CN109002915A - Photovoltaic plant short term power prediction technique based on Kmeans-GRA-Elman model - Google Patents
Photovoltaic plant short term power prediction technique based on Kmeans-GRA-Elman model Download PDFInfo
- Publication number
- CN109002915A CN109002915A CN201810769372.9A CN201810769372A CN109002915A CN 109002915 A CN109002915 A CN 109002915A CN 201810769372 A CN201810769372 A CN 201810769372A CN 109002915 A CN109002915 A CN 109002915A
- Authority
- CN
- China
- Prior art keywords
- day
- sample
- predicted
- kth
- parameter
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 25
- 238000012549 training Methods 0.000 claims abstract description 20
- 238000013528 artificial neural network Methods 0.000 claims abstract description 17
- 241000196324 Embryophyta Species 0.000 claims description 30
- 210000002569 neuron Anatomy 0.000 claims description 14
- 238000010606 normalization Methods 0.000 claims description 14
- 230000005855 radiation Effects 0.000 claims description 13
- 238000009792 diffusion process Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 238000007619 statistical method Methods 0.000 claims description 5
- 241000208340 Araliaceae Species 0.000 claims description 4
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 4
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 4
- 238000010219 correlation analysis Methods 0.000 claims description 4
- 235000008434 ginseng Nutrition 0.000 claims description 4
- 230000007613 environmental effect Effects 0.000 claims description 3
- 230000006835 compression Effects 0.000 claims description 2
- 238000007906 compression Methods 0.000 claims description 2
- 230000001932 seasonal effect Effects 0.000 abstract description 3
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 10
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 10
- 238000010248 power generation Methods 0.000 description 9
- 238000012545 processing Methods 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000012706 support-vector machine Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- PEDCQBHIVMGVHV-UHFFFAOYSA-N Glycerine Chemical compound OCC(O)CO PEDCQBHIVMGVHV-UHFFFAOYSA-N 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
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 238000000556 factor analysis Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling 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
- G06Q10/04—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
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Game Theory and Decision Science (AREA)
- Primary Health Care (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The photovoltaic plant short term power prediction technique based on Kmeans-GRA-Elman model that the present invention relates to a kind of, comprising: correspond to the meteorologic parameter of period on acquisition photovoltaic plant history daily generated output and weather station daily;Data are pre-processed;It combines improvement Kmeans algorithm to cluster the sample on the day before first day to day to be predicted in history day using six statistical indicators, classification number is determined according to silhouette coefficient;The central point for calculating each cluster Meteorological Characteristics value, judges classification belonging to day to be predicted;Determine the similar day and best similar day of day to be predicted;Determine Elman neural network parameter;Obtain training pattern;The generated output of day to be predicted will be predicted in the parameter sample combination of best similar day and the meteorologic parameter of day to be predicted input training pattern.The present invention can be improved the precision and accuracy of different weather condition short term power prediction of the photovoltaic plant under Various Seasonal.
Description
Technical field
The invention belongs to the short term power Predicting Techniques of photovoltaic plant, are based on Kmeans-GRA- more particularly to one kind
The photovoltaic plant short term power prediction technique of Elman model.
Background technique
In recent years, with the development of the social economy, energy shortage and problem of environmental pollution receive the height of various circles of society
Pay attention to, develop and uses renewable energy and have become the important channel for solving energy and environmental problem.In addition, as electric power needs
The drawbacks such as the growth asked, power grid scale constantly expand, and tradition is extensive, high concentration degree generation investment is at high cost, operation difficulty is big
It is increasingly prominent.In this context, photovoltaic power generation is grown rapidly under the attention of countries in the world.It is horizontal for photovoltaic power generation technology,
Country proposes requirements at the higher level, wherein what is be most difficult to is exactly the prediction of photovoltaic power generation.Since the output of photovoltaic generating system is by too
The influence of positive irradiation intensity and weather conditions, so that photovoltaic generating system output has both randomness and fluctuation.Therefore, photovoltaic is sent out
Electricity is a uncontrolled source to bulk power grid, and randomness and fluctuation can cause impact to influence on power grid.Power grid is in order to full
The demand of sufficient user and the safety for guaranteeing power grid, can formulate corresponding scheduling strategy and plan, and the prediction of photovoltaic power generation can
Effectively power grid is helped to make a plan, conducive to the scheduling of power grid, and can coordinate in real time photo-voltaic power supply and normal power supplies it
Between relationship, promote power grid security, stable operation.
Currently, predicting power of photovoltaic plant method can be divided mainly into indirect prediction method and two kinds of direct forecast methods.Wherein,
Predicted method is connect firstly the need of prediction intensity of solar radiation, then according to intensity of solar radiation predicted value indirect predictions photovoltaic power generation system
The generated output of system.This method needs accurate weather forecast information, and must repeatedly model, and prediction process complexity is cumbersome, difficult
With practical application.Direct forecast methods are by carrying out statistical to photovoltaic generating system history output power and relevant weather factor
Analysis, finds out the relationship of photovoltaic power generation system output power and historical power and meteorologic factor, establishes predicting power of photovoltaic plant
Model, current photovoltaic power prediction mainly use direct forecast methods.
Common direct forecast methods have artificial neural network (artificial neural network, ANN), Ma Erke
Husband's chain, time series method, support vector machines (support vector machines, SVM) scheduling algorithm.Wherein using the widest
General is based on the intelligent algorithm based on ANN and SVM.Wherein the prediction algorithm based on SVM can be solved preferably small
Sample situation, precision is higher, but when carrying out parameter optimization to it using optimization algorithm, training time for needing to grow very much.And it is refreshing
Although easily falling into local minimum through network, due to it is higher fitting with generalization ability and the training time it is shorter, in comparison,
It is better than support vector machines in terms of estimated performance, also has been achieved for more successfully applying at present.
