CN104992238B - A kind of Methods of electric load forecasting based on typical daily load characteristic - Google Patents
A kind of Methods of electric load forecasting based on typical daily load characteristic Download PDFInfo
- Publication number
- CN104992238B CN104992238B CN201510354856.3A CN201510354856A CN104992238B CN 104992238 B CN104992238 B CN 104992238B CN 201510354856 A CN201510354856 A CN 201510354856A CN 104992238 B CN104992238 B CN 104992238B
- Authority
- CN
- China
- Prior art keywords
- user
- load
- data
- curve
- industry
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 36
- 230000005611 electricity Effects 0.000 claims abstract description 16
- 230000008859 change Effects 0.000 claims abstract description 9
- 238000005457 optimization Methods 0.000 claims description 11
- 230000009977 dual effect Effects 0.000 claims description 7
- 238000012706 support-vector machine Methods 0.000 claims description 7
- 238000012549 training Methods 0.000 claims description 5
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 abstract description 3
- 230000002159 abnormal effect Effects 0.000 abstract description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000015556 catabolic process Effects 0.000 abstract description 2
- 238000004458 analytical method Methods 0.000 description 10
- 230000002354 daily effect Effects 0.000 description 8
- 241001269238 Data Species 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 239000011435 rock Substances 0.000 description 3
- 235000001674 Agaricus brunnescens Nutrition 0.000 description 2
- 230000003466 anti-cipated effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 241000894006 Bacteria Species 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- 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
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention discloses a kind of Methods of electric load forecasting based on typical daily load characteristic, including：The Power system load data of each user in setting regions setting industry is gathered, and data are pre-processed；Obtain the typical day load curve and industry integral load curve of each user；The coefficient correlation between each user's typical day load curve is asked for respectively；It is classified according to the electricity consumption rule of each user；Electric load variation tendency and industry integral load change uniformity highest user are chosen, the overall Power system load data of industry is predicted by the electricity consumption Behavior law of the user.Beneficial effect of the present invention：Can realize from the exception of certain custom power load can derive the abnormal prediction of exception even whole society's electric load of certain specific industry, the accuracy of load forecast result is improved, the powerful guarantee for carry out reasonable economic load dispatching for power system, reducing production cost, prevent electric grid large area power cut or mains breakdown from providing.
Description
Technical field
The present invention relates to electric load analysis technical field, especially a kind of electric load based on typical daily load characteristic
Forecasting Methodology.
Background technology
With expanding economy, power consumption rises year by year, the diversification and complication of electric load, electric load is determined
Plan is theoretical, management means and technical support propose higher requirement, but is still suffered from current work many unreasonable, not to the utmost
Such as the problem of people's will：
1. the analysis of pair electric load lays particular emphasis on macroscopic perspective
Analysis of the country about electric load can only accomplish the ex-post analysis to Power system load data at present, can only be rough
Overall load rule change is held from macroscopic aspect, it is impossible to do specific point for certain industry, the part throttle characteristics of certain client
Analysis.
2. the intelligence degree of load forecast is inadequate
The prediction on electric load is generally using display data, form as Main Means at present, and staff often needs
To use working experience of itself etc. to carry out secondary operation to data, just can effectively be judged；Such prediction mode, one
Aspect, load forecast result and practical experience, the theoretical water equality of analysis personnel have very big relation；On the other hand, when
When data volume is larger, it can largely increase the amount of labour of prognosticator, and prediction result can not be effectively comprehensive
Covering state of market.
3. the early warning mechanism of electric load change is unsound
Electric load change early warning both domestic and external often lays particular emphasis on macroscopical, middle sight angle at present, becomes from economy change, industry
Change is angularly predicted, and, often misses optimal early warning opportunity, or even be only capable of being divided afterwards due to data acquisition not in time
Analysis, the suggesting effect of early warning can not be played.
