CN107919664A - A kind of feature tag with Running State defines method - Google Patents
A kind of feature tag with Running State defines method Download PDFInfo
- Publication number
- CN107919664A CN107919664A CN201711088107.6A CN201711088107A CN107919664A CN 107919664 A CN107919664 A CN 107919664A CN 201711088107 A CN201711088107 A CN 201711088107A CN 107919664 A CN107919664 A CN 107919664A
- Authority
- CN
- China
- Prior art keywords
- cluster
- mrow
- msub
- running state
- subset
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E60/00—Enabling technologies; Technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Power Engineering (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present invention relates to a kind of feature tag with Running State to define method, including:Key parameter of the reflection with Running State is collected, carries out the Pre-Evaluation of operating status, low volume data is chosen and defines label value, as subset;The clustering to all power distribution stations characterized by reflecting the key parameter with Running State;According to subset, cluster result is assessed using entropy, selects optimal clustering cluster;According to clustering cluster and subset, to no label platform area definition label value;For the taiwan area that can not be delimited, new subset is found according to existing cluster.The present invention is by having collected key parameter of the reflection with Running State, a small amount of taiwan area is marked, as subset data, taiwan area is clustered using clustering algorithm, cluster result is weighed using subset afterwards, feature tag definition is carried out to unmarked taiwan area, greatly improves the efficiency that feature tag defines.
Description
Technical field
The present invention relates to power equipment assessment technique field, and in particular to a kind of feature tag definition with Running State
Method.
Background technology
With rapid economic development, urbanization process accelerates, and people's living standard is continuously improved, production and living electricity consumption
Demand constantly expands.Distribution network is the key link of user power utilization, is played an important role in power grid.With distribution network planning
Mould is increasing, and user is also higher and higher for the reliability requirement of power supply, fortune of the Distribution Network Equipment as supplying power allocation service
The crucial composition of battalion, its operation conditions just seem especially important.The operating status of distribution is assessed by the way of artificial, no
Substantial amounts of man power and material is only needed, and when to being assessed with Running State, there are efficiency is low and real time problems.
The content of the invention
It is an object of the invention to provide a kind of feature tag with Running State to define method, to the operation shape of distribution
State is assessed, and greatly improves the efficiency and real-time of assessment, is found distribution operation problem in time, is taken appropriate measures,
To ensureing that distribution stable operation is of great significance.
To achieve the above object, present invention employs following technical scheme:
A kind of feature tag with Running State defines method, includes the following steps:
(1) key parameter of the reflection with Running State is collected, carries out the Pre-Evaluation of operating status, low volume data is chosen and determines
Adopted label value, as subset;
(2) the clustering to all power distribution stations characterized by reflecting the key parameter with Running State;
(3) according to subset, cluster result is assessed using entropy, selects optimal clustering cluster;
(4) according to clustering cluster and subset, to no label platform area definition label value;
(5) for the taiwan area that can not be delimited, new subset is found according to existing cluster, redirects and performs step (2), Zhi Daosuo
There is taiwan area to have label value.
In such scheme, in step (1), the key parameter includes taiwan area scale, number of users, the time limit, rate of load condensate, unusual fluctuation
Rate, three-phase imbalance and heavy-overload situation.
In such scheme, in step (2), to all distribution characterized by the key parameter for matching somebody with somebody Running State by reflection
Taiwan area cluster, and specifically includes following steps:
(21) cluster centre is found for point to be clustered;
(22) each point is calculated to the distance of cluster centre, and by each point cluster into the cluster nearest from the point;
(23) distance average of all the points in each cluster is calculated, and using the average value as new cluster centre;
(24) step (22), (23) are performed repeatedly, until cluster centre is no longer moved or clustered on a large scale number
Untill reaching requirement.
It is described according to subset in step (3) in such scheme, cluster result is assessed using entropy, selection is most
Good clustering cluster, specifically includes following steps:
(31) probability that each clustering cluster includes seed words is calculated, calculation formula is as follows:
Wherein, pijRepresent that for occurring classification in cluster i be that label is k seed probability, miRepresent for the taiwan area that cluster i has
Number, mikRepresent that in cluster i be k seed words numbers comprising class label;
(32) entropy for clustering each cluster is calculated:
Wherein, SiRepresent total entropy of cluster i;
(33) total entropy of cluster result is calculated:
Wherein, S represents the entropy according to seed words, and m represents to participate in all taiwan areas sum of training, and L represents the cluster of cluster
Value.