Wherein, Elman neural network is compared with traditional BP neural network, and more one receive feedback letter from hidden layer
Number, for remember hidden layer neuron previous moment output valve undertaking layer so that network to historical data have sensibility,
Increase the ability of network itself processing multidate information.Therefore Elman neural network more better than BP neural network is selected herein
As prediction model.In order to enable model more can Accurate Prediction go out the hair of different weather situation of the photovoltaic plant under Various Seasonal
Electrical power, needs to establish corresponding model according to different Meteorological Characteristics, and by the history day closest with day to be predicted
As the input of model, the precision of prediction can be thus greatly improved.Therefore, the present invention is by using Kmeans++ algorithm knot
GRA algorithm is closed, the training sample and survey of the similar day sample and best similar day sample of day to be predicted respectively as model are found
This input of sample.So can be realized the power generation to photovoltaic plant using Kmeans-GRA-Elman model is improved based on mixing
Power is quick and precisely predicted.
Currently, there is not yet Kmeans-GRA-Elman algorithm will be improved based on mixing in the document and patent published
Research applied to the prediction of photovoltaic plant short term power.
Summary of the invention
In view of this, the object of the present invention is to provide a kind of photovoltaics for improving Kmeans-GRA-Elman model based on mixing
Power station short term power prediction technique.According to six statistical indicators (mean powers, standard deviation in the statistical analysis after normalization
Difference, the coefficient of variation, the coefficient of skew, coefficient of kurtosis and general power) first day in history day is arrived to pre- using Kmeans++ algorithm
Sample on the day before survey day is clustered, and according to silhouette coefficient, determines most suitable cluster classification number.Then day to be predicted is calculated
Meteorologic parameter characteristic value and it is each cluster sample set meteorologic parameter characteristic value central point Euclidean distance, determine to be predicted
Classification belonging to day.GRA algorithm cluster belonging to day to be predicted is utilized then according to the meteorologic parameter characteristic value of day to be predicted
Its similar day and best similar day are chosen in sample set.With the generated output at each moment on the day of similar day sample, the same day
The meteorologic parameters such as illumination, environment temperature, humidity and wind speed and meteorologic parameter one day after be input, when one day after each
The generated output at quarter is to export prediction model of the training based on Elman neural network.With each of the best similar day of this models coupling
The generated output at a moment, meteorologic parameter and the meteorologic parameter of day to be predicted carry out photovoltaic generation power as the input of model
Prediction, predict the generated output at each moment of day to be predicted.
The present invention is realized using following scheme: a kind of photovoltaic plant short term power based on Kmeans-GRA-Elman model
Prediction technique comprising following steps:
Step S1: the meteorology of period is corresponded on acquisition photovoltaic plant history daily generated output and weather station daily
Parameter is combined in conjunction with daily meteorology-power parameter sample is obtained;
Step S2: daily meteorology-power parameter sample combination is pre-processed, abnormal data is removed and carries out normalizing
Change processing;
Step S3: it is combined using six statistical indicators in the statistical analysis after normalization and improves Kmeans algorithm to history
Sample in day on the day before first day to day to be predicted is clustered, and determines classification number according to silhouette coefficient;
Step S4: according to the Meteorological Characteristics value of each cluster sample set;Determine cluster centre position;Utilize Euclidean distance
Judge classification belonging to day to be predicted;
Step S5: according to the Meteorological Characteristics value combination grey correlation analysis GRA algorithm of day to be predicted in the same cluster sample
The similar day and best similar day of day to be predicted are determined in this set;
Step S6: the parameter of Elman neural network is determined;
Step S7: the parameter sample combined training Elman neural network of similar day, constantly modification hidden layer neuron are utilized
Number obtains training pattern;
Step S8: will be in the parameter sample combination of best similar day and the meteorologic parameter of day to be predicted input training pattern
The generated output of day to be predicted is predicted, the output power value at day to be predicted at each moment is obtained.
The advantage of the invention is that can in advance relatively accurately prediction photovoltaic plant following one day per every other hour
The generated output at each moment.Historical sample is clustered with Kmeans++ algorithm, in conjunction with GRA algorithm belonging to the day to be predicted
It clusters and determines its similar day and best similar day in classification, predicted by Elman neural network model, further increase light
The precision and accuracy of the short-term power generation power prediction of overhead utility different weather condition under Various Seasonal.
Detailed description of the invention
Fig. 1 is flow diagram of the invention.
Fig. 2 is the power prediction result 1 of experimental group of the embodiment of the present invention and control group.
Fig. 3 is prediction of each moment error curve 1 of experimental group of the embodiment of the present invention and control group.
Fig. 4 is the power prediction result 2 of experimental group of the embodiment of the present invention and control group.
Fig. 5 is prediction of each moment error curve 2 of experimental group of the embodiment of the present invention and control group.
Fig. 6 is the power prediction result 3 of experimental group of the embodiment of the present invention and control group.
Fig. 7 is prediction of each moment error curve 3 of experimental group of the embodiment of the present invention and control group.
Fig. 8 is the power prediction result 4 of experimental group of the embodiment of the present invention and control group.
Fig. 9 is prediction of each moment error curve 4 of experimental group of the embodiment of the present invention and control group.
Specific embodiment
The present invention will be further described with reference to the accompanying drawings and embodiments.
