CN108985570A - Load prediction method and system - Google Patents
Load prediction method and system Download PDFInfo
- Publication number
- CN108985570A CN108985570A CN201810629211.XA CN201810629211A CN108985570A CN 108985570 A CN108985570 A CN 108985570A CN 201810629211 A CN201810629211 A CN 201810629211A CN 108985570 A CN108985570 A CN 108985570A
- Authority
- CN
- China
- Prior art keywords
- prediction
- load
- forecasting method
- load forecasting
- library
- 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 103
- 238000013277 forecasting method Methods 0.000 claims abstract description 68
- 238000013507 mapping Methods 0.000 claims abstract description 48
- 238000011161 development Methods 0.000 claims description 10
- 238000012216 screening Methods 0.000 claims description 9
- 238000005457 optimization Methods 0.000 claims description 7
- 238000013139 quantization Methods 0.000 claims description 7
- 239000002131 composite material Substances 0.000 claims description 6
- 230000003044 adaptive effect Effects 0.000 claims description 5
- 238000013461 design Methods 0.000 claims description 3
- 230000035945 sensitivity Effects 0.000 abstract description 2
- 238000004451 qualitative analysis Methods 0.000 abstract 1
- 238000012549 training Methods 0.000 description 4
- 240000002853 Nelumbo nucifera Species 0.000 description 3
- 235000006508 Nelumbo nucifera Nutrition 0.000 description 3
- 235000006510 Nelumbo pentapetala Nutrition 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000011156 evaluation Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 241001123248 Arma Species 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000013178 mathematical model Methods 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 244000131316 Panax pseudoginseng Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000004836 empirical method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000001556 precipitation Methods 0.000 description 1
- 238000011002 quantification Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 239000004575 stone Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
- G06Q10/06375—Prediction of business process outcome or impact based on a proposed change
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a load forecasting method and a system thereof, wherein the method comprises the following steps: establishing a relevant factor mapping database according to relevant factors required by a prediction strategy, wherein the relevant factors at least comprise weather relevant factors; determining a plurality of corresponding load prediction methods according to a plurality of load change rules, and establishing a load prediction method library; and establishing a comprehensive prediction model according to the prediction strategy and the load prediction method library. The system is used for realizing the method. The invention fully exploits the sensitivity of load change to weather related factor change and gives qualitative analysis result; in the load prediction process, the influence expression of weather-related parameters in the prediction method on weather factors is sufficient and reasonable.
Description
Technical field
The present invention relates to Techniques for Prediction of Electric Loads fields, and in particular to a kind of load forecasting method and its system.
Background technique
Load prediction is explored and is used from the constraints such as the variation of known electric load and meteorology influential on this
Inner link between electric load and major influence factors and development and change rule make the following power load preparatory pre-
It surveys.For accurately demand of the prediction markets to this commodity of electric power, improve the approach of precision of prediction first is that in prediction process
In try the influence to prediction result of meter and various correlative factors (such as meteorologic factor).And prediction technique general at present is difficult
Preferable precision of prediction is obtained, because of the characteristics of these methods do not account for this area's electric load.
Summary of the invention
The present invention for load prediction does not consider location electric load in the prior art the characteristics of and can not obtain compared with
The problem of good precision of prediction, provides a kind of load forecasting method and its system, sufficiently excavates load variations for meteorological related
The sensitivity of factor variation provides qualitatively analysis result;During load prediction, ginseng relevant to meteorology in prediction technique
The influence expression of several pairs of meteorologic factors is sufficiently, rationally.
For this purpose, first aspect present invention embodiment provides a kind of load forecasting method, include the following steps:
The correlative factor according to needed for predicting strategy establishes Correlative Factor Mapping database, and the correlative factor includes at least
Meteorological correlative factor;
Corresponding a variety of load forecasting methods are determined according to a variety of load variations rules, and establish load forecasting method library;
Comprehensive Model is established according to predicting strategy and load forecasting method library.