(34) cluster result for selecting entropy maximum is optimal clustering cluster.
As shown from the above technical solution, the present invention defines method based on the feature tag with Running State, by having received
Key parameter of the collection reflection with Running State, is marked a small amount of taiwan area, as subset data, using clustering algorithm pair
Taiwan area is clustered, and cluster result is weighed using subset afterwards, carries out feature tag definition to unmarked taiwan area, greatly
The big efficiency for improving feature tag and defining.
Brief description of the drawings
Fig. 1 is flow chart of the method for the present invention;
Fig. 2 is the method flow diagram of the specific embodiment of the invention.
Embodiment
The present invention will be further described below in conjunction with the accompanying drawings:
A kind of feature tag definition being directed to Running State, comprises the following steps:
S1:Key parameter of the reflection with Running State is collected, carries out the Pre-Evaluation of operating status, low volume data is chosen and determines
Adopted label value, as subset, it is uneven which includes taiwan area scale, number of users, the time limit, rate of load condensate, unusual fluctuation rate, three-phase
Weighing apparatus, heavy-overload situation;Feature is pre-processed, carries out feature selecting.A small amount of data are chosen, is defined and marked according to characteristic value
Label, form subset.
The tag definition of the subset is
C=(C1,C2,…Ck) (1)
Wherein, C represents it is classification, and k represents the species of classification.
S2:Taiwan area clusters, the gathering to all power distribution stations characterized by reflecting the key parameter with Running State
Class:The step uses K-Means clustering algorithms, which has large data sets higher efficiency and be scalability, specifically
Step is:
S21:Cluster centre is found for point to be clustered;
S22:Each point is calculated to the distance of cluster centre, by each point cluster into the cluster nearest from the point;
S23:The distance average of all the points in each cluster is calculated, and using this average value as new cluster centre;
S24:Step S22 and S23 are performed repeatedly, are reached until cluster centre is no longer moved or clustered on a large scale number
Untill requiring.
S3:According to subset, cluster result is assessed using entropy, selects optimal clustering cluster, specific calculating process
It is as follows:
S31:The probability that each clustering cluster includes seed words is calculated, calculation formula is as follows:
Wherein, pijRepresent that for occurring classification in cluster i be that label is k seed probability, seed words miRepresent the platform that cluster i has
The number in area, mikRepresent that in cluster i be k seed words numbers comprising class label.
S32:Calculate the entropy for clustering each cluster:
Wherein, SiRepresent total entropy of cluster i;
S33:Calculate total entropy of cluster result:
Wherein, S represents the entropy according to seed words, and m is all taiwan areas sum for participating in training, and L is the cluster value of cluster.
S34:The cluster result for selecting entropy maximum is optimal clustering cluster.
S4:According to clustering cluster and subset, to no label platform area definition label value, training obtains the cluster knot of entropy maximum
Fruit, to each cluster, selection includes the most classification of seed words, and as the class label of the cluster, signature is carried out to taiwan area.
Ci=max (ni1,ni2…nik) (5)
Wherein, nkIt is number of the k seed words in cluster i to represent class label, chooses the value of maximum as the label value for changing cluster.
S5:For the taiwan area that can not be delimited, new subset, jump procedure S1, until all are found according to existing cluster
There is label value in area.For not including seed words cluster in clustering cluster, the point near such central point is chosen as new subset, and
Assign new feature tag, repeat step S2, S3, S4, until all taiwan areas have feature tag.
Cluster result is weighed using subset, feature tag definition is carried out to unmarked taiwan area, is greatly improved
The efficiency that feature tag defines.Timely to be paid close attention to and be overhauled for the abnormal taiwan area of mark, have found that it is likely that deposit in time
The defects of, improve with network operation stability.