The content of the invention
The purpose of the present invention is exactly to solve the above problems, there is provided a kind of load forecast based on typical daily load characteristic
The overall electric load variation tendency of industry can be more accurately predicted in method, this method by individual electricity consumption rule, for
The formulation of electricity plan provides reliable theoretical foundation.
To achieve the above object, the present invention uses following technical proposals, including：
A kind of Methods of electric load forecasting based on typical daily load characteristic, comprises the following steps：
(1) electric power of each user in setting regions setting industry is gathered by power regulation Central Grid EMS
Load data, deposit Power system load data storehouse, and data are pre-processed；
(2) selection of typical day is carried out using one-class support vector machines, and obtain the typical day load curve of each user with
And industry integral load curve；
(3) coefficient correlation between each user's typical day load curve is asked for respectively, the association each customer charge curve
Characteristic is analyzed；
(4) according to the coefficient correlation between each user, it is classified according to the electricity consumption rule of each user, it is determined that per a kind of
The load curve of user；
(5) in the load curve of every a kind of user, electric load variation tendency and industry integral load change one are chosen
Cause property highest user, the overall Power system load data of industry is predicted by the electricity consumption Behavior law of the user.
The method pre-processed in the step (1) to data includes：
Initial data 1-1) is read from Power system load data storehouse, the data are associated according to critical field, carried
Take required data, and authority data form；
1-2) screen and reject bad data, completion is carried out to the data of missing；
1-3) wrong data is repaired, data are smoothed.
The specific method of the selection of progress typical day is in the step (2)：
Electric load sample 2-1) is chosen, sample is clustered using support vector machine method, builds object function；
The dual problem of object function 2-2) is asked for using Lagrangian Arithmetic；
2-3) by QP optimization methods solution, this dual problem obtains optimization solution；
The center of circle c value 2-4) is obtained according to optimization solution, i.e., typical day data.
The step 2-1) in structure object function be specially：
Constraints：||xi-c||2≤R2+ξi；ξi>=0, i=1,2 ..., l；
Wherein, R is hyperspherical radius, ξiFor penalty, xiFor training sample, l is sample total, and c is hypersphere ball
The heart, coefficient ν are setting value.
The step 2-2) in ask for the dual problem of object function using Lagrangian Arithmetic and be specially：
Constraints：
Wherein, αi、αjRespectively Lagrangian, xi、xjRespectively training sample, i=1,2 ..., l, j=1,
2,…,l。
The step 2-4) specific method be：
The optimization is solved into α and brings formula into：
The value for trying to achieve center of circle C is typical day data.
The method analyzed in the step (3) the associate feature each customer charge curve is：
The coefficient correlation between n user's typical day load curve is asked for respectively, and the diagonal entry for being as a result stored in n*n is
In 1 symmetrical matrix；Off-diagonal element represents the coefficient correlation of two users' typical day load curve；
When the linear relationship of two user data strengthens, coefficient correlation tends to 1 or -1；When the increase of user data
When another user data also increases, coefficient correlation is more than 0；When a user data increases and another user data is reduced, phase
Relation number is less than 0；When two user data independence, coefficient correlation 0.
The specific method that the coefficient correlation between each user's typical day load curve is asked in the step (3) is：
Wherein, Xi、YiThe respectively Power system load data of two of which user；Respectively two of which user
The average value of Power system load data；N is Power system load data number, is positive integer.
The specific method of the step (4) is：
A certain user is chosen as user is referred to, the user that a certain setting value is more than with user's correlation coefficient value is playbacked
It is a kind of；In a kind of user of division, choose and be used as such with the load curve with reference to user correlation coefficient value highest user
The load curve of user.
The beneficial effects of the invention are as follows：
By the inventive method, the load variations rule and industry integral load changing rule that can find out which user have
There is higher uniformity, and then the load variations that industry entirety can be better anticipated by holding individual load variations rule become
Gesture.