Further, it includes: logical that the correlative factor according to needed for predicting strategy, which establishes Correlative Factor Mapping database,
Each department and correlative factor feature are crossed, the influence quantization using mapping function to the load of each factor, each day carries out clustering;
According to difference of the different affecting factors on mapping function, Correlative Factor Mapping database is established according to data base system design.
Further, the new short-term load forecasting method of load forecasting method library real-time reception and it is stored in load prediction
In method base, and the existing prediction technique in load forecasting method library is improved.
It is further, described that establish Comprehensive Model to load forecasting method according to predicting strategy include: according to related
Each department in factor mapping database, correlative factor feature predicting strategy, filter out different prediction model collection;It is pre- in load
It surveys on the basis of a variety of prediction techniques of method base, establishes Comprehensive Model.
Further, described on the basis of a variety of prediction techniques in load forecasting method library, establish Comprehensive Model packet
It includes: the prediction result of prediction techniques various in load forecasting method library is combined, establish and be based on several Individual forecast methods
Comprehensive Model, the objective law that the Comprehensive Model develops according to different regions load, optimization calculate difference
The weight of prediction technique.
Further, the combined mode includes optimizing to a variety of prediction models, optimizes the weight of various models, is determined
The history confidence level of forecast sample, is then combined prediction result in prediction model.
Further, the Comprehensive Model of the foundation based on several Individual forecast methods includes: according to prediction object
And its characteristic of relevant historical data carries out adaptive training, the suitable method mould of automatic screening from load forecasting method library
Type, by selection and the error analysis of method it can be concluded that being suitble to the Comprehensive Model of prediction object.
Further, the load correlative factor in the prediction model can customize, and modify the parameter of various prediction models,
The maps values in Correlative Factor Mapping library are adjusted, the structure of various models is defined, customized different predicting strategy verifies mould
Calculating process during type prediction.
Second aspect of the present invention embodiment provides a kind of load prediction system for load forecasting method described in first aspect
System, the system comprises:
For storing the Correlative Factor Mapping database of each department and correlative factor feature, for the negative of Storage Estimation method
He Yucefangfaku and Comprehensive Model.
Further, the Correlative Factor Mapping database includes analysis module, and the analysis module is used for using mapping
Influence quantization of the function to the load of each factor, each day carries out clustering;
The load forecasting method library includes addition modified module, and the addition modified module is used to receive new short-term bear
Lotus prediction technique and the existing prediction technique in load forecasting method library is improved, modify the parameter of various prediction models, adjusted
Save Correlative Factor Mapping library in maps values, define various models structure and customized different predicting strategy;
The Comprehensive Model includes screening module, composite module and computing module, and the screening module is used for various regions
Area, correlative factor feature predicting strategy, filter out different prediction model collection;The composite module is used for load prediction side
The prediction result of various prediction techniques is combined in Faku County;The computing module is used for the visitor developed according to different regions load
Rule is seen, optimization calculates the weight of different prediction techniques.
Compared with prior art, implement present invention method and its system has the following beneficial effects:
1) embodiment of the present invention is using a variety of prediction models as a result, Comprehensive Model is realized, by various prediction techniques
Prediction result carry out optimal combination, as a result, according to each department, every profession and trade the characteristics of, construct different integrated forecasting moulds
Type makes precision of prediction be further enhanced.
2) embodiment of the present invention uses the characteristics of each department, each season power load demand, selects prediction technique, and construction is pre-
Strategy is surveyed, fitting and the precision of prediction of Comprehensive Model is used to provide the function of automatic modeling for user, is solved previous
The problem of load prediction software package requirement user setting Model Weight.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is a kind of load forecasting method flow chart described in the embodiment of the present invention;
Fig. 2 is load forecasting method library schematic diagram described in the embodiment of the present invention;
Fig. 3 is load prediction system structure diagram described in the embodiment of the present invention.
Specific embodiment
In being described below, for illustration and not for limitation, the tool of such as particular system structure, technology etc is proposed
Body details understands the embodiment of the present invention to cut thoroughly.However, it will be clear to one skilled in the art that there is no these specific
The present invention also may be implemented in the other embodiments of details.In other situations, it omits to well-known system, system, electricity
The detailed description of road and method, in case unnecessary details interferes description of the invention.
