CN107247995A - Transmission line of electricity running status association rule mining and Forecasting Methodology based on Bayesian model - Google Patents
Transmission line of electricity running status association rule mining and Forecasting Methodology based on Bayesian model Download PDFInfo
- Publication number
- CN107247995A CN107247995A CN201610866011.7A CN201610866011A CN107247995A CN 107247995 A CN107247995 A CN 107247995A CN 201610866011 A CN201610866011 A CN 201610866011A CN 107247995 A CN107247995 A CN 107247995A
- Authority
- CN
- China
- Prior art keywords
- transmission line
- running status
- electricity running
- frequent
- association rule
- 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
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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Computational Mathematics (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Probability & Statistics with Applications (AREA)
- Algebra (AREA)
- Artificial Intelligence (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of transmission line of electricity running status association rule mining based on Bayesian model and Forecasting Methodology, it is characterised in that comprises the following steps：(1) frequent 1 item collection L1 is generated based on the item collection I related to transmission line of electricity running status, frequent 1 item collection L1 is connected two-by-two to generate the item collection C2 of candidate 2, frequent 2 item collection L2 is generated based on the item collection C2 of candidate 2；(2) Bayesian network is constructed based on frequent 2 item collection L2；(3) correlation rule R is generated according to Bayesian network；(4) the corresponding supports of the correlation rule R and confidence level are calculated, correlation rule R Strong association rule is obtained；(5) correlation rule R Strong association rule and its corresponding confidence level prediction transmission line of electricity running status is combined.Correlation rule between each parameter related to transmission line of electricity running status can be applied in transmission line status prediction by the inventive method, so as to improve the precision and accuracy of prediction.
Description
Technical field
It is based on the present invention relates to a kind of transmission line of electricity running status association rule mining and Forecasting Methodology, more particularly to one kind
The transmission line of electricity running status association rule mining and Forecasting Methodology of Bayesian model.
Background technology
With the high speed development of national economy, society's electricity consumption demand increasingly increases, and electricity shortage has become a warp
The limiting factor of Ji development, it is especially prominent in the economically developed Yangtze River Delta, Pearl River Delta area.At present, the development of power network substantially falls
After expanding economy, the construction of power supply is lagged behind so that the deficiency of transmitting capacity of the electric wire netting seems more prominent.In addition power network is built
If being restricted by factors such as environment, costs so that construction period length, development speed are limited, tend not to meet the need of transmission of electricity
Ask.Therefore, accelerating outside Electric Power Network Planning construction, by technological transformation and upgrading, excavating electric network transportation ability potentiality, improve defeated
Electrical efficiency becomes the direction of a research.
Shadow of the running status of transmission line of electricity by many-sided condition such as meteorological condition, loading condition and wire gauge
Ring.Analyzed and predicted by choosing suitable forecast model, and to key influence factor, shape can be run to transmission line of electricity
State carries out more rational assessment, and raising efficiency makes great sense in being run to actual production.Entered using a large amount of historical datas
Row analysis, the variation tendency for excavating and predicting certain parameter is a key content of big data analysis.Existing power transmission line operation
Trend prediction method is confined to the analysis to single parameter mostly, not in view of the correlation between many reference amounts.In addition, by
In the complexity and dynamic of transmission line of electricity running environment, analyzing prediction to the systematic parameter in continuous time section can not be accurate
Reflect the state change of circuit.
Therefore, it is desirable to obtain a kind of transmission line of electricity running status association rule mining and Forecasting Methodology, this method can be by
Correlation rule between each parameter related to transmission line of electricity running status is applied in transmission line status forecast analysis assessment,
So as to be conducive to improving the precision predicted.In addition, this method can to the association rule mining of multisystem parameter, ambient parameter,
So as to improve the accuracy predicted transmission line status.
The content of the invention
It is an object of the invention to provide a kind of transmission line of electricity running status association rule mining based on Bayesian model and
Correlation rule between each parameter related to transmission line of electricity running status can be applied to power transmission line by Forecasting Methodology, this method
In line state prediction, so as to be conducive to improving the precision and accuracy predicted.
According to foregoing invention purpose, the present invention proposes a kind of transmission line of electricity running status association based on Bayesian model
Rule digging and Forecasting Methodology, it is characterised in that comprise the following steps：
(1) frequent 1- item collections L1 is generated based on the item collection I related to transmission line of electricity running status, connects frequent 1- two-by-two
Collect L1 to generate candidate 2- item collection C2, frequent 2- item collections L2 is generated based on candidate's 2- item collections C2；
(2) Bayesian network is constructed based on frequent 2- item collections L2；
(3) correlation rule R is generated according to the Bayesian network；
(4) the corresponding supports of the correlation rule R and confidence level are calculated, correlation rule R Strong association rule is obtained；
(5) with reference to the Strong association rule of the correlation rule R and its corresponding confidence level prediction transmission line of electricity running status.
For ease of understanding, principle of the present invention is illustrated first：
Bayesian network (BayesianNetwork, BN) provides one to uncertainty knowledge for association rule mining
The framework with reasoning is represented, possesses the ability for remaining each node state of reasoning of being set out with any node.
Bayesian network is the model that probability statistics and graph theory are combined, and represents the net of dependence and independence between stochastic variable
Network model, is that uncertain problem and challenge provide intuitively method for expressing.Bayesian network is directed acyclic graph, uses arc
To represent the dependence between each variable, probability distribution table represents the power of dependence between variable, and general based on posteriority
The Bayes' theorem of rate, can carry out tight reasoning and calculation.Present invention application Bayesian network is to transmission line of electricity and power transmission line
The excavation of incidence relation between the related each parameter of road running status.
Bayesian network is mainly made up of two parts, respectively describe the directed acyclic graph of dependence structure and description according to
The strong and weak conditional probability table of the relation of relying, directed acyclic graph is from node with being constituted to side.Each node corresponds to a random change
Amount, in the present invention for describing each parameter related to transmission line of electricity running status；Each between the correspondence stochastic variable of side
Dependence, arrow embodies the directionality of cause and effect, and the dependence that conditional probability table is then described between node is strong and weak.
The chain type rule of Bayesian network is provided according to conditional probability, with stochastic variable collection X={ X1,X2,…,XnExemplified by,
Joint probability is：
Bayesian network has a critical nature, and node is when its father node is determined, the node condition is independently of all
Non- father node.Based on this property.Joint probability can using abbreviation as：
Wherein, Parents (Xi) represent nodes XiFather node joint, probable value can find from conditional probability table.
Association rule mining can be found that in database the contact between data, and these contacts be it is unknown in advance, no
It can be obtained by logical operation and statistical method, not be the own attribute of data in itself, but based on appearance simultaneously between data
Feature.
Association rule mining problem can be as follows with formalized description：
Affairs：If I={ i1,i2,…,imIt is the set that m disparity items is constituted, ik(k=1,2 ..., m) it is referred to as project
(Item).DB is the transaction database for I, and each affairs T is the set of one group of attribute in I, i.e.,And have one only
One identifier TID.
Item collection X support Support (X)：Item collection X support represents X importance, and X can be known frequent
Item collection.Support item collection X number of transactions to be referred to as item collection X support number in transaction database, be designated as X.Count.If | DB | it is affairs
The sum of database, then item collection X support be designated as:
Support (X)=X.Count/ | DB | (3)
Minimum support:It was found that the minimum support of task, the item collection for meeting minimum support just can be in correlation rule
Occur, referred to as frequent item set, otherwise referred to as weakness collection.
