zones after the fault by topological analysis;
2. judge the character of action protection in the dead electricity zone by information such as remote measurement, remote signalling and protection type, actuation times, if main protection action, with its object of protection as suspected fault equipment, if back-up protection is moved or can't be judged protective nature, all electrical equipments in its reserve scope as suspected fault equipment, are dwindled the fault zone; The dead electricity regional record of not protecting information is the fault zone;
3. by expert system the suspect device in the fault zone is made fundamental analysis, if having only an equipment to be confirmed as suspected fault equipment this moment, then at this moment phylogenetic is simple fault; When if the corresponding various faults of the failure symptom that obtains may or can't utilize expert system rule to make a strategic decision, the analysis that enters complex fault;
4. backward reasoning: suppose that respectively suspected fault equipment concentrates all electrical equipment malfunctions, consider the situation of combined fault simultaneously, set up information source according to the action protection character of judging in 2. then; If judge it is the main protection action, when supposing equipment failure, only consider it to be isolated by the main protection of suspect device; If actual is back-up protection action, when supposing equipment failure, backward reasoning is isolated it to back-up protection thus; If protective nature is complicated or can't judge protective nature, according to main protection or back-up protection action fault isolation is carried out the fault hypothesis respectively, infer the protection, the switch motion sequence that cause regional dead electricity, setting up one group of failure symptom combination of sets is information source;
5. the tripping probability calculation of the prior probability of bonding apparatus fault, protection and switch goes out the prior probability of each information source, according to the failure symptom information that receives, utilizes the information transmission theory then, calculates the part self-information amount or the posterior probability bar of information source;
6. with the ascending arrangement of condition self-information amount of information source, the posterior probability that is about to information source is according to descending arrangement, come most possible generation of fault hypothesis of top information source correspondence, whether correct operation and information exist in transmission course and lose or the situation of error code can to judge protection and switch simultaneously;
Being explained as follows of main protection, back-up protection, operating time of protection:
Main protection: satisfy system stability and safety equipment requirement, can excise the protection of protected equipment and line fault selectively with prestissimo.The i.e. protection of snap action when device fails;
Back-up protection: other link damages, isolating switch tripping in the loop of protective device tripping, protection, working power is undesired and even disappearance etc. happens occasionally, and causes main protection can not excise fault fast, at this moment needs back-up protection to excise fault; Promptly when main protection is failure to actuate, move again, generally having time-delay to judge that whether main protection move;
Operating time of protection: the time that starts to the actuating signal outlet by protective device.
3. combine with expert system according to the described information theory of claim 1 and carry out the method for power system failure diagnostic; it is characterized in that; described foundation is based on the failure diagnosis information TRANSFER MODEL of actual channel; the theory of handling the error code of communicating by letter in expert system and the information theory is combined; clear easy to protecting in the fault diagnosis; the isolating switch tripping is carried out quantitative description with the uncertainty in communicating by letter; combine with expert system; improved diagnosis speed; and have practicality and an extendability; with unified index---quantity of information sorts to suspected fault; reduce the risk of false dismissal and false-alarm, and under the situation that larger interference exists, drawn rational diagnostic result; has certain fault-tolerant and error correction; handle the uncertainty of remote signalling, instruct the yardman to carry out fault diagnosis and fault handling, help realizing online application.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200910093007A CN101661070A (en) | 2009-09-25 | 2009-09-25 | Method for conducting power system fault diagnosis by combining information theory with expert system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200910093007A CN101661070A (en) | 2009-09-25 | 2009-09-25 | Method for conducting power system fault diagnosis by combining information theory with expert system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101661070A true CN101661070A (en) | 2010-03-03 |
Family
ID=41789208
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN200910093007A Pending CN101661070A (en) | 2009-09-25 | 2009-09-25 | Method for conducting power system fault diagnosis by combining information theory with expert system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101661070A (en) |