In order to illustrate technical solutions according to the invention, it is illustrated below by specific embodiment combination attached drawing.
It in actual operation, is extremely complex model to the processing of weather condition, this, which is primarily due to meteorological description, is
One group has different dimensions and the object of different descriptions (such as: weather pattern temperature, humidity, precipitation, wind-force, wind speed), is difficult
Unified mathematical expression and use are carried out in the mathematical model of system again.Therefore the embodiment of the present invention proposes a kind of load prediction
Method and its system should be set to all kinds of different factor values in one comparable same section (such as: between 0~1).
A kind of load forecasting method process is provided for first aspect present invention embodiment as shown in Figure 1, this method includes such as
Lower step:
S101 correlative factor according to needed for predicting strategy establishes Correlative Factor Mapping database, and the correlative factor is at least
Including meteorological correlative factor;
S102 determines corresponding a variety of load forecasting methods according to a variety of load variations rules, and establishes load forecasting method
Library;
S103 establishes Comprehensive Model according to predicting strategy and load forecasting method library.
Further, the correlative factor according to needed for predicting strategy described in step S101 establishes Correlative Factor Mapping data
Library includes: that the influence quantization by each department and correlative factor feature, using mapping function to the load of each factor, each day carries out
Clustering;According to difference of the different affecting factors on mapping function, Correlative Factor Mapping is established according to data base system design
Database.
Specifically, thought of the present embodiment using the clustering in pattern-recognition, with a mapping function to it is each because
Influence quantization of the element to the load of each day, such as: (Thursday, it is fine, 35 degrees Celsius ...)=> (0.25,0.12,0.88 ...).Together
When, the influence mapping value meteorology to load considers with the influence unification of various other correlative factors, and sets according to data base system
Meter forms Correlative Factor Mapping database, is exemplified below:
Table 1
It is designed by Correlative Factor Mapping database, achieves following advantage in terms of solving loading effects factor:
The sampling key point for deleting mapping mapping function can flexibly be increased;
Value is after the mapping of mapping mapping function can flexibly be modified to obtain reasonable mapping result;
Table 2
In conclusion establishing Correlative Factor Mapping database, it is established that such as weather all kinds of correlative factors are to load shadow
Loud numerical Evaluation and application system, it may be considered that the meteorologic factor studied in detail, and one can be considered simultaneously
As property classification indicators, such as working day/day off, normal day/festivals or holidays, etc..The present embodiment Correlative Factor Mapping data
Difference of the different affecting factors on mapping function is reflected in library.According to the basic principle of pattern-recognition, using clustering method
Difference degree caused by difference due to correlative factor between day and history day to be predicted is described.And in meteorological data
Using analysis means such as correlation analysis, distance analysis, principal component analysis in processing, the recurrence mould based on Study on Relative Factors is established
Type and regression model based on principal component analysis establish the processing of a full set of correlative factor and mathematics based on Correlative Factor Mapping library
Quantification theory and method.
Further, in step S102, the new short-term load forecasting method of load forecasting method library real-time reception is simultaneously
It is stored in load forecasting method library, and the existing prediction technique in load forecasting method library is improved.Specifically, various
Mathematical model prediction method is the foundation stone of load prediction system work.Any method is a kind of ideal mould mathematically
Type can only have preferable fitting and prediction effect to the rule of development of a certain load.Different regions, are not gone together at different time
The load variations rule of industry is different, and is difficult to describe all load variations rules with one or more of prediction models.Therefore,
The present invention establishes load forecasting method library, and a variety of prediction models meet the multifarious demand of the load rule of development.
Wherein, load forecasting method library is as shown in table 3-4:
Normal day prediction technique is as shown in table 3:
Table 3
Festivals or holidays prediction technique is as shown in table 4:
Table 4
There are many method of short-term load forecasting, such as multiple regression, spectrum analysis, arma modeling, Artificial Neural Network
(ANN) etc..To sum up can be mainly divided into following a few classes:
(1) merely with the method for load development rules: such as arma modeling.
(2) method of the load rule of development in conjunction with meteorologic factor: such as ANN method.