Regular confidence level：Regular confidence level represents the reliability standard of rule.For correlation rule:
Note X is preceding paragraph, and Y is consequent, then the confidence level of R correlation rules is：
Confidence (R)=Support (X ∪ Y)/Support (X) (5)
If R confidence level is more than min confidence, referred to as Strong association rule, if less than min confidence, it is referred to as weak
Correlation rule.
Conventional association rules mining algorithm has Apriori, AprioriTid and AprioriHybrid algorithm at present.Its
Middle Apriori algorithm finds frequent item set, cycle calculations until being produced without new frequent item set by scan database, but by
It is generally in large scale in database, support consumption ample resources is iterated to calculate each time.AprioriTid methods, which are used, gradually to be subtracted
Few candidate data storehouse calculates the support number of candidate's strong point collection, improves computational efficiency.AprioriHybrid algorithms are preceding two
The combination of person, can use AprioriTid algorithms, otherwise using Apriori algorithm in candidate data storehouse during graftabl.
The step of Apriori association rule minings, is broadly divided into two parts：
1) all frequent item sets that identification target data is concentrated, i.e. support are not less than the item collection of setting minimum support；
2) correlation rule that all confidence levels are not less than setting min confidence is constructed from above-mentioned frequent item set.
Apriori algorithm is after the higher frequent item set (such as frequent 2- item collections) of dimension is produced, it is necessary to scan again
Database is to calculate support, and the association rule mining method based on Bayesian network can avoid this from scanning again, with shellfish
The form of this network of leaf represents the frequent 2- item collections that single pass database is produced, and finds out correlation rule accordingly.
Association rule mining method based on Bayesian network assumes obtained Bayesian network by n node and l bars side
Constitute, travel through the directed acyclic graph complexity for O (n+n2+l).Because data volume is far longer than nodes, therefore, pass through the party
The method efficiency of method generation correlation rule is higher than the method that correlation rule is directly obtained by Apriori algorithm.
Correlation rule is introduced to the neutral net of prediction transmission line of electricity running status so that focus more on the association between data
Property, so as to lift precision of prediction and the degree of accuracy.
In transmission line of electricity running status association rule mining and Forecasting Methodology of the present invention based on Bayesian model,
The project of the item collection I includes each parameter related to transmission line of electricity running status.It was found from principles above, frequent 1- item collections L1
Characterize important parameter.Important contact between frequently 2- item collections L2 characterizes important parameter two-by-two, therefore can be based on frequent
2- item collections L2 constructs Bayesian network.The present invention excavates all frequent item sets using mining algorithm, is found using Bayesian network
All correlation rules, and the support and confidence level of every correlation rule are calculated, so as to avoid after frequent item set dimension increase
Still need the huge amount of calculation that scan database is brought.The present invention introduces pattra leaves on the basis of traditional association rule digging method
This network, transmission line status prediction is applied to by the correlation rule between each parameter related to transmission line of electricity running status
In, so as to be conducive to improving the precision and accuracy predicted.
Further, transmission line of electricity running status association rule mining of the present invention based on Bayesian model and pre-
In survey method, step is also included between the step (1) and step (2)：Screen frequent 2- item collections L2.
In such scheme, screening frequent 2- item collections L2 method can be：When item collection X, Y has following relation
Support(X∪Y)≈Support(X)·Support(Y) (6)
It is separate then to think item collection X, Y, and correlation rule is not meaningful, should be rejected from frequent 2- item collections.
Further, transmission line of electricity running status association rule mining of the present invention based on Bayesian model and pre-
In survey method, in the step (1), using Apriori algorithm, AprioriTid algorithms and AprioriHybrid algorithms
One of them generates frequent 1- item collections L1 by scanning item collection I.
Further, transmission line of electricity running status association rule mining of the present invention based on Bayesian model and pre-
In survey method, in the step (1), using Apriori algorithm, AprioriTid algorithms and AprioriHybrid algorithms
One of them generates frequent 2- item collections L2 by scanning candidate's 2- item collections C2.
