CN109473985A - One kind being based on BP neural network smart grid distribution method - Google Patents
One kind being based on BP neural network smart grid distribution method Download PDFInfo
- Publication number
- CN109473985A CN109473985A CN201910039304.1A CN201910039304A CN109473985A CN 109473985 A CN109473985 A CN 109473985A CN 201910039304 A CN201910039304 A CN 201910039304A CN 109473985 A CN109473985 A CN 109473985A
- Authority
- CN
- China
- Prior art keywords
- load
- data
- neural network
- carried out
- distribution method
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
-
- 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
- H02J3/003—Load forecast, e.g. methods or systems for forecasting future load demand
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Power Engineering (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The present invention relates to distribution correlative technology fields, specifically, being a kind of based on BP neural network smart grid distribution method, comprising the following steps: 1) pre-process to collected data;2) feature selecting is carried out to electricity consumption data concentrated load feature vector, reduces the complexity of redundancy and characteristic model, improves the operation efficiency of algorithm;3) load prediction is carried out, guarantees the reliability of power supply, improves power quality;4) load smart allocation is carried out.Feature selecting should be carried out to electricity consumption data concentrated load feature vector based on BP neural network smart grid distribution method, and reduce the complexity of redundancy and characteristic model, improve the operation efficiency of algorithm, and, load prediction is carried out, guarantees the reliability of power supply, improves power quality.
Description
Technical field
The present invention relates to distribution correlative technology fields, specifically, being a kind of based on BP neural network smart grid distribution
Method.
Background technique
Distribution is the link for being directly connected and distributing to user electric energy with user in the power system.Distribution system is by distribution
Electric substation, high-tension distributing line, distribution transformer, low-voltage distributing line and corresponding control protection equipment composition.
But in existing distribution method, the complexity of redundancy and characteristic model is excessively high, mentions the operation efficiency of algorithm
It is low, cause power quality low.
Summary of the invention
The purpose of the present invention is to provide one kind to be based on BP neural network smart grid distribution method, to solve existing distribution
In method, the complexity of redundancy and characteristic model is excessively high, and the operation efficiency for mentioning algorithm is low, leads to the problem that power quality is low.
To achieve the above object, the present invention provides the following technical solutions a kind of based on BP neural network smart grid distribution side
Method, comprising the following steps: 1) collected data are pre-processed;2) electricity consumption data concentrated load feature vector is carried out special
Sign selection, reduces the complexity of redundancy and characteristic model, improves the operation efficiency of algorithm;3) load prediction is carried out, guarantees to supply
The reliability of electricity improves power quality;4) load smart allocation is carried out.
Preferably, the data processing in the step 1), since electricity consumption data will not mutate immediately in a short time
Between continuity on axis can carry out lateral comparison to load data therefore when screening to load data, in addition it can
The previous day, load data a few days ago in the same time is used to be compared in conjunction with longitudinal comparison.
Preferably, the data processing in the step 1) comes from different sensors due to constituting feature vector in data set,
It also needs there are dimension difference and data value range are different, therefore after carrying out data screening by characteristic vector data collection normalizing
Change.
Preferably, the feature selecting in the step 2, if born all since load characteristic vector data is excessive
Lotus characteristic is analyzed, and " dimension disaster " inherently occurs, because it is necessary to carry out feature selecting.
Preferably, the load prediction in the step 3), the feature obtained after feature selecting such as average load, same
Several load values etc. that working day adjoins can be used as the input of BP neural network, export as the load prediction at one day corresponding moment
Value.
Compared with prior art, the invention has the following advantages: BP neural network smart grid distribution side should be based on
Method carries out feature selecting to electricity consumption data concentrated load feature vector, reduces the complexity of redundancy and characteristic model, improves and calculates
The operation efficiency of method, also, load prediction is carried out, guarantee the reliability of power supply, improves power quality.
Detailed description of the invention
Fig. 1 is flow chart of the invention.
Specific embodiment
Below in conjunction with the embodiment of the present invention and attached drawing, the technical solution in the present invention is described in further detail,
The embodiment is only for explaining the present invention, does not constitute and limits to protection scope of the present invention.
