CN110276501A - The prediction technique and device of the electricity price of short term power trade market - Google Patents
The prediction technique and device of the electricity price of short term power trade market Download PDFInfo
- Publication number
- CN110276501A CN110276501A CN201910579144.XA CN201910579144A CN110276501A CN 110276501 A CN110276501 A CN 110276501A CN 201910579144 A CN201910579144 A CN 201910579144A CN 110276501 A CN110276501 A CN 110276501A
- Authority
- CN
- China
- Prior art keywords
- data set
- price
- electricity price
- training
- electricity
- 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
- 230000005611 electricity Effects 0.000 title claims abstract description 152
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000012549 training Methods 0.000 claims abstract description 116
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 35
- 238000003860 storage Methods 0.000 claims abstract description 9
- 238000007637 random forest analysis Methods 0.000 claims description 22
- 238000005457 optimization Methods 0.000 claims description 15
- 238000011160 research Methods 0.000 claims description 15
- 238000012360 testing method Methods 0.000 claims description 12
- 238000013480 data collection Methods 0.000 claims description 2
- 230000005055 memory storage Effects 0.000 claims description 2
- 241001269238 Data Species 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 238000004590 computer program Methods 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 230000014759 maintenance of location Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000013078 crystal Substances 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 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
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0201—Market modelling; Market analysis; Collecting market data
- G06Q30/0206—Price or cost determination based on market factors
-
- 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)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Development Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Water Supply & Treatment (AREA)
- Data Mining & Analysis (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses the prediction technique of the electricity price of short term power trade market, device, computer readable storage medium and electronic equipments, method includes: the training dataset and forecast sample data set for obtaining short term power trade market, and the training dataset includes multiple history electricity price data and the corresponding multiple influence factor data of each history electricity price data；According to a variety of learning algorithms and the training dataset, Ji Yucemoxingji is determined, and determine the corresponding electricity price data set of the base prediction model collection；The training electricity price data set is to determine Price Forecasting；According to the forecast sample data set and the Price Forecasting forecasted electricity market price.According to the technical solution of the present invention, the electricity price of prediction short term power trade market that can be more accurate.
Description
Technical field
The present invention relates to energy technology field more particularly to the prediction techniques and dress of the electricity price of short term power trade market
It sets.
Background technique
Electric Price Forecasting is one of Research on electricity price prediction, and Electric Price Forecasting has certain predictability.
Currently, the main sample data set by obtaining in short term power trade market, according to sample data set and same
Learning algorithm carries out integrated study, determines Price Forecasting, and pass through Price Forecasting forecasted electricity market price.
But the influence factor for influencing electricity price is more, each influence factor is different to the influence degree of electricity price, single study
Algorithm may not be able to accurately reflect the relationship of influence factor and electricity price, and causing may not be able to be accurately pre- by the above method
Survey the electricity price of short term power trade market.
Summary of the invention
The present invention provides a kind of prediction technique of the electricity price of short term power trade market, device, computer-readable storages
Medium and electronic equipment, the electricity price of prediction short term power trade market that can be more accurate.
In a first aspect, the present invention provides a kind of prediction techniques of the electricity price of short term power trade market, comprising:
The training dataset and forecast sample data set of short term power trade market are obtained, the training dataset includes going through
History electricity price data and the corresponding influence factor data of the history electricity price data；
According to a variety of learning algorithms and the training dataset, Ji Yucemoxingji is determined, and determine that the base predicts mould
The corresponding electricity price data set of type collection；
The training electricity price data set is to determine Price Forecasting；
According to the forecast sample data set and the Price Forecasting forecasted electricity market price.
Preferably,
It is described to determine Ji Yucemoxingji according to a variety of learning algorithms and the training dataset, and determine that the base is pre-
Survey the corresponding electricity price data set of Models Sets, comprising:
The training dataset is divided into the first training dataset and the second training dataset according to default hyper parameter；
Ji Yucemoxingji is determined according to first training dataset and a variety of learning algorithms；
The predictive data set of second training dataset is determined according to the Ji Yucemoxingji；
Electricity is formed using each history electricity price data that the predictive data set and second training data are concentrated
Valence data set.
Preferably,
The training electricity price data set is to determine Price Forecasting, comprising:
According to the Random Forest model training electricity price data set, the Random Forest model after training is determined, and will be described
Random Forest model after training is determined as Price Forecasting.
