CN108364236A - A kind of data Method of Stochastic of sunlight irradiation intensity - Google Patents
A kind of data Method of Stochastic of sunlight irradiation intensity Download PDFInfo
- Publication number
- CN108364236A CN108364236A CN201810107185.4A CN201810107185A CN108364236A CN 108364236 A CN108364236 A CN 108364236A CN 201810107185 A CN201810107185 A CN 201810107185A CN 108364236 A CN108364236 A CN 108364236A
- Authority
- CN
- China
- Prior art keywords
- irradiation intensity
- data
- sunlight irradiation
- stochastic
- day
- 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
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
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Abstract
The invention discloses a kind of data Method of Stochastic of sunlight irradiation intensity, carry out data acquisition for the sunlight irradiation intensity of website, are pre-processed to gathered data, establish sunlight irradiation intensity data library；To treated, sunlight irradiation intensity data carries out feature extraction；Daily feature is clustered using clustering algorithm, each class corresponds to a kind of sunlight irradiation weather condition of the suitable photovoltaic generation degree of characterization, respectively fine day, cloudy day, rainy day, cloudy four kinds of weather patterns；The result of cluster is constituted into time series as unit of day, obtain it is all kinds of between probability transfer matrix；Based on probability transfer matrix and sunlight irradiation intensity data library data stochastic simulation is realized using randomizer.The present invention can utilize existing sunlight irradiation intensity data, and feature extraction and cluster are carried out to it, to simulate magnanimity sunlight irradiation intensity data, play an important roll for the mass data analysis of photovoltaic generation sunlight irradiation condition.
Description
Technical field
The invention belongs to photovoltaic generation meteorological datas to emulate field, be related to a kind of random mould of data of sunlight irradiation intensity
Quasi- method.
Background technology
Solar energy resources are abundant at present, and new energy development potentiality is huge.And since photovoltaic generation is contributed obviously by sunlight
Irradiation intensity influences, so photovoltaic generation is contributed with uncertain and randomness, this gives electric system after large-scale grid connection
More uncertain factors are brought, the planning of electric system, simulation analysis, Optimum utilization etc. are had an important influence on.
Therefore, for sunlight irradiation intensity, this meteorological condition is analyzed, the prediction that can be contributed to photovoltaic generation
Important evidence is provided, to realizing that the maximization consumption of clean energy resource is of great significance.And current sunlight irradiation intensity
Method of Stochastic is difficult to accurately portray and describe the fluctuation condition of photovoltaic generation output, and the present invention proposes a kind of sun thus
The data Method of Stochastic of light intensity provides data for the analysis that photovoltaic generation is contributed and supports.
Invention content
It is an object of the invention to overcome the above-mentioned prior art, provide a kind of data of sunlight irradiation intensity with
Machine analogy method simulates sunlight spoke based on the sunlight irradiation intensity data that website detects in longer period of time
According to the method for intensity data.
In order to achieve the above objectives, the present invention is achieved by the following scheme：
A kind of data Method of Stochastic of sunlight irradiation intensity, includes the following steps：
1) the irradiation intensity data more than 1 year are obtained by the sunlight irradiation intensity collection to website；
2) the irradiation intensity data of acquisition are pre-processed, it is strong to obtain irradiation for the excessive abnormal point removal of the amount of will deviate from
Degrees of data library；
3) feature extraction is carried out to irradiation intensity database, using daily feature vector as input, utilizes clustering algorithm
Clustering is carried out, obtains four classes cluster as a result, respectively fine day, cloudy day, rainy day and cloudy four kinds of weather patterns, correspondence are different
Photovoltaic generation sunlight irradiation weather condition；
4) daily cluster result is constituted into time series, obtain it is all kinds of between probability transfer matrix；
5) it is strong to simulate sunlight irradiation using randomizer by irradiation intensity database and probability transfer matrix
Degrees of data.
The present invention, which further improves, to be：
Selected characteristic includes in step 2)：For daily data xN, N is the quantity of daily sampled point, mean value：
Standard deviation：
Kurtosis：
The degree of bias：
Maximum value：
H=(Max (x (i)) |I=1~N)
Cymomotive force：
MFIN=AFMN×RFCN。
Cymomotive force MFINFor same day average fluctuation margin AFMNWith reversed fluctuation number RFCNProduct.
Average fluctuation margin AFMNFor the sum of the absolute value of amplitude of variation at same day whole consecutive number strong point.
Reversed fluctuation number RFCNThere is reversed total degree for the same day all adjacent fluctuations.
By daily cluster result, makeup time sequence in step 4), wherein cluster result is weather pattern label；When
Between in sequence adjacent label on the two constitute two tuples：
(i, j) i, j ∈ { fine day, cloudy day, the rainy day, cloudy }
The frequency for counting each two tuple, is calculated the probability P (i, j) to each weather pattern of next day, and generating probability turