CN109191011A - A kind of DC bushing state evaluating method and device based on Apriori algorithm - Google Patents
A kind of DC bushing state evaluating method and device based on Apriori algorithm Download PDFInfo
- Publication number
- CN109191011A CN109191011A CN201811190797.0A CN201811190797A CN109191011A CN 109191011 A CN109191011 A CN 109191011A CN 201811190797 A CN201811190797 A CN 201811190797A CN 109191011 A CN109191011 A CN 109191011A
- Authority
- CN
- China
- Prior art keywords
- bushing
- characteristic quantity
- data
- stage
- quantity data
- 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
- 238000000034 method Methods 0.000 title claims abstract description 41
- 230000007547 defect Effects 0.000 claims abstract description 31
- 230000007257 malfunction Effects 0.000 claims abstract description 21
- 230000008569 process Effects 0.000 claims abstract description 4
- 238000012360 testing method Methods 0.000 claims description 22
- 230000008439 repair process Effects 0.000 claims description 17
- 238000012549 training Methods 0.000 claims description 17
- 239000012212 insulator Substances 0.000 claims description 7
- 230000002159 abnormal effect Effects 0.000 claims description 6
- 230000005856 abnormality Effects 0.000 claims description 6
- 230000007613 environmental effect Effects 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 6
- 238000011156 evaluation Methods 0.000 claims description 3
- 230000002950 deficient Effects 0.000 claims 1
- 238000004458 analytical method Methods 0.000 description 24
- 230000008859 change Effects 0.000 description 7
- 238000007689 inspection Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000009412 basement excavation Methods 0.000 description 3
- 238000005065 mining Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000010606 normalization Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 238000000844 transformation Methods 0.000 description 3
- 241001269238 Data Species 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000009413 insulation Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000556 factor analysis Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000004868 gas analysis Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000005502 phase rule Effects 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
- 238000004804 winding Methods 0.000 description 1
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
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
Abstract
The present invention provides a kind of DC bushing state evaluating method based on Apriori algorithm, comprising: the operation phase according to locating for DC bushing collects the corresponding characteristic quantity data of the DC bushing and pre-processes the characteristic quantity data；According to predetermined DC bushing status assessment model, determine DC bushing status information corresponding with the characteristic quantity data of the pretreated DC bushing, the DC bushing status assessment model is to determine based on Apriori algorithm according to DC bushing defect malfunction history data.DC bushing state evaluating method provided by the invention based on Apriori algorithm is according to the operation data of converter transformer valve side sleeve and direct-current wall bushing, the state of casing is assessed, so that it is determined that the fault mode that may occur in the casing current locating operation phase, to realize the preparatory failure reminded or forecast may occur.
Description
Technical field
The present invention relates to electrical equipment technical fields, and more particularly, to a kind of direct current based on Apriori algorithm
Casing state evaluating method and device.
Background technique
In high voltage DC power transmission converter station, converter power transformer (abbreviation change of current change) value side bushing and direct-current wall bushing point
Be not connected to converter valve with exchange side line road.Converter transformer valve side sleeve and direct-current wall bushing are in alternating current-direct current mixing line commutation
The key position in area, any of operation, which is broken down, can all directly result in direct current locking, while influence AC and DC power grid
Operational safety.
In current fortune inspection work, do not accomplish to remind in advance for converter transformer valve side sleeve and direct-current wall bushing failure or
Forecast, can only be by the way of subsequent emergent management.
Summary of the invention
For cannot remind or forecast in advance at present converter transformer valve side sleeve and the failure of direct-current wall bushing this problem,
The present invention provides a kind of DC bushing state evaluating method based on Apriori algorithm, to assess converter transformer valve side set in time
The operating status of pipe and direct-current wall bushing realizes the failure that prompting in advance or forecast may occur.
The present invention provides a kind of DC bushing state evaluating method based on Apriori algorithm, comprising the following steps:
Step S10: the operation phase according to locating for DC bushing collects the corresponding characteristic quantity data of the DC bushing simultaneously
Pre-process the characteristic quantity data；
Step S20: according to predetermined DC bushing status assessment model, the determining and pretreated direct current set
The corresponding DC bushing status information of the characteristic quantity data of pipe, the DC bushing status assessment model are calculated based on Apriori
Method determines that the DC bushing status information includes: that casing SF6 gas is let out according to DC bushing defect malfunction history data
Leakage, bushing insulator housing corona discharge, casing oil leak and casing dielectric loss are abnormal.