The present invention provides a kind of technical solution: one kind is based on BP neural network smart grid distribution method, including following step
It is rapid: 1) collected data to be pre-processed;2) feature selecting is carried out to electricity consumption data concentrated load feature vector, reduced superfluous
The complexity of remaining and characteristic model improves the operation efficiency of algorithm;3) load prediction is carried out, guarantees the reliability of power supply, mentions
High power quality;4) load smart allocation is carried out;
Data processing in step 1), due to the continuity that electricity consumption data will not mutate on i.e. time shaft in a short time,
Therefore when screening to load data, lateral comparison can be carried out to load data, in addition it can be in conjunction with longitudinal comparison
Load data using the previous day, a few days ago in the same time is compared, in this way in the laterally and longitudinally comparison of time, Ke Yigeng
Add and clearly data are analyzed, can find electrical problem in time, ensure that the electricity consumption reliability and electricity consumption of users
Quality;
Data processing in step 1), due to constitute data set in feature vector come from different sensors, there are dimension difference with
And data value range is different, therefore also needs to normalize characteristic vector data collection after carrying out data screening, after screening,
Again by characteristic vector data centralized planning, on the one hand have the advantages that big data era planning application, on the other hand can adopt
The data of collection deviate big data, when something goes wrong, find in time;
Feature selecting in step 2, if divided all load profiles since load characteristic vector data is excessive
Inherently " dimension disaster " occurs for analysis, because it is necessary to carry out feature selecting, exhaustive analysis electricity consumption data reduces redundancy and spy
The complexity for levying model, improves the operation efficiency of algorithm;
Load prediction in step 3), the feature obtained after feature selecting such as average load, same working day are adjoined several
A load value etc. can be used as the input of BP neural network, export as the predicted load at one day corresponding moment, then to predicted value
It is compared with actual value, when deviation is larger, can find that in time user data goes wrong, problem can be made rapidly
Reaction and countermeasure, to ensure the normal electricity consumption of users.
Basic principles and main features and advantage of the invention have been shown and described above.The technical staff of the industry should
Understand, the present invention is not limited to the above embodiments, and the above embodiments and description only describe originals of the invention
Reason, without departing from the spirit and scope of the present invention, various changes and improvements may be made to the invention, these changes and improvements
It all fall within the protetion scope of the claimed invention.The claimed scope of the invention is by appended claims and its equivalent circle
It is fixed.
Claims (5)
1. one kind is based on BP neural network smart grid distribution method, comprising the following steps:
1) collected data are pre-processed;
2) feature selecting is carried out to electricity consumption data concentrated load feature vector, reduces the complexity of redundancy and characteristic model, mentions
The operation efficiency of high algorithm;
3) load prediction is carried out, guarantees the reliability of power supply, improves power quality;
4) load smart allocation is carried out.
2. according to claim 1 be based on BP neural network smart grid distribution method, it is characterised in that the step 1)
In data processing, due to the continuity that electricity consumption data will not mutate on i.e. time shaft in a short time, to negative
When lotus data are screened, lateral comparison can be carried out to load data, on the day before being used in addition it can combination longitudinal comparison,
A few days ago load data in the same time is compared.
3. according to claim 1 be based on BP neural network smart grid distribution method, it is characterised in that the step 1)
In data processing, due to constitute data set in feature vector come from different sensors, there are dimension difference and data value models
Enclose it is different, therefore carry out data screening after also need to normalize characteristic vector data collection.
4. according to claim 1 be based on BP neural network smart grid distribution method, it is characterised in that the step 2
In feature selecting, if all load profiles analyzed since load characteristic vector data is excessive, inherently
Occur " dimension disaster ", because it is necessary to carry out feature selecting.