The present embodiment provides a kind of photovoltaic plant short term power for improving Kmeans-GRA-Elman model based on mixing is pre-
Survey method, flow diagram are as shown in Figure 1.Specifically includes the following steps:
Step S1: the meteorology of period is corresponded on acquisition photovoltaic plant history daily generated output and weather station daily
Parameter, meteorologic parameter include the meteorological factors such as illumination, environment temperature, humidity, wind speed, in conjunction with obtaining daily meteorology-power
The combination of parameter sample;
Step S2: daily meteorology-power parameter sample combination is pre-processed, abnormal data is removed and carries out normalizing
Change processing;
Step S3: it is combined using six statistical indicators in the statistical analysis after normalization and improves Kmeans algorithm to history
Sample in day on the day before first day to day to be predicted is clustered, and determines classification number according to silhouette coefficient;
Step S4: according to the Meteorological Characteristics value of each cluster sample set, the center of each cluster Meteorological Characteristics value is calculated
Point judges classification belonging to day to be predicted using Euclidean distance;
Step S5: according to Meteorological Characteristics value combination grey correlation analysis (GRA) algorithm of day to be predicted in the same cluster
The similar day and best similar day of day to be predicted are determined in sample set;
Step S6: input layer number m, the node in hidden layer b and output layer number of nodes p of Elman neural network are determined
And the every weight and threshold value of initialization Elman neural network;
Step S7: it using the parameter sample combined training Elman neural network of similar day, is constantly modified by " trial and error procedure "
Hidden layer neuron number, obtains training pattern;
Step S8: will be in the parameter sample combination of best similar day and the meteorologic parameter of day to be predicted input training pattern
The generated output of day to be predicted is predicted, the output power value at available day at each moment to be predicted.
Preferably, acquiring the alice springs light that photovoltaic plant used by data is Australia in the present embodiment
Overhead utility, the photovoltaic plant are made of the photovoltaic panel that 22 rated values are 250W, and the rated value of photovoltaic array is 5.5KW,
It is generated electricity by way of merging two or more grid systems by inverter.
In the present embodiment, meteorology described in the step S1-power parameter sample combination includes photovoltaic plant history
The meteorologic parameter of period is corresponded on daily generated output and weather station daily.The parameter sample combination is denoted as (Pki,
Gki, Dki, Tki, Wki, Hki);Wherein, k is the serial number on the date of sample collection, indicates number of days, the integer for arriving K for 1, and K is greater than 1
Integer;At the time of i is sample collection in one day, moment number is indicated, the integer for arriving I for 1, I is the integer greater than 1;PkiFor kth
The power parameter sample at i-th of moment in the combination of its parameter sample;GkiFor i-th of moment in the combination of kth day parameter sample
Global horizontal radiation parameter sample;DkiFor the diffusion levels radiation parameter sample at i-th of moment in the combination of kth day parameter sample
This;TkiFor the environment temperature parameter sample at i-th of moment in the combination of kth day parameter sample;WkiFor kth day parameter sample group
The wind speed parameter sample at i-th of moment in conjunction;HkiFor the relative humidity ginseng at i-th of moment in the combination of kth day parameter sample
Numerical example.
In the present embodiment, sample is pre-processed in the step S2, main includes removal abnormal data and normalizing
Change processing.Removal abnormal data refers to that those days of negative or apparent error occurs in removal historical data.It is normalized specific
Method are as follows: the same parameter sample same moment is mapped in section [0,1] using scale compression method, with power sample P
=(P1i,P2i... Pki... PKi) for, specific mapping equation are as follows:
In formula, y ' indicates the data obtained after normalization, PimaxIndicate the maximum value in i-th of moment of data group P,
PiminIndicate the minimum value in i-th of moment of data group P.
In the present embodiment, six statistical indicators in the statistical analysis after normalization are utilized described in the step S3
The sample on the day before first day to day to be predicted in history day is clustered in conjunction with Kmeans algorithm is improved, according to silhouette coefficient
Determine classification number.Six statistical indicators be denoted as (σk, cvk, skk, kurk, Psumk), wherein k is the day of sample collection
The serial number of phase indicates number of days, the integer for arriving K for 1.For the mean power parameter sample of kth day, σkFor the standard deviation of kth day
Poor parameter sample, cvkFor the coefficient of variation parameter sample of kth day, skkFor the coefficient of skew parameter sample of kth day, kurkFor kth
It coefficient of kurtosis parameter sample, PsumkFor the aggregate power parameters sample of kth day.It is combined after six statistical indicators are normalized
Kmeans++ algorithm is clustered, and determines classification number according to silhouette coefficient s.The cluster situation of s > 0.45 is chosen as suitable poly-
Class result.Wherein, the specific formula of the calculating of the calculating, normalization and silhouette coefficient of parameters sample is as follows.With pre-
Survey September 14th (Australian spring) in 2017,26 days 2 months (Australian summer) in 2018, on March 30th, 2018
For (Australian autumn), on July 29th, 2017 (Australian winter), the best silhouette coefficient that calculates and
Corresponding classification number is as shown in table 1.
In formula,σ, cv, sk, kur, Psum respectively indicate daily mean power, standard deviation, the coefficient of variation, skewness
Coefficient, coefficient of kurtosis and general power.I indicates each moment of sample collection in one day, and I indicates one day total moment.
In formula, x ' indicates the data obtained after normalization, xminAnd xmaxIndicate the minimum value and maximum value of sample array,
yminTake -1, ymaxTake 1.
In formula, s (i) indicates that silhouette coefficient, i indicate that the sample in each cluster sample set, a (i) indicate that sample i is arrived
With the average value of the dissimilar degree of other points in cluster, the minimum of the average dissimilar degree of b (i) expression sample i to other clusters
Value.
The silhouette coefficient and classification number that the day to be predicted of table 1 clusters
2017-9-14 (spring) | 2018-2-26 (summer) | 2018-3-30 (autumn) | 2017-7-29 (winter) | |
Silhouette coefficient | 0.4868 | 0.4925 | 0.4960 | 0.4768 |
Classification number | 3 | 3 | 3 | 3 |
In the present embodiment, the Meteorological Characteristics value according to each cluster sample set is needed in step S4, is calculated each poly-
The central point of class Meteorological Characteristics value judges classification belonging to day to be predicted using Euclidean distance.The Meteorological Characteristics value is denoted as
(Gkmax, Gkmin, Dkmax,Dkmin,Tkmax,Tkmin,Wkmax,Wkmin,Hkmax,Hkmin), wherein k is the sequence on the date of sample collection
Number, indicate number of days, the integer for arriving N for 1.GkmaxAnd GkminFor minimum and maximum global horizontal radiation parameter sample, DkmaxAnd Dkmin
For minimum and maximum diffusion levels radiation parameter sample, TkmaxAnd TkminFor minimum and maximum environment temperature parameter sample, WkmaxWith
WkminFor minimum and maximum wind speed parameter sample, HkmaxAnd HkminFor minimum and maximum relative humidity pa sample.It calculates each poly-
The average value of each characteristic value of class determines cluster centre position.Day to be predicted is calculated using Euclidean distance formula and is each gathered
Day to be predicted is attributed in the smallest cluster by the distance of class central point.Euclidean distance formula is as follows:
In formula, d0cIndicate the Euclidean distance of day to be predicted Yu each cluster, x0Indicate the Meteorological Characteristics value of day to be predicted,
Indicate c-th of cluster centre point.
In the present embodiment, the Meteorological Characteristics value (identical with step S4) according to day to be predicted is needed in the step S5
The similar day of day to be predicted and best similar is determined in the same cluster sample set in conjunction with grey correlation analysis (GRA) algorithm
Day.The degree of association for calculating each sample in day to be predicted and the same cluster sample set, is greater than some threshold value for the degree of association
Date is determined as similar day, the degree of association maximum that day in last 10 days of the cluster sample set is determined as best similar
Day.To predict September 14th (Australian spring) in 2017,26 days 2 months (Australian summer) in 2018,2018 years 3
For month 30 (Australian autumn), on July 29th, 2017 (Australian winter), according to weather forecast obtain this 4
The meteorologic parameter of day is as shown in table 2.The degree of association that this 4 days with respective same category sample are calculated separately according to table 2, by the degree of association
Date greater than some threshold value is determined as similar day, and the degree of association maximum that day in last 10 days of the cluster sample set is true
It is set to best similar day, this 4 days threshold value and best similar day and corresponding best correlation are as shown in table 3.The degree of association
Calculation method is shown below.As shown in Table 3, in the present embodiment, 2017 days September 14th (Australian spring) to be predicted
Best similar day be on September 7th, 2017, optimal relevance angle value be 0.8788.On 2 26th, 2018 day to be predicted, (Australia was big
The summer of Leah) best similar day be on 2 25th, 2018, optimal relevance angle value be 0.8311.2018 3 day to be predicted
The best similar day of the moon (Australian autumn) on the 30th is on March 23rd, 2018, and optimal relevance angle value is 0.9354.2017
The best similar day on July 29, in (Australian winter) is on July 26th, 2017, and optimal relevance angle value is 0.9113.
In formula, riIndicate the degree of association, n indicates characteristic value number;ξiIndicate the incidence coefficient of sample and day to be predicted;It is public
Formula is as follows:
In formula, y (n) indicates the Meteorological Characteristics value after day normalization to be predicted, xi(n) gas after the normalization of history day is indicated
As characteristic value, ρ indicates that resolution ratio, ρ can take 0.5, n to indicate characteristic value number.
2 to be predicted days meteorologic parameters of table
Determine the threshold value and best similar day and corresponding best correlation of similar day the day to be predicted of table 3
Date | 2017-9-14 | 2018-2-26 | 2018-3-30 | 2017-7-29 |
Threshold value | 0.8 | 0.75 | 0.9 | 0.8 |
Best similar day | 2017-9-7 | 2018-2-25 | 2018-3-23 | 2017-7-26 |
Best correlation | 0.8788 | 0.8311 | 0.9354 | 0.9113 |
In the present embodiment, Elman neural network described in the step S6 is specifically configured to: input layer number is
23 neurons, node in hidden layer are 10 neurons, and output layer number of nodes is 11 neurons, the number of iterations 2000
Secondary, other parameters use default setting.Wherein, the formula of hidden layer neuron number is as follows:
In formula, b indicates hidden layer node number, and m indicates input layer number, and p indicates output layer node number.A takes 1
~10.
In the present embodiment, date day to be predicted according to selection is needed in the step S7, it is similar by day to be predicted
The parameter sample combination on the same day and corresponding meteorologic parameter one day after are as input in day sample set, one day after per hour
Output power value as export training Elman neural network model.The input and output combination is denoted as (Pk,
Tkmax,Tkmin,T(k+1)max,T(k+1)min,Pk+1), wherein k is sample collection
The serial number on date indicates number of days, the integer for arriving K for 1.Preceding 13 variables indicate the input variable in input and output combination, finally
One variable indicates output variable.The wherein P in input variablekFor the power parameter sample at each moment of kth day,WithFor the average daily global horizontal radiation parameter sample of kth day and kth+1 day,WithFor kth day and kth+1 day
Average daily diffusion levels radiation parameter sample, TkmaxAnd T(k+1)maxFor kth day and kth+1 day maximum environmental temperature parameter sample,
TkminAnd T(k+1)minFor kth day and kth+1 day minimum environment temperature parameter sample,WithFor kth day and kth+1 day
Average daily wind speed parameter sample,WithFor kth day and kth+1 day average daily relative humidity pa sample, output variable
Pk+1For the output power parameter sample hourly at kth+1 day each moment.Hidden layer can be constantly modified by " trial and error procedure "
Neuron number obtains training pattern.
In the present embodiment, it is needed the combination of the parameter sample of best similar day and day to be predicted in the step S8
Meteorologic parameter utilizes trained model predicting per generated output every other hour to day to be predicted as input.Institute
The input sample setting stated is identical as step S7 setting.