(3) other methods.First kind method only counts historical load data, analyzed, operation, and for other
Relevant information especially the weather information that short term is affected is not accounted for, make the precision of prediction of normal day without
Method further increases, and special weather day but will cause error excessive.This is because being only to reflect very well by historical data
Its following development trend, influence of the meteorologic factor to short term is very big and fails to embody in the algorithm.Second class method
Meteorologic factor is considered, is generally compensated using empirical method using rough weather condition, or is entered as related member
Neural network model calculates, but since the information of use is very little and relevant way is weaker, as a result sometimes unsatisfactory.In addition, should
Class method is not related to the factor other than meteorology generally, and the mode of meter and meteorologic factor is also not flexible.In fact, load prediction
The correlative factor of consideration is absolutely not only meteorologic factor, and should include: day classification (normal day, National Day, Spring Festival etc.);Week
Type (Monday~Sunday);Date is poor (number of days between two days apart);Day weather pattern (fine, negative etc.);Max. daily temperature, day
Mean temperature, Daily minimum temperature;Daily rainfall;Humidity;Wind speed;Etc..With the development of science and technology, it is possible to newly increase it
His correlative factor.Therefore, the present invention considers various correlative factors (being not only meteorologic factor), and prognosticator's construction can be instructed new
Short-term load forecasting method store to load forecasting method library, can also be to existing predictions various in load forecasting method library
Method is transformed, and makes it possible to the influence of meter and various factors.
Further, in the step S103, Comprehensive Model packet is established according to predicting strategy and load forecasting method
It includes: according to each department in Correlative Factor Mapping database, the predicting strategy of correlative factor feature, filtering out different prediction models
Collection;In load forecasting method library on the basis of a variety of prediction techniques, Comprehensive Model is established.
Further, described on the basis of a variety of prediction techniques in load forecasting method library, establish Comprehensive Model packet
It includes: the prediction result of prediction techniques various in load forecasting method library is combined, establish and be based on several Individual forecast methods
Comprehensive Model, the objective law that the Comprehensive Model develops according to different regions load, optimization calculate difference
The weight of prediction technique.As a result, according to each department, every profession and trade the characteristics of, different Comprehensive Models can be constructed, made pre-
Precision is surveyed to be further enhanced.
Further, the combined mode includes optimizing to a variety of prediction models, optimizes the weight of various models, makes pre-
It surveys precision and reaches highest, determine the history confidence level of forecast sample in prediction model, reflect prediction model as far as possible negative in the recent period
The rule of lotus variation, is then combined prediction result.
Further, the Comprehensive Model of the foundation based on several Individual forecast methods includes: according to prediction object
And its characteristic of relevant historical data carries out adaptive training, the suitable method mould of automatic screening from load forecasting method library
Type, by selection and the error analysis of method it can be concluded that being suitble to the Comprehensive Model of prediction object;To further increase
The convenience and model prediction accuracy of system.
Specifically, the adaptive training uses the strategy of virtual prognostication, generated in dress, realized load curve at the beginning of system
Large variation, precision of prediction presentation start study and adaptation process when being decreased obviously automatically, can also usually manually start.
Meanwhile in study, adaptation process, in terms of specific prediction technique error evaluation, empty prediction result error evaluation is used
Method counteracts the deficiency of error of fitting assessment method.Therefore, the present embodiment load forecasting method library adapts to the whole network differently
Area, Various Seasonal load character, improve prediction accuracy provide the foundation and ensure.
Further, all prediction techniques in load forecasting method library, the predicting strategy in Correlative Factor Mapping database
With prediction process be all to user be open state.User can be modified with the load correlative factor in customized prediction model
The parameter of various prediction models adjusts the maps values in Correlative Factor Mapping library, can define the structure of various models, can make by oneself
The different predicting strategy of justice, verifies the calculating process in model predictive process.Sufficient space is provided for user, user is passed through
It tests and is organically combined with forecasting system, to improve and ensure that precision of prediction.