Specifically, the method,
The operation phase includes: factory stage, the stage of putting into operation, depot repair stage；
The characteristic quantity data is that real-time measurement obtains, the number of the physical quantity of the operating condition including multiple reflection casings
The numerical value of the physical quantity of value and the environmental condition of multiple reflection casings.
Specifically, the method,
The operation phase according to locating for DC bushing collects the corresponding characteristic quantity data of the DC bushing, comprising:
If DC bushing is in the factory stage, the characteristic quantity number that the DC bushing is tested in the factory stage is collected
According to；
If DC bushing is in and puts into operation the stage, the characteristic quantity number that the DC bushing is tested in the factory stage is collected
According to the characteristic quantity data tested in the stage of putting into operation；
If DC bushing is in the depot repair stage, the feature that the DC bushing is tested in the factory stage is collected
Amount data, the characteristic quantity data tested in the stage of putting into operation and the characteristic quantity data tested in the factory stage.
Specifically, the method, the pretreatment characteristic quantity data include:
It, will be for current operation phase, the normal characteristic quantity data note of index according to the determining each metrics-thresholds of pre-selection
Record is 1；The characteristic quantity data of Indexes Abnormality is recorded as 0.
Specifically, the method, further includes:
DC bushing status assessment model is determined based on Apriori algorithm according to DC bushing defect malfunction history data
Step:
Collect the defect malfunction history data and the pretreatment historical data of DC bushing, the DC bushing be in
Under any operation phase: factory stage, the stage of putting into operation or depot repair stage；The DC bushing comes from multiple change of current transformations
Device；
The defect malfunction history data includes: the corresponding characteristic quantity data of the DC bushing described in any operation phase,
The corresponding DC bushing status information of the DC bushing described in any operation phase；
The pretreated historical data is divided into training data and test data；
Determine that the Apriori algorithm excavation training data obtains under the support threshold and confidence threshold value for meeting setting
The Strong association rule arrived；
Obtain the amendment Strong association rule determined after human expert modifies for the Strong association rule；
The amendment Strong association rule is applied to the test data, with the covering of the determination amendment Strong association rule
Degree；
Determine that coverage is strong regular for DC bushing state not less than preset value in the amendment Strong association rule
Assessment models.
DC bushing state evaluating method provided by the invention based on Apriori algorithm according to converter transformer valve side sleeve and
The operation data of direct-current wall bushing assesses the state of casing, so that it is determined that in the casing current locating operation phase
The fault mode that may occur, to realize the failure that prompting in advance or forecast may occur.
Detailed description of the invention
By reference to the following drawings, exemplary embodiments of the present invention can be more fully understood by:
Fig. 1 is a kind of schematic diagram of method of one embodiment of the invention；
Fig. 2 is a kind of schematic diagram of device of one embodiment of the invention.
Specific embodiment
Exemplary embodiments of the present invention are introduced referring now to the drawings, however, the present invention can use many different shapes
Formula is implemented, and is not limited to the embodiment described herein.
In order to improve the efficiency of fortune inspection work, fault occurrence reason is analyzed, authorities, which collect, converter transformer valve side sleeve
The defect report for the various faults being had occurred and that with direct-current wall bushing.In defect report (in general, involved in a report extremely
A few casing fault case) in, operating condition and environmental condition to converter transformer valve side sleeve and direct-current wall bushing carry out
It summarizes and analysis.
Specifically, the DC bushing to come into operation in different geographic regions is related in defect report, is also related to from not
Same manufacturer is served in the DC bushing of different voltages rank, and data are comprehensive, can cover the overwhelming majority in current use
The type of DC bushing.
It should be noted that although converter transformer valve side sleeve is different with the installation site of direct-current wall bushing, specific structure
It is not exactly the same, it is essentially identical in view of the operation characteristic and fault mode of the two, in the text, the two is unified for " direct current set
Pipe ".The description or explanation for being related to " DC bushing " all in following discussion, it is believed that be suitable for converter transformer valve side sleeve and
Direct-current wall bushing.