Preferably,
It is described according to the forecast sample data set and the Price Forecasting forecasted electricity market price, comprising:
According to default test data set, optimize the hyper parameter of the Price Forecasting；
According to the super ginseng of the forecast sample data set, the Price Forecasting and the Price Forecasting of optimization
Number forecasted electricity market price.
Second aspect, the present invention provides a kind of prediction meanss of the electricity price of short term power trade market, comprising:
Module is obtained, for obtaining the training dataset and forecast sample data set of short term power trade market, the instruction
Practicing data set includes history electricity price data and the corresponding influence factor data of the history electricity price data；
Data set determining module, for determining Ji Yucemoxingji according to a variety of learning algorithms and the training dataset,
And determine the corresponding electricity price data set of the base prediction model collection；
Model determining module, for training the electricity price data set to determine Price Forecasting；
Research on electricity price prediction module, for according to the forecast sample data set and the Price Forecasting forecasted electricity market price.
Preferably,
The data set determining module, comprising: division unit, model determination unit, the first data set determination unit and
Two data set determination units；Wherein,
The division unit, for according to preset hyper parameter by the training dataset be divided into the first training dataset and
Second training dataset；
The model determination unit, for determining that base predicts mould according to first training dataset and a variety of learning algorithms
Type collection；
The first data set determination unit, for determining second training dataset according to the Ji Yucemoxingji
Predictive data set；
The second data set determination unit, for what is concentrated using the predictive data set and second training data
Each history electricity price data form electricity price data set.
Preferably,
The model determining module, for determining after training according to the Random Forest model training electricity price data set
Random Forest model, and the Random Forest model after the training is determined as Price Forecasting.
Preferably,
The Research on electricity price prediction module, comprising: optimization unit and predicting unit；Wherein,
The optimization unit, for optimizing the hyper parameter of the Price Forecasting according to test data set is preset；
The predicting unit, for according to the forecast sample data set, the Price Forecasting and optimization
The hyper parameter forecasted electricity market price of Price Forecasting.
The third aspect, the present invention provides a kind of computer readable storage mediums, including execute instruction, when electronic equipment
When executing instruction described in processor execution, the processor executes the method as described in any in first aspect.
Fourth aspect, the present invention provides a kind of electronic equipment, including processor and are stored with the storage executed instruction
Device, when executing instruction described in the processor executes memory storage, the processor is executed as in first aspect
Any method.
The present invention provides a kind of prediction technique of the electricity price of short term power trade market, device, computer-readable storages
Medium and electronic equipment, training dataset and default sample data set of this method by acquisition short term power trade market, instruction
Practicing data set includes multiple history electricity price data and the corresponding multiple influence factor data of each history electricity price data, so
Afterwards, it according to a variety of learning algorithms and training dataset, determines Ji Yucemoxingji, and determines the corresponding electricity price of base prediction model collection
Data set, training electricity price data set, later, can be according to forecast sample data set and Research on electricity price prediction to determine Price Forecasting
Model prediction electricity price.In conclusion the Price Forecasting determined has comprehensively considered a variety of learning algorithms, Research on electricity price prediction is improved
The precision and generalization ability of model, the electricity price of prediction short term power trade market that can be more accurate.
Further effect possessed by above-mentioned non-usual preferred embodiment adds hereinafter in conjunction with specific embodiment
With explanation.
Detailed description of the invention
It in order to illustrate the embodiments of the present invention more clearly or existing technical solution, below will be to embodiment or the prior art
Attached drawing needed in description is briefly described, it should be apparent that, the accompanying drawings in the following description is only in the present invention
The some embodiments recorded without any creative labor, may be used also for those of ordinary skill in the art
To obtain other drawings based on these drawings.
Fig. 1 is that a kind of process of the prediction technique of the electricity price for short term power trade market that one embodiment of the invention provides is shown
It is intended to；
Fig. 2 is that a kind of structure of the prediction meanss of the electricity price for short term power trade market that one embodiment of the invention provides is shown
It is intended to；
Fig. 3 is the structure of the prediction meanss of the electricity price for another short term power trade market that one embodiment of the invention provides
Schematic diagram；
Fig. 4 is the structure of the prediction meanss of the electricity price for another short term power trade market that one embodiment of the invention provides
Schematic diagram；
Fig. 5 is the structural schematic diagram for a kind of electronic equipment that one embodiment of the invention provides.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with specific embodiment and accordingly
Technical solution of the present invention is clearly and completely described in attached drawing.Obviously, described embodiment is only a part of the invention
Embodiment, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art are not making wound
Every other embodiment obtained under the premise of the property made labour, shall fall within the protection scope of the present invention.