Further, transmission line of electricity running status association rule mining of the present invention based on Bayesian model and pre-
In survey method, in the step (5), correlation rule R and its corresponding support and confidence level are applied to neural network prediction
Transmission line of electricity running status.
Further, above-mentioned transmission line of electricity running status association rule mining and Forecasting Methodology based on Bayesian model
In, the neutral net is wavelet neural network.
Wavelet neural network is the product that neutral net is combined with both wavelet analysises technology.Neutral net is to imitate big
One model of cerebral nervous system, a large amount of neurons are connected with each other, the class self-adaptation nonlinear dynamical system formed is combined.
Small echo is then the new technology of T/F analysis field.Wavelet neural network has time-frequency domain characteristic, with reference to neutral net, make
The excitation function of neutral net hidden layer is substituted with wavelet function, the connection of wavelet basis function and whole network, small echo is constituted
The diversity of function also allows when predicting inhomogeneous data to be selected according to different Wavelet Properties.Wavelet Neural Network
The shift factor of network, contraction-expansion factor can be determined in advance, it is to avoid network is absorbed in local minimum, also shows on the whole more
Strong learning ability, Function approximation capabilities, pattern classification ability and network generalization, adds precision of prediction.
Further, transmission line of electricity running status association rule mining of the present invention based on Bayesian model and pre-
In survey method, project in the item collection I related to transmission line of electricity running status include environment temperature, line load, wind speed,
Some items in strain section circuit mean temperature, strain section ruling span sag and strain section tension force.
Further, above-mentioned transmission line of electricity running status association rule mining and Forecasting Methodology based on Bayesian model
In, the project in the item collection I related to transmission line of electricity running status also includes environment temperature, line load, wind speed, strain insulator
The rate of change of some in section circuit mean temperature, strain section ruling span sag and strain section tension force.
Transmission line of electricity running status association rule mining and Forecasting Methodology of the present invention based on Bayesian model, its
Advantage and beneficial effect include：
(1) that the correlation rule between each parameter related to transmission line of electricity running status is applied into transmission line status is pre-
Survey in analysis and evaluation, so as to be conducive to improving the precision predicted.
(2) to the association rule mining of multisystem parameter, ambient parameter, so as to improve the standard predicted transmission line status
True property.
(3) all correlation rules are found using Bayesian network, so as to be stilled need after avoiding frequent item set dimension increase
The huge amount of calculation that scan database is brought.
Brief description of the drawings
Fig. 1 is transmission line of electricity running status association rule mining and prediction side of the present invention based on Bayesian model
A kind of schematic flow sheet of the method under embodiment.
Fig. 2 is the ambient temperature data in the corresponding embodiments of the method for Fig. 1.
Fig. 3 is the line load data in the corresponding embodiments of the method for Fig. 1.
Fig. 4 is the air speed data in the corresponding embodiments of the method for Fig. 1.
Fig. 5 is the strain section circuit average temperature data in the corresponding embodiments of the method for Fig. 1.
Fig. 6 is the strain section ruling span sag data in the corresponding embodiments of the method for Fig. 1.
Fig. 7 is the strain section tension data in the corresponding embodiments of the method for Fig. 1.
Fig. 8 is the Bayesian network schematic diagram in the corresponding embodiments of the method for Fig. 1.
Fig. 9 is the strain section span sag predicated error correction data in the corresponding embodiments of the method for Fig. 1.
Figure 10 is the strain section circuit mean temperature predicated error correction data in the corresponding embodiments of the method for Fig. 1.
Embodiment
Below in conjunction with Figure of description and specific embodiment to the transmission of electricity of the present invention based on Bayesian model
Circuit running status association rule mining and Forecasting Methodology are described in further detail.
Fig. 1 illustrates the transmission line of electricity running status association rule mining of the present invention based on Bayesian model and pre-
A kind of flow of the survey method under embodiment.
