CN115034671A - Secondary system information fault analysis method based on association rule and cluster - Google Patents
Secondary system information fault analysis method based on association rule and cluster Download PDFInfo
- Publication number
- CN115034671A CN115034671A CN202210773125.2A CN202210773125A CN115034671A CN 115034671 A CN115034671 A CN 115034671A CN 202210773125 A CN202210773125 A CN 202210773125A CN 115034671 A CN115034671 A CN 115034671A
- Authority
- CN
- China
- Prior art keywords
- fault
- secondary system
- association rules
- information
- association
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention discloses a secondary system information fault analysis method based on association rules and clustering, which belongs to the technical field of secondary system evaluation of a power system transformer substation. According to the invention, the transformer substation secondary system logic loop model and the analysis method are established by utilizing the massive information and the structural characteristics of the transformer substation secondary system, so that the speed and the accuracy of fault diagnosis of the secondary system are improved.
Description
Technical Field
The invention relates to the technical field of evaluation of secondary systems of power system substations, in particular to a secondary system information fault analysis method based on association rules and clustering.
Background
The intelligent substation is an important composition content for the construction of the intelligent power grid. The secondary system of the intelligent substation realizes the monitoring, control and protection of the primary system, and the secondary system is in a fault state at the moment because the failure of certain secondary equipment causes the failure of a plurality of functions to normally operate. Therefore, new indexes are needed to be introduced to represent the reliability of the secondary system and the disturbance of the secondary system caused by the failure of the secondary equipment. The failure probability and the failure consequence of the system are expressed, so that the secondary system reliability evaluation can provide more comprehensive understanding for the operation and maintenance of the protection system, and the design, operation and maintenance of the protection system are more credible and reliable.
At present, methods adopted by reliability evaluation mainly include a Markov model method and a fault tree analysis method, and in addition, an efficacy analysis method, a probability method and the like, and with the progress of computer technology, the application range of Monte Carlo random simulation or other sampling algorithms is wider and wider. Specifically, in the power system, researchers complete the evaluation of the reliability of the power system by using a Monte Carlo algorithm based on hypercube sampling, and use the modeling method to research the reliability of power grid planning, grid structure, power supply configuration, operation mode, maintenance plan and other aspects, and on the basis of the reliability of elements, the reliability of a certain system is calculated as a basic target, weak elements of the system are analyzed, and measures for improving the reliability are planned to be high-level engineering application or academic research. These concepts can be used as references for secondary system reliability evaluation. As for the data-driven approach, there are two studies currently: on the one hand, data analysis for the secondary circuit and on the other hand, abnormal parameter identification for the secondary equipment. Data analysis of such textual, rather than numeric, information is not uncommon.
Disclosure of Invention
In view of the above, the present invention provides a secondary system information fault analysis method based on association rules and clustering, which can improve the speed and accuracy of secondary system fault diagnosis, aiming at the defects of the prior art.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows: the secondary system information fault analysis method based on association rules and clustering comprises the following steps:
s1: traversing a secondary system message information and fault device information data set by using an Apriori algorithm, mining the relevance of the secondary system message information and the fault device information data set, describing the relevance relation between the message information and the fault device of the secondary system, extracting the relevance rule between the message information and the fault device of the secondary system, and realizing fault diagnosis of the secondary system;
s2: and performing k-means clustering on the extracted association rules to obtain the most appropriate category number, and directly checking fault devices in the association rules and fault devices of the association rules of the same category when message information in some association rules appears again, thereby realizing efficient fault checking.
Further, in S1, the extracted association rules include message information and failure device information with high support degree, confidence degree, and promotion degree, and specific data information of the support degree, confidence degree, and promotion degree corresponding to the association rules is calculated, that is, each association rule corresponds to a set of three-dimensional data.
Further, in S2, the association rules with similar support degree, confidence degree and lifting degree are classified into a category, and all the association rules are classified by using the three-dimensional data, i.e., the similarity between the association rules can be quantified.
Further, in S1, after the station personnel obtains the association rule, the station personnel is used as a reference basis, that is, when there are multiple pieces of message information appearing at the same time, if the message information is included in a certain association rule and a fault does not appear yet, the station personnel performs a troubleshooting on a potential fault device mapped by the association rule in advance, so as to implement a secondary system fault diagnosis.
Further, in S1, the process of extracting the association rule between the message information and the secondary system failure device by using Apriori algorithm is as follows:
1) the Apriori algorithm scans the information of the fault devices of all secondary systems and the message information in the same time period to obtain each item and generate C1; then counting each item, and then deleting unsatisfied items from the C1 according to the minimum support degree, the minimum confidence degree and the minimum promotion degree, so far obtaining a frequent 1 item set L1;
2) according to a pruning strategy, a set generated by the L1 generates a set C2 of candidate 2 items, then fault device information of all secondary systems and message information in the same time period are scanned, item number statistics is carried out on the C2, and similarly, items which do not meet requirements are deleted from the C2 according to the minimum support degree, the minimum confidence degree and the minimum promotion degree, so that a frequent 2 item set L2 is obtained;
3) obtaining a frequent 3 item set L3, a frequent 4 item set L4, … and a frequent k item set Lk according to the method of the step 2);
4) and judging whether k is the maximum, if so, generating an association rule Lk, and if not, adding 1 to the value of k and returning to the step 3) for circulation.
Further, in S2, k-means clustering is performed on the extracted association rules:
the distance function of the k-means algorithm is chosen as follows:
wherein x is 1 ，x 2 ，x 3 ，y 1 ，y 2 ，y 3 Respectively representing three-dimensional data between the two groups of association rules; d (x, y) represents the Euclidean distance between the two sets of association rules;
selecting a landed coefficient for a scoring function of the clustering effect as follows:
wherein TP is true positive rate, FP is false positive rate, and FN is false negative rate.
Further, in step 1), the Apriori algorithm scans the message information in the same time period to be the message information before the actual failure occurs.
Further, in step 1), the set minimum support degree, minimum confidence degree and minimum promotion degree are respectively 0.5%, 5% and 1.
Compared with the prior art, the invention has the following beneficial effects:
the method comprises the steps of solving association rules of fault alarm information and fault devices in a secondary circuit by using an Apriori algorithm, realizing logic circuit fault diagnosis, carrying out k-means clustering on the extracted association rules, obtaining the most appropriate category number, and verifying the effectiveness of the method by using real operation data, wherein the association rules of the same category are in potential connection, so that technical guidance is provided for substation workers.
According to the invention, only by analyzing the secondary system historical message information, the corresponding association rules are screened out, and the association rules are clustered, when some reason items in the association rules appear again, the fault devices in the association rules and the fault devices in the association rules of the same type can be directly checked, so that the checking efficiency is improved, the checking time is shortened, and the speed and the accuracy of secondary system fault diagnosis are improved.
Drawings
FIG. 1 is a flow chart of the Apriori algorithm in an embodiment of the present invention;
FIG. 2 is a flow chart of association rule classification of the K-means algorithm in the embodiment of the present invention;
FIG. 3 is a two-dimensional display diagram of the association rule best classification normalization in the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the drawings of the embodiments of the present invention. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the described embodiments of the invention, are within the scope of the invention.
The secondary system information fault analysis method based on association rules and clustering is characterized by comprising the following steps of:
s1: traversing a secondary system message information and fault device information data set by using an Apriori algorithm, mining the relevance of the secondary system message information and the fault device information data set, describing the relevance relation between the message information and the fault device of the secondary system, extracting the relevance rule between the message information and the fault device of the secondary system, and realizing fault diagnosis of the secondary system;
for such algorithms of correlation analysis, a set containing different elements is usually defined as an item set, wherein the number of the elements is called the length of the item set and is denoted by k. This set of terms is called the k-term set. The sample used to correlate the analysis becomes a sample set, which is a subset of the set of items. The indexes used in the text to measure the quality of the correlation analysis result are support degree, confidence degree and promotion degree, wherein the support degree represents the probability of the item set { X, Y } appearing in the total set { I }, and is as follows:
s(X→Y)＝P(X∪Y)/P(I)
confidence, which refers to the degree of confidence in the association rule, represents the probability that Y occurs if the precondition X occurs, as follows:
c(X→Y)＝P(Y|X)＝P(Y∪X)/P(X)
the promotion degree is the ratio of the probability of containing Y under the condition of containing X to the probability of containing Y under the condition of not containing X, and is as follows:
l(X→Y)＝P(Y|X)/P(Y)
the mining process of the association rule by using the Apriori algorithm is mainly used for reducing the search space by using the prior property of the frequent item set, so that the search space is easy to know that the corresponding non-empty subset of the frequent item set is also the frequent item set. For the message information of the secondary system, the sub-items included in the screened association rule are always frequent item sets. And successively screening through support degree, confidence degree and promotion degree thresholds until the frequent k item set can not be found out any more. Finding the frequent item set Lk in the kth iteration requires one complete database scan.
As shown in fig. 1, in S1, the process of extracting the association rule between the message information and the secondary system failure device by using Apriori algorithm includes:
1) the Apriori algorithm scans the information of the fault devices of all secondary systems and the message information in the same time period to obtain each item and generate C1; then counting each item, and then deleting the unsatisfied items from the C1 according to the minimum support degree, the minimum confidence degree and the minimum promotion degree, so far obtaining a frequent 1 item set L1;
2) according to a pruning strategy, a set generated by the L1 generates a set C2 of candidate 2 items, then fault device information of all secondary systems and message information in the same time period are scanned, item number statistics is carried out on the C2, and similarly, items which do not meet requirements are deleted from the C2 according to the minimum support degree, the minimum confidence degree and the minimum promotion degree, so that a frequent 2 item set L2 is obtained;
3) obtaining a frequent 3 item set L3, a frequent 4 item set L4, … and a frequent k item set Lk according to the method of the step 2);
4) judging whether k is the maximum, if so, generating an association rule Lk, and if not, adding 1 to the value of k and returning to the step 3) for circulation.
In step 1), the Apriori algorithm scans the message information in the same time period as the message information before the actual failure.
In S1, the extracted association rules include message information and failure device information with high support degree, confidence degree, and promotion degree, and specific data information of the support degree, confidence degree, and promotion degree corresponding to the association rules is calculated, that is, each association rule corresponds to a set of three-dimensional data.
The station personnel who work can be as the reference foundation after obtaining important association rule, when having many alarm information to appear simultaneously, if contain in a certain association rule, and the trouble has not appeared yet, can carry out in advance to the potential fault device that the association rule maps at this moment, thereby realize secondary system failure diagnosis, reduce the loss.
S2: and performing k-means clustering on the extracted association rules to obtain the most appropriate category number, and directly checking fault devices in the association rules and fault devices of the association rules of the same category when message information in some association rules appears again, thereby realizing efficient fault checking.
In S2, the association rules with similar support, confidence and lift are classified into one category, and all the association rules are classified by the three-dimensional data, i.e., the similarity between the association rules can be quantified.
As shown in fig. 2, in S2, k-means clustering is performed on the extracted association rules:
the distance function of the k-means algorithm is chosen as follows:
wherein x is 1 ，x 2 ，x 3 ，y 1 ，y 2 ，y 3 Respectively representing three-dimensional data between the two groups of association rules; d (x, y) represents the Euclidean distance between the two sets of association rules;
selecting a Lande coefficient for a scoring function of the clustering effect as follows:
wherein TP is true positive rate, FP is false positive rate, and FN is false negative rate.
The schematic data of the invention is from a certain transformer substation, and the actual data comprises the alarm content when the substation fails and the actual failure device information in the same time period. These data are processed uniformly, as shown below:
table 1 schematic diagram of alarm information and message data of fault device
Table 2 important parameter settings
Results display
And (4) displaying (part) the screened association rules, and verifying by a worker that the following results are in accordance with reality:
the clustering diagram is shown in FIG. 3.
