CN106529733A - Distributed photovoltaic output prediction input variable dimensionality reduction method based on Gamma Test and NSGA-II - Google Patents
Distributed photovoltaic output prediction input variable dimensionality reduction method based on Gamma Test and NSGA-II Download PDFInfo
- Publication number
- CN106529733A CN106529733A CN201611026469.8A CN201611026469A CN106529733A CN 106529733 A CN106529733 A CN 106529733A CN 201611026469 A CN201611026469 A CN 201611026469A CN 106529733 A CN106529733 A CN 106529733A
- Authority
- CN
- China
- Prior art keywords
- input variable
- data
- variable
- nsga
- input
- 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 36
- 238000012360 testing method Methods 0.000 title claims abstract description 24
- 238000004458 analytical method Methods 0.000 claims description 16
- 230000001186 cumulative effect Effects 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 7
- 230000005855 radiation Effects 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 230000005021 gait Effects 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 2
- 238000010219 correlation analysis Methods 0.000 claims description 2
- 238000000354 decomposition reaction Methods 0.000 claims description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims 1
- 230000002068 genetic effect Effects 0.000 abstract description 3
- 238000007781 pre-processing Methods 0.000 abstract 1
- 238000010248 power generation Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- 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
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Game Theory and Decision Science (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Development Economics (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Public Health (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Photovoltaic Devices (AREA)
Abstract
A distributed photovoltaic output prediction input variable dimensionality reduction method based on Gamma Test and NSGA-II is provided which comprises step 1 of obtaining an original input variable; a step 2 of preprocessing data; a step 3 of constructing a new variable factor; a step 4 of computing the influence factor of the input variable, wherein the computing comprises using the preprocessed data in the step 2 and the new variable introduced in the step 3 as input variables, estimating the noise variance of the input variables by using Gamma Test, secondly, constructing a fitness function and computing the influence factor of the input variable by using NSGA-II multi-objective genetic algorithm; a step 5 of constructing an input feature vector subjected to dimensionality reduction: determining an input variable subjected to dimensionality reduction according to the influence factor of the input variable in the step 4, and constructing the input feature vector. The method effectively reduces the input variable dimensionality and improves prediction accuracy.
Description
Technical field
The prediction field the invention belongs to distributed photovoltaic is exerted oneself, it is more particularly to a kind of to be based on Gamma Test and NSGA-
The distributed photovoltaic of II is exerted oneself and predicts input variable dimension reduction method.
Background technology
At present, photovoltaic is exerted oneself and still suffers from obvious discontinuity and uncertainty.When distributed photovoltaic system be incorporated to it is public
During electrical network, the disturbance of photovoltaic system may affect the stability of electrical network, and this management and running and Electrical Safety to electrical network is produced
Huge challenge.In order to eliminate the impact of this respect, ground to having done in the prediction of exerting oneself of photovoltaic generating system in a large number both at home and abroad
Study carefully.However, the realization of photovoltaic power generation output forecasting technology also there are problems that it is more.
In order to improve the precision of prediction of model, it is necessary to which introducing is comprehensively input into the factor as far as possible.But the excessive input factor
The problems such as high variable space dimension, information redundancy can be produced again, affects the precision of forecast model.
Therefore, dimensionality reduction is carried out to the input variable of forecasting system how, is urgent need to resolve with the precision for improving forecast model
Problem.
The content of the invention
In order to overcome existing photovoltaic system exert oneself prediction mode input variable dimension is higher, precision of prediction is relatively low not
Foot, the present invention provide it is a kind of it is effective reduce input variable dimension, lifted precision of prediction based on Gamma Test and NSGA-II
Distributed photovoltaic is exerted oneself and predicts input variable dimension reduction method.