Further, the present embodiment carries out comprehensive error analysis to a variety of prediction technique acquired results, to prediction result
It makes an appraisal, forms comprehensive prediction result accordingly, more relevantly reflect the rule of development of load.All error analysis results
It is stored in the message file that user specifies, for consulting at any time comprising
Variance analysis: point-by-point deviation/maximum deviation/minimum deflection/average deviation/total deviation square;
Residual analysis: point-by-point residual error/maximum residul difference/least residual/mean residual/residue square;
Return Difference analysis: poor/maximum Return Difference/minimum Return Difference of pointwise recurrent/average Return Difference/recurrence square;
Relative error analysis: point-by-point relative error/maximum/minimum relative error/average relative error/relative error is flat
Side;
Fitting precision analysis: the index of correlation/residual standard deviation/coefficient of dispersion;
Gray system error analysis: posteriority difference ratio/small error possibility/point-by-point incidence coefficient/grey relational grade;
Model significance analysis: F statistic/model F threshold value of model;
Index of correlation significance analysis: actual value/index of correlation threshold value of the index of correlation, and
Confidence interval analysis: threshold value/confidence interval of confidence level/correspondence confidence level.
Second aspect of the present invention embodiment as shown in Figure 2 provides a kind of for the negative of load forecasting method described in first aspect
The structural schematic diagram of lotus forecasting system, the system comprises:
For storing the Correlative Factor Mapping database of each department and correlative factor feature, for the negative of Storage Estimation method
He Yucefangfaku and Comprehensive Model.
Further, the Correlative Factor Mapping database includes analysis module 1, and the analysis module 1, which is used to use, reflects
It penetrates influence quantization of the function to the load of each factor, each day and carries out clustering;
The load forecasting method library includes addition modified module 2, and the addition modified module 2 is used to receive new short-term
Load forecasting method and the existing prediction technique in load forecasting method library is improved, modifies the parameter of various prediction models,
Adjust Correlative Factor Mapping library in maps values, define various models structure and customized different predicting strategy;
The Comprehensive Model includes screening module 3, composite module 4 and computing module 5, and the screening module 3 is used for
Each department, correlative factor feature predicting strategy, filter out different prediction model collection;The composite module 4 is used for load
The prediction result of various prediction techniques is combined in prediction technique library;The computing module 5 is used for according to different regions load
The objective law of development, optimization calculate the weight of different prediction techniques.
The embodiment of the present invention fully demonstrates local, load change at that time for the experience adjustments by adaptive training, user
Law, the higher Comprehensive Model combination of prediction accuracy, prognosticator can easily be defined as an application scheme.
And take one by one property assumed name claim (such as: 2 days prediction schemes after Jinan City's heavy rain on June 15).Such scheme recording strategy, side
Just forecast and decision person is compared different predicting strategies, obtains optimal predicting strategy.
It should be noted that system embodiment is corresponding with embodiment of the method, therefore other portions that system embodiment is not described in detail
The related content that may refer to embodiment of the method is divided to obtain, details are not described herein again.
By the description of above embodiments, implements present invention method and its system has the following beneficial effects:
1) embodiment of the present invention is using a variety of prediction models as a result, Comprehensive Model is realized, by various prediction techniques
Prediction result carry out optimal combination, as a result, according to each department, every profession and trade the characteristics of, construct different integrated forecasting moulds
Type makes precision of prediction be further enhanced.
2) embodiment of the present invention uses the characteristics of each department, each season power load demand, selects prediction technique, and construction is pre-
Strategy is surveyed, fitting and the precision of prediction of Comprehensive Model is used to provide the function of automatic modeling for user, is solved previous
The problem of load prediction software package requirement user setting Model Weight.
Wherein, any process described otherwise above in flow chart or herein or method description are construed as, table
Show the module for including the steps that one or more codes for realizing specific logical function or the executable instruction of process, piece
Section or part, and the range of the preferred embodiment of the present invention includes other realization, wherein can not be by shown or discussion
Sequence, including according to related function by it is basic simultaneously in the way of or in the opposite order, to execute function, this should be by this
The embodiment person of ordinary skill in the field of invention is understood.