As shown in figure 1, the transmission line of electricity running status association rule mining based on Bayesian model under the embodiment
And Forecasting Methodology, it comprises the following steps：
Step 110：In database, frequent 1- item collections L1 is generated based on the item collection I related to transmission line of electricity running status,
Frequent 1- item collections L1 is connected two-by-two to generate candidate 2- item collection C2, and frequent 2- item collections L2 is generated based on candidate's 2- item collections C2.
In the step, the project in item collection I includes environment temperature i1, line load i2, wind speed i3, strain section circuit be averaged
Temperature i4, strain section ruling span sag i5And strain section tension force i6.In addition, also including variation of ambient temperature rate i7, circuit bear
Lotus rate of change i8, wind speed rate of change i9, strain section circuit average ramp rate i10, strain section ruling span sag rate of change i11
And strain section tension variation rate i12.Above-mentioned rate of change is the rate of change in the unit time, and the unit interval is h in the present embodiment
(hour).Above-mentioned project i1-i6Specific data as shown in Fig. 2-Fig. 7, include 1118 of certain 500kV transmission line of electricity operation respectively
Data point.
In the step, all frequent item sets are found by Multiple-Scan database using Apriori algorithm.Setting is minimum
Support is 0.35.In first time is scanned, item collection I is scanned, the support of all single projects is calculated, takes wherein that support is not
Item collection less than minimum support constitutes frequent 1- item collections L1.In second scans, ergodic data storehouse, calculates candidate 2- again
The support of each element in item collection C2, takes wherein support to construct frequent 2- item collections L2 not less than the item collection of minimum support.
The method of above-mentioned calculating support is：Its support Support (X) is calculated as follows by taking item collection X as an example：
Support (X)=X.Count/ | DB |
Wherein, support item collection X number of transactions to be referred to as item collection X support number, be designated as X.Count, | DB | it is transaction database
Sum.
Step 120：In database, frequent 2- item collections L2 is screened.
In the step, screening frequent 2- item collections L2 method is：When item collection X, Y has following relation
Support(X∪Y)≈Support(X)·Support(Y) (6)
It is separate then to think item collection X, Y, and correlation rule is not meaningful, should be rejected from frequent 2- item collections.
Step 130：Bayesian network is constructed based on frequent 2- item collections L2.
In the step, by step 120 screening after frequent 2- item collections L2 based on, using frequent 2- item collections L2 project as
Node, L2 is side, constitutes Bayesian network.The Bayesian network is as shown in Figure 8.
Step 140：The Bayesian network generation correlation rule R according to Fig. 8.
In the step, correlation rule R is expressed as
Step 150：The corresponding supports of the correlation rule R and confidence level are calculated, correlation rule R strong association rule are obtained
Then.
In the step, for correlation rule R, note X is preceding paragraph, and Y is consequent, then correlation rule R confidence level
Confidence (R) computational methods are：
Confidence (R)=Support (X ∪ Y)/Support (X)
If Confidence (R) is more than min confidence (min confidence is set in the present embodiment as 0.7), it is determined as
Correlation rule R Strong association rule.
In the present embodiment, wind speed i3, wind speed rate of change i9With strain section ruling span sag i5Between, environment temperature i1, line
Road load i2With strain section circuit mean temperature i4Between there is High relevancy, be correlation rule R Strong association rule, such as the institute of table 1
Show.
The correlation rule R of table 1. Strong association rule
Step 160：With reference to above-mentioned correlation rule R Strong association rule and its prediction transmission line of electricity operation of corresponding confidence level
State.
In the step, correlation rule R and its corresponding support and confidence level are applied to wavelet neural network prediction defeated
Electric line running status.
Respectively by wind speed i3, wind speed rate of change i9, strain section ruling span sag i5Inputted as wavelet neural network, it is resistance to
Open section ruling span sag i5Exported as neutral net, respectively by environment temperature i1, line load i2It is average with strain section circuit
Temperature i4Inputted as neutral net, strain section circuit mean temperature i4As output, obtained correlation rule probability is added (i.e.