As described in Figure 1, the embodiment of the invention provides a kind of prediction techniques of the electricity price of short term power trade market, including
Following each step:
Step 101, the training dataset and forecast sample data set of short term power trade market, the training data are obtained
Collection includes multiple history electricity prices and the corresponding multiple influence factor data of each history electricity price；
Step 102, according to a variety of learning algorithms and the training dataset, Ji Yucemoxingji is determined, and described in determination
The corresponding electricity price data set of base prediction model collection；
Step 103, the training electricity price data set is to determine Price Forecasting；
Step 104, according to the forecast sample data set and the Price Forecasting forecasted electricity market price.
Embodiment as shown in Figure 1, the training dataset and default sample that this method passes through acquisition short term power trade market
Notebook data collection, training dataset include multiple history electricity price data and the corresponding multiple influences of each history electricity price data because
Then prime number evidence according to a variety of learning algorithms and training dataset, determines Ji Yucemoxingji, and determine Ji Yucemoxingji
Corresponding electricity price data set, training electricity price data set, later, can be according to forecast sample data sets to determine Price Forecasting
And Price Forecasting forecasted electricity market price.In conclusion the Price Forecasting determined has comprehensively considered a variety of learning algorithms, improve
The precision and generalization ability of Price Forecasting, the electricity price of prediction short term power trade market that can be more accurate.
Specifically, short term power trade market refers specifically to refer to power purchase object and sale of electricity object in electricity transaction market
Zhou Qian, corresponding bid information was submitted a few days ago or before hour, then the operator in electricity transaction market uses market clearing algorithm
Determine conclusion of the business electricity price.The Research on electricity price prediction of short term power trade market includes electricity transaction market environment next week Research on electricity price prediction, a few days ago
Research on electricity price prediction and Research on electricity price prediction before hour, wherein Day-ahead Electricity Price Forecasting Using is that short term power trade market Research on electricity price prediction is most common
Predicting means, technical solution provided in an embodiment of the present invention are particularly suitable for Day-ahead Electricity Price Forecasting Using.
Specifically, influence history electricity price data influence factor data include but is not limited to the moment, temperature value, humidity value,
Gas price data, festivals or holidays, historical load etc..
It is described according to a variety of learning algorithms and the training dataset in one embodiment of the invention, determine that base predicts mould
Type collection, and determine the corresponding electricity price data set of the base prediction model collection, comprising:
The training dataset is divided into the first training dataset and the second training dataset according to default hyper parameter；
Ji Yucemoxingji is determined according to first training dataset and a variety of learning algorithms；
The predictive data set of second training dataset is determined according to the Ji Yucemoxingji；
Electricity is formed using each history electricity price data that the predictive data set and second training data are concentrated
Valence data set.
In the embodiment, training dataset is divided by the first training dataset and the second training number according to default hyper parameter
According to collection, for example, the data number of the first training dataset and the second training dataset can be 4:1, herein, learning framework is
Stacking learning framework, presetting hyper parameter is 2 layers, then, according to a variety of learning algorithms the first training dataset of training, is determined
The corresponding feature (influence factor) of each base prediction model of Ji Yucemoxingji, Ji Yucemoxingji may be different, base prediction
The model quantity of Models Sets is hyper parameter, specifically can be in conjunction with the data volume of actual training dataset and the changing rule of electricity price
Determine the quantity of base prediction model；Every group of influence factor data are concentrated for the second training data, i.e., by each history electricity price number
It is determined as one group of influence factor data according to corresponding multiple influence factor data, it is pre- that this group of influence factor data are substituted into each base
It surveys in model, determines the corresponding prediction data of each base prediction model respectively, later, utilize each prediction data and the second training
Multiple history electricity price data in data form electricity price data set, i.e. electricity price data set includes true electricity price data and forecasted electricity market price
Data, it is clear that, there are corresponding relationships with forecasted electricity market price data for the true electricity price data of every group of influence factor data.