It is corresponding, using all data samples as training sample, best similar day is determined using GRA algorithm, it is best similar
Day sample regards control group as the experiment that test sample inputs;It is clustered using Kmeans++ algorithm, it is true in conjunction with GRA algorithm
Determine similar day and best similar day, similar day sample is inputted as training sample, best similar day sample as test sample
Experimental group is regarded in experiment.Arbitrarily spring, summer, autumn, the winter each 1 day is selected to be tested as day to be predicted, each operation 10 times, experimental group
With the prediction result of control group and prediction of each moment error curve as shown in Fig. 2-Fig. 9, model error index value such as 4 institute of table
Show.Wherein, RMSE is root-mean-square error, and MAPE is mean absolute percentage error, R2For the coefficient of determination, when t is that program is run
Between.RMSE,MAPE,R2Calculation formula it is as follows.The RMSE of day in September, 2017 to be predicted (Australian spring) on the 14th
For 3.5098kW, MAPE 2.4556%, R2For 0.9964, t 58.5124s.On 2 26th, 2018 day (the big benefit of Australia to be predicted
Sub- summer) RMSE be 7.5822kW, MAPE 4.1643%, R2For 0.9883, t 56.4642s.Day 2018 to be predicted
The RMSE on March 30, in (Australian autumn) is 4.3614kW, MAPE 2.7878%, R2It is for 0.9953, t
27.5976s.(Australian winter) RMSE on July 29 2017 day to be predicted be 2.4634kW, MAPE 2.0861%,
R2For 0.9988, t 52.7167s.The MAPE error in four seasons is within 4.5%, R2Reach 0.988 or more.
Preferably, the MAPE average value of experimental group is 2.8735%, and 5.5105% than control group mentions from the point of view of average value
About 3 percentage points high, the RMSE average value of experimental group is 4.4792kW, and the 7.7248kW than control group reduces 3.2456kW, real
Test the R of group2Average value is 0.9947, and 0.0086 is improved than the 0.9861 of control group, and the t average value of experimental group is 48.8227s,
72.7582s than control group reduces 23.9355s.From the point of view of mean square deviation, the MAPE mean square deviation of experimental group is 0.9070%, is compared
About 1 percentage point small according to the 1.9158% of group, the RMSE mean square deviation of experimental group is 2.2095kW, smaller than the 3.1168kW of control group
0.9073kW, the R of experimental group2Mean square deviation is 0.0045, and 0.0099 than control group is small by 0.0054.The MAPE of experimental group is average
Value, RMSE average value and 3 mean square deviations are respectively less than control group, and R2Average value is greater than control group.Therefore, it is improved based on mixing
The photovoltaic plant short term power prediction technique precision of Kmeans-GRA-Elman model is higher, and faster, prediction effect is more preferable for speed.
Wherein
In formula, Pf,iIndicate photovoltaic plant output power predicted value, Pm,iIndicate photovoltaic plant output power measured value,
Indicate that photovoltaic plant day gross output measured value, N indicate the sampling number of photovoltaic power station power generation period.
4 model error index of table
The foregoing is merely presently preferred embodiments of the present invention, all equivalent changes done according to scope of the present invention patent with
Modification, is all covered by the present invention.
Claims (9)
1. a kind of photovoltaic plant short term power prediction technique based on Kmeans-GRA-Elman model, it is characterised in that: including
Following steps:
Step S1: the meteorological ginseng of period is corresponded on acquisition photovoltaic plant history daily generated output and weather station daily
Number is combined in conjunction with daily meteorology-power parameter sample is obtained;
Step S2: daily meteorology-power parameter sample combination is pre-processed, abnormal data is removed and place is normalized
Reason;
Step S3: it is combined using six statistical indicators in the statistical analysis after normalization and improves Kmeans algorithm in history day
Sample on the day before first day to day to be predicted is clustered, and determines classification number according to silhouette coefficient;
Step S4: according to the Meteorological Characteristics value of each cluster sample set;Determine cluster centre position;Judged using Euclidean distance
Classification belonging to day to be predicted;
Step S5: according to the Meteorological Characteristics value combination grey correlation analysis GRA algorithm of day to be predicted in the same cluster sample set
The similar day and best similar day of day to be predicted are determined in conjunction;
Step S6: the parameter of Elman neural network is determined;
Step S7: using the parameter sample combined training Elman neural network of similar day, hidden layer neuron is constantly modified
Number, obtains training pattern;
Step S8: it will be treated in the parameter sample combination of best similar day and the meteorologic parameter of day to be predicted input training pattern
The generated output of prediction day is predicted, the output power value at day to be predicted at each moment is obtained.
2. the photovoltaic plant short term power prediction technique according to claim 1 based on Kmeans-GRA-Elman model,
It is characterized by: meteorology described in the step S1-power parameter sample combination is denoted as (Pki, Gki, Dki, Tki, Wki, Hki);
Wherein, k is the serial number on the date of sample collection, indicates number of days, the integer for arriving K for 1, and K is the integer greater than 1;I is sample in one day
At the time of this acquisition, moment number is indicated, the integer for arriving I for 1, I is the integer greater than 1;PkiFor in the combination of kth day parameter sample
The power parameter sample at i-th of moment;GkiFor the global horizontal radiation parameter at i-th of moment in the combination of kth day parameter sample
Sample;DkiFor the diffusion levels radiation parameter sample at i-th of moment in the combination of kth day parameter sample;TkiFor kth day parameter
The environment temperature parameter sample at i-th of moment in sample combination;WkiFor i-th of moment in the combination of kth day parameter sample
Wind speed parameter sample;HkiFor the relative humidity pa sample at i-th of moment in the combination of kth day parameter sample.