Each physical quantity being related in following list is perhaps related to converter transformer valve side sleeve or is related to direct current set through walls
Pipe, or both is directed to.Those skilled in the art can differentiate and confirm, which is not described herein again.
Casing fault case is analyzed it is found that the reason of causing inside pipe casing insulation fault is numerous, such as inside pipe casing overheat, office
Portion's electric discharge, air pressure reduction, mechanical failure, dampness etc. will cause casing failure, and many factors influence each other, and carry out cannula-like
State assessment is difficult with fault diagnosis.
On the other hand, currently, smart grid can support operation scene in real time or the fortune of measuring device periodically
Row data, these data can be directly or after treatment as equipment running status assessment or the features of fault diagnosis
Amount.Such as, the following operation data of DC bushing: casing dielectric dissipation factor, capacitance, internal gas decomposition etc. can be obtained.
These operation datas can be used as the data source for carrying out casing status assessment and basis.
As shown in Figure 1, the DC bushing status assessment side provided by one embodiment of the present invention based on Apriori algorithm
Method, comprising the following steps:
Step S10: the operation phase according to locating for DC bushing collects the corresponding characteristic quantity data of the DC bushing simultaneously
Pre-process the characteristic quantity data；
Step S20: according to predetermined DC bushing status assessment model, the determining and pretreated direct current set
The corresponding DC bushing status information of the characteristic quantity data of pipe, the DC bushing status assessment model are calculated based on Apriori
Method determines that the DC bushing status information includes: that casing SF6 gas is let out according to DC bushing defect malfunction history data
Leakage, bushing insulator housing corona discharge, casing oil leak and casing dielectric loss are abnormal.
The DC bushing state evaluating method based on Apriori algorithm is estimated in advance by machine learning and expertise
The operating status for counting converter transformer valve side sleeve or direct-current wall bushing these two types equipment, using having merged human expert's knowledge and gone through
The casing state assessment models of history operation data assess casing operating status, it is ensured that the conclusion of assessment be it is credible and
Effectively.
The DC bushing state evaluating method based on Apriori algorithm provides necessary technology for reply failure in time
It supports, effectively supplementary AC and DC grid can transport inspection work, help to ensure that safe operation；Power generation thing can also be drawn
Therefore give a lesson to, it takes precautions against similar accident and occurs again, improve power grid security production level.
Be construed as DC bushing status information, namely participate in training fault mode include the leakage of casing SF6 gas,
Casing SF6 micro-water content exception, bushing insulator housing corona discharge, bushing insulator housing external flashover, casing oil leak, set
Pipe internal discharge, casing current-carrying connecting component abnormal heating, casing dielectric loss exception etc..
Specifically, in the method, the operation phase includes: factory stage, the stage of putting into operation, depot repair stage；
The characteristic quantity data is that real-time measurement obtains, the numerical value of physical quantity of the operating conditions including multiple reflection casings and multiple
Reflect the numerical value of the physical quantity of the environmental condition of casing.
Specifically, the method, the operation phase according to locating for DC bushing, it is corresponding to collect the DC bushing
Characteristic quantity data, comprising:
If DC bushing is in the factory stage, the characteristic quantity number that the DC bushing is tested in the factory stage is collected
According to；
If DC bushing is in and puts into operation the stage, the characteristic quantity number that the DC bushing is tested in the factory stage is collected
According to the characteristic quantity data tested in the stage of putting into operation；
If DC bushing is in the depot repair stage, the feature that the DC bushing is tested in the factory stage is collected
Amount data, the characteristic quantity data tested in the stage of putting into operation and the characteristic quantity data tested in the factory stage.
Specifically, the method, the pretreatment characteristic quantity data include:
It, will be for current operation phase, the normal characteristic quantity data note of index according to the determining each metrics-thresholds of pre-selection
Record is 1；The characteristic quantity data of Indexes Abnormality is recorded as 0.