It should be noted that can be selected from multiple learning algorithms during determining base prediction model, herein,
The selection of multiple learning algorithms follows and different principles, for example, a variety of different typess of models of selection (linear model with it is non-thread
Property model, reduce buggy model and reduce Tobin's mean variance model etc.), or different hyper parameters of selection same model, or using different
Random seed selects learning algorithm so that learning algorithm has randomness, the multiplicity of the learning algorithm of guarantee base prediction model
Property, correspondingly, the precision and generalization ability of Price Forecasting can be improved, herein, a variety of learning algorithms can be in conjunction with practical
The data volume of training dataset and the changing rule of electricity price data determine a variety of learning algorithms.Learning algorithm includes but is not limited to
Neural network algorithm and decision Tree algorithms etc..
For example, it is illustrated with an influence factor data instance, giving sample total is N, the first training data
Integrate as X_train_1={ (xi, yi), i=1,2 ..., m }, the second sample data set is X_train_2={ (xj, yj), j=m+
1, m+2 ..., N }, wherein xiCharacterize the influence factor data of i-th of first training datas, xjCharacterize j-th of second training datas
Influence factor data, yiCharacterize the electricity price value of i-th of first training datas, yjCharacterize the electricity price of j-th of second training datas
Value, then, trains n basic mode type according to the first training dataset X-train_1, the second training dataset X_train_2
In influence factor data n basic mode type to be entered obtains predicted value Z respectivelyJ, n, ZjnIt characterizes j-th of second training datas and substitutes into n-th
The predicted value of a basic mode type, is indicated by following table:
Electricity price data in above-mentioned prediction data and X_train_2 are formed into electricity price data set { (Z togetherJ, 1, ZJ, 2...,
ZJ, n, yj), j=m+1, m+2 ..., N }, above-mentioned electricity price data set is trained using random forest or other machines learning algorithm,
To determine Price Forecasting, the predicted value that is obtained using the Price Forecasting.It will be apparent that above-mentioned example is one
In the case where influence factor data, in practical business scene, influence factor data can be multiple, for example, can be the moment,
The influence factor data of the other influences predicted values such as temperature value, humidity value, gas price data, festivals or holidays.
In one embodiment of the invention, the training electricity price data set is to determine Price Forecasting, comprising:
According to the Random Forest model training electricity price data set, the Random Forest model after training is determined, and will be described
Random Forest model after training is determined as Price Forecasting.
Specifically, using Random Forest model training electricity price data set, that is, it can determine the Random Forest model after training, and
Random Forest model after training is determined as Price Forecasting, determining Price Forecasting precision with higher and general
Change ability.It is of course also possible to use other model training electricity price data sets.
It should be noted that the integrated study frame of above-mentioned determining Price Forecasting is 2 layers based on heterogeneous classifier
Stacking learning framework, first layer Ji Yucemoxingji, the second layer are Random Forest model.Heterogeneous classifier can be improved
The precision and generalization ability of Price Forecasting.
It is described that electricity is predicted according to the forecast sample data set and the Price Forecasting in one embodiment of the invention
Valence, comprising:
According to default test data set, optimize the hyper parameter of the Price Forecasting；
According to the super ginseng of the forecast sample data set, the Price Forecasting and the Price Forecasting of optimization
Number forecasted electricity market price.
In the embodiment, in order to further ensure the precision and generalization ability of Price Forecasting, need to survey by default
The hyper parameter for trying data set optimization Price Forecasting, later, according to forecast sample data set, Price Forecasting and optimization
The hyper parameter of Price Forecasting can forecasted electricity market price.
It should be noted that test set can test Price Forecasting, the quality of Price Forecasting is evaluated,
Correspondingly, can use test optimization electricity price test model when Price Forecasting is when the performance of test set is not good enough, with
Optimize the hyper parameter of electricity price test model.