Confidence level) matrix is neutral net hidden layer, takes at first 1000 points as network training data, and predicts following 4 days and amount to 96
Data point strain section ruling span sag i5With strain section circuit mean temperature i4, neutral net every 12 hours are 12 data points
Update once, and calculate its relative error.
Fig. 9 shows strain section ruling span sag i5Predicated error correction data, Figure 10 shows that strain section circuit is put down
Equal temperature i4Predicated error correction data.
As shown in figure 9, usage history data take neural network prediction strain section ruling span sag i5Obtained prediction
Error B is introduced after correlation rule between 7.8% to 16%, prediction strain section ruling span sag i5Obtained prediction is missed
Poor A drops between 1.5%-10.3%.Equally, as shown in Figure 10, the strain section circuit mean temperature introduced after correlation rule
i4Predict the outcome C and usage history data take neural network prediction strain section circuit mean temperature i4Obtained predicated error D
Compared to there is obvious reduction, maximum relative error is no more than 13%.Therefore, introduce correlation rule and considerably improve precision of prediction,
Demonstrate validity of the association rule mining in the prediction of transmission line of electricity running status.
It should be noted that listed above is only specific embodiment of the invention, it is clear that implement the invention is not restricted to more than
Example, the similar change for having many therewith.If those skilled in the art directly exports or joined from present disclosure
All deformations expected, all should belong to protection scope of the present invention.
Claims (8)
1. a kind of transmission line of electricity running status association rule mining and Forecasting Methodology based on Bayesian model, it is characterised in that
Comprise the following steps：
(1) frequent 1- item collections L1 is generated based on the item collection I related to transmission line of electricity running status, frequent 1- item collections L1 is connected two-by-two
To generate candidate 2- item collection C2, frequent 2- item collections L2 is generated based on candidate's 2- item collections C2；
(2) Bayesian network is constructed based on frequent 2- item collections L2；
(3) correlation rule R is generated according to the Bayesian network；
(4) the corresponding supports of the correlation rule R and confidence level are calculated, correlation rule R Strong association rule is obtained；
(5) with reference to the Strong association rule of the correlation rule R and its corresponding confidence level prediction transmission line of electricity running status.
2. transmission line of electricity running status association rule mining and prediction side as claimed in claim 1 based on Bayesian model
Method, it is characterised in that also include step between the step (1) and step (2)：Screen frequent 2- item collections L2.
3. transmission line of electricity running status association rule mining and prediction side as claimed in claim 1 based on Bayesian model
Method, it is characterised in that in the step (1), is calculated using Apriori algorithm, AprioriTid algorithms and AprioriHybrid
One of method generates frequent 1- item collections L1 by scanning item collection I.
4. transmission line of electricity running status association rule mining and prediction side as claimed in claim 1 based on Bayesian model
Method, it is characterised in that in the step (1), is calculated using Apriori algorithm, AprioriTid algorithms and AprioriHybrid
One of method generates frequent 2- item collections L2 by scanning candidate's 2- item collections C2.
5. transmission line of electricity running status association rule mining and prediction side as claimed in claim 1 based on Bayesian model
Method, it is characterised in that in the step (5), neutral net is applied to by correlation rule R and its corresponding support and confidence level
Predict transmission line of electricity running status.
6. transmission line of electricity running status association rule mining and prediction side as claimed in claim 5 based on Bayesian model
Method, it is characterised in that the neutral net is wavelet neural network.
7. transmission line of electricity running status association rule mining and prediction side as claimed in claim 1 based on Bayesian model
Method, it is characterised in that project in the item collection I related to transmission line of electricity running status include environment temperature, line load,
Some items in wind speed, strain section circuit mean temperature, strain section ruling span sag and strain section tension force.