The inventive method can be realized to custom power information on load from rough to the analysis to become more meticulous, realized from certain use
The exception of family electric load can derive the abnormal prediction of exception even whole society's electric load of certain specific industry, improve
The accuracy of load forecast result, the timely early warning of electric load can be accomplished, rationally economic adjust is carried out for power system
The powerful guarantee spending, reduce production cost, preventing electric grid large area power cut or mains breakdown from providing.
Brief description of the drawings
Fig. 1 is sample data pretreatment process figure of the present invention.
Embodiment
The present invention is further described below in conjunction with the accompanying drawings.
A kind of Methods of electric load forecasting based on typical daily load characteristic, comprises the following steps：
Step 1：Setting regions is gathered by power regulation Central Grid EMS and sets each user in industry
Power system load data, deposit Power system load data storehouse, and data are pre-processed；
Flow chart of data processing is lost as shown in figure 1, having partial data during information gathering communication storage at present
Situation occur, the data division do not lost causes the disorder of the data format in database there is also certain bad data,
So to carry out data prediction, data prediction, which is divided into following four step, to be carried out：
1. authority data form：
The data in database are imported into text document first, 96 point datas then read in text document are preserved into square
The form of battle array.The form of authority data, line number and its corresponding Customs Assigned Number are provided respectively, and the measurement of missing is mended 0.Need
It should be noted that the stage main purpose is authority data form, only the measurement of missing is temporarily replaced with 0, not
From substantial completion metric data.
2. reject bad data：
Main purpose is to screen and reject data corresponding to the too many user for being difficult to rationally repair of factor data missing, because
It is that almost no any measurement, the analysis to problem can not extend efficient help this user.96 point datas are converted first
For 24 point datas, to eliminate the influence of metric data missing to a certain extent.24 point datas are lacked still greater than 80% or counted
Rejected according to the too high user data of decentralization.
3. completion missing data：
For shortage of data lesser extent, most of missing is the user of in a few days data division missing, can be directly by inserting
The method of value carries out completion to data.If user data missing degree is heavier, or there is a situation where to lack all day, then carry out laterally
And longitudinal comparison, take the method for completion in proportion to carry out completion, i.e., the performance number for the data period having according to this day is with being somebody's turn to do
User corresponds to the ratio of period power average value, will in a few days lack part completion with reference to the average daily load curve of user.Completion lacks
The step of losing metric data is as follows：
1) all data are weighted with average, one typical day load curve for being used for completion data of selection；
2) according to typical day load curve, according to longitudinal comparison and lateral comparison, missing is measured into preliminary completion.
4. the bad measurement of amendment：
Metric data is handled using the permanent delay optimal smoothing in Kalman filtering, reduces bad measuring band
Influence.Permanent delay optimal smoothing is on the time point for lagging newest one Fixed Time Interval N of observation time, provides and is
A kind of method of system state optimization estimation.
Step 2：The selection of typical day is carried out using one-class support vector machines, and the typical daily load for obtaining each user is bent
Line and industry integral load curve；
The selection of typical day is carried out with branching algorithm one-class support vector machines, replaces hyperplane with hypersphere to divide number
According to the initial problem of object function is：
By setup parameter 0≤ν≤1, make hyperspherical radius and carried out between training sample number that it can be included
It is compromise；When ν is small, (v is the number between 0-1) is put data inside ball into as far as possible, when ν is big, compressed ball as far as possible
Size.Wherein, R represents hyperspherical radius, and hypersphere radius surface is smaller, and the extensive risk of classification is smaller, to what is be likely to occur
Bad data resistance ability is strong, still, may produce the phenomenon for owing study；L is sample total；C is the hypersphere centre of sphere；V tables
Show the compromise between supporting vector and wrong point of vector, hypersphere radius surface can be controlled to a certain extent；xiRepresent training sample
This, ξiFor penalty.
This problem is solved using Lagrangian：
Then
Bring into obtain dual problem be：
By QP optimization methods solution, this dual problem obtains optimization solution α.
Bring α into values that formula (3.4) can obtain center of circle c, that is, wait to ask typical day.αi、ajRepresent Lagrange multiplier.L is sample
This total amount；V represents the compromise between supporting vector and wrong point of vector, can control hypersphere radius surface to a certain extent；xi、
xjRepresent sample.
It can be seen that one-class support vector machines major function is the cluster to sample set, and mould is changed by adjustment parameter v
The structure risk of type, so as to be traded off among empiric risk and confidence risk, and the selection of typical day is also a kind of in itself
Cluster, if regarding the metric data of every day as a sample, sample is exactly a bit in 24 dimension spaces, super using one
Sphere allows it to cover all sample points as far as possible, then the center of circle of hypersphere i.e. the center of metric data, that is, allusion quotation
Place where type day data.
Step 3：The coefficient correlation between each user's typical day load curve is asked for respectively, each customer charge curve
Associate feature is analyzed；
By the typical day load curve of each user, electricity consumption rule between user and user can be analyzed and the sector is total
Associate feature between body electricity consumption rule.The coefficient correlation between n user's typical day load curve is asked for respectively, as a result can be stored in
N*n diagonal entry is in 1 symmetrical matrix.Off-diagonal element represents the coefficient correlation of two users' typical day load curve.
The similitude of each user's typical day load curve can be more intuitively observed by drawing coefficient correlation curve.
When the linear relationship of two user data strengthens, coefficient correlation tends to 1 or -1；When the increase of user data
When another user data also increases, coefficient correlation is more than 0；When a user data increases and another user data is reduced, phase
Relation number is less than 0；When two user data independence, coefficient correlation 0.
The specific method for asking for the coefficient correlation between each user's typical day load curve is：
Wherein, Xi、YiThe respectively Power system load data of two of which user；Respectively two of which user
The average value of Power system load data；N is Power system load data number, is positive integer.
Step 4：According to the coefficient correlation between each user, it is classified according to the electricity consumption rule of each user, it is determined that often
The load curve of a kind of user；
A certain user is chosen as user is referred to, the user that a certain setting value is more than with user's correlation coefficient value is playbacked
It is a kind of；In a kind of user of division, choose and be used as such with the load curve with reference to user correlation coefficient value highest user
The load curve of user.
By taking the huge rock shop town huge rock shop of user Haiyang City as an example, coefficient associated therewith has the effect peak life of user Shandong more than 0.95
Thing Science and Technology Co., Ltd., Qingdao profit source of students edible mushroom Co., Ltd, Weifang Linhai Biotechnology Co., Ltd., Jining City
Bacteria Co., Ltd. of Lexmark, Shandong Chen Yang Bacteria Co., Ltd.s, Pepsi spy edible mushroom Specialty Co-operative Organization of Jinxiang County, Zaozhuang City's hat space
Agricultural science and technology Co., Ltd, Shandong Fu He Bacteria Technology Co., Ltd.s.It can be seen that with Haiyang huge rock shop user typical case's daily load
The high user of curve correlation coefficient is mostly the enterprise for being engaged in mushroom production, and can be playbacked a kind of user.
Step 5：In the load curve of every a kind of user, choose electric load variation tendency and become with industry integral load
Change uniformity highest user, the overall Power system load data of industry is predicted by the electricity consumption Behavior law of the user.
By the analysis to coefficient correlation, we can find out the electricity consumption rule and industry entirety electricity consumption rule of which user
With higher uniformity, and then the overall electricity consumption trend of industry can be better anticipated by holding individual electricity consumption rule.
Although above-mentioned the embodiment of the present invention is described with reference to accompanying drawing, model not is protected to the present invention
The limitation enclosed, one of ordinary skill in the art should be understood that on the basis of technical scheme those skilled in the art are not
Need to pay various modifications or deformation that creative work can make still within protection scope of the present invention.
Claims (5)
1. a kind of Methods of electric load forecasting based on typical daily load characteristic, it is characterized in that, comprise the following steps：
(1) electric load of each user in setting regions setting industry is gathered by power regulation Central Grid EMS
Data, deposit Power system load data storehouse, and data are pre-processed；
(2) selection of typical day is carried out using branching algorithm one-class support vector machines, and the typical daily load for obtaining each user is bent
Line and industry integral load curve；
The specific method for carrying out the selection of typical day is：
Electric load sample 2-1) is chosen, sample is clustered using support vector machine method, builds object function；
The object function of structure is specially：
Constraints：||xi-c||2≤R2+ξi；ξi≥0；
The dual problem of object function 2-2) is asked for using Lagrangian Arithmetic；Specially：
Constraints：
2-3) by QP optimization methods solution, this dual problem obtains optimization solution；
The hypersphere centre of sphere c value 2-4) is obtained according to optimization solution, i.e., typical day data:
The optimization is solved into αiBring formula into：
The value for trying to achieve hypersphere centre of sphere c is typical day data；
Wherein, R is hyperspherical radius, ξiFor penalty, l is sample total, and c is the hypersphere centre of sphere, and coefficient ν is setting value；
αi、αjRespectively Lagrangian, xi、xjRespectively training sample, i=1,2 ..., l, j=1,2 ..., l；L is glug
Bright day function；
(3) coefficient correlation between each user's typical day load curve is asked for respectively, the associate feature each customer charge curve
Analyzed；
(4) according to the coefficient correlation between each user, it is classified according to the electricity consumption rule of each user, it is determined that per a kind of user
Load curve；
(5) in the load curve of every a kind of user, electric load variation tendency and industry integral load change uniformity are chosen
Highest user, the overall Power system load data of industry is predicted by the electricity consumption Behavior law of the user.
2. a kind of Methods of electric load forecasting based on typical daily load characteristic as claimed in claim 1, it is characterized in that, it is described
The method pre-processed in step (1) to data includes：
Initial data 1-1) is read from Power system load data storehouse, the initial data is associated according to critical field, carried
Take required data, and authority data form；
1-2) screen and reject bad data, completion is carried out to the data of missing；
1-3) wrong data is repaired, data are smoothed.
3. a kind of Methods of electric load forecasting based on typical daily load characteristic as claimed in claim 1, it is characterized in that, it is described
The method analyzed in step (3) the associate feature each customer charge curve is：
The coefficient correlation between n user's typical day load curve is asked for respectively, and the diagonal entry for being as a result stored in n*n is 1
In symmetrical matrix；Off-diagonal element represents the coefficient correlation of two users' typical day load curve；
When the linear relationship of two user data strengthens, coefficient correlation tends to 1 or -1；When a user data increase is another
When user data also increases, coefficient correlation is more than 0；When a user data increases and another user data is reduced, phase relation
Number is less than 0；When two user data independence, coefficient correlation 0.
4. a kind of Methods of electric load forecasting based on typical daily load characteristic as claimed in claim 1, it is characterized in that, it is described
The specific method that the coefficient correlation between each user's typical day load curve is asked in step (3) is：
Wherein, Xi、YiThe respectively Power system load data of two of which user；The respectively electric load of the two users
The average value of data；N is Power system load data number, is positive integer.
5. a kind of Methods of electric load forecasting based on typical daily load characteristic as claimed in claim 1, it is characterized in that, it is described
The specific method of step (4) is：
A certain user is chosen as user is referred to, user's playback one of a certain setting value will be more than with user's correlation coefficient value
Class；In a kind of user of division, choose with the load curve with reference to user correlation coefficient value highest user as such use
The load curve at family.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510354856.3A CN104992238B (en) | 2015-06-24 | 2015-06-24 | A kind of Methods of electric load forecasting based on typical daily load characteristic |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510354856.3A CN104992238B (en) | 2015-06-24 | 2015-06-24 | A kind of Methods of electric load forecasting based on typical daily load characteristic |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104992238A CN104992238A (en) | 2015-10-21 |
| CN104992238B true CN104992238B (en) | 2018-04-06 |
Family
ID=54304050
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510354856.3A Active CN104992238B (en) | 2015-06-24 | 2015-06-24 | A kind of Methods of electric load forecasting based on typical daily load characteristic |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104992238B (en) |
Families Citing this family (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105337277B (en) * | 2015-11-25 | 2017-03-29 | 国家电网公司 | The computational methods that power distribution network project yet to be built is affected on rate of qualified voltage |
| CN105719194B (en) * | 2016-01-11 | 2020-02-11 | 河南许继仪表有限公司 | Automatic optimization-searching type power load classification method |
| CN106042963B (en) * | 2016-06-17 | 2019-02-05 | 广东顺德中山大学卡内基梅隆大学国际联合研究院 | The cooperative optimization method and system of electrified transportation network and electric system |
| CN106651015A (en) * | 2016-12-12 | 2017-05-10 | 国网上海市电力公司 | Method for predicting typical day load of power grid by double sides approximate process |
| CN108205721B (en) * | 2016-12-16 | 2021-11-02 | 国家电网公司 | Spline interpolation typical daily load curve selecting device based on clustering |
| CN107545374A (en) * | 2017-09-15 | 2018-01-05 | 中国电力科学研究院 | A kind of daily load curve choosing method and system |
| CN107910864A (en) * | 2017-10-11 | 2018-04-13 | 长沙理工大学 | A kind of 220kV busbar equivalent load volatility forecast methods based on Information Granulating and support vector machines |
| CN108169621A (en) * | 2017-12-05 | 2018-06-15 | 国电南瑞科技股份有限公司 | Taiwan area power-off event complementing method based on support vector machines |
| CN107833153B (en) * | 2017-12-06 | 2020-11-03 | 广州供电局有限公司 | Power grid load missing data completion method based on k-means clustering |
| CN108596416B (en) * | 2017-12-29 | 2021-07-16 | 南方电网科学研究院有限责任公司 | Method and system for acquiring electric load reference quantity |
| CN108228862A (en) * | 2018-01-12 | 2018-06-29 | 上海市建筑科学研究院 | A kind of method for processing abnormal data and system of energy consumption monitoring platform |
| CN109102103A (en) * | 2018-06-26 | 2018-12-28 | 上海鲁班软件股份有限公司 | A kind of multi-class energy consumption prediction technique based on Recognition with Recurrent Neural Network |
| CN109085423B (en) * | 2018-09-20 | 2020-07-03 | 广东石油化工学院 | Load switch event detection method and system |
| CN109324285B (en) * | 2018-09-20 | 2020-07-03 | 广东石油化工学院 | Load switch event detection method and system |
| CN109657867B (en) * | 2018-12-25 | 2023-04-07 | 广州汇电云联互联网科技有限公司 | A-type unit output curve decomposition method |
| CN109783716B (en) * | 2019-01-21 | 2020-02-21 | 贵州黔源电力股份有限公司 | Automatic extraction and processing system and method for power supply and demand data |
| CN111598302B (en) * | 2020-04-19 | 2023-09-05 | 华电郑州机械设计研究院有限公司 | AP-TS-SVR model-based thermal power plant short-term industrial heat load prediction method |
| CN113326303A (en) * | 2021-05-31 | 2021-08-31 | 广东电网有限责任公司 | User electricity consumption behavior analysis method and related device |
| CN117195580B (en) * | 2023-09-21 | 2024-06-11 | 国网江苏省电力有限公司南通供电分公司 | Intelligent planning method and system for power distribution network |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104091293A (en) * | 2014-07-30 | 2014-10-08 | 国家电网公司 | ower grid long-term load characteristic predication method based on variation of electricity consumption structure |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9460478B2 (en) * | 2012-12-17 | 2016-10-04 | Arizona Board Of Regents On Behalf Of Arizona State University | System and method for wind generation forecasting |