Specifically, the method, further includes:
DC bushing status assessment model is determined based on Apriori algorithm according to DC bushing defect malfunction history data
Step:
Collect the defect malfunction history data and the pretreatment historical data of DC bushing, the DC bushing be in
Under any operation phase: factory stage, the stage of putting into operation or depot repair stage；The DC bushing comes from multiple change of current transformations
Device；
The defect malfunction history data includes: the corresponding characteristic quantity data of the DC bushing described in any operation phase,
The corresponding DC bushing status information of the DC bushing described in any operation phase；
The pretreated historical data is divided into training data and test data；
Determine that the Apriori algorithm excavation training data obtains under the support threshold and confidence threshold value for meeting setting
The Strong association rule arrived；
Obtain the amendment Strong association rule determined after human expert modifies for the Strong association rule；
The amendment Strong association rule is applied to the test data, with the covering of the determination amendment Strong association rule
Degree；
Determine that coverage is strong regular for DC bushing state not less than preset value in the amendment Strong association rule
Assessment models.
This based on the DC bushing state evaluating method based on Apriori algorithm by defect report failure cause, therefore
The defects of hindering sign, fault degree, data structured formed the characteristic quantity data that can characterize DC bushing by data mining
Strong association rule between fault mode, is corrected by human expert, and test verifying, so that finally formed direct current set
The coverage original text of tubulose state assessment models, effectively, regular suitable scale carries out convenient for subsequent in the different operation phase rule
In DC bushing status assessment, determine the status information of DC bushing.
As shown in Fig. 2, the DC bushing state evaluation device based on Apriori algorithm of one embodiment of the invention, packet
It includes:
Characteristic quantity data obtains module 100, is used for:
The operation phase according to locating for DC bushing collects the corresponding characteristic quantity data of the DC bushing and pre-processes institute
State characteristic quantity data；
DC bushing status information determining module 200, is used for:
According to predetermined DC bushing status assessment model, the determining feature with the pretreated DC bushing
The corresponding DC bushing status information of data is measured, the DC bushing status assessment model is based on Apriori algorithm according to straight
Flow casing imperfection malfunction history data determine, the DC bushing status information include: casing SF6 gas leakage, casing it is exhausted
Edge housing corona discharge, casing oil leak and casing dielectric loss are abnormal.
Specifically, in the device, the operation phase includes: factory stage, the stage of putting into operation, depot repair stage；
The characteristic quantity data is that real-time measurement obtains, the numerical value of physical quantity of the operating conditions including multiple reflection casings and multiple
Reflect the numerical value of the physical quantity of the environmental condition of casing.
Specifically, in the device, the characteristic quantity data obtains module 100, is specifically used for:
If DC bushing is in the factory stage, the characteristic quantity number that the DC bushing is tested in the factory stage is collected
According to；
If DC bushing is in and puts into operation the stage, the characteristic quantity number that the DC bushing is tested in the factory stage is collected
According to the characteristic quantity data tested in the stage of putting into operation；
If DC bushing is in the depot repair stage, the feature that the DC bushing is tested in the factory stage is collected
Amount data, the characteristic quantity data tested in the stage of putting into operation and the characteristic quantity data tested in the factory stage.
Specifically, in the device, the characteristic quantity data obtains module 100, is specifically also used to:
It, will be for current operation phase, the normal characteristic quantity data note of index according to the determining each metrics-thresholds of pre-selection
Record is 1；The characteristic quantity data of Indexes Abnormality is recorded as 0.
Specifically, the device, further includes:
DC bushing status assessment model determining module 300, for being based on according to DC bushing defect malfunction history data
Apriori algorithm determines DC bushing status assessment model:
Collect the defect malfunction history data and the pretreatment historical data of DC bushing, the DC bushing be in
Under any operation phase: factory stage, the stage of putting into operation or depot repair stage；The DC bushing comes from multiple change of current transformations
Device；
The defect malfunction history data includes: the corresponding characteristic quantity data of the DC bushing described in any operation phase,
The corresponding DC bushing status information of the DC bushing described in any operation phase；
The pretreated historical data is divided into training data and test data；
Determine that the Apriori algorithm excavation training data obtains under the support threshold and confidence threshold value for meeting setting
The Strong association rule arrived；
Obtain the amendment Strong association rule determined after human expert modifies for the Strong association rule；
The amendment Strong association rule is applied to the test data, with the covering of the determination amendment Strong association rule
Degree；
Determine that coverage is strong regular for DC bushing state not less than preset value in the amendment Strong association rule
Assessment models.
