CN101888087A - Method for realizing distributed super-short-term area load forecasting in distribution network terminal - Google Patents
Method for realizing distributed super-short-term area load forecasting in distribution network terminal Download PDFInfo
- Publication number
- CN101888087A CN101888087A CN2010101866427A CN201010186642A CN101888087A CN 101888087 A CN101888087 A CN 101888087A CN 2010101866427 A CN2010101866427 A CN 2010101866427A CN 201010186642 A CN201010186642 A CN 201010186642A CN 101888087 A CN101888087 A CN 101888087A
- Authority
- CN
- China
- Prior art keywords
- distribution network
- network terminal
- load
- short
- output
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 238000012549 training Methods 0.000 claims abstract description 32
- 238000013528 artificial neural network Methods 0.000 claims abstract description 21
- 230000001537 neural effect Effects 0.000 claims description 27
- 210000002569 neuron Anatomy 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000036316 preload Effects 0.000 claims description 2
- 238000012546 transfer Methods 0.000 claims description 2
- 230000003044 adaptive effect Effects 0.000 abstract description 2
- 238000013507 mapping Methods 0.000 abstract 1
- 230000006870 function Effects 0.000 description 5
- 230000006872 improvement Effects 0.000 description 5
- 230000005611 electricity Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000007774 longterm Effects 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 108010022579 ATP dependent 26S protease Proteins 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000010977 unit operation Methods 0.000 description 1
Images
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
- Supply And Distribution Of Alternating Current (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a method for realizing distributed super-short-term area load forecasting in a distribution network terminal, which comprises the following steps of: (a) constructing an input vector and an output vector of a neural network program of the distribution network terminal; (b) initializing a neural network according to different area load types of a feeder line section of the distribution network terminal; (c) acquiring a training sample by using historical load data; (d) training the neural network by using the sample; and (e) obtaining a super-short-term area load curve by using the trained neural network program. By utilizing a nonlinear mapping relationship between the input and output of the neural network and adaptive learning capacity, and according to the different area load types of the feeder line section of the distribution network terminal, a corresponding neural network structure and an initial value are selected in the distribution network terminal, the distributed load forecasting is realized, and the forecast accuracy is improved.
Description
Technical field
The present invention relates to the power technology field, especially a kind of method of super-short-term area load prediction.
Background technology
Power industry standard DL/T 721-2000 " distribution automation system terminal unit " is defined as the distribution automation system terminal unit: the distribution automation system terminal unit is the general designation of various distribution terminal units, distribution transformer terminal unit and the pressure monitoring unit equipment such as (power distribution automation and management system substations) that is used for the power distribution network distribution loop.Adopt communication port to finish functions such as data acquisition and distant place control.
The elementary cell of neural net is neuron models.Neuron models are mathematicization models of the 26S Proteasome Structure and Function of simulation biological neuron, generally are the nonlinear information process units of the single output of input more than.
The citation form of neural net has feedforward network, feedback network, the type that mutually combines network and heterogeneous network.Use more neural network model Hopfield network, BP network, Blotaman network, ART network are arranged.The Hopfield network is most typical feedback network model, is mainly used in restriction and optimizes and associative memory.The BP network is a counterpropagation network, and it is a kind of multilayer feedforward network, can be used for language identification and adaptive control.The Blotaman network is typical stochastic network model, is mainly used in pattern recognition.The ART network is a kind of self organizing network model, is mainly used in pattern recognition.
The learning algorithm of neural net is divided into two big classes: teacher learning and teacherless learning are arranged.Have teacher learning to be meant in the neural metwork training process, have the network output of an expectation all the time, the distance between desired output and actual the output is as error metrics and be used to adjust the network link weight coefficients.The teacherless learning is meant that there is not the output valve of an expectation in network, thereby does not have direct control information.