Embodiment described above is only that the preferred embodiment of the present invention is described, not to the model of the present invention
Enclose and be defined, on the premise of design spirit of the present invention is not departed from, technical side of the those of ordinary skill in the art to the present invention
The various modifications and improvement that case is made, should all fall into the protection domain that claims of the present invention determines.
Claims (4)
1. a kind of feature tag with Running State defines method, it is characterised in that includes the following steps:
(1) key parameter of the reflection with Running State is collected, carries out the Pre-Evaluation of operating status, chooses low volume data definition mark
Label value, as subset;
(2) the clustering to all power distribution stations characterized by reflecting the key parameter with Running State;
(3) according to subset, cluster result is assessed using entropy, selects optimal clustering cluster;
(4) according to clustering cluster and subset, to no label platform area definition label value;
(5) for the taiwan area that can not be delimited, new subset is found according to existing cluster, redirects and performs step (2), until all
There is label value in area.
2. the feature tag according to claim 1 with Running State defines method, it is characterised in that:In step (1),
The key parameter includes taiwan area scale, number of users, the time limit, rate of load condensate, unusual fluctuation rate, three-phase imbalance and heavy-overload situation.
3. the feature tag according to claim 1 with Running State defines method, it is characterised in that:In step (2),
The clustering to all power distribution stations characterized by reflecting the key parameter with Running State, specifically includes following step
Suddenly:
(21) cluster centre is found for point to be clustered;
(22) each point is calculated to the distance of cluster centre, and by each point cluster into the cluster nearest from the point;
(23) distance average of all the points in each cluster is calculated, and using the average value as new cluster centre;
(24) step (22), (23) are performed repeatedly, are reached until cluster centre is no longer moved or clustered on a large scale number
Untill it is required that.
4. the feature tag according to claim 1 with Running State defines method, it is characterised in that:In step (3),
It is described according to subset, cluster result is assessed using entropy, optimal clustering cluster is selected, specifically includes following steps:
(31) probability that each clustering cluster includes seed words is calculated, calculation formula is as follows:
<mrow>
<msub>
<mi>p</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>=</mo>
<mfrac>
<msub>
<mi>m</mi>
<mrow>
<mi>i</mi>
<mi>k</mi>
</mrow>
</msub>
<msub>
<mi>m</mi>
<mi>i</mi>
</msub>
</mfrac>
</mrow>
Wherein, pijRepresent that for occurring classification in cluster i be that label is k seed probability, miRepresent the number for the taiwan area that cluster i has,
mikRepresent that in cluster i be k seed words numbers comprising class label;
(32) entropy for clustering each cluster is calculated:
<mrow>
<msub>
<mi>S</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>k</mi>
</munderover>
<msub>
<mi>p</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>&times;</mo>
<mi>l</mi>
<mi>o</mi>
<mi>g</mi>
<mn>2</mn>
<mrow>
<mo>(</mo>
<msub>
<mi>p</mi>
<mrow>
<mi>i</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
</mrow>
Wherein, SiRepresent total entropy of cluster i;
(33) total entropy of cluster result is calculated:
<mrow>
<mi>S</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>L</mi>
</munderover>
<mfrac>
<msub>
<mi>m</mi>
<mi>i</mi>
</msub>
<mi>m</mi>
</mfrac>
<msub>
<mi>S</mi>
<mi>i</mi>
</msub>
</mrow>
Wherein, S represents the entropy according to seed words, and m represents to participate in all taiwan areas sum of training, and L represents the cluster value of cluster.