The DC bushing state evaluation device and the DC bushing based on Apriori algorithm based on Apriori algorithm
The technical solution of state evaluating method is identical, and technical effect is identical, and which is not described herein again.
Above method and device are explained in detail below in conjunction with specific data.
The Life cycle of DC bushing (direct-current wall bushing and converter transformer valve side sleeve) substantially dispatch from the factory the stage, put into operation
Stage and this 3 stages of depot repair.In in each stage, all have obtain casing operation data channel and therefore and obtain
Operation data.
Specifically, in the factory stage, operation data is measured by casing delivery test or commissioning test；In the stage of putting into operation,
Operation data (such as table of the casing under the operation data of charging operation state and live power failure is obtained by smart grid equipment
1 and table 2 shown in)；In the depot repair stage, operation data (as shown in table 3) is obtained on shop test platform.
Table 1 puts into operation the method that operation data is obtained under stage casing charging operation state
| UV corona analysis |
| Measurement terminal voltage analysis |
| Infrared temperature analysis |
| The analysis of SF6 gas pressure |
| Outboard leak oil analysis |
| Main equipment oil chromatogram analysis |
| The alarm of direct-current wall bushing status monitoring and protection action analysis |
| Put down anti-status monitoring alarm and protection action analysis |
| The change of current becomes status monitoring alarm and protection action analysis |
Table 2 puts into operation the method that operation data is obtained under stage casing power failure operating status
| Visual examination analysis |
| Bushing insulator jacket internal inspection |
| Oil middle part sorting is looked into after casing is removed |
| Direct current resistance m easurem ent after casing is removed |
| The analysis of main equipment winding resistance |
| The analysis of casing dielectric dissipation factor |
| The analysis of bottom shielding of bushing dielectric dissipation factor |
| The analysis of capacitance of bushing amount |
| The analysis of casing insulation resistance |
| Main equipment oil chromatogram analysis |
| The analysis of SF6 slip |
| SF6 gas analysis |
The method that 3 failure of table (defect) casing returns acquisition operation data after factory
| The analysis of shelf depreciation PRPD chromatogram characteristic |
| The analysis of casing D.C. resistance |
| The analysis of casing D.C. resistance |
| Dielectric and magnetic analysis |
| Dielectric dissipation factor analysis |
| The test analysis of capacitance |
| Electric discharge type analysis |
From these operation datas, the characteristic quantity data of casing can be extracted, capacitance of bushing amount and set when including factory
Pipe dielectric dissipation factor, each characteristic quantity under stage charging operation state of putting into operation, each spy to put into operation under stage scene power failure
Sign amount, failure (defect) casing return each characteristic quantity after factory.Each characteristic quantity that casing is in the different operation phase is as shown in table 4.
4 casing of table is in each characteristic quantity of different operation phase
When it is implemented, arranging accident defect report and correlation test data and defect/fault model to collect data；
Following pretreatment is carried out to the data of collection: by defect/fault case Data Discretization, if a certain item casing operating status
Indexes Abnormality, then it is discrete to turn to " 1 "；It is discrete to turn to " 0 " if index is normal.
The data volume being related in view of 3 stages of casing is very big, pre- using " 1 " " 0 " this qualitative and non-quantitation data
Processing mode, it is possible to reduce data volume accelerates the speed of rule digging in subsequent mining process；On the other hand, this discrete
The mode of change makes the value of characteristic quantity only there are two state, is conducive to be further reduced data volume.
On the other hand, data are handled by the way of " 1 " " 0 " this two states, also realizes normalization simultaneously.?
Carry out the modeling of data-driven and when model training, do not have well-regulated data train the rule come can because the dimension of data not
Together, the magnitude differences of data are larger, and multiple rule occur, are unfavorable for model training.By to input data or output number
According to normalized is done, the degree of polymerization of treated training result is more preferable, regular more closer, the training result that training obtains
It is more meaningful.
Showing the normalization result obtained after the threshold value comparison of the value of the characteristic quantity of casing and index is given in table 5
Example.