Based on design identical with embodiment of the present invention method, referring to FIG. 2, the embodiment of the invention also provides a kind of short
The prediction meanss of the electricity price in phase electricity transaction market, comprising:
Module 201 is obtained, it is described for obtaining the training dataset and forecast sample data set of short term power trade market
Training dataset includes multiple history electricity price data and the corresponding multiple influence factor numbers of each history electricity price data
According to；
Data set determining module 202, for determining base prediction model according to a variety of learning algorithms and the training dataset
Collection, and determine the corresponding electricity price data set of the base prediction model collection；
Model determining module 203, for training the electricity price data set to determine Price Forecasting；
Research on electricity price prediction module 204, for according to the forecast sample data set and the Price Forecasting forecasted electricity market price.
Referring to FIG. 3, in one embodiment of the invention, the data set determining module 202, comprising: division unit 2021,
Model determination unit 2022, the first data set determination unit 2023 and the second data set determination unit 2024；Wherein,
The division unit 2021, for the training dataset to be divided into the first training data according to default hyper parameter
Collection and the second training dataset；
The model determination unit 2022, for determining that base is pre- according to first training dataset and a variety of learning algorithms
Survey Models Sets；
The first data set determination unit 2023, for determining the second training number according to the Ji Yucemoxingji
According to the predictive data set of collection；
The second data set determination unit 2024, for utilizing the predictive data set and second training dataset
In each history electricity price data form electricity price data set.
In one embodiment of the invention, the model determining module 203, for according to the Random Forest model training electricity
Valence data set, the Random Forest model after determining training, and the Random Forest model after the training is determined as Research on electricity price prediction
Model.
Referring to FIG. 4, in one embodiment of the invention, the Research on electricity price prediction module 204, comprising: optimization unit 2041 and
Predicting unit 2042；Wherein,
The optimization unit 2041, for optimizing the hyper parameter of the Price Forecasting according to test data set is preset；
The predicting unit 2042, for according to the forecast sample data set, the Price Forecasting and optimization
The hyper parameter forecasted electricity market price of the Price Forecasting.
Fig. 5 is the structural schematic diagram of a kind of electronic equipment provided in an embodiment of the present invention.In hardware view, the electronic equipment
Including processor 501 and it is stored with the memory 502 executed instruction, optionally further comprising internal bus 503 and network interface
504.Wherein, memory 502 may include memory 5021, such as high-speed random access memory (Random-Access
Memory, RAM), it is also possible to it further include nonvolatile memory 5022 (non-volatile memory), for example, at least 1 magnetic
Disk storage etc.；Processor 501, network interface 504 and memory 502 can be connected with each other by internal bus 503, inside this
Bus 503 can be ISA (Industry Standard Architecture, industry standard architecture) bus, PCI
(Peripheral Component Interconnect, Peripheral Component Interconnect standard) bus or EISA (Extended
Industry Standard Architecture, expanding the industrial standard structure) bus etc.；Internal bus 503 can be divided into ground
Location bus, data/address bus, control bus etc., only to be indicated with a four-headed arrow in Fig. 5, it is not intended that only convenient for indicating
There are a bus or a type of bus.Certainly, which is also possible that hardware required for other business.Work as place
It manages device 501 and executes when executing instruction of the storage of memory 502, processor 501 executes the side in any one embodiment of the invention
Method, and at least for executing method as shown in Figure 1.
In a kind of mode in the cards, processor reads corresponding execute instruction to interior from nonvolatile memory
It is then run in depositing, can also obtain from other equipment and execute instruction accordingly, to form a kind of short-term electricity on logic level
The prediction meanss of the electricity price of power trade market.What processor execution memory was stored executes instruction, to pass through the execution executed
A kind of prediction technique of the electricity price of the short term power trade market provided in any embodiment of the present invention is realized in instruction.
Processor may be a kind of IC chip, the processing capacity with signal.During realization, the above method
Each step can be completed by the instruction of the integrated logic circuit of the hardware in processor or software form.Above-mentioned processing
Device can be general processor, including central processing unit (Central Processing Unit, CPU), network processing unit
(Network Processor, NP) etc.；Can also be digital signal processor (Digital Signal Processor,
DSP), specific integrated circuit (Application Specific Integrated Circuit, ASIC), field programmable gate
Array (Field-Programmable GateArray, FPGA) either other programmable logic device, discrete gate or crystal
Pipe logical device, discrete hardware components.It may be implemented or execute the disclosed each method in the embodiment of the present invention, step and patrol
Collect block diagram.General processor can be microprocessor or the processor is also possible to any conventional processor etc..