(34) cluster result for selecting entropy maximum is optimal clustering cluster.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711088107.6A CN107919664A (en) | 2017-11-07 | 2017-11-07 | A kind of feature tag with Running State defines method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711088107.6A CN107919664A (en) | 2017-11-07 | 2017-11-07 | A kind of feature tag with Running State defines method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107919664A true CN107919664A (en) | 2018-04-17 |
Family
ID=61895260
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711088107.6A Pending CN107919664A (en) | 2017-11-07 | 2017-11-07 | A kind of feature tag with Running State defines method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107919664A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103107535A (en) * | 2013-01-17 | 2013-05-15 | 中国电力科学研究院 | Comprehensive evaluation method on safety of grid structure based on entropy weight method |
CN105022021A (en) * | 2015-07-08 | 2015-11-04 | 国家电网公司 | State discrimination method for gateway electrical energy metering device based on the multiple agents |
CN105574165A (en) * | 2015-12-17 | 2016-05-11 | 国家电网公司 | Power grid operation monitoring information identification and classification method based on clustering |
CN105866725A (en) * | 2016-04-20 | 2016-08-17 | 国网上海市电力公司 | Method for fault classification of smart electric meter based on cluster analysis and cloud model |
CN106356906A (en) * | 2016-10-13 | 2017-01-25 | 国网山东省电力公司威海供电公司 | Energy conservation control method and system based on microgrid |
-
2017
- 2017-11-07 CN CN201711088107.6A patent/CN107919664A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103107535A (en) * | 2013-01-17 | 2013-05-15 | 中国电力科学研究院 | Comprehensive evaluation method on safety of grid structure based on entropy weight method |
CN105022021A (en) * | 2015-07-08 | 2015-11-04 | 国家电网公司 | State discrimination method for gateway electrical energy metering device based on the multiple agents |
CN105574165A (en) * | 2015-12-17 | 2016-05-11 | 国家电网公司 | Power grid operation monitoring information identification and classification method based on clustering |
CN105866725A (en) * | 2016-04-20 | 2016-08-17 | 国网上海市电力公司 | Method for fault classification of smart electric meter based on cluster analysis and cloud model |
CN106356906A (en) * | 2016-10-13 | 2017-01-25 | 国网山东省电力公司威海供电公司 | Energy conservation control method and system based on microgrid |
Non-Patent Citations (1)
Title |
---|
余振华等: "基于聚类的电网运行监控信息辨识分类方法研究", 《湖北电力》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102629106B (en) | Water supply control method and water supply control system | |
CN109461025A (en) | A kind of electric energy substitution potential customers' prediction technique based on machine learning | |
CN109508823B (en) | Scene analysis method based distributed power supply planning method in power distribution network | |
CN103136585B (en) | Weighting Voronoi diagram substation planning method based on chaotic and genetic strategy | |
CN109634121A (en) | More parent genetic algorithm air source heat pump multiobjective optimization control methods based on radial basis function neural network | |
CN106487005A (en) | A kind of Electric power network planning method considering T-D tariff | |
CN109980700A (en) | A kind of distributed generation resource multi-objection optimization planning method, apparatus and equipment | |
CN109784589B (en) | Power grid business expansion user access optimization method, device and equipment | |
CN108573326A (en) | Video server site selecting method based on genetic-ant colony hybrid algorithm | |
CN110619351A (en) | Vegetable and bird stager site selection method based on improved k-means algorithm | |
CN105225021B (en) | The optimum choice method of power distribution network project yet to be built | |
CN110854891B (en) | Power distribution network pre-disaster resource allocation method and system | |
CN107944747A (en) | A kind of low-voltage platform area evaluation method based on improvement principal component analysis | |
CN109492874A (en) | A kind of decision-making technique of three levels power distribution network investment decision system | |
CN105488589A (en) | Genetic simulated annealing algorithm based power grid line loss management evaluation method | |
CN108197805A (en) | A kind of electric energy and natural gas competitiveness integrated evaluating method that planning guidance is substituted available for electric energy | |
CN109003010A (en) | Evaluation method for investment benefit of power grid project | |
CN109214546A (en) | A kind of Power Short-Term Load Forecasting method based on improved HS-NARX neural network | |
CN113450031A (en) | Method and device for selecting intelligent energy consumption service potential transformer area of residents | |
CN107748940A (en) | A kind of energy conservation potential Quantitative prediction methods | |
Chen et al. | Many-objective optimal power dispatch strategy incorporating temporal and spatial distribution control of multiple air pollutants | |
CN108122173A (en) | A kind of conglomerate load forecasting method based on depth belief network | |
Xia et al. | Improving the performance of stock trend prediction by applying GA to feature selection | |
CN107220905A (en) | The distribution transformer running status method for prewarning risk that a kind of logic-based is returned | |
CN105701717B (en) | Power distribution network interaction scheme compilation method based on improved genetic 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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180417 |
|
RJ01 | Rejection of invention patent application after publication |