Although the embodiments of the present invention has been shown and described above, it is to be understood that above-described embodiment is example
Property, it is not considered as limiting the invention, those skilled in the art within the scope of the invention can be to above-mentioned
Embodiment is changed, modifies, replacement and variant.
Claims (10)
1. a kind of load forecasting method, which comprises the steps of:
The correlative factor according to needed for predicting strategy establishes Correlative Factor Mapping database, and the correlative factor includes at least meteorology
Correlative factor;
Corresponding a variety of load forecasting methods are determined according to a variety of load variations rules, and establish load forecasting method library;
Comprehensive Model is established according to predicting strategy and load forecasting method library.
2. load forecasting method as described in claim 1, which is characterized in that the correlative factor according to needed for predicting strategy
Establishing Correlative Factor Mapping database includes: by each department and correlative factor feature, using mapping function to each factor, each day
Load influence quantization carry out clustering;According to difference of the different affecting factors on mapping function, according to data base system
Correlative Factor Mapping database is established in design.
3. load forecasting method as described in claim 1, which is characterized in that load forecasting method library real-time reception is new
Short-term load forecasting method is simultaneously stored in load forecasting method library, and is carried out to the existing prediction technique in load forecasting method library
It improves.
4. load forecasting method as described in claim 1, which is characterized in that described according to predicting strategy and load forecasting method
Establishing Comprehensive Model includes: according to the predicting strategy of each department in Correlative Factor Mapping database, correlative factor feature, sieve
Select different prediction model collection;In load forecasting method library on the basis of a variety of prediction techniques, Comprehensive Model is established.
5. load forecasting method as claimed in claim 4, which is characterized in that described in a variety of prediction sides in load forecasting method library
On the basis of method, establishing Comprehensive Model includes: to carry out the prediction result of prediction techniques various in load forecasting method library
The Comprehensive Model based on several Individual forecast methods is established in combination, and the Comprehensive Model is according to different regions load
The objective law of development, optimization calculate the weight of different prediction techniques.
6. load forecasting method as claimed in claim 5, which is characterized in that the combined mode includes to a variety of prediction moulds
Type optimization, optimize the weight of various models, determine the history confidence level of forecast sample in prediction model, then to prediction result into
Row combination.
7. load forecasting method as claimed in claim 6, which is characterized in that the foundation is based on several Individual forecast methods
Comprehensive Model include: according to prediction object and its relevant historical data characteristic carried out from load forecasting method library it is adaptive
It should train, the suitable method model of automatic screening, by selection and the error analysis of method it can be concluded that being suitble to prediction object
Comprehensive Model.
8. load forecasting method as claimed in claim 7, which is characterized in that the load correlative factor in the prediction model can
It is customized, the parameter of various prediction models is modified, the maps values in Correlative Factor Mapping library is adjusted, defines the knot of various models
Structure, customized different predicting strategy verify the calculating process in model predictive process.
9. a kind of load prediction system for realizing load forecasting method described in claim 1, which is characterized in that the system
Include:
For store each department and correlative factor feature Correlative Factor Mapping database, for Storage Estimation method load it is pre-
Survey method base and Comprehensive Model.