The embodiment of the invention also provides a kind of computer readable storage mediums, including execute instruction, when electronic equipment
When processor executes instruction, the processor executes the method provided in any one embodiment of the invention.The electronics is set
It is standby specifically to can be electronic equipment as shown in Figure 5；Execute instruction be a kind of short term power trade market electricity price prediction dress
Set corresponding computer program.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method or computer program product.
Therefore, the form that complete hardware embodiment, complete software embodiment or software and hardware combine can be used in the present invention.
Various embodiments are described in a progressive manner in the present invention, same and similar part between each embodiment
It may refer to each other, each embodiment focuses on the differences from other embodiments.Implement especially for device
For example, since it is substantially similar to the method embodiment, so being described relatively simple, related place is referring to embodiment of the method
Part illustrates.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the boiler that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or boiler intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including described want
There is also other identical elements in the process, method of element, commodity or boiler.
The above description is only an embodiment of the present invention, is not intended to restrict the invention.For those skilled in the art
For, the invention may be variously modified and varied.All any modifications made within the spirit and principles of the present invention are equal
Replacement, improvement etc., should be included within scope of the presently claimed invention.
Claims (10)
1. a kind of prediction technique of the electricity price of short term power trade market characterized by comprising
The training dataset and forecast sample data set of short term power trade market are obtained, the training dataset includes history electricity
Valence mumber evidence and the corresponding influence factor data of the history electricity price data；
According to a variety of learning algorithms and the training dataset, Ji Yucemoxingji is determined, and determine the Ji Yucemoxingji
Corresponding electricity price data set；
The training electricity price data set is to determine Price Forecasting；
According to the forecast sample data set and the Price Forecasting forecasted electricity market price.
2. the method according to claim 1, wherein
It is described to determine Ji Yucemoxingji according to a variety of learning algorithms and the training dataset, and determine that the base predicts mould
The corresponding electricity price data set of type collection, comprising:
The training dataset is divided into the first training dataset and the second training dataset according to default hyper parameter；
Ji Yucemoxingji is determined according to first training dataset and a variety of learning algorithms；
The predictive data set of second training dataset is determined according to the Ji Yucemoxingji；
Electricity price number is formed using each history electricity price data that the predictive data set and second training data are concentrated
According to collection.
3. the method according to claim 1, wherein
The training electricity price data set is to determine Price Forecasting, comprising:
According to the Random Forest model training electricity price data set, Random Forest model after determining training, and by the training
Random Forest model afterwards is determined as Price Forecasting.
4. method according to any one of claims 1 to 3, which is characterized in that
It is described according to the forecast sample data set and the Price Forecasting forecasted electricity market price, comprising:
According to default test data set, optimize the hyper parameter of the Price Forecasting；
Hyper parameter according to the forecast sample data set, the Price Forecasting and the Price Forecasting of optimization is pre-
Survey electricity price.
5. a kind of prediction meanss of the electricity price of short term power trade market characterized by comprising
Module is obtained, for obtaining the training dataset and forecast sample data set of short term power trade market, the trained number
It include history electricity price data and the corresponding influence factor data of the history electricity price data according to collection；
Data set determining module, for determining Ji Yucemoxingji according to a variety of learning algorithms and the training dataset, and really
Determine the corresponding electricity price data set of the base prediction model collection；
Model determining module, for training the electricity price data set to determine Price Forecasting；
Research on electricity price prediction module, for according to the forecast sample data set and the Price Forecasting forecasted electricity market price.
6. device according to claim 5, which is characterized in that
The data set determining module, comprising: division unit, model determination unit, the first data set determination unit and the second number
According to collection determination unit；Wherein,
The division unit, for the training dataset to be divided into the first training dataset and second according to default hyper parameter
Training dataset；
The model determination unit, for determining base prediction model according to first training dataset and a variety of learning algorithms
Collection；
The first data set determination unit, for determining the pre- of second training dataset according to the Ji Yucemoxingji
Measured data collection；
The second data set determination unit, it is each for being concentrated using the predictive data set and second training data
The history electricity price data form electricity price data set.
7. device according to claim 5, which is characterized in that
The model determining module, for determining random after training according to the Random Forest model training electricity price data set
Forest model, and the Random Forest model after the training is determined as Price Forecasting.