The technical solution adopted for the present invention to solve the technical problems is:
A kind of distributed photovoltaic based on Gamma Test and NSGA-II is exerted oneself and predicts input variable dimension reduction method, described
Dimension reduction method is comprised the following steps:
Step 1. is originally inputted variable
The exert oneself variable that is originally inputted of prediction of distributed photovoltaic includes historical power sequence and meteorological data sequence, described to go through
, from photovoltaic plant parameters of electric power detection means, the meteorological data sequence is from photovoltaic weather station for history power data sequence;
Step 2. data prediction
Data prediction is carried out to being originally inputted variable, denoising is carried out to data first, secondly, data are returned
One change is processed;
Step 3. constructs new Variable Factors
New Variable Factors, including assembly temperature, wavelet analysis data and clearness index are introduced, the assembly temperature refers to
The temperature of surface of photovoltaic cell panel, the wavelet analysis data are referred to and are decomposed into photovoltaic plant power output using wavelet analysis method
Low frequency component and high fdrequency component, the clearness index refer to that the actual emanations for inciding horizontal plane and the theory under the conditions of clear sky are radiated
Ratio;
Step 4. calculates the factor of influence of input variable
Using in step 2 in the data and step 3 of pretreatment the new variables that introduces as input variable, initially with
Gamma Test estimate that to the noise variance of input variable next constructs fitness function, using NSGA-II multiple targets
Genetic algorithm is calculated to the factor of influence of input variable;
Input feature value after step 5. construction dimensionality reduction.
Input variable after dimensionality reduction is determined according to the factor of influence of input variable in step 4, input feature value is constructed.
Further, in the step 4, the noise variance step by Gamma Test algorithm construction input variables is as follows:
First, calculate input space midpoint xiWith its k-th closest dot spacing from mean value be,
Wherein, the scope of k is 1≤k≤kmax, xN[i,k]Represent and xiClosest vector, the scope of i is 1≤i≤M;
Secondly, calculating the average distance corresponded between output data is,
Wherein, yN[i,k]It is xiThe corresponding output of k-th closest point;
Finally, constructing linear equation is,
γM(k)=Γ+A δM(k)
Wherein Γ represents the noise variance of input variable.
Further, in the step 4, constructing fitness function is,
Wherein, gΓAnd gAIt is penalty term, is the function of Γ and A respectively, wΓAnd wAIt is designated value, xi' it is input vector.
Further, in the step 5, it is that 0 input variable is rejected first to factor of influence, secondly, will affects
The factor is 1 input variable combination, constructs input feature value, realizes input variable dimensionality reduction.
In the step 1, historical power sequence includes direct solar radiation degree, scattering radiancy, horizontal integrated radiant emittance, inclined plane
Integrated radiant emittance and power output, meteorological data sequence include environment temperature, ambient humidity, atmospheric pressure, air visibility, wind
Speed, wind direction, weather pattern.
In the step 2, the denoising to being originally inputted variable contains data filling and data correction, using linear
Interpolation is filled to missing data, abnormal meteorological data is modified using correlation analysis method.
In the step 2, the data in each input variable are normalized by the following method:
First, variable will be originally inputted and will be expressed as M={ M1,M2,...,Mi, secondly, by the sample under each input variable
Data accumulation is obtainedGenerate cumulative Number Sequence M'={ M1',M2',...,Mk', finally it is normalized place
Reason
In above formula, i is input variable species, M be sample it is cumulative after numerical value, MmaxMaximum after adding up for sample,
MminFor minimum of a value of the sample after cumulative, Y is the numerical value after normalization, YmaxFor the minimum of a value after normalization, YminAfter normalization
Maximum.
In the step 3, the assembly temperature is photovoltaic panel surface temperature, and assembly temperature expression formula is, T=Ta+kGt;
Wherein TaIt is environment temperature, k is empirical coefficient, GtIt is t inclined plane integrated radiant emittance.
In the step 3, introducing wavelet analysis data is decomposed into the power output of photovoltaic plant with wavelet analysis method
Low frequency component and high fdrequency component, i.e., by selecting suitable small echo, reconstruct each frequency component, the high fdrequency component after reconstruct carried out
Classification, such that it is able to be directed to the suitable input variable of different weather type selecting.
In the step 3, the clearness index refers to the theory under the conditions of the actual emanations and clear sky that incide horizontal plane
The ratio of radiation, expression formula is,
Wherein, GtTo incide the actual emanations degree of horizontal plane,Incide the actual emanations of horizontal plane.
Beneficial effects of the present invention are mainly manifested in:
1st, present invention achieves distributed photovoltaic is exerted oneself in prediction input variable dimensionality reduction, it is to avoid the excessive input factor
The problems such as information redundancy of generation, complicated modeling, so as to improve the precision of photovoltaic power generation output forecasting model.