Cited By (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102346440A (en) * | 2010-07-28 | 2012-02-08 | 霍尼韦尔国际公司 | Automatic determination of success of using a computerized decision support system |
CN102788933A (en) * | 2012-09-25 | 2012-11-21 | 河北省电力公司 | Method for utilizing automatic switch-out controller to search power grid low-current single-phase ground fault |
CN103177186A (en) * | 2013-03-21 | 2013-06-26 | 国家电网公司 | Electric return circuit fault probability forecasting method |
CN103278715A (en) * | 2013-05-16 | 2013-09-04 | 广东电网公司 | Electrical equipment testing method |
CN103278746A (en) * | 2013-05-31 | 2013-09-04 | 华北电力大学 | Implementation method of grid fault diagnosis system considering uncertain information |
CN103293421A (en) * | 2013-06-07 | 2013-09-11 | 云南电力调度控制中心 | Power grid fault diagnostic model and diagnostic method thereof |
CN103324845A (en) * | 2013-06-13 | 2013-09-25 | 国电南瑞科技股份有限公司 | Intelligent substation switch state estimation method based on uncertainty reasoning |
CN103743963A (en) * | 2012-11-29 | 2014-04-23 | 浙江省电力公司 | Electric power system fault diagnosis method based on electrical quantity and state quantity |
CN104297637A (en) * | 2014-10-31 | 2015-01-21 | 国家电网公司 | Power system fault diagnosis method comprehensively using electricity amount and timing sequence information |
CN104360237A (en) * | 2014-11-21 | 2015-02-18 | 国家电网公司 | Intelligent on-line distinguishing method for main equipment failures of regional power grid |
CN104410168A (en) * | 2014-12-16 | 2015-03-11 | 柳涛 | Method for identifying power distribution network by remote signaling based on closing probability of switch |
CN104749493A (en) * | 2015-05-04 | 2015-07-01 | 江苏省电力公司苏州供电公司 | Grid fault equipment analyzing and reasoning method based on rule tree |
CN104897980A (en) * | 2015-03-15 | 2015-09-09 | 国家电网公司 | Method of evaluating longitudinal differential protection device abnormal action probabilities caused by communication delay error code |
CN105528679A (en) * | 2015-12-11 | 2016-04-27 | 谭焕玲 | Power system fault diagnosis method |
CN104198854B (en) * | 2014-08-29 | 2016-09-28 | 国家电网公司 | A kind of online Branch Type Contingency screening and ranking method |
CN106646030A (en) * | 2016-11-15 | 2017-05-10 | 国家电网公司 | Power grid fault diagnosis method and device based on multiple data sources and expert rule base |
CN107423414A (en) * | 2017-07-28 | 2017-12-01 | 西安交通大学 | A kind of process industry complex electromechanical systems fault source tracing method based on information transmission model |
CN107657375A (en) * | 2017-09-25 | 2018-02-02 | 国网上海市电力公司 | A kind of method for electric network fault judgement, verification and fault incidence analysis |
CN107846016A (en) * | 2017-11-16 | 2018-03-27 | 中国南方电网有限责任公司 | A kind of Distribution Network Failure localization method and equipment based on Bayes and Complex event processing |
CN109063839A (en) * | 2018-09-26 | 2018-12-21 | 北京航天自动控制研究所 | A kind of complicated sign construction method of the simulation temporal logic of expert system |
CN109726214A (en) * | 2018-12-26 | 2019-05-07 | 国电南瑞科技股份有限公司 | A kind of equipment fault disposition aid decision-making method and system |
CN114142438A (en) * | 2021-11-18 | 2022-03-04 | 广东电网有限责任公司电力调度控制中心 | Analysis method based on station bus voltage loss and power restoration |
-
2009
- 2009-09-25 CN CN200910093007A patent/CN101661070A/en active Pending
Cited By (30)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102346440A (en) * | 2010-07-28 | 2012-02-08 | 霍尼韦尔国际公司 | Automatic determination of success of using a computerized decision support system |
CN102788933A (en) * | 2012-09-25 | 2012-11-21 | 河北省电力公司 | Method for utilizing automatic switch-out controller to search power grid low-current single-phase ground fault |
CN103743963A (en) * | 2012-11-29 | 2014-04-23 | 浙江省电力公司 | Electric power system fault diagnosis method based on electrical quantity and state quantity |
CN103177186A (en) * | 2013-03-21 | 2013-06-26 | 国家电网公司 | Electric return circuit fault probability forecasting method |
CN103177186B (en) * | 2013-03-21 | 2016-05-11 | 国家电网公司 | A kind of electric loop probability of malfunction Forecasting Methodology |
CN103278715A (en) * | 2013-05-16 | 2013-09-04 | 广东电网公司 | Electrical equipment testing method |
CN103278715B (en) * | 2013-05-16 | 2015-09-16 | 广东电网公司 | Power equipment test method |
CN103278746B (en) * | 2013-05-31 | 2015-12-09 | 华北电力大学 | Consider the implementation method of the electric network failure diagnosis system of uncertain information |
CN103278746A (en) * | 2013-05-31 | 2013-09-04 | 华北电力大学 | Implementation method of grid fault diagnosis system considering uncertain information |
CN103293421A (en) * | 2013-06-07 | 2013-09-11 | 云南电力调度控制中心 | Power grid fault diagnostic model and diagnostic method thereof |
CN103324845B (en) * | 2013-06-13 | 2016-12-07 | 国电南瑞科技股份有限公司 | A kind of intelligent substation switch state estimation method based on uncertain inference |
CN103324845A (en) * | 2013-06-13 | 2013-09-25 | 国电南瑞科技股份有限公司 | Intelligent substation switch state estimation method based on uncertainty reasoning |
CN104198854B (en) * | 2014-08-29 | 2016-09-28 | 国家电网公司 | A kind of online Branch Type Contingency screening and ranking method |