8. transmission line of electricity running status association rule mining and prediction side as claimed in claim 7 based on Bayesian model
Method, it is characterised in that the project in the item collection I related to transmission line of electricity running status also includes environment temperature, circuit and born
The change of some in lotus, wind speed, strain section circuit mean temperature, strain section ruling span sag and strain section tension force
Rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610866011.7A CN107247995A (en) | 2016-09-29 | 2016-09-29 | Transmission line of electricity running status association rule mining and Forecasting Methodology based on Bayesian model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610866011.7A CN107247995A (en) | 2016-09-29 | 2016-09-29 | Transmission line of electricity running status association rule mining and Forecasting Methodology based on Bayesian model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107247995A true CN107247995A (en) | 2017-10-13 |
Family
ID=60016139
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610866011.7A Pending CN107247995A (en) | 2016-09-29 | 2016-09-29 | Transmission line of electricity running status association rule mining and Forecasting Methodology based on Bayesian model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107247995A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108647808A (en) * | 2018-04-11 | 2018-10-12 | 济南大学 | A kind of manufacturing parameter Optimization Prediction method, apparatus, equipment and storage medium |
| CN108876138A (en) * | 2018-06-12 | 2018-11-23 | 国网天津市电力公司 | A kind of supervisory control of substation information leakage prison detection method based on big data analysis |
| CN109753519A (en) * | 2018-12-29 | 2019-05-14 | 成都信息工程大学 | A kind of Meteorological Services service discovering method excavated based on strong and weak dependency rule |
| CN111156946A (en) * | 2020-03-26 | 2020-05-15 | 郑州工程技术学院 | Accurate measuring device for cable sag |
| CN111859301A (en) * | 2020-07-23 | 2020-10-30 | 广西大学 | Data reliability evaluation method based on improved Apriori algorithm and Bayesian network inference |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102609528A (en) * | 2012-02-14 | 2012-07-25 | 云南大学 | Frequent mode association sorting method based on probabilistic graphical model |
| CN103391185A (en) * | 2013-08-12 | 2013-11-13 | 北京泰乐德信息技术有限公司 | Cloud security storage and processing method and system for rail transit monitoring data |
| CN103592587A (en) * | 2013-12-02 | 2014-02-19 | 国家电网公司 | Partial discharge diagnosis method based on data mining |
| CN104239437A (en) * | 2014-08-28 | 2014-12-24 | 国家电网公司 | Power-network-dispatching-oriented intelligent warning analysis method |
| CN105740245A (en) * | 2014-12-08 | 2016-07-06 | 北京邮电大学 | Frequent item set mining method |
| CN105843210A (en) * | 2016-03-22 | 2016-08-10 | 清华大学 | Power transformer defect information data mining method |
-
2016
- 2016-09-29 CN CN201610866011.7A patent/CN107247995A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102609528A (en) * | 2012-02-14 | 2012-07-25 | 云南大学 | Frequent mode association sorting method based on probabilistic graphical model |
| CN103391185A (en) * | 2013-08-12 | 2013-11-13 | 北京泰乐德信息技术有限公司 | Cloud security storage and processing method and system for rail transit monitoring data |
| CN103592587A (en) * | 2013-12-02 | 2014-02-19 | 国家电网公司 | Partial discharge diagnosis method based on data mining |
| CN104239437A (en) * | 2014-08-28 | 2014-12-24 | 国家电网公司 | Power-network-dispatching-oriented intelligent warning analysis method |
| CN105740245A (en) * | 2014-12-08 | 2016-07-06 | 北京邮电大学 | Frequent item set mining method |
| CN105843210A (en) * | 2016-03-22 | 2016-08-10 | 清华大学 | Power transformer defect information data mining method |
Non-Patent Citations (1)
| Title |
|---|
| GEHAO SHENG 等: "A Novel Association Rule Mining Method of Big Data for Power Transformers State Parameters Based on Probabilistic Graph Model", 《IEEE TRANSACTIONS ON SMART GRID》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108647808A (en) * | 2018-04-11 | 2018-10-12 | 济南大学 | A kind of manufacturing parameter Optimization Prediction method, apparatus, equipment and storage medium |