2nd, foundation of the HFS characteristic value that the method is obtained by the use of wavelet analysis as classification, can with simplified model,
Avoid weather forecast not in time, the puzzlement that weather data imperfection is caused.It is used as by introducing radiancy theoretical calculation data outer
The amount of changing, reduces the dependence to the quality of data in historical data and data environment.It is additionally by correlation factor secondary calculating, main
To include Multiple Time Scales clearness index, assembly temperature amendment data and each Wavelet Component.Realization makes full use of data and drop
The purpose of low data degree of coupling.
3rd, the method can be analyzed to the input variable under different weather type, calculated simple, be can be used for chi in short-term
The photovoltaic power generation output forecasting research of degree.
Description of the drawings
Fig. 1 is that being exerted oneself based on the distributed photovoltaic of Gamma Test and NSGA-II for the present invention predicts input variable dimensionality reduction
Method flow diagram.
Fig. 2 is the NSGA-II algorithm flow charts that the present invention is adopted.
Fig. 3 is the photovoltaic power generation output forecasting modeling procedure figure that the present invention is adopted.
Specific embodiment
The invention will be further described below in conjunction with the accompanying drawings.
With reference to Fig. 1~Fig. 3, a kind of distributed photovoltaic based on Gamma Test and NSGA-II is exerted oneself prediction input variable
Dimension reduction method, comprises the following steps:
Step 1, is originally inputted variable
This example selects the Ashland photovoltaic generating system (latitudes positioned at Ore.:42.19 °, longitude:
122.70 °, height above sea level 595m) used as algorithm research object, its total capacity is 15kW.The time scale of prediction is 1 hour, that is, predict
The hourly average power of distributed output of power station.From the historical power number of the parameters of electric power detection device records of photovoltaic generating system
According to having:Direct solar radiation degree, scattering radiancy, horizontal integrated radiant emittance, inclined plane integrated radiant emittance, power output, from photovoltaic weather station
The meteorological data of record has:Environment temperature, ambient humidity, atmospheric pressure, air visibility, wind speed, wind direction, weather pattern.
Step 2, data prediction
Data in step 1 are pre-processed, including denoising and normalized.
First, denoising is carried out to data.Missing data is filled using linear interpolation method, using correlation point
Analysis method is modified to abnormal meteorological data.
Secondly, it is pending data normalization is interval to [- 1,1] by normalized method to being originally inputted variable.
Step 3, constructs new Variable Factors
Introduce new Variable Factors, including assembly temperature, wavelet analysis data and clearness index.
First, construct assembly temperature.Assembly temperature is photovoltaic panel surface temperature, and assembly temperature expression formula is,
T=Ta+kGt
Wherein, TaIt is environment temperature, k is empirical coefficient, GtIt is t inclined plane integrated radiant emittance.
Secondly, wavelet structure analyze data.The photovoltaic plant power output of pretreatment is passed through with step 2 as input, is adopted
Three layers of decomposition are carried out with wavelet analysis method, the high fdrequency component and low frequency component of power output sequence is decomposited, and by reconstructing
To weather associated eigenvalue.Weather conditions can substantially be reflected from high fdrequency component, under the conditions of sunny, HFS is put down very much
Surely, and when weather cloud amount is larger, HFS is changed greatly.
Again, construct clearness index.Clearness index refers to the theory under the conditions of the actual emanations and clear sky that incide horizontal plane
The ratio of radiation, expression formula is,
Wherein, GtTo incide the actual emanations degree of horizontal plane,Incide the actual emanations of horizontal plane.
Step 4, calculates the factor of influence of input variable
The new variables introduced in the data and step 3 of pretreatment using in step 2 first, is adopted as input variable
To the noise variance of input variable, Gamma Test estimate that method is as follows:
Calculate input space midpoint xiWith its k-th closest dot spacing from mean value be,
Wherein, the scope of k is 1≤k≤kmax, xN[i,k]Represent and xiClosest vector, the scope of i is 1≤i≤M.
Calculating the average distance corresponded between output data is,
Wherein, yN[i,k]It is xiThe corresponding output of k-th closest point.