According to the invention, only the secondary system historical message information is analyzed, the corresponding association rules are screened out, the association rules are clustered, and when some reason items in the association rules appear again, the fault devices in the association rules and the fault devices in the association rules of the same type can be directly checked, so that the checking efficiency is improved, and the checking time is shortened.
While the foregoing is directed to the preferred embodiment of the present invention, it will be appreciated by those skilled in the art that various changes and modifications may be made therein without departing from the principles of the invention as set forth in the appended claims.
Claims (8)
1. The secondary system information fault analysis method based on association rules and clustering is characterized by comprising the following steps of:
s1: traversing a secondary system message information and fault device information data set by using an Apriori algorithm, mining the relevance in the secondary system message information and fault device information data set, describing the relevance relation between the message information and a secondary system fault device, extracting the relevance rule between the message information and the secondary system fault device, and realizing secondary system fault diagnosis;
s2: and performing k-means clustering on the extracted association rules to obtain the most appropriate category number, and when message information in some association rules appears again, directly checking fault devices in the association rules and fault devices of the association rules of the same category to realize efficient fault checking.
2. The secondary system information fault analysis method based on association rules and clustering of claim 1, wherein: in S1, the extracted association rules include message information and failure device information with high support degree, confidence degree, and promotion degree, and specific data information of the support degree, confidence degree, and promotion degree corresponding to the association rules is calculated, that is, each association rule corresponds to a set of three-dimensional data.
3. The secondary system information fault analysis method based on association rules and clustering according to claim 2, wherein: in S2, the association rules with similar support, confidence and lift are classified into one category, and all the association rules are classified by the three-dimensional data, i.e., the similarity between the association rules can be quantified.
4. The secondary system information fault analysis method based on association rules and clustering of claim 1, wherein: in S1, after the station personnel obtains the association rule, the station personnel is used as a reference basis, that is, when a plurality of pieces of message information simultaneously appear, if the message information is included in a certain association rule and a fault does not appear yet, the station personnel performs in advance a troubleshooting on a potential fault device mapped by the association rule, thereby implementing a secondary system fault diagnosis.
5. The association rule and cluster-based secondary system information fault analysis method of claim 1, wherein: in S1, the flow of extracting the association rule between the message information and the secondary system failure device using Apriori algorithm is as follows:
1) the Apriori algorithm scans the information of the fault devices of all secondary systems and the message information in the same time period to obtain each item and generate C1; then counting each item, and then deleting the unsatisfied items from the C1 according to the minimum support degree, the minimum confidence degree and the minimum promotion degree, so far obtaining a frequent 1 item set L1;
2) according to a pruning strategy, a set generated by the L1 generates a set C2 of candidate 2 items, then fault device information of all secondary systems and message information in the same time period are scanned, item number statistics is carried out on the C2, and similarly, items which do not meet requirements are deleted from the C2 according to the minimum support degree, the minimum confidence degree and the minimum promotion degree, and a frequent 2 item set L2 is obtained;
3) obtaining a frequent 3 item set L3, a frequent 4 item set L4, … and a frequent k item set Lk according to the method of the step 2);
4) judging whether k is the maximum, if so, generating an association rule Lk, and if not, adding 1 to the value of k and returning to the step 3) for circulation.
6. The secondary system information fault analysis method based on association rules and clustering of claim 1, wherein: in S2, k-means clustering is carried out on the extracted association rules:
the distance function of the k-means algorithm is chosen as follows:
wherein x is 1 ，x 2 ，x 3 ，y 1 ，y 2 ，y 3 Respectively representing three-dimensional data between the two groups of association rules; d (x, y) represents the Euclidean distance between the two sets of association rules;
selecting a Lande coefficient for a scoring function of the clustering effect as follows:
wherein TP is true positive rate, FP is false positive rate, and FN is false negative rate.
7. The association rule and cluster-based secondary system information fault analysis method of claim 5, wherein: in step 1), the Apriori algorithm scans the message information at the same time interval to obtain the message information before the actual fault occurs.