Load prediction is the many factors such as operation characteristic, increase-volume decision-making, natural conditions and social influence according to system, study or utilize the mathematical method of loading in cover system ground processing past and future, satisfying under the condition of certain required precision, determining the load data of following certain particular moment.The core of the mathematical theory of load prediction is that how to obtain the historical variations rule of forecasting object and be subjected to the prediction of some power system loads be one of important process of administrative departments such as power system dispatching, electricity consumption, plan and planning.Load prediction accurately helps economy, reasonably arranges the start and stop of the inner generating set of electrical network, keeps the security and stability of operation of power networks, reduces unnecessary rotation idle capacity; Help the management of power use, rationally arrange power system operating mode and unit maintenance plan, guarantee the ordinary production and the household electricity of society; Help economizing on coal, fuel-economizing and reduction cost of electricity-generating, improve the economic benefit and the social benefit of electric power system; Help formulating rational power supply construction plan, the installation of the following new generating set of decision and size, the when and where of installed capacity; Help rationally arranging the increase-volume and the reconstruction of electrical network, and the construction and the development of decision electrical network.The load prediction problem relates to power system planning and design, the economy of power system operation, reliability and fail safe, and many aspects such as power market transaction have become an important research field in operation of modern power industry system and the management.
Load prediction can be divided into ultrashort phase, short-term, medium and long term according to the difference of time bar: 1. ultra-short term is meant that following 1h is with interior load prediction, under the security monitoring state, the predicted value that needs 5～10s or 1～5min, the preventative control and the state of emergency are handled the predicted value that needs 10min to 1h.2. short-term load forecasting is meant daily load prediction and all load predictions, be respectively applied for and arrange day operation plan and all operation plans, comprise definite startup and shutdown of units, the coordination of extreme misery electricity, Tie line Power, load economical distribution, reservoir regulation and overhaul of the equipments etc., to short-term forecast, need fully research network load Changing Pattern, analysis load changes the relation of correlation factor, particularly weather conditions, day type etc. and short-term load variations.3. the load prediction in mid-term is meant month extremely load prediction in year, mainly is to determine unit operation mode and equipment rebuilding plan etc.4. long-term load prediction be meant following 3～5 years in addition the longer time section in load prediction, mainly be Electric Power Network Planning department according to development and national economy with to the demand of electric load, the perspective long-term plan of electric network reconstruction of being done and enlarging work.Centering, long-term load prediction will be studied the influence of the national economic development, national policy etc. especially.
Ultra-short term will become the basis of AGC (automatic generation control) practicability, also be simultaneously the required essential information of dynamic economic dispatch and electricity market operation.The characteristics of ultra-short term are on-line operations, and the up-to-date information on load that will obtain is used to predict next load constantly.The ultra-short term cycle is short, and the computational speed of requirement forecast method is very fast, does not generally consider the influence of meteorological condition simultaneously, because for the time interval of ultra-short term, meteorological variation is not obvious.
And compare with the system loading prediction, regional load prediction has following characteristics:
1, the prediction radix is smaller, and this is because bus load is far smaller than system loading;
2, because service area's intra domain user is more fixing, so the power structure of bus load is fairly simple and stable;
3, owing to the influence of user behavior in the power supply area, bus load is easy to generate sudden change, and stability is poor, and more burr is arranged;
4, the data inaccuracy that is accumulated, and usually contain the bigger outlier of error;
5, be subjected to the influence of meteorologic factor may be bigger;
6, the tendency of load variations is not obvious;
7, the load curve diversity ratio between the different buses is bigger.
Present load prediction mainly is to carry out the system loading prediction in main website.Along with the expansion of scale of power and the increase of electric network composition complexity, the data that main website handles sharply increase, and the ultra-short term cycle is short, requires computational speed very fast, and main website is caused very big burden.The influencing factor of load prediction simultaneously is numerous, is difficult to same model the load of zones of different be made prediction comparatively accurately.
Summary of the invention
Technical problem to be solved by this invention provides a kind of method that realizes the distributed super-short-term area load prediction in distribution network terminal, overcome ultra-short term real-time and big data quantity and cause the main website burden, and be difficult to the load of zones of different be made the problem of prediction comparatively accurately with same model.
For solving the problems of the technologies described above, technical scheme of the present invention is:
A kind of method that realizes the distributed super-short-term area load prediction in distribution network terminal is characterized in that: may further comprise the steps:
(a) input vector and the output vector of structure distribution network terminal neural network procedure;
(b) according to the zones of different load type initialization neural net of distribution network terminal place feeder line section;
(c) the historical load data of utilizing distribution network terminal to write down obtain the training sample of neural net;
(d) utilize training sample that neural net is trained;
(e) neural network procedure after distribution network terminal utilization training is carried out load prediction.
As improvement, in the described step (a), choose the distribution network terminal region same date type a few days ago, the previous day the first five constantly, the previous day preceding two constantly, previous moment, the current time of the previous day and the same day of the previous day the first five constantly, the same day preceding two constantly, the load data of the previous moment on the same day is as input vector, the current time load of choosing the same day is as output vector.
As improvement, in the described step (b), determine the neural network structure of terminal to comprise hidden layer quantity, the hidden layer node number according to regional load type; Determine the neural net initial weight and the threshold value of terminal according to regional load type.
As improvement, in the described step (c), obtain training sample and learning sample contrast, and sample data is carried out normalized according to the regional historical load data of distribution network terminal place feeder line section.
As improvement, described step (d) may further comprise the steps:
(d1) to the error criterion assignment;
(d2) input training sample calculates intermediate layer, each neuron output of output layer;
(d3) calculation expectation output and actual output error;
(d4) back transfer error is adjusted the weights that are connected in intermediate layer and input layer and output layer and intermediate layer;
(d5) error meets the demands and just finishes training, and error does not meet the demands with regard to repetition training.As improvement, described step (e) may further comprise the steps:
(e1) the regional load data of distribution network terminal record place feeder line section;
(e2) will work as preload and the historical load data are sent into neural net;
(e3) drawing next by neural network procedure loads constantly;
(e4) next is loaded constantly and the historical load data are sent into neural net;
(e5) upgrade load curve.
The beneficial effect that the present invention is compared with prior art brought is:
The present invention realizes the distributed super-short-term load prediction in distribution network terminal, main target is the zone load of prediction distribution network terminal place feeder line section, the power structure of load is fairly simple and stable, be to carry out ultra-short term simultaneously, so when neural metwork training input choose a few days ago, the load data of the first five moment of the previous day, preceding two moment, previous moment, current time and the first five moment on the same day, preceding two moment, previous moment, the time interval is 3min.Consider different date types to effects of load, can be divided into working day and day off the date.Refer to working day to Friday, refer to Saturday, Sunday and festivals or holidays day off normal Monday.At the zones of different load type of different distribution network terminal place feeder line section, it can be divided into shopping centre, industrial area, residential block etc. simultaneously.
Description of drawings
Fig. 1 is the flow chart of steps of Forecasting Methodology of the present invention;
Fig. 2 is steps d neural network procedure training flow chart of the present invention;
Fig. 3 carries out the load prediction flow chart for distribution network terminal of the present invention.
Embodiment
The invention will be further described below in conjunction with Figure of description.
As shown in Figure 1, as one embodiment of the present of invention, distribution network terminal input vector when neural metwork training is chosen a few days ago, the previous day the first five constantly, the previous day preceding two constantly, previous moment, the current time of the previous day and the same day of the previous day the first five constantly, the same day preceding two constantly, the load data of the previous moment on the same day, the time interval is 3min.Output vector is a current time load value on the same day.Ultra-short term is the time interval with 3min, the zone load of prediction 1h.Date type is divided into working day and day off, refers to working day to Friday, refer to Saturday, Sunday and festivals or holidays day off normal Monday.At the zones of different load type of different distribution network terminal place feeder line section, it can be divided into shopping centre, industrial area, residential block etc. simultaneously.
As shown in Figure 2, the present invention can adopt the BP neural net, based on the error propagation algorithm, must draw the output layer error by training sample, adjusts neural net by every layer error propagation, changes output error, so that output error is less than the assigned error index.
During the neural network training program, middle layer node number and the learning rate set in the program all can influence network convergence speed, promptly frequency of training are impacted.If the middle layer node number very little, network may can not train at all or network performance very poor; If the middle layer node number is too many, though the systematic error of network is reduced, net training time is prolonged, on the other hand, training is absorbed in local minimum point easily and can not get optimum point, also occurs easily " over-fitting ".Learning rate is value between 0 and 1, and value makes unstable networks more easily, and the less meeting of value makes the training time long.
During the neural network procedure initialization, determine the zones of different load type of distribution network terminal place feeder line section earlier, as shopping centre, industrial area, residential block etc., determine the neural network structure of terminal again according to regional load type, comprise hidden layer quantity, the hidden layer node number, last neural net initial weight and the threshold value of determining terminal according to regional load type.