5. according to claim 1 be based on BP neural network smart grid distribution method, it is characterised in that the step 3)
In load prediction, several load values etc. that the feature obtained after feature selecting such as average load, same working day are adjoined
The input that can be used as BP neural network exports as the predicted load at one day corresponding moment.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910039304.1A CN109473985A (en) | 2019-01-16 | 2019-01-16 | One kind being based on BP neural network smart grid distribution method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910039304.1A CN109473985A (en) | 2019-01-16 | 2019-01-16 | One kind being based on BP neural network smart grid distribution method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109473985A true CN109473985A (en) | 2019-03-15 |
Family
ID=65678735
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910039304.1A Pending CN109473985A (en) | 2019-01-16 | 2019-01-16 | One kind being based on BP neural network smart grid distribution method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109473985A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110880152A (en) * | 2019-12-20 | 2020-03-13 | 国网湖北省电力公司咸宁供电公司 | Power supply method for intelligent power distribution network |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1478075A2 (en) * | 2003-05-13 | 2004-11-17 | Siemens Power Transmission & Distribution, Inc. | Very short term load predictor |
CN102722759A (en) * | 2012-05-17 | 2012-10-10 | 河海大学 | Method for predicting power supply reliability of power grid based on BP neural network |
CN103295081A (en) * | 2013-07-02 | 2013-09-11 | 上海电机学院 | Electrical power system load prediction method based on back propagation (BP) neural network |
CN108320046A (en) * | 2017-12-27 | 2018-07-24 | 安徽机电职业技术学院 | Short-term electric load prediction modeling method |
-
2019
- 2019-01-16 CN CN201910039304.1A patent/CN109473985A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1478075A2 (en) * | 2003-05-13 | 2004-11-17 | Siemens Power Transmission & Distribution, Inc. | Very short term load predictor |
CN102722759A (en) * | 2012-05-17 | 2012-10-10 | 河海大学 | Method for predicting power supply reliability of power grid based on BP neural network |
CN103295081A (en) * | 2013-07-02 | 2013-09-11 | 上海电机学院 | Electrical power system load prediction method based on back propagation (BP) neural network |
CN108320046A (en) * | 2017-12-27 | 2018-07-24 | 安徽机电职业技术学院 | Short-term electric load prediction modeling method |
Non-Patent Citations (3)
Title |
---|
杨姚: ""智能小区用户侧智能配电策略"", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
陈宇杰 等: ""基于BP神经网络的区域配电网中期电力负荷预测"", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
鲍正江 等: ""一种基于神经网络的电力负荷预测方法"", 《浙江电力》 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110880152A (en) * | 2019-12-20 | 2020-03-13 | 国网湖北省电力公司咸宁供电公司 | Power supply method for intelligent power distribution network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101551446B (en) | Energy feedback power source aging system | |
CN101976827A (en) | Method for identifying candidate prevention and control measure of self-adaptive external-environmental electric power system | |
CN113191575B (en) | Optimization method, device, terminal and storage medium for power distribution network maintenance power failure mode | |
CN114977202B (en) | Intelligent kitchen electrical equipment network distribution method and system | |
CN110435485A (en) | A kind of progress control method, device and medium applied to inversion feedback device | |
CN109473985A (en) | One kind being based on BP neural network smart grid distribution method | |
US11791631B2 (en) | Dynamic computation and control of distributed assets at the edge of a power grid | |
CN109615117A (en) | A kind of coal based on ARIMA model changes electric Methods of electric load forecasting | |
CN116029491B (en) | Power dispatching management system and control method thereof | |
CN111130116A (en) | Scheduling operation power flow checking method based on key topology change item identification | |
CN110649712A (en) | Urban power grid operation visual monitoring system based on electricity consumption information acquisition system | |
CN102890450A (en) | On-line transformer economic operation optimization platform | |
CN116340323A (en) | Electric power working room message collecting device and collecting method | |
CN110687379A (en) | Topology-configurable non-invasive building electrical equipment monitoring and analyzing system | |
KR20200048744A (en) | Methdo and system for controlling distribution network including numerous distributed energy resource | |
CN106887848B (en) | Voltage power-less real-time control method based on Fuzzy Pattern Recognition | |
CN112202845B (en) | Distribution electricity service oriented edge computing gateway load system, analysis method and distribution system thereof | |
CN104901315B (en) | Power grid unloading method and system thereof | |
CN112560210A (en) | Method for adjusting a power grid structure, associated device and computer program product | |
CN108242800B (en) | A kind of differential protecting method and device of layer-specific access formula extra-high voltage direct-current system | |
CN111342502A (en) | Forsmark effect suppression method and device based on bidirectional converter | |
CN110676939A (en) | Metering and fault protection boundary switch controller for intelligent power distribution network | |
CN114548485B (en) | Big data application system for power production | |
CN117650628B (en) | Energy efficiency management system based on self-adaptive preconditioning scene | |
CN108123436A (en) | Voltage out-of-limit prediction model based on principal component analysis and multivariate regression 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: 20190315 |
|
RJ01 | Rejection of invention patent application after publication |