| CN108647808B (en) * | 2018-04-11 | 2022-03-29 | 济南大学 | Production parameter optimization prediction method, device, equipment and storage medium |
| CN108876138A (en) * | 2018-06-12 | 2018-11-23 | 国网天津市电力公司 | A kind of supervisory control of substation information leakage prison detection method based on big data analysis |
| CN109753519A (en) * | 2018-12-29 | 2019-05-14 | 成都信息工程大学 | A kind of Meteorological Services service discovering method excavated based on strong and weak dependency rule |
| CN111156946A (en) * | 2020-03-26 | 2020-05-15 | 郑州工程技术学院 | Accurate measuring device for cable sag |
| CN111859301A (en) * | 2020-07-23 | 2020-10-30 | 广西大学 | Data reliability evaluation method based on improved Apriori algorithm and Bayesian network inference |
| CN111859301B (en) * | 2020-07-23 | 2024-02-02 | 广西大学 | Data reliability evaluation method based on improved Apriori algorithm and Bayesian network reasoning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107247995A (en) | Transmission line of electricity running status association rule mining and Forecasting Methodology based on Bayesian model | |
| CN106649479B (en) | Transformer state association rule mining method based on probability graph | |
| CN110619420A (en) | Attention-GRU-based short-term residential load prediction method | |
| Wu et al. | Small perturbations are enough: Adversarial attacks on time series prediction | |
| CN103778569A (en) | Distributed generation island detection method based on meta learning | |
| CN114547974A (en) | Dynamic soft measurement modeling method based on input variable selection and LSTM neural network | |
| Sundararajan et al. | Regression and generalized additive model to enhance the performance of photovoltaic power ensemble predictors | |
| CN105825430A (en) | Heterogeneous social network-based detection method | |
| Qiu et al. | Fuzzy time series forecasting model based on automatic clustering techniques and generalized fuzzy logical relationship | |
| Jaiswal et al. | Distributed fog computing architecture for real-time anomaly detection in smart meter data | |
| Kumar et al. | Cloud-based electricity consumption analysis using neural network | |
| Wan et al. | Hydrological big data prediction based on similarity search and improved BP neural network | |
| Vijayan | Energy consumption prediction in low energy buildings using machine learning and artificial intelligence for energy efficiency | |
| Dhanya et al. | Data mining for evolving fuzzy association rules for predicting monsoon rainfall of India | |
| Dai et al. | Nonlinear inversion for electrical resistivity tomography based on chaotic DE-BP algorithm | |
| CN110349050B (en) | Intelligent electricity stealing criterion method and device based on power grid parameter key feature extraction | |
| Pürlü et al. | Estimating the distributed generation unit sizingand itseffects on the distribution system by using machine learning methods | |
| Zhao et al. | Optimisation algorithm for decision trees and the prediction of horizon displacement of landslides monitoring | |
| Gao et al. | Personalized context-aware collaborative filtering based on neural network and slope one | |
| WO2009082042A1 (en) | Nonlinear time series prediction method for weighted average defuzzification based on newfm | |
| Chen et al. | [Retracted] Prediction of Concrete Compressive Strength Based on the BP Neural Network Optimized by Random Forest and ISSA | |
| Peng et al. | Meteorological satellite operation prediction using a BiLSTM deep learning model | |
| Niu et al. | PSO-Markov residual correction method based on Verhulst-Fourier prediction model | |
| Cao et al. | Design of river water quality assessment and prediction algorithm | |
| Chu | [Retracted] Design and Application of BP Neural Network Optimization Method Based on SIWSPSO Algorithm |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20171013 |
|
| WD01 | Invention patent application deemed withdrawn after publication |