The training sample of neural metwork training comes from the history area load data of distribution network terminal place feeder line section, needs sample data is carried out normalized after drawing training sample.
The BP neural net comprises input layer, intermediate layer and output layer, is a kind of error back propagation network.Its basic thought is a least square method, adopts the gradient search technology, so that the mean square error minimum of the real output value of network and desired value.Input layer input a few days ago, the previous day the first five constantly, the previous day preceding two constantly, previous moment, the current time of the previous day and the same day of the previous day the first five constantly, the same day preceding two constantly, the previous moment load data on the same day, intermediate layer and input layer initial is connected weights and draws according to the regional load type of terminal, the node in intermediate layer is the weighted sum of input layer output, and the excitation function of node adopts S (Sigmoid) type function.
The input of the node of output layer is the weighted sum of middle layer node output, and output layer and intermediate layer initial is connected the regional load type of weights by terminal and draws.With the output result of output layer be the same day current time load make comparisons with the desired output of teacher's sample, as do not meet and then change backpropagation over to, error signal is returned along original connecting path, by revising the neuronic weight coefficient of each layer, make switch that the output result who obtains on the output layer node is each distribution network terminal place and the error signal minimum between circuit breaker deciliter state and the desired output.
Get the study target function during neural network learning:
In the formula, y
d(t) be the desirable system output of current time; Y (t) is the actual output of current neural net.
Right for each sample data, begin to calculate the output valve of each node from the input node, and then bring into use backpropagation to calculate the partial derivative of all implicit nodes from output node by propagated forward, the broad sense learning rules are:
In the formula: η is a learning rate
Calculate the output layer generalized error during study earlier, calculate feedback error and adjust the output layer weight coefficient by the output layer generalized error then, adjust the input layer weight coefficient by feedback error again.
After a sample was finished the adjustment of network weight coefficient, it was right to send into another sample mode again, carries out similar study, up to the training study of finishing all samples.
Calculate the neural network procedure output error after the training study of sample is finished, and make comparisons,, withdraw from training if less than specification error then preserve the weight coefficient value of neural net with the error of setting.If output error is greater than specification error then continue all sample training.
As shown in Figure 3, distribution network terminal is gathered feeder line section zone, the terminal place load of current time, and then in the historical load data of record, choose a few days ago, the previous day preceding four constantly, the previous moment of the previous day, the current time of the previous day, the previous day back one constantly and the same day preceding four constantly, the load data of the previous moment on the same day, as the input of neural network procedure, the time interval is 3min together.
Distribution network terminal draws next prediction load data constantly by neural net, will predict that again load data and relevant historical load data import as neural net.Ultra-short term is the time interval with 3min, and prediction is the load of 1h in the future.
Distribution network terminal online real time collecting current time zone load data upgrades load prediction curve.
The present invention handles the super-short-term area load prediction with neural network procedure in distribution network terminal, overcome ultra-short term real-time and big data quantity and caused main website over-burden, and be difficult to the load of zones of different be made the problem of predicting comparatively accurately with same model.By distinguishing the zones of different load type of distribution network terminal place feeder line section, in distribution network terminal, choose corresponding neural network structure and initial value, realized distributed load prediction, improved prediction accuracy.
Claims (6)
1. method that realizes the distributed super-short-term area load prediction in distribution network terminal is characterized in that: may further comprise the steps:
(a) input vector and the output vector of structure distribution network terminal neural network procedure;
(b) according to the zones of different load type initialization neural net of distribution network terminal place feeder line section;
(c) the historical load data of utilizing distribution network terminal to write down obtain the training sample of neural net;
(d) utilize training sample that neural net is trained;
(e) neural network procedure after distribution network terminal utilization training is carried out load prediction.
2. a kind of method that in distribution network terminal, realizes the distributed super-short-term area load prediction according to claim 1, it is characterized in that: in the described step (a), choose the distribution network terminal region same date type a few days ago, the previous day the first five constantly, the previous day preceding two constantly, previous moment, the current time of the previous day and the same day of the previous day the first five constantly, the same day preceding two constantly, the load data of the previous moment on the same day is as input vector, the current time load of choosing the same day is as output vector.
3. a kind of method that in distribution network terminal, realizes the distributed super-short-term area load prediction according to claim 1, it is characterized in that: in the described step (b), determine the neural network structure of terminal to comprise hidden layer quantity, the hidden layer node number according to regional load type; Determine the neural net initial weight and the threshold value of terminal according to regional load type.
4. a kind of method that in distribution network terminal, realizes the distributed super-short-term area load prediction according to claim 1, it is characterized in that: in the described step (c), regional historical load data according to distribution network terminal place feeder line section obtain training sample and learning sample contrast, and sample data is carried out normalized.
5. a kind of method that in distribution network terminal, realizes the distributed super-short-term area load prediction according to claim 1, it is characterized in that: described step (d) may further comprise the steps:
(d1) to the error criterion assignment;
(d2) input training sample calculates intermediate layer, each neuron output of output layer;
(d3) calculation expectation output and actual output error;
(d4) back transfer error is adjusted the weights that are connected in intermediate layer and input layer and output layer and intermediate layer;
(d5) error meets the demands and just finishes training, and error does not meet the demands with regard to repetition training.
6. a kind of method that in distribution network terminal, realizes the distributed super-short-term area load prediction according to claim 1, it is characterized in that: described step (e) may further comprise the steps:
(e1) the regional load data of distribution network terminal record place feeder line section;
(e2) will work as preload and the historical load data are sent into neural net;
(e3) drawing next by neural network procedure loads constantly;
(e4) next is loaded constantly and the historical load data are sent into neural net;
(e5) upgrade load curve.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101866427A CN101888087A (en) | 2010-05-21 | 2010-05-21 | Method for realizing distributed super-short-term area load forecasting in distribution network terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010101866427A CN101888087A (en) | 2010-05-21 | 2010-05-21 | Method for realizing distributed super-short-term area load forecasting in distribution network terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101888087A true CN101888087A (en) | 2010-11-17 |
Family
ID=43073861
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010101866427A Pending CN101888087A (en) | 2010-05-21 | 2010-05-21 | Method for realizing distributed super-short-term area load forecasting in distribution network terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101888087A (en) |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799953A (en) * | 2012-07-16 | 2012-11-28 | 安徽省电力公司池州供电公司 | Bus load prediction method based on stacked generalization training strategy |
| CN102982393A (en) * | 2012-11-09 | 2013-03-20 | 山东电力集团公司聊城供电公司 | Online prediction method of electric transmission line dynamic capacity |
| CN103295075A (en) * | 2013-04-01 | 2013-09-11 | 沈阳航空航天大学 | Ultra-short-term power load forecasting and early warning method |
| CN103530700A (en) * | 2013-10-11 | 2014-01-22 | 国家电网公司 | Comprehensive urban distribution network saturated load forecasting method |
| CN104598985A (en) * | 2014-12-12 | 2015-05-06 | 国家电网公司 | Power load forecasting method |
| CN104850918A (en) * | 2015-06-02 | 2015-08-19 | 国网山东省电力公司经济技术研究院 | Node load prediction method taking power grid topology constraints into consideration |
| CN105162151A (en) * | 2015-10-22 | 2015-12-16 | 国家电网公司 | Intelligent energy storage system grid-connected real-time control method based on artificial fish swarm algorithm |
| CN105787606A (en) * | 2016-03-24 | 2016-07-20 | 国网辽宁省电力有限公司电力科学研究院 | Power dispatching online trend early warning system based on ultra short term load prediction |
| CN106451434A (en) * | 2016-11-03 | 2017-02-22 | 国网浙江省电力公司电力科学研究院 | Power distribution network voltage determination method and device based on neural network algorithm |
| CN106599417A (en) * | 2016-11-30 | 2017-04-26 | 中国电力科学研究院 | Method for identifying urban power grid feeder load based on artificial neural network |
| CN106845673A (en) * | 2016-12-14 | 2017-06-13 | 国网北京市电力公司 | The method of supplying power to and device of power system |
| CN107609667A (en) * | 2017-07-20 | 2018-01-19 | 国网山东省电力公司电力科学研究院 | Heating load forecasting method and system based on Box_cox conversion and UFCNN |
| CN107977737A (en) * | 2017-11-19 | 2018-05-01 | 国网浙江省电力公司信息通信分公司 | Distribution transformer load Forecasting Methodology based on mxnet frame depth neutral nets |
| CN108845492A (en) * | 2018-05-23 | 2018-11-20 | 上海电力学院 | A kind of AGC system Intelligent predictive control method based on CPS evaluation criterion |
| CN109325668A (en) * | 2018-09-04 | 2019-02-12 | 北京国电通网络技术有限公司 | The regulation method and apparatus of energy resource system |
| CN109345028A (en) * | 2018-10-25 | 2019-02-15 | 国家电网有限公司 | Power grid data processing method and device based on Situation Awareness |
| CN110322369A (en) * | 2019-07-03 | 2019-10-11 | 厦门理工学院 | A kind of building load optimum combination determines method, terminal device and storage medium |
| CN110781947A (en) * | 2019-10-22 | 2020-02-11 | 北京交通大学 | Power load prediction model training and power load prediction method and device |
| CN111079914A (en) * | 2018-10-19 | 2020-04-28 | 中科寒武纪科技股份有限公司 | Operation method, system and related product |
| CN111160625A (en) * | 2019-12-10 | 2020-05-15 | 中铁电气化局集团有限公司 | Power load prediction method, power load prediction device, computer equipment and storage medium |
| CN111461409A (en) * | 2020-03-10 | 2020-07-28 | 国网山西省电力公司经济技术研究院 | Abnormal value processing method for medium and long-term load data |
| CN111555363A (en) * | 2020-04-09 | 2020-08-18 | 广西大学 | AGC real-time control strategy based on deep learning under big data environment |
| CN111709566A (en) * | 2020-06-09 | 2020-09-25 | 信雅达***工程股份有限公司 | Bank branch business prediction and scheduling method |
| CN112564105A (en) * | 2020-12-15 | 2021-03-26 | 深圳供电局有限公司 | Automatic power grid dispatching method and system |
| CN113131476A (en) * | 2021-04-28 | 2021-07-16 | 南方电网科学研究院有限责任公司 | Power load prediction method |
| CN113515896A (en) * | 2021-08-06 | 2021-10-19 | 红云红河烟草(集团)有限责任公司 | Data missing value filling method for real-time cigarette acquisition |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101288089A (en) * | 2005-07-28 | 2008-10-15 | 西门子电力输送及配电有限公司 | Load prediction based on-line and off-line training of neural networks |
| CN101414366A (en) * | 2008-10-22 | 2009-04-22 | 西安交通大学 | Method for forecasting electric power system short-term load based on method for improving uttermost learning machine |
| CN101702537A (en) * | 2009-11-10 | 2010-05-05 | 深圳市科陆电子科技股份有限公司 | Method for processing failures on adaptive basis in terminal of distribution network |
-
2010
- 2010-05-21 CN CN2010101866427A patent/CN101888087A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101288089A (en) * | 2005-07-28 | 2008-10-15 | 西门子电力输送及配电有限公司 | Load prediction based on-line and off-line training of neural networks |
| CN101414366A (en) * | 2008-10-22 | 2009-04-22 | 西安交通大学 | Method for forecasting electric power system short-term load based on method for improving uttermost learning machine |
| CN101702537A (en) * | 2009-11-10 | 2010-05-05 | 深圳市科陆电子科技股份有限公司 | Method for processing failures on adaptive basis in terminal of distribution network |
Non-Patent Citations (1)
| Title |
|---|
| 姚晓林: "《基于超短期负荷预测的补偿电容器优化投切》", 《山东大学硕士学位论文》, 31 December 2006 (2006-12-31), pages 18 - 30 * |
Cited By (41)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102799953A (en) * | 2012-07-16 | 2012-11-28 | 安徽省电力公司池州供电公司 | Bus load prediction method based on stacked generalization training strategy |
| CN102799953B (en) * | 2012-07-16 | 2016-06-29 | 国家电网公司 | Bus load Forecasting Methodology based on the extensive Training strategy of stacking |
| CN102982393A (en) * | 2012-11-09 | 2013-03-20 | 山东电力集团公司聊城供电公司 | Online prediction method of electric transmission line dynamic capacity |
| CN102982393B (en) * | 2012-11-09 | 2016-08-10 | 山东电力集团公司聊城供电公司 | A kind of on-line prediction method of electric transmission line dynamic capacity |
| CN103295075A (en) * | 2013-04-01 | 2013-09-11 | 沈阳航空航天大学 | Ultra-short-term power load forecasting and early warning method |
| CN103295075B (en) * | 2013-04-01 | 2016-09-28 | 沈阳航空航天大学 | A kind of ultra-short term load forecast and method for early warning |
| CN103530700A (en) * | 2013-10-11 | 2014-01-22 | 国家电网公司 | Comprehensive urban distribution network saturated load forecasting method |
| CN103530700B (en) * | 2013-10-11 | 2016-06-29 | 国家电网公司 | Urban distribution network saturation loading Comprehensive Prediction Method |
| CN104598985B (en) * | 2014-12-12 | 2017-08-25 | 国家电网公司 | Methods of electric load forecasting |
| CN104598985A (en) * | 2014-12-12 | 2015-05-06 | 国家电网公司 | Power load forecasting method |
| CN104850918A (en) * | 2015-06-02 | 2015-08-19 | 国网山东省电力公司经济技术研究院 | Node load prediction method taking power grid topology constraints into consideration |
| CN104850918B (en) * | 2015-06-02 | 2018-05-01 | 国网山东省电力公司经济技术研究院 | A kind of node load Forecasting Methodology counted and power network topology constrains |
| CN105162151A (en) * | 2015-10-22 | 2015-12-16 | 国家电网公司 | Intelligent energy storage system grid-connected real-time control method based on artificial fish swarm algorithm |
| CN105787606A (en) * | 2016-03-24 | 2016-07-20 | 国网辽宁省电力有限公司电力科学研究院 | Power dispatching online trend early warning system based on ultra short term load prediction |
| CN106451434A (en) * | 2016-11-03 | 2017-02-22 | 国网浙江省电力公司电力科学研究院 | Power distribution network voltage determination method and device based on neural network algorithm |
| CN106451434B (en) * | 2016-11-03 | 2019-04-02 | 国网浙江省电力公司电力科学研究院 | A kind of distribution network voltage based on neural network algorithm determines method and device |
| CN106599417A (en) * | 2016-11-30 | 2017-04-26 | 中国电力科学研究院 | Method for identifying urban power grid feeder load based on artificial neural network |
| CN106845673A (en) * | 2016-12-14 | 2017-06-13 | 国网北京市电力公司 | The method of supplying power to and device of power system |
| CN106845673B (en) * | 2016-12-14 | 2020-12-11 | 国网北京市电力公司 | Power supply method and device of power system |
| CN107609667A (en) * | 2017-07-20 | 2018-01-19 | 国网山东省电力公司电力科学研究院 | Heating load forecasting method and system based on Box_cox conversion and UFCNN |
| CN107609667B (en) * | 2017-07-20 | 2020-08-18 | 国网山东省电力公司电力科学研究院 | Heat supply load prediction method and system based on Box _ cox transformation and UFCNN |
| CN107977737A (en) * | 2017-11-19 | 2018-05-01 | 国网浙江省电力公司信息通信分公司 | Distribution transformer load Forecasting Methodology based on mxnet frame depth neutral nets |
| CN107977737B (en) * | 2017-11-19 | 2021-06-08 | 国网浙江省电力公司信息通信分公司 | Distribution transformation load prediction method based on mxnet frame deep neural network |
| CN108845492A (en) * | 2018-05-23 | 2018-11-20 | 上海电力学院 | A kind of AGC system Intelligent predictive control method based on CPS evaluation criterion |
| CN109325668A (en) * | 2018-09-04 | 2019-02-12 | 北京国电通网络技术有限公司 | The regulation method and apparatus of energy resource system |
| CN111079914A (en) * | 2018-10-19 | 2020-04-28 | 中科寒武纪科技股份有限公司 | Operation method, system and related product |
| CN111079914B (en) * | 2018-10-19 | 2021-02-09 | 中科寒武纪科技股份有限公司 | Operation method, system and related product |
| CN109345028A (en) * | 2018-10-25 | 2019-02-15 | 国家电网有限公司 | Power grid data processing method and device based on Situation Awareness |
| CN109345028B (en) * | 2018-10-25 | 2021-07-13 | 国家电网有限公司 | Situation awareness-based power grid data processing method and device |
| CN110322369B (en) * | 2019-07-03 | 2021-10-15 | 厦门理工学院 | Building load optimal combination determination method, terminal device and storage medium |
| CN110322369A (en) * | 2019-07-03 | 2019-10-11 | 厦门理工学院 | A kind of building load optimum combination determines method, terminal device and storage medium |
| CN110781947A (en) * | 2019-10-22 | 2020-02-11 | 北京交通大学 | Power load prediction model training and power load prediction method and device |
| CN111160625A (en) * | 2019-12-10 | 2020-05-15 | 中铁电气化局集团有限公司 | Power load prediction method, power load prediction device, computer equipment and storage medium |
| CN111461409A (en) * | 2020-03-10 | 2020-07-28 | 国网山西省电力公司经济技术研究院 | Abnormal value processing method for medium and long-term load data |
| CN111555363A (en) * | 2020-04-09 | 2020-08-18 | 广西大学 | AGC real-time control strategy based on deep learning under big data environment |
| CN111709566A (en) * | 2020-06-09 | 2020-09-25 | 信雅达***工程股份有限公司 | Bank branch business prediction and scheduling method |
| CN112564105A (en) * | 2020-12-15 | 2021-03-26 | 深圳供电局有限公司 | Automatic power grid dispatching method and system |
| CN113131476A (en) * | 2021-04-28 | 2021-07-16 | 南方电网科学研究院有限责任公司 | Power load prediction method |
| CN113131476B (en) * | 2021-04-28 | 2022-06-14 | 南方电网科学研究院有限责任公司 | Power load prediction method |
| CN113515896A (en) * | 2021-08-06 | 2021-10-19 | 红云红河烟草(集团)有限责任公司 | Data missing value filling method for real-time cigarette acquisition |
| CN113515896B (en) * | 2021-08-06 | 2022-08-09 | 红云红河烟草(集团)有限责任公司 | Data missing value filling method for real-time cigarette acquisition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101888087A (en) | Method for realizing distributed super-short-term area load forecasting in distribution network terminal | |
| Macedo et al. | Demand side management using artificial neural networks in a smart grid environment | |
| Raghav et al. | Analytic hierarchy process (AHP)–swarm intelligence based flexible demand response management of grid-connected microgrid | |
| Dehalwar et al. | Electricity load forecasting for Urban area using weather forecast information | |
| AU2017368470B2 (en) | System and method for dynamic energy storage system control | |
| Mamlook et al. | A fuzzy inference model for short-term load forecasting | |
| Su et al. | A systematic data-driven Demand Side Management method for smart natural gas supply systems | |
| Polimeni et al. | Development and experimental validation of hierarchical energy management system based on stochastic model predictive control for Off-grid Microgrids | |
| CN108122068A (en) | A kind of power distribution network risk-averse retailer method and system | |
| CN104008432A (en) | Micro-grid short-term load forecasting method based on EMD-KELM-EKF | |
| CN103093288A (en) | Partition power grid bus load prediction system based on weather information | |
| Acakpovi et al. | Time Series Prediction of Electricity Demand Using Adaptive Neuro‐Fuzzy Inference Systems | |
| CN104573877A (en) | Power distribution network equipment demand prediction and quantitative method and system | |
| Capuno et al. | Very Short‐Term Load Forecasting Using Hybrid Algebraic Prediction and Support Vector Regression | |
| Matijašević et al. | A systematic review of machine learning applications in the operation of smart distribution systems | |
| Kabir et al. | Deep reinforcement learning-based two-timescale Volt-VAR control with degradation-aware smart inverters in power distribution systems | |
| Salgado et al. | A short-term bus load forecasting system | |
| Kim et al. | Economical energy storage systems scheduling based on load forecasting using deep learning | |
| Hua et al. | Digital twin based reinforcement learning for extracting network structures and load patterns in planning and operation of distribution systems | |
| CN113887809A (en) | Power distribution network supply and demand balance method, system, medium and computing equipment under double-carbon target | |
| Sperstad et al. | Optimal power flow methods and their application to distribution systems with energy storage: a survey of available tools and methods | |
| Saha et al. | Short-Term Electricity Consumption Forecasting: Time-Series Approaches | |
| Makanju et al. | Machine Learning Approaches for Power System Parameters Prediction: A Systematic Review | |
| Gao et al. | Substation Load Characteristics and Forecasting Model for Large-scale Distributed Generation Integration | |
| Guo et al. | Load prediction of multi‐type electric vehicle charging stations based on secondary decomposition and feature selection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20101117 |