The index and normalization example of 5 casing operating status of table
Pretreated data are divided into two parts, a portion accounts for the 70% of total amount of data, as training set；It is remaining
30% part as test set.
Minimal confidence threshold is set, the characteristic quantity and failure (defect) mould of casing operation are excavated using Apriori algorithm
Strong rule between formula.The method for excavating to obtain strong rule using Apriori algorithm, it is known to those skilled in the art, this
In repeat no more.
Further, the validity of the correlation rule formed is confirmed by human expert.When it is implemented, for less than 40%
The correlation rule of support, human expert are to retain or delete according to business scenario and experience confirmation.
Determining effective correlation rule by human expert, there may be certain redundancy rules, in order to advanced optimize shape
State assessment models determine that effective correlation rule carries out test verifying to human expert using test set.
Specifically, human expert is determined that effective correlation rule is applied in test set, to determine that human expert determines
The coverage of effective correlation rule；And determine that coverage is not less than the strong rule of preset value in the amendment Strong association rule
It is then DC bushing status assessment model.
The strong rule as DC bushing status assessment model finally determined is as shown in table 6.
6 rule digging result of table
To sum up, the characteristic quantity that the present invention is run using casing is independent variable, using failure (defect) mode as dependent variable, uses
Rule mining algorithms, Mining Frequent Itemsets Based, and using the threshold filtering of min confidence and minimum support, form high confidence level
Strong rule.
The present invention is described by reference to a small amount of embodiment above.However, it is known in those skilled in the art,
As defined by subsidiary Patent right requirement, in addition to the present invention other embodiments disclosed above equally fall in this hair
In bright range.
Claims (10)
1. a kind of DC bushing state evaluating method based on Apriori algorithm, which comprises the following steps:
Step S10: the operation phase according to locating for DC bushing collects the corresponding characteristic quantity data of the DC bushing and locates in advance
Manage the characteristic quantity data；
Step S20: determining and the pretreated DC bushing according to predetermined DC bushing status assessment model
The corresponding DC bushing status information of characteristic quantity data, the DC bushing status assessment model are based on Apriori algorithm root
It is determined according to DC bushing defect malfunction history data, the DC bushing status information includes: the leakage of casing SF6 gas, set
Bushing insulator housing corona discharge, casing oil leak and casing dielectric loss are abnormal.
2. the method according to claim 1, wherein
The operation phase includes: factory stage, the stage of putting into operation, depot repair stage；
The characteristic quantity data be real-time measurement obtain, including it is multiple reflection casings operating conditions physical quantity numerical value and
The numerical value of the physical quantity of the environmental condition of multiple reflection casings.
3. according to the method described in claim 2, it is characterized in that,
The operation phase according to locating for DC bushing collects the corresponding characteristic quantity data of the DC bushing, comprising:
If DC bushing is in the factory stage, the characteristic quantity data that the DC bushing is tested in the factory stage is collected；
Put into operation the stage if DC bushing is in, collect characteristic quantity data that the DC bushing was tested in the factory stage and
In the characteristic quantity data that the stage of putting into operation tests；
If DC bushing is in the depot repair stage, the characteristic quantity number that the DC bushing is tested in the factory stage is collected
According to, the characteristic quantity data tested in the stage of putting into operation and the characteristic quantity data tested in the factory stage.
4. the method according to claim 1, wherein the pretreatment characteristic quantity data includes:
According to the determining each metrics-thresholds of pre-selection, by for the current operation phase, the normal characteristic quantity data of index is recorded as
1；The characteristic quantity data of Indexes Abnormality is recorded as 0.
5. the method according to claim 1, wherein further include:
The step of DC bushing status assessment model is determined based on Apriori algorithm according to DC bushing defect malfunction history data
It is rapid:
The defect malfunction history data and the pretreatment historical data, the DC bushing for collecting DC bushing are in following
One operation phase: factory stage, the stage of putting into operation or depot repair stage；The DC bushing comes from multiple converter power transformers；
The defect malfunction history data includes: the corresponding characteristic quantity data of the DC bushing described in any operation phase, in office
The corresponding DC bushing status information of DC bushing described in one operation phase；
The pretreated historical data is divided into training data and test data；
Determine that Apriori algorithm excavates what the training data obtained under the support threshold and confidence threshold value for meeting setting
Strong association rule；
Obtain the amendment Strong association rule determined after human expert modifies for the Strong association rule；
The amendment Strong association rule is applied to the test data, with the coverage of the determination amendment Strong association rule；
Determine that coverage is strong regular for DC bushing status assessment not less than preset value in the amendment Strong association rule
Model.