3. the photovoltaic plant short term power prediction technique according to claim 1 based on Kmeans-GRA-Elman model,
It is characterized by: normalized in the step S2 method particularly includes: using scale compression method that same parameter sample is same
A moment is mapped in section [0,1].
4. the photovoltaic plant short term power prediction technique according to claim 1 based on Kmeans-GRA-Elman model,
It is characterized by: six statistical indicators described in the step S3 be denoted as ( σk,cvk,skk,kurk,Psumk), wherein k is
The serial number on the date of sample collection indicates number of days, the integer for arriving K for 1, and K is the integer greater than 1;For the average function of kth day
Rate parameter sample, σkFor the standard deviation parameter sample of kth day, cvkFor the coefficient of variation parameter sample of kth day, skkFor kth day
Coefficient of skew parameter sample, kurkFor the coefficient of kurtosis parameter sample of kth day, PsumkFor the aggregate power parameters sample of kth day;
It combines Kmeans++ algorithm to cluster after six statistical indicators are normalized, classification number is determined according to silhouette coefficient s;
The cluster situation of s > 0.45 is chosen as suitable cluster result, wherein calculating, normalization and the wheel of parameters sample
The specific formula of the calculating of wide coefficient is as follows:
In formula,σ, cv, sk, kur, Psum respectively indicate daily mean power, standard deviation, the coefficient of variation, the coefficient of skew,
Coefficient of kurtosis and general power, i indicate each moment of sample collection in one day, and I indicates one day total moment;
In formula, x ' indicates the data obtained after normalization, xminAnd xmaxIndicate the minimum value and maximum value of sample array, yminTake-
1, ymaxTake 1;
In formula, s (i) indicates that sample silhouette coefficient, a (i) indicate that other put being averaged for dissimilar degree in the sample to same cluster
Value, b (i) indicate the sample to other clusters it is average dissmilarity degree minimum value.
5. the photovoltaic plant short term power prediction technique according to claim 1 based on Kmeans-GRA-Elman model,
It is characterized by: step S4 comprising the following specific steps
Step S41: the Meteorological Characteristics value is denoted as
(Gkmax, Gkmin, Dkmax,Dkmin,Tkmax,Tkmin,Wkmax,Wkmin,Hkmax,Hkmin), wherein k is the date of sample collection
Serial number indicates number of days, the integer for arriving K for 1, and K is the integer greater than 1;GkmaxAnd GkminJoin for minimum and maximum global horizontal radiation
Numerical example, DkmaxAnd DkminFor minimum and maximum diffusion levels radiation parameter sample, TkmaxAnd TkminFor minimum and maximum environment temperature
Spend parameter sample, WkmaxAnd WkminFor minimum and maximum wind speed parameter sample, HkmaxAnd HkminFor minimum and maximum relative humidity ginseng
Numerical example;
Step S42: the average value of each each characteristic value of cluster is calculated, that is, determines cluster centre position;
Step S43: day to be predicted is calculated at a distance from each cluster centre point using Euclidean distance formula, day to be predicted is belonged to
For in the smallest cluster, Euclidean distance formula is as follows:
In formula, d0cIndicate the Euclidean distance of day to be predicted Yu each cluster, x0Indicate the Meteorological Characteristics value of day to be predicted,It indicates
C-th of cluster centre point.
6. the photovoltaic plant short term power prediction technique according to claim 1 based on Kmeans-GRA-Elman model,
It is characterized by: step S5 comprising the following specific steps
Step S1: the degree of association of each sample in day to be predicted and the same cluster sample set, the calculating side of the degree of association are calculated
Method is shown below:
In formula, riIndicate the degree of association, n indicates characteristic value number;ξiIndicate the incidence coefficient of sample and day to be predicted;Its formula is such as
Shown in lower:
In formula, y (n) indicates the Meteorological Characteristics value after day normalization to be predicted, xi(n) meteorology after indicating the normalization of history day is special
Value indicative, ρ indicate that resolution ratio, n indicate characteristic value number.
7. the photovoltaic plant short term power prediction technique according to claim 1 based on Kmeans-GRA-Elman model,
It is characterized by: Elman neural network described in step S6 is specifically configured to: input layer number is 23 neurons, hidden
Number containing node layer is 10 neurons, and output layer number of nodes is 11 neurons, and the number of iterations is 2000 times, and other parameters use
Default setting;Wherein, the formula of hidden layer neuron number is as follows:
In formula, b indicates hidden layer node number, and m indicates input layer number, and p indicates output layer node number;A is constant,
Take 1~10.
8. the photovoltaic plant short term power prediction technique according to claim 1 based on Kmeans-GRA-Elman model,
It is characterized by: step S7 comprising the following specific steps
Step S71: input and output combination is denoted as (Pk,Tkmax,Tkmin, k+1,
T(k+1)max,T(k+1)min,Pk+1), wherein k is the serial number on the date of sample collection, indicates number of days, arrives K's for 1
Integer, K are the integer greater than 1;Preceding 13 variables indicate the input variable in input and output combination, the last one variable indicates defeated
Variable out;The wherein P in input variablekFor the power parameter sample at each moment of kth day,WithFor kth day and
Kth+1 day average daily global horizontal radiation parameter sample,WithFor kth day and kth+1 day average daily diffusion levels spoke
Penetrate parameter sample, TkmaxAnd T(k+1)maxFor kth day and kth+1 day maximum environmental temperature parameter sample, TkminAnd T(k+1)minFor
Kth day and kth+1 day minimum environment temperature parameter sample,WithFor kth day and kth+1 day average daily wind speed ginseng
Numerical example,WithFor kth day and kth+1 day average daily relative humidity pa sample, output variable Pk+1It is kth+1 day
Each moment output power parameter sample hourly;
Step S72: hidden layer neuron number is constantly modified by trial and error procedure, obtains training pattern.
9. the photovoltaic plant short term power prediction technique according to claim 8 based on Kmeans-GRA-Elman model,
It is characterized by: being needed in step S8 using the combination of the parameter sample of best similar day and the meteorologic parameter of day to be predicted as defeated
Enter, utilizes trained model predicting per generated output every other hour to day to be predicted;Input sample setting and step
Rapid S7 setting is identical.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810769372.9A CN109002915B (en) | 2018-07-13 | 2018-07-13 | Photovoltaic power station short-term power prediction method based on Kmeans-GRA-Elman model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810769372.9A CN109002915B (en) | 2018-07-13 | 2018-07-13 | Photovoltaic power station short-term power prediction method based on Kmeans-GRA-Elman model |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109002915A true CN109002915A (en) | 2018-12-14 |
CN109002915B CN109002915B (en) | 2022-03-18 |
Family
ID=64599682
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810769372.9A Active CN109002915B (en) | 2018-07-13 | 2018-07-13 | Photovoltaic power station short-term power prediction method based on Kmeans-GRA-Elman model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109002915B (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109543911A (en) * | 2018-11-29 | 2019-03-29 | 中国农业科学院农业信息研究所 | A kind of solar radiation prediction technique and system |
CN109711609A (en) * | 2018-12-15 | 2019-05-03 | 福州大学 | Photovoltaic plant output power predicting method based on wavelet transformation and extreme learning machine |
CN110009037A (en) * | 2019-04-03 | 2019-07-12 | 中南大学 | A kind of engineering wind speed Forecasting Approach for Short-term and system based on physical message coupling |
CN110148068A (en) * | 2019-05-23 | 2019-08-20 | 福州大学 | One kind being based on meteorological data similarity analysis and LSTM neural fusion photovoltaic plant ultra-short term power forecasting method |
CN110163437A (en) * | 2019-05-23 | 2019-08-23 | 太原理工大学 | Day-ahead photovoltaic power generation power prediction method based on DPK-means |
CN110348648A (en) * | 2019-08-02 | 2019-10-18 | 国网电子商务有限公司 | A kind of predicting power of photovoltaic plant method and device |
CN111091139A (en) * | 2019-11-18 | 2020-05-01 | 特变电工西安电气科技有限公司 | Photovoltaic prediction method, device and equipment for similar day clustering and readable storage medium |
CN112508246A (en) * | 2020-11-26 | 2021-03-16 | 国电南瑞科技股份有限公司 | Photovoltaic power generation power prediction method based on similar days |
CN112668611A (en) * | 2020-12-08 | 2021-04-16 | 湖南工业大学 | Short-term photovoltaic power generation power prediction method based on Kmeans and CEEMD-PE-LSTM |
CN112927097A (en) * | 2021-01-29 | 2021-06-08 | 国网辽宁省电力有限公司阜新供电公司 | Photovoltaic power generation short-term prediction method based on GRA-ABC-Elman model |
CN113313298A (en) * | 2021-05-21 | 2021-08-27 | 上海电力大学 | Feature selection-based hybrid Kmeans-GRA-SVR photovoltaic power generation power prediction method |
CN115660132A (en) * | 2022-08-05 | 2023-01-31 | 科大数字(上海)能源科技有限公司 | Photovoltaic power generation power prediction method and system |
CN116432874A (en) * | 2023-06-14 | 2023-07-14 | 青岛鼎信通讯科技有限公司 | Distributed photovoltaic power prediction method based on characteristic power |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106447098A (en) * | 2016-09-22 | 2017-02-22 | 许昌许继软件技术有限公司 | Photovoltaic ultra-short period power predicting method and device |
CN106779129A (en) * | 2015-11-19 | 2017-05-31 | 华北电力大学(保定) | A kind of Short-Term Load Forecasting Method for considering meteorologic factor |
-
2018
- 2018-07-13 CN CN201810769372.9A patent/CN109002915B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106779129A (en) * | 2015-11-19 | 2017-05-31 | 华北电力大学(保定) | A kind of Short-Term Load Forecasting Method for considering meteorologic factor |
CN106447098A (en) * | 2016-09-22 | 2017-02-22 | 许昌许继软件技术有限公司 | Photovoltaic ultra-short period power predicting method and device |
Non-Patent Citations (4)
Title |
---|
KUK YEOL BAE: "Hourly Solar Irradiance Prediction Based on Support Vector Machine and Its Error Analysis", 《IEEE TRANSAC TIONS ON POWER SYSTEMS》 * |
XIAOYU ZHANG 等: "Short-term forecasting of wind power generation Based on the Similar Day and Elman Neural Network", 《2015 IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE》 * |
XIUYUAN YANG 等: "Day-ahead forecasting of photovoltaic output power with similar cloud space fusion based on incomplete historical data mining", 《APPLIED ENERGY》 * |
李伟 等: "基于气象因子权重相似日的短期光伏功率预测", 《广东电力》 * |
Cited By (20)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109543911A (en) * | 2018-11-29 | 2019-03-29 | 中国农业科学院农业信息研究所 | A kind of solar radiation prediction technique and system |
CN109543911B (en) * | 2018-11-29 | 2020-09-18 | 中国农业科学院农业信息研究所 | Sunlight radiation prediction method and system |
CN109711609A (en) * | 2018-12-15 | 2019-05-03 | 福州大学 | Photovoltaic plant output power predicting method based on wavelet transformation and extreme learning machine |
CN109711609B (en) * | 2018-12-15 | 2022-08-12 | 福州大学 | Photovoltaic power station output power prediction method based on wavelet transformation and extreme learning machine |
CN110009037A (en) * | 2019-04-03 | 2019-07-12 | 中南大学 | A kind of engineering wind speed Forecasting Approach for Short-term and system based on physical message coupling |
CN110009037B (en) * | 2019-04-03 | 2020-10-27 | 中南大学 | Short-term engineering wind speed prediction method and system based on physical information coupling |