CN104297637A (en) * | 2014-10-31 | 2015-01-21 | 国家电网公司 | Power system fault diagnosis method comprehensively using electricity amount and timing sequence information |
CN104360237A (en) * | 2014-11-21 | 2015-02-18 | 国家电网公司 | Intelligent on-line distinguishing method for main equipment failures of regional power grid |
CN104410168B (en) * | 2014-12-16 | 2016-09-28 | 柳涛 | A kind of method of power distribution network remote signalling identification based on switch Guan Bi probability |
CN104410168A (en) * | 2014-12-16 | 2015-03-11 | 柳涛 | Method for identifying power distribution network by remote signaling based on closing probability of switch |
CN104897980A (en) * | 2015-03-15 | 2015-09-09 | 国家电网公司 | Method of evaluating longitudinal differential protection device abnormal action probabilities caused by communication delay error code |
CN104897980B (en) * | 2015-03-15 | 2017-11-03 | 国家电网公司 | A kind of improper action probability evaluation method of failure of longitudinal differential protecting equipment caused by communication delay error code |
CN104749493A (en) * | 2015-05-04 | 2015-07-01 | 江苏省电力公司苏州供电公司 | Grid fault equipment analyzing and reasoning method based on rule tree |
CN105528679A (en) * | 2015-12-11 | 2016-04-27 | 谭焕玲 | Power system fault diagnosis method |
CN106646030A (en) * | 2016-11-15 | 2017-05-10 | 国家电网公司 | Power grid fault diagnosis method and device based on multiple data sources and expert rule base |
CN107423414A (en) * | 2017-07-28 | 2017-12-01 | 西安交通大学 | A kind of process industry complex electromechanical systems fault source tracing method based on information transmission model |
CN107657375A (en) * | 2017-09-25 | 2018-02-02 | 国网上海市电力公司 | A kind of method for electric network fault judgement, verification and fault incidence analysis |
CN107846016A (en) * | 2017-11-16 | 2018-03-27 | 中国南方电网有限责任公司 | A kind of Distribution Network Failure localization method and equipment based on Bayes and Complex event processing |
CN109063839A (en) * | 2018-09-26 | 2018-12-21 | 北京航天自动控制研究所 | A kind of complicated sign construction method of the simulation temporal logic of expert system |
CN109063839B (en) * | 2018-09-26 | 2020-08-14 | 北京航天自动控制研究所 | Complex symptom construction method of simulation temporal logic of expert system |
CN109726214A (en) * | 2018-12-26 | 2019-05-07 | 国电南瑞科技股份有限公司 | A kind of equipment fault disposition aid decision-making method and system |
CN114142438A (en) * | 2021-11-18 | 2022-03-04 | 广东电网有限责任公司电力调度控制中心 | Analysis method based on station bus voltage loss and power restoration |
CN114142438B (en) * | 2021-11-18 | 2023-12-01 | 广东电网有限责任公司电力调度控制中心 | Analysis method based on station bus voltage loss and power restoration |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101661070A (en) | Method for conducting power system fault diagnosis by combining information theory with expert system | |
Yu et al. | An UHV grid security and stability defense system: Considering the risk of power system communication | |
CN110210095B (en) | Power distribution network reliability index calculation method based on mixed integer linear programming | |
CN102854461B (en) | The probability forecasting method that a kind of switchgear breaks down and system thereof | |
CN102354977B (en) | Feeder fault handling method | |
CN103354390A (en) | Distributed intelligent feed line automation control method | |
CN101938117A (en) | Current differential protection method for smart distribution network | |
CN103872660B (en) | Network type protection method based on closed-loop mode operation of power network | |
CN104269809B (en) | Method for on-line verification of relay protection setting value of regional power grid | |
CN104767182A (en) | Fault processing method for overhead line type power distribution network | |
CN103293421A (en) | Power grid fault diagnostic model and diagnostic method thereof | |
CN103354355B (en) | A kind of micro-capacitance sensor networking guard method and device | |
CN205620503U (en) | Regional trouble fault -tolerance automatic location system of distribution network feeder | |
CN106771848B (en) | A kind of electric network failure diagnosis method based on the fusion of topological primitive information | |
CN109116181A (en) | A kind of fault detector failure judgment method based on maximum probability | |
CN105548801B (en) | Fault Locating Method suitable for power distribution network operation with closed ring | |
CN105093063A (en) | Online power grid fault diagnosis method based on multisource data characteristic unit combination judgment | |
CN104267294A (en) | IGIS fault online diagnostic method based on rough set theory and Petri network | |
CN111999605A (en) | Power distribution network fault tolerance judgment method and device based on fault correlation analysis | |
CN105334417A (en) | Portable spare power automatic switching check device | |
CN105259860B (en) | A kind of master-slave mode intelligent power device method of automatic configuration | |
CN113131448A (en) | Intelligent distributed feeder line method and device based on 5G communication | |
CN105759165B (en) | Power distribution automation main station functionization evaluating method based on feeder fault condition diagnosing | |
CN102005819A (en) | Fault isolation method of electric distribution network | |
CN107976598B (en) | Method for testing loss of SOE signal after line fault of intelligent substation |
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 |
Open date: 20100303 |