Constructing linear equation is,
γM(k)=Γ+A δM(k)
Wherein Γ represents the noise variance of input variable.
Secondly, construct fitness function.Using following fitness function:
Wherein, gΓAnd gAIt is penalty term, is the function of Γ and A respectively, wΓAnd wAIt is designated value, xi' it is input vector.
Again, the factor of influence of input variable is calculated using NSGA-II multi-objective genetic algorithms.
Calculate and find, factor of influence is that 1 variable has:Direct solar radiation degree, horizontal integrated radiant emittance, hour angle, the little wavelength-division of power
Analysis data, inclined plane integrated radiant emittance, wind speed, power output, clearness index, factor of influence is that 0 variable has:Assembly temperature, ring
Border temperature, ambient humidity, atmospheric pressure, air visibility, wind direction, weather pattern.
Step 5, constructs the input feature value after dimensionality reduction
Input variable after dimensionality reduction is determined according to the factor of influence of input variable in step 4, input feature value is constructed.
Step 6, sets up photovoltaic power generation output forecasting model
According to step 5 construction input feature value as sample sequence, sample is entered using SVM supporting vector machine models
Row training, obtains photovoltaic power generation output forecasting model.The basic process of its prediction is as follows:
First, dependent variable and fruit variable are selected according to model hypothesis, secondly, data is pre-processed, the 3rd, intersection is tested
The optimal parameter that card selection is returned, the 4th, using optimal parameter training SVM, the 5th, prediction is fitted, prediction index is obtained.
Error analysis is carried out to predicting the outcome, is found:The method reality current weather mutation degree is missed with respect to root mean square when relatively low
Difference is respectively 9.7%, 9.1%, 7.8%;Mutation degree is respectively 13.54%, 13.36% when higher, 13.87% prediction essence
Degree.
Finally, in addition it is also necessary to it is noted that listed above is only a specific embodiment of the invention.Obviously, the present invention
Above example is not limited to, there can also be many deformations.One of ordinary skill in the art can be straight from present disclosure
The all deformations derived or associate are connect, protection scope of the present invention is considered as.
Claims (10)
1. a kind of distributed photovoltaic based on Gamma Test and NSGA-II is exerted oneself and predicts input variable dimension reduction method, its feature
It is:The dimension reduction method is comprised the following steps:
Step 1. is originally inputted variable
The exert oneself variable that is originally inputted of prediction of distributed photovoltaic includes historical power sequence and meteorological data sequence, the history work(
, from photovoltaic plant parameters of electric power detection means, the meteorological data sequence is from photovoltaic weather station for rate data sequence;
Step 2. data prediction
Data prediction is carried out to being originally inputted variable, denoising is carried out to data first, secondly, data are normalized
Process;
Step 3. constructs new Variable Factors
New Variable Factors, including assembly temperature, wavelet analysis data and clearness index are introduced, the assembly temperature refers to photovoltaic
The temperature of battery plate surface, the wavelet analysis data refer to photovoltaic plant power output is decomposed into low frequency using wavelet analysis method
Component and high fdrequency component, the clearness index refer to that the actual emanations for inciding horizontal plane and the theory under the conditions of clear sky radiate it
Than;
Step 4. calculates the factor of influence of input variable
The new variables introduced in the data and step 3 for pretreatment through using in step 2 as input variable, initially with Gamma
Test is estimated to the noise variance of input variable, secondly, constructs fitness function, calculated using NSGA-II multi-objective Genetics
Method is calculated to the factor of influence of input variable;
Input feature value after step 5. construction dimensionality reduction
Input variable after dimensionality reduction is determined according to the factor of influence of input variable in step 4, input feature value is constructed.
2. as claimed in claim 1 being exerted oneself based on the distributed photovoltaic of Gamma Test and NSGA-II predicts input variable drop
Dimension method, it is characterised in that:In the step 4, by the noise variance step of Gamma Test algorithm construction input variables such as
Under:
First, calculate input space midpoint xiWith its k-th closest dot spacing from mean value be,
Wherein, the scope of k is 1≤k≤kmax, xN[i,k]Represent and xiClosest vector, the scope of i is 1≤i≤M;
Secondly, calculating the average distance corresponded between output data is,
Wherein, yN[i,k]It is xiThe corresponding output of k-th closest point;
Finally, constructing linear equation is,
γM(k)=Γ+A δM(k)
Wherein Γ represents the noise variance of input variable.