CN110148068A (en) * | 2019-05-23 | 2019-08-20 | 福州大学 | One kind being based on meteorological data similarity analysis and LSTM neural fusion photovoltaic plant ultra-short term power forecasting method |
CN110163437A (en) * | 2019-05-23 | 2019-08-23 | 太原理工大学 | Day-ahead photovoltaic power generation power prediction method based on DPK-means |
CN110163437B (en) * | 2019-05-23 | 2022-05-13 | 太原理工大学 | Day-ahead photovoltaic power generation power prediction method based on DPK-means |
CN110348648A (en) * | 2019-08-02 | 2019-10-18 | 国网电子商务有限公司 | A kind of predicting power of photovoltaic plant method and device |
CN111091139A (en) * | 2019-11-18 | 2020-05-01 | 特变电工西安电气科技有限公司 | Photovoltaic prediction method, device and equipment for similar day clustering and readable storage medium |
CN111091139B (en) * | 2019-11-18 | 2024-04-19 | 特变电工西安电气科技有限公司 | Photovoltaic prediction method, device and equipment for similar day clustering and readable storage medium |
CN112508246A (en) * | 2020-11-26 | 2021-03-16 | 国电南瑞科技股份有限公司 | Photovoltaic power generation power prediction method based on similar days |
CN112668611A (en) * | 2020-12-08 | 2021-04-16 | 湖南工业大学 | Short-term photovoltaic power generation power prediction method based on Kmeans and CEEMD-PE-LSTM |
CN112668611B (en) * | 2020-12-08 | 2024-02-02 | 湖南工业大学 | Kmeans and CEEMD-PE-LSTM-based short-term photovoltaic power generation power prediction method |
CN112927097A (en) * | 2021-01-29 | 2021-06-08 | 国网辽宁省电力有限公司阜新供电公司 | Photovoltaic power generation short-term prediction method based on GRA-ABC-Elman model |
CN113313298A (en) * | 2021-05-21 | 2021-08-27 | 上海电力大学 | Feature selection-based hybrid Kmeans-GRA-SVR photovoltaic power generation power prediction method |
CN115660132A (en) * | 2022-08-05 | 2023-01-31 | 科大数字(上海)能源科技有限公司 | Photovoltaic power generation power prediction method and system |
CN115660132B (en) * | 2022-08-05 | 2024-01-30 | 科大数字(上海)能源科技有限公司 | Photovoltaic power generation power prediction method and system |
CN116432874A (en) * | 2023-06-14 | 2023-07-14 | 青岛鼎信通讯科技有限公司 | Distributed photovoltaic power prediction method based on characteristic power |
Also Published As
Publication number | Publication date |
---|---|
CN109002915B (en) | 2022-03-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109002915A (en) | Photovoltaic plant short term power prediction technique based on Kmeans-GRA-Elman model | |
CN104573879B (en) | Photovoltaic plant based on optimal similar day collection goes out force prediction method | |
CN109685257A (en) | A kind of photovoltaic power generation power prediction method based on Support vector regression | |
CN109086928A (en) | Photovoltaic plant realtime power prediction technique based on SAGA-FCM-LSSVM model | |
CN102663513B (en) | Utilize the wind power combined prediction modeling method of grey relational grade analysis | |
CN109902874A (en) | A kind of micro-capacitance sensor photovoltaic power generation short term prediction method based on deep learning | |
CN109858673A (en) | A kind of photovoltaic generating system power forecasting method | |
CN106446494B (en) | Honourable power forecasting method based on wavelet packet-neural network | |
Oudjana et al. | Short term photovoltaic power generation forecasting using neural network | |
CN107528350B (en) | A kind of wind power output typical scene generation method adapting to long -- term generation expansion planning | |
CN110334870B (en) | Photovoltaic power station short-term power prediction method based on gated cyclic unit network | |
CN109978284B (en) | Photovoltaic power generation power time-sharing prediction method based on hybrid neural network model | |
CN110443405A (en) | A kind of built photovoltaic power station power generation amount forecasting system and method | |
CN104463356A (en) | Photovoltaic power generation power prediction method based on multi-dimension information artificial neural network algorithm | |
CN108876013A (en) | One kind being based on best similar day and Elman neural fusion photovoltaic plant short term power prediction technique | |
CN109886567A (en) | A kind of short-term load forecasting method considering sendible temperature and radiation intensity | |
CN109034464A (en) | A kind of method that short-term photovoltaic generating system power prediction and result are checked | |
CN109242180A (en) | Long-medium term power load forecasting method and system | |
CN110909310A (en) | Photovoltaic short-term power generation capacity prediction method and system based on model parameter optimization | |
Luo et al. | Short-term photovoltaic generation forecasting based on similar day selection and extreme learning machine | |
CN110070227A (en) | Migration neural network power prediction method suitable for grid-connected photovoltaic power generation | |
CN112307675B (en) | Neural network-based temperature-sensitive load separation identification method and system | |
Mellit | Sizing of photovoltaic systems: a review | |
Salgado et al. | Hybrid fuzzy clustering neural networks to wind power generation forecasting | |
Madhiarasan et al. | Review of forecasters application to solar irradiance forecasting |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20230330 Address after: 327, Building 1, No.1 Xinzhen Road, Xinwei Town, Xiang'an District, Xiamen City, Fujian Province, 361000 Patentee after: Hengchao Construction Engineering Group Co.,Ltd. Address before: 350108 No. 2 Xueyuan Road, New District of Fuzhou University, Minhou County, Fuzhou City, Fujian Province Patentee before: FUZHOU University |
|
TR01 | Transfer of patent right |