6. a kind of DC bushing state evaluation device based on Apriori algorithm characterized by comprising
Characteristic quantity data obtains module, is used for:
The operation phase according to locating for DC bushing collects the corresponding characteristic quantity data of the DC bushing and pre-processes the spy
Sign amount data；
DC bushing status information determining module, is used for:
According to predetermined DC bushing status assessment model, the determining characteristic quantity number with the pretreated DC bushing
According to corresponding DC bushing status information, the DC bushing status assessment model is based on Apriori algorithm according to direct current set
What defective tube malfunction history data determined, the DC bushing status information includes: the leakage of casing SF6 gas, bushing insulator
Housing corona discharge, casing oil leak and casing dielectric loss are abnormal.
7. the apparatus according to claim 1, which is characterized in that
The operation phase includes: factory stage, the stage of putting into operation, depot repair stage；
The characteristic quantity data be real-time measurement obtain, including it is multiple reflection casings operating conditions physical quantity numerical value and
The numerical value of the physical quantity of the environmental condition of multiple reflection casings.
8. device according to claim 7, which is characterized in that
The characteristic quantity data obtains module, is specifically used for:
If DC bushing is in the factory stage, the characteristic quantity data that the DC bushing is tested in the factory stage is collected；
Put into operation the stage if DC bushing is in, collect characteristic quantity data that the DC bushing was tested in the factory stage and
In the characteristic quantity data that the stage of putting into operation tests；
If DC bushing is in the depot repair stage, the characteristic quantity number that the DC bushing is tested in the factory stage is collected
According to, the characteristic quantity data tested in the stage of putting into operation and the characteristic quantity data tested in the factory stage.
9. device according to claim 6, which is characterized in that
The characteristic quantity data obtains module, is specifically also used to:
According to the determining each metrics-thresholds of pre-selection, by for the current operation phase, the normal characteristic quantity data of index is recorded as
1；The characteristic quantity data of Indexes Abnormality is recorded as 0.
10. device according to claim 6, which is characterized in that further include:
DC bushing status assessment model determining module, for being based on Apriori according to DC bushing defect malfunction history data
Algorithm determines DC bushing status assessment model:
The defect malfunction history data and the pretreatment historical data, the DC bushing for collecting DC bushing are in following
One operation phase: factory stage, the stage of putting into operation or depot repair stage；The DC bushing comes from multiple converter power transformers；
The defect malfunction history data includes: the corresponding characteristic quantity data of the DC bushing described in any operation phase, in office
The corresponding DC bushing status information of DC bushing described in one operation phase；
The pretreated historical data is divided into training data and test data；
Determine that Apriori algorithm excavates what the training data obtained under the support threshold and confidence threshold value for meeting setting
Strong association rule；
Obtain the amendment Strong association rule determined after human expert modifies for the Strong association rule；
The amendment Strong association rule is applied to the test data, with the coverage of the determination amendment Strong association rule；
Determine that coverage is strong regular for DC bushing status assessment not less than preset value in the amendment Strong association rule
Model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811190797.0A CN109191011A (en) | 2018-10-12 | 2018-10-12 | A kind of DC bushing state evaluating method and device based on Apriori algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811190797.0A CN109191011A (en) | 2018-10-12 | 2018-10-12 | A kind of DC bushing state evaluating method and device based on Apriori algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109191011A true CN109191011A (en) | 2019-01-11 |
Family
ID=64948308