10. load prediction system as claimed in claim 9, which is characterized in that the Correlative Factor Mapping database includes point
Module is analysed, the analysis module is used for the influence quantization using mapping function to the load of each factor, each day and carries out clustering;
The load forecasting method library includes addition modified module, and the addition modified module is pre- for receiving new short term
Survey method and the existing prediction technique in load forecasting method library is improved, modify the parameter of various prediction models, adjusts phase
Maps values in the factor mapping library of pass, define various models structure and customized different predicting strategy;
The Comprehensive Model includes screening module, composite module and computing module, and the screening module is used for each department, phase
The predicting strategy of pass factor feature filters out different prediction model collection;The composite module is used for load forecasting method library
In the prediction results of various prediction techniques be combined;The computing module is used for the objective rule developed according to different regions load
Rule, optimization calculate the weight of different prediction techniques.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810629211.XA CN108985570A (en) | 2018-08-17 | 2018-08-17 | Load prediction method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810629211.XA CN108985570A (en) | 2018-08-17 | 2018-08-17 | Load prediction method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108985570A true CN108985570A (en) | 2018-12-11 |
Family
ID=64540603
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810629211.XA Pending CN108985570A (en) | 2018-08-17 | 2018-08-17 | Load prediction method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108985570A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110071502A (en) * | 2019-04-24 | 2019-07-30 | 广东工业大学 | A kind of calculation method of short-term electric load prediction |
CN110222877A (en) * | 2019-05-17 | 2019-09-10 | 华中科技大学 | A kind of load prediction system and load forecasting method based on customized neural network |
CN110288127A (en) * | 2019-05-31 | 2019-09-27 | 武汉烽火富华电气有限责任公司 | A kind of energy big data processing method |
CN110728401A (en) * | 2019-10-10 | 2020-01-24 | 郑州轻工业学院 | Short-term power load prediction method of neural network based on squirrel and weed hybrid algorithm |
CN111242393A (en) * | 2020-03-09 | 2020-06-05 | 南京中电科能技术有限公司 | Comprehensive short-term power load prediction method based on self-adaptive weight |
CN111461448A (en) * | 2020-04-09 | 2020-07-28 | 云南电网有限责任公司大理供电局 | SageMath-based load prediction model configuration method and computer program product |
CN111695739A (en) * | 2020-06-17 | 2020-09-22 | 广东电网有限责任公司计量中心 | Load prediction method, system and equipment |
CN112561192A (en) * | 2020-12-23 | 2021-03-26 | 上海亿边科技有限公司 | AI artificial intelligence based power load prediction system |
CN113256016A (en) * | 2021-06-04 | 2021-08-13 | 广东电网有限责任公司珠海供电局 | Load prediction method and system based on virtual power plant operation |
CN113327174A (en) * | 2021-04-19 | 2021-08-31 | 国网河北省电力有限公司电力科学研究院 | Distribution transformer load prediction method |
CN113487447A (en) * | 2021-04-19 | 2021-10-08 | 国网河北省电力有限公司电力科学研究院 | Power distribution network load prediction and line loss cause analysis method based on big data |
US20220043180A1 (en) * | 2018-09-28 | 2022-02-10 | Aquanty Inc. | Method and system of real-time simulation and forecasting in a fully-integrated hydrologic environment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103065201A (en) * | 2012-12-19 | 2013-04-24 | 福建省电力有限公司 | Electric load prediction method used for electric power based on factors of temperature and festivals and holidays |
CN103123665A (en) * | 2012-07-31 | 2013-05-29 | 上海交通大学 | Short-term power load forecasting method based on fuzzy clustering similar day |
CN104200277A (en) * | 2014-08-12 | 2014-12-10 | 南方电网科学研究院有限责任公司 | Medium-and-long-term power load prediction model establishment method |
CN105552902A (en) * | 2016-01-25 | 2016-05-04 | 中国电力科学研究院 | Super-short-term forecasting method for load of terminal of power distribution network based on real-time measurement of feeder end |
-
2018
- 2018-08-17 CN CN201810629211.XA patent/CN108985570A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103123665A (en) * | 2012-07-31 | 2013-05-29 | 上海交通大学 | Short-term power load forecasting method based on fuzzy clustering similar day |
CN103065201A (en) * | 2012-12-19 | 2013-04-24 | 福建省电力有限公司 | Electric load prediction method used for electric power based on factors of temperature and festivals and holidays |
CN104200277A (en) * | 2014-08-12 | 2014-12-10 | 南方电网科学研究院有限责任公司 | Medium-and-long-term power load prediction model establishment method |
CN105552902A (en) * | 2016-01-25 | 2016-05-04 | 中国电力科学研究院 | Super-short-term forecasting method for load of terminal of power distribution network based on real-time measurement of feeder end |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20220043180A1 (en) * | 2018-09-28 | 2022-02-10 | Aquanty Inc. | Method and system of real-time simulation and forecasting in a fully-integrated hydrologic environment |
CN110071502A (en) * | 2019-04-24 | 2019-07-30 | 广东工业大学 | A kind of calculation method of short-term electric load prediction |
CN110071502B (en) * | 2019-04-24 | 2022-08-12 | 广东工业大学 | Calculation method for short-term power load prediction |
CN110222877A (en) * | 2019-05-17 | 2019-09-10 | 华中科技大学 | A kind of load prediction system and load forecasting method based on customized neural network |
CN110288127A (en) * | 2019-05-31 | 2019-09-27 | 武汉烽火富华电气有限责任公司 | A kind of energy big data processing method |
CN110728401A (en) * | 2019-10-10 | 2020-01-24 | 郑州轻工业学院 | Short-term power load prediction method of neural network based on squirrel and weed hybrid algorithm |
CN111242393A (en) * | 2020-03-09 | 2020-06-05 | 南京中电科能技术有限公司 | Comprehensive short-term power load prediction method based on self-adaptive weight |
CN111461448A (en) * | 2020-04-09 | 2020-07-28 | 云南电网有限责任公司大理供电局 | SageMath-based load prediction model configuration method and computer program product |
CN111695739A (en) * | 2020-06-17 | 2020-09-22 | 广东电网有限责任公司计量中心 | Load prediction method, system and equipment |
CN112561192A (en) * | 2020-12-23 | 2021-03-26 | 上海亿边科技有限公司 | AI artificial intelligence based power load prediction system |
CN113327174A (en) * | 2021-04-19 | 2021-08-31 | 国网河北省电力有限公司电力科学研究院 | Distribution transformer load prediction method |
CN113487447A (en) * | 2021-04-19 | 2021-10-08 | 国网河北省电力有限公司电力科学研究院 | Power distribution network load prediction and line loss cause analysis method based on big data |
CN113256016A (en) * | 2021-06-04 | 2021-08-13 | 广东电网有限责任公司珠海供电局 | Load prediction method and system based on virtual power plant operation |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108985570A (en) | Load prediction method and system | |
CN109615226B (en) | Operation index abnormity monitoring method | |
CN110363347A (en) | The method of neural network prediction air quality based on decision tree index | |
Vile et al. | Predicting ambulance demand using singular spectrum analysis | |
CN106485262A (en) | A kind of bus load Forecasting Methodology | |
CN110555561A (en) | Medium-and-long-term runoff ensemble forecasting method | |
CN110119845A (en) | A kind of application method of track traffic for passenger flow prediction | |
CN111932402A (en) | Short-term power load bidirectional combined prediction method based on similar day and LSTM | |
Aly et al. | Short-term forecasting for urban water consumption | |
CN104156786A (en) | Non-working day maximum daily load forecasting system considering meteorological multi-factor influences | |
CN113962745A (en) | Sales prediction method and system based on prophet model and big data | |
KR101860457B1 (en) | Method for analyzing weather affect and apparatus for executing the method | |
CN105678406A (en) | Short-term load prediction method based on cloud model | |
CN113326654A (en) | Method and device for constructing gas load prediction model | |
CN112613637A (en) | Method and device for processing charging load | |
CN115097548B (en) | Sea fog classification early warning method, device, equipment and medium based on intelligent prediction | |
CN112465266A (en) | Bus load prediction accuracy analysis method and device and computer equipment | |
CN112785089A (en) | Agent service configuration method and device, electronic equipment and storage medium | |
CN112651534B (en) | Method, device and storage medium for predicting resource supply chain demand | |
CN108154259B (en) | Load prediction method and device for heat pump, storage medium, and processor | |
CN117422374A (en) | Canteen intelligent purchasing method based on big data technology, electronic equipment and medium | |
US20200110196A2 (en) | Decision-making support method for issuing warnings and selection of mitigation actions parameterized by weather-climate decision index based on user preferences | |
CN116050636A (en) | Output prediction method, device, equipment and medium of photovoltaic power station | |
JPH09233700A (en) | Method of evaluating reliability on estimation of day maximum demand power | |
CN105976069A (en) | Regionalism-based prediction system and method for short-term power load of grid region at Guigang |
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: 20181211 |
|
RJ01 | Rejection of invention patent application after publication |