3. as claimed in claim 2 being exerted oneself based on the distributed photovoltaic of Gamma Test and NSGA-II predicts input variable drop
Dimension method, it is characterised in that:In the step 4, constructing fitness function is,
Wherein, gΓAnd gAIt is penalty term, is the function of Γ and A respectively, wΓAnd wAIt is designated value, x 'iFor input vector.
4. being exerted oneself based on the distributed photovoltaic of Gamma Test and NSGA-II as described in one of claims 1 to 3 predicts input
Variable dimension reduction method, it is characterised in that:In the step 5, it is that 0 input variable is rejected first to factor of influence, secondly,
By the input variable combination that factor of influence is 1, input feature value is constructed, input variable dimensionality reduction is realized.
5. being exerted oneself based on the distributed photovoltaic of Gamma Test and NSGA-II as described in one of claims 1 to 3 predicts input
Variable dimension reduction method, it is characterised in that:In the step 1, historical power sequence includes direct solar radiation degree, scattering radiancy, water
Flat integrated radiant emittance, inclined plane integrated radiant emittance and power output, meteorological data sequence include environment temperature, ambient humidity, air
Pressure, air visibility, wind speed, wind direction, weather pattern.
6. being exerted oneself based on the distributed photovoltaic of Gamma Test and NSGA-II as described in one of claims 1 to 3 predicts input
Variable dimension reduction method, it is characterised in that:In the step 2, the denoising to being originally inputted variable contain data filling and
Data correction, is filled to missing data using linear interpolation method, abnormal meteorological data is entered using correlation analysis method
Row amendment.
7. as claimed in claim 6 being exerted oneself based on the distributed photovoltaic of Gamma Test and NSGA-II predicts input variable drop
Dimension method, it is characterised in that:In the step 2, the data in each input variable are normalized into place by the following method
Reason:
First, variable will be originally inputted and will be expressed as M={ M1,M2,...,Mi, secondly, by the sample data under each input variable
Add upGenerate cumulative Number Sequence M'={ M1',M2',...,Mk', finally it is normalized
In above formula, i is input variable species, M be sample it is cumulative after numerical value, MmaxFor maximum of the sample after cumulative, MminFor
Minimum of a value after sample is cumulative, Y is the numerical value after normalization, YmaxFor the minimum of a value after normalization, YminFor after normalization most
It is big to be worth.
8. being exerted oneself based on the distributed photovoltaic of Gamma Test and NSGA-II as described in one of claims 1 to 3 predicts input
Variable dimension reduction method, it is characterised in that:In the step 3, the assembly temperature be photovoltaic panel surface temperature, assembly temperature table
Up to formula it is, T=Ta+kGt;Wherein TaIt is environment temperature, k is empirical coefficient, GtIt is t inclined plane integrated radiant emittance.
9. as claimed in claim 8 being exerted oneself based on the distributed photovoltaic of Gamma Test and NSGA-II predicts input variable drop
Dimension method, it is characterised in that:In the step 3, it is the output with wavelet analysis method by photovoltaic plant to introduce wavelet analysis data
Power Decomposition is low frequency component and high fdrequency component, i.e., by selecting suitable small echo, reconstruct each frequency component, to the height after reconstruct
Frequency component is classified, such that it is able to be directed to the suitable input variable of different weather type selecting.