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811190797.0A Pending CN109191011A (en) | 2018-10-12 | 2018-10-12 | A kind of DC bushing state evaluating method and device based on Apriori algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109191011A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120173301A1 (en) * | 2011-01-04 | 2012-07-05 | International Business Machines Corporation | System and method for failure association analysis |
| CN106022950A (en) * | 2016-05-06 | 2016-10-12 | 中国电力科学研究院 | Power distribution network secondary equipment type identification method and system |
| CN106526352A (en) * | 2016-09-30 | 2017-03-22 | 中国电力科学研究院 | Method and system for determining power transformer fault types |
| CN107862459A (en) * | 2017-11-10 | 2018-03-30 | 国网浙江省电力公司电力科学研究院 | Metering equipment state evaluation method and system based on big data |
-
2018
- 2018-10-12 CN CN201811190797.0A patent/CN109191011A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120173301A1 (en) * | 2011-01-04 | 2012-07-05 | International Business Machines Corporation | System and method for failure association analysis |
| CN106022950A (en) * | 2016-05-06 | 2016-10-12 | 中国电力科学研究院 | Power distribution network secondary equipment type identification method and system |
| CN106526352A (en) * | 2016-09-30 | 2017-03-22 | 中国电力科学研究院 | Method and system for determining power transformer fault types |
| CN107862459A (en) * | 2017-11-10 | 2018-03-30 | 国网浙江省电力公司电力科学研究院 | Metering equipment state evaluation method and system based on big data |
Non-Patent Citations (2)
| Title |
|---|
| 李志超 等: ""基于集对分析和综合赋权的电力变压器套管绝缘状态评估"", 《高压电器》 * |
| 黄灿 等: ""基于气象数据挖掘的输变电设备故障预警"", 《新型工业化》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
|---|---|---|
| CN103454516B (en) | Intelligent transformer substation secondary equipment health state diagnostic method | |
| CN110727662A (en) | Low-voltage transformer area user phase identification method and system based on correlation analysis | |
| CN111008485B (en) | Neural network-based multi-parameter life prediction method for three-phase alternating current asynchronous motor | |
| EP2680017A1 (en) | A method of early detection of feeder lines with a high-ohm ground fault in compensated power networks | |
| CN1862278A (en) | Method and system for predicting remaining life for motors featuring on-line insulation condition monitor | |
| CN103513125A (en) | Integrated intelligent diagnosis system and method of above-220KV transformers | |
| CN111898092A (en) | Cable running state evaluation decision analysis method and system | |
| CN108344917A (en) | The cross interconnected cable on-line fault diagnosis method of 110kV crosslinked polyethylenes based on method of loci | |
| CN112305319B (en) | Method and system for online monitoring of converter transformer valve side sleeve parameters | |
| CN109239516A (en) | Merge the transformer fault layering diagnostic method of a variety of intelligent diagnostics models | |
| CN106295240A (en) | A kind of Condition Assessment for Power Transformer method based on fuzzy mathematics | |
| CN109342882A (en) | Cable fault on-line monitoring method based on multi-source Eigenvalues analysis | |
| Hernandez et al. | Selecting a condition monitoring system for enhancing effectiveness of power transformer maintenance | |
| US20190340708A1 (en) | Method for the usage planning of an electrical system for supplying energy | |
| KR102260550B1 (en) | Facility health monitoring method by measuring the electric circuit constant inside the power facility in operation | |
| Dhote et al. | Improvement in transformer diagnosis by DGA using fuzzy logic | |
| Milosavljevic et al. | Integrated transformer health estimation methodology based on Markov chains and evidential reasoning | |
| CN109191011A (en) | A kind of DC bushing state evaluating method and device based on Apriori algorithm | |
| KR101988670B1 (en) | Device for state estimation of power distribution system | |
| Singh et al. | A Fuzzy Modeling Technique to Assist Submersible Inspection Robot for Internal Inspection of Transformers | |
| Chang et al. | Fuzzy theory-based partial discharge technique for operating state diagnosis of high-voltage motor | |
| Mokhlis et al. | Fault location estimation for distribution system using simulated voltage sags data | |
| Montanari | Envisaging links between fundamental research in electrical insulation and electrical asset management | |
| KR102419753B1 (en) | Facility health monitoring method by measuring the electric circuit constant inside the power facility in operation | |
| Tippannavar et al. | Smart Transformer-An Analysis of Recent Technologies for Monitoring Transformer |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190111 |
|
| RJ01 | Rejection of invention patent application after publication |