8. according to the device any in claim 5 to 7, which is characterized in that
The Research on electricity price prediction module, comprising: optimization unit and predicting unit；Wherein,
The optimization unit, for optimizing the hyper parameter of the Price Forecasting according to test data set is preset；
The predicting unit, for according to the forecast sample data set, the Price Forecasting and the electricity price of optimization
The hyper parameter forecasted electricity market price of prediction model.
9. a kind of computer readable storage medium, including execute instruction, executed when the processor of electronic equipment described in execute instruction
When, the processor executes the method as described in any in Claims 1-4.
10. a kind of electronic equipment including processor and is stored with the memory executed instruction, described in processor execution
When executing instruction described in memory storage, the processor executes the method as described in any in Claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910579144.XA CN110276501A (en) | 2019-06-28 | 2019-06-28 | The prediction technique and device of the electricity price of short term power trade market |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910579144.XA CN110276501A (en) | 2019-06-28 | 2019-06-28 | The prediction technique and device of the electricity price of short term power trade market |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110276501A true CN110276501A (en) | 2019-09-24 |
Family
ID=67962575
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910579144.XA Pending CN110276501A (en) | 2019-06-28 | 2019-06-28 | The prediction technique and device of the electricity price of short term power trade market |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110276501A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110888857A (en) * | 2019-10-14 | 2020-03-17 | 平安科技（深圳）有限公司 | Data label generation method, device, terminal and medium based on neural network |
| CN112308335A (en) * | 2020-11-12 | 2021-02-02 | 南方电网能源发展研究院有限责任公司 | Short-term electricity price prediction method and device based on xgboost algorithm |
-
2019
- 2019-06-28 CN CN201910579144.XA patent/CN110276501A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110888857A (en) * | 2019-10-14 | 2020-03-17 | 平安科技（深圳）有限公司 | Data label generation method, device, terminal and medium based on neural network |
| CN110888857B (en) * | 2019-10-14 | 2023-11-07 | 平安科技（深圳）有限公司 | Data tag generation method, device, terminal and medium based on neural network |
| CN112308335A (en) * | 2020-11-12 | 2021-02-02 | 南方电网能源发展研究院有限责任公司 | Short-term electricity price prediction method and device based on xgboost algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20160239368A1 (en) | Systems and methods for detecting anomalies | |
| US11551244B2 (en) | Nowcasting abstracted census from individual customs transaction records | |
| TWI690865B (en) | Transaction volume prediction method and device | |
| Chan et al. | Performance optimization of a leagility inspired supply chain model: a CFGTSA algorithm based approach | |
| CN110400022A (en) | Self-help teller machine cash dosage prediction technique and device | |
| CN109992866A (en) | Training method, device, readable medium and the electronic equipment of load forecasting model | |
| CN110276501A (en) | The prediction technique and device of the electricity price of short term power trade market | |
| CN110222871A (en) | The selection method and device of similar day for load forecast | |
| CN110069676A (en) | Keyword recommendation method and device | |
| CN107748938A (en) | A kind of electric power demand forecasting method based on Vector Autoression Models | |
| CN110599307A (en) | Commodity recommendation method and device | |
| CN109670648A (en) | The training of multi-energy data prediction model, the prediction technique of multi-energy data and device | |
| CN110135655A (en) | It is a kind of for determine energy source station operation control strategy method and apparatus | |
| CN107644047A (en) | Tag Estimation generation method and device | |
| CN109559218A (en) | A kind of determination method, apparatus traded extremely and storage medium | |
| CN109726861A (en) | A kind of optimization method and system of electric power application data | |
| CN106127551A (en) | Item recommendation method and device | |
| CN110288250A (en) | Pricing method, device, readable medium and the electronic equipment of electricity price | |
| CN116050987A (en) | Prediction method and device for material replenishment, storage medium and processor | |
| CN110390434A (en) | The method and device of Electric Price Forecasting | |
| CN109360041A (en) | Trade company's methods of exhibiting, device, electronic equipment and storage medium | |
| TWM620099U (en) | Server for predicting category tag of browsed website address | |
| Azadeh et al. | Optimum production planning of a dedicated remanufacturing process by an integrated simulation-based optimisation approach | |
| CN110110892A (en) | Power consumer load forecasting model selection method and device | |
| CN108648016A (en) | The audient's model method for digging and device of bank finance product |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190924 |