10. as claimed in claim 8 being exerted oneself based on the distributed photovoltaic of Gamma Test and NSGA-II predicts input variable drop
Dimension method, it is characterised in that:In the step 3, the clearness index refers to the actual emanations and clear sky bar for inciding horizontal plane
The ratio of the theoretical radiation under part, expression formula is,
Wherein, GtTo incide the actual emanations degree of horizontal plane,Incide the actual emanations of horizontal plane.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611026469.8A CN106529733A (en) | 2016-11-22 | 2016-11-22 | Distributed photovoltaic output prediction input variable dimensionality reduction method based on Gamma Test and NSGA-II |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611026469.8A CN106529733A (en) | 2016-11-22 | 2016-11-22 | Distributed photovoltaic output prediction input variable dimensionality reduction method based on Gamma Test and NSGA-II |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106529733A true CN106529733A (en) | 2017-03-22 |
Family
ID=58352674
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611026469.8A Pending CN106529733A (en) | 2016-11-22 | 2016-11-22 | Distributed photovoltaic output prediction input variable dimensionality reduction method based on Gamma Test and NSGA-II |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106529733A (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104866919A (en) * | 2015-05-20 | 2015-08-26 | 天津大学 | Multi-target planning method for power grid of wind farms based on improved NSGA-II |
CN105631528A (en) * | 2015-09-22 | 2016-06-01 | 长沙理工大学 | NSGA-II and approximate dynamic programming-based multi-objective dynamic optimal power flow solving method |
-
2016
- 2016-11-22 CN CN201611026469.8A patent/CN106529733A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104866919A (en) * | 2015-05-20 | 2015-08-26 | 天津大学 | Multi-target planning method for power grid of wind farms based on improved NSGA-II |
CN105631528A (en) * | 2015-09-22 | 2016-06-01 | 长沙理工大学 | NSGA-II and approximate dynamic programming-based multi-objective dynamic optimal power flow solving method |
Non-Patent Citations (2)
Title |
---|
VALDES JULIO等: "Evolutionary Computation Methods for Helicopter Loads Estimation", 《2011 IEEE CONGRESS OF EVOLUTIONARY COMPUTATION》 * |
吴江等: "基于NMF-SVM的光伏系统发电功率短期预测模型", 《华东电力》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Han et al. | Mid-to-long term wind and photovoltaic power generation prediction based on copula function and long short term memory network | |
Han et al. | Quantitative evaluation method for the complementarity of wind–solar–hydro power and optimization of wind–solar ratio | |
CN103996082B (en) | A kind of intensity of solar radiation Forecasting Methodology theoretical based on dual random | |
CN106295899B (en) | Wind power probability density Forecasting Methodology based on genetic algorithm Yu supporting vector quantile estimate | |
CN104050514B (en) | A kind of long-term trend Forecasting Methodology of the wave significant wave height based on analyze data again | |
CN109902874A (en) | A kind of micro-capacitance sensor photovoltaic power generation short term prediction method based on deep learning | |
CN107563565B (en) | A kind of short-term photovoltaic decomposition prediction technique considering Meteorology Factor Change | |
CN107507097A (en) | A kind of short-term wind power prediction method | |
CN103683274B (en) | Regional long-term wind power generation capacity probability prediction method | |
CN103023065B (en) | Wind power short-term power prediction method based on relative error entropy evaluation method | |
CN104376384B (en) | A kind of maximum daily load prediction system of typhoon day analyzed based on electric power big data | |
CN106779154A (en) | Area distribution formula photovoltaic power generation output forecasting method based on satellite cloud picture | |
CN108985521A (en) | Output power predicting method, device, equipment and the medium of photovoltaic generating system | |
CN103559561A (en) | Super-short-term prediction method of photovoltaic power station irradiance | |
CN103489038A (en) | Photovoltaic ultra-short-term power prediction method based on LM-BP neural network | |
CN110263984A (en) | Ultra-short term net load prediction technique based on phase space reconfiguration and deep neural network | |
CN114707688A (en) | Photovoltaic power ultra-short-term prediction method based on satellite cloud chart and space-time neural network | |
CN106548270A (en) | A kind of photovoltaic plant power anomalous data identification method and device | |
CN105678397A (en) | Short-term photovoltaic power prediction method based on improved EMD algorithm and Elman algorithm | |
CN107480781A (en) | The nuclear accident Source Term Inversion method of neutral net adaptive Kalman filter | |
CN117390894B (en) | Method for predicting extreme storm tide level | |
CN111242355A (en) | Photovoltaic probability prediction method and system based on Bayesian neural network | |
CN112381282B (en) | Photovoltaic power generation power prediction method based on width learning system | |
CN110222411A (en) | It is a kind of based on mRMR-DBN algorithm without measure area's solar radiation evaluation method | |
CN113159102A (en) | Multi-time-scale photovoltaic power prediction method and system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | 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: 20170322 |
|
RJ01 | Rejection of invention patent application after publication |