-
2015
- 2015-06-24 CN CN201510354856.3A patent/CN104992238B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104091293A (en) * | 2014-07-30 | 2014-10-08 | 国家电网公司 | ower grid long-term load characteristic predication method based on variation of electricity consumption structure |
Non-Patent Citations (1)
| Title |
|---|
| 基于最小二乘支持向量机的短期负荷预测方法及应用研究;耿艳;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20090115(第01期);第5-45页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104992238A (en) | 2015-10-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104992238B (en) | A kind of Methods of electric load forecasting based on typical daily load characteristic | |
| CN104992239B (en) | A kind of trade power consumption gauge rule Forecasting Methodology based on related coefficient | |
| Chang | Energy intensity, target level of energy intensity, and room for improvement in energy intensity: An application to the study of regions in the EU | |
| CN106777244A (en) | A kind of power customer electricity consumption behavior analysis method and system | |
| CN106410781B (en) | A kind of power consumer demand response Potential Determining Method | |
| Liu et al. | Factor allocation, economic growth and unbalanced regional development in China | |
| CN110109971A (en) | A kind of low-voltage platform area user power utilization Load Characteristic Analysis method | |
| CN108710992A (en) | A kind of demand response participates in the appraisal procedure of power grid peak load shifting effect | |
| Zhou et al. | Multiobjective model of time-of-use and stepwise power tariff for residential consumers in regulated power markets | |
| Karimi et al. | Location decision for foreign direct investment in ASEAN countries: A TOPSIS approach | |
| CN104376402B (en) | A kind of load classification and integrated approach based on frequency-domain index | |
| CN108428055B (en) | Load clustering method considering load longitudinal characteristics | |
| Sedghamiz et al. | A game theory approach for conjunctive use optimization model based on virtual water concept | |
| CN110797872B (en) | Method, device, equipment and storage medium for configuring energy storage capacity of user side | |
| CN105574325B (en) | A kind of medium-term and long-term electricity demand forecasting method of combination demographic indicator | |
| CN109754123A (en) | The distance weighted positioned alternate method of rotation centerline that feeder line supply district divides | |
| CN109242218A (en) | Method and device for predicting wind power in plateau mountain area based on SVM | |
| Ogawa | Population aging and policy options for a sustainable future: the case of Japan | |
| CN107392678A (en) | A kind of electricity pricing method and system | |
| CN105654194B (en) | A kind of power network development index measurement method method | |
| CN108205761A (en) | A kind of multi-layer power sales data analysis monitors system | |
| CN105811414B (en) | It is a kind of to use power grid short term power prediction technique and device | |
| CN106251217A (en) | Portfolio Optimization method based on historical analogy method WCVaR Risk Model | |
| Alba et al. | Oil shocks and monetary policy rules in emerging economies | |
| CN105406463B (en) | A kind of evaluation method of adaptive load type distribution transformer performance driving economy |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 250002 Wang Yue Road, Ji'nan City, Shandong Province, No. 2000 Patentee after: ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER Co. Patentee after: STATE GRID CORPORATION OF CHINA Address before: 250002 Wang Yue Road, Ji'nan City, Shandong Province, No. 2000 Patentee before: ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER Co. Patentee before: State Grid Corporation of China |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210409 Address after: No. 150, Jinger Road, Daguanyuan, Shizhong District, Jinan City, Shandong Province Patentee after: Shandong Electric Power Marketing Center Patentee after: ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER Co. Patentee after: STATE GRID CORPORATION OF CHINA Address before: 250002 Wang Yue Road, Ji'nan City, Shandong Province, No. 2000 Patentee before: ELECTRIC POWER RESEARCH INSTITUTE OF STATE GRID SHANDONG ELECTRIC POWER Co. Patentee before: STATE GRID CORPORATION OF CHINA |