8. The secondary system information fault analysis method based on association rules and clustering of claim 5, wherein: in the step 1), the set minimum support degree, minimum confidence degree and minimum promotion degree are respectively 0.5%, 5% and 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210773125.2A CN115034671A (en) | 2022-07-01 | 2022-07-01 | Secondary system information fault analysis method based on association rule and cluster |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210773125.2A CN115034671A (en) | 2022-07-01 | 2022-07-01 | Secondary system information fault analysis method based on association rule and cluster |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115034671A true CN115034671A (en) | 2022-09-09 |
Family
ID=83129447
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210773125.2A Pending CN115034671A (en) | 2022-07-01 | 2022-07-01 | Secondary system information fault analysis method based on association rule and cluster |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115034671A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116720324A (en) * | 2023-05-15 | 2023-09-08 | 中铁第四勘察设计院集团有限公司 | Traction substation key equipment fault early warning method and system based on prediction model |
| CN117114454A (en) * | 2023-10-25 | 2023-11-24 | 南京中鑫智电科技有限公司 | DC sleeve state evaluation method and system based on Apriori algorithm |
-
2022
- 2022-07-01 CN CN202210773125.2A patent/CN115034671A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116720324A (en) * | 2023-05-15 | 2023-09-08 | 中铁第四勘察设计院集团有限公司 | Traction substation key equipment fault early warning method and system based on prediction model |
| CN117114454A (en) * | 2023-10-25 | 2023-11-24 | 南京中鑫智电科技有限公司 | DC sleeve state evaluation method and system based on Apriori algorithm |
| CN117114454B (en) * | 2023-10-25 | 2024-01-23 | 南京中鑫智电科技有限公司 | DC sleeve state evaluation method and system based on Apriori algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107835087B (en) | Automatic extraction method of alarm rule of safety equipment based on frequent pattern mining | |
| CN115034671A (en) | Secondary system information fault analysis method based on association rule and cluster | |
| CN105677791B (en) | For analyzing the method and system of the operation data of wind power generating set | |
| CN111259947A (en) | Power system fault early warning method and system based on multi-mode learning | |
| CN112910859B (en) | Internet of things equipment monitoring and early warning method based on C5.0 decision tree and time sequence analysis | |
| CN111709361A (en) | Unmanned aerial vehicle inspection data processing method for power transmission line | |
| CN113570200B (en) | Power grid running state monitoring method and system based on multidimensional information | |
| CN104281525B (en) | A kind of defect data analysis method and the method utilizing its reduction Software Testing Project | |
| CN111984788B (en) | Electric power system violation management method and device and electric power equipment | |
| CN105607631B (en) | The weak fault model control limit method for building up of batch process and weak fault monitoring method | |
| CN111126820A (en) | Electricity stealing prevention method and system | |
| CN112685459A (en) | Attack source feature identification method based on K-means clustering algorithm | |
| CN110826735A (en) | Electric power SCADA intelligent multidimensional query and maintenance method | |
| CN115186935B (en) | Electromechanical device nonlinear fault prediction method and system | |
| CN115658772A (en) | Unmanned aerial vehicle photovoltaic inspection data asset management method and system | |
| CN116011982A (en) | Online monitoring method and system for breakage of grinding roller of coal mill | |
| CN114880380A (en) | Method for realizing power grid alarm data association traceability system based on density clustering and self-organizing network | |
| CN109754159B (en) | Method and system for extracting information of power grid operation log | |
| Li | Practice of machine learning algorithm in data mining field | |
| Yang et al. | The relationship between management and control indexes of distribution network construction based on Apriori algorithm | |
| Yang et al. | Design and Application of Big Data Technology Management for the Analysis System of High Speed Railway Operation Safety Rules | |
| Luan et al. | A fast outlier detection for categorical datasets | |
| CN109190217B (en) | Message mask simulation test system of elastic regulation and control platform | |
| Feng et al. | RESEARCH ON RELIABILITY EVALUATION METHOD OF ARTIFICIAL INTELLIGENCE SOFTWARE IN NUCLEAR POWER PLANTS | |
| CN117390546A (en) | Multimode database fusion calculation model for instant anti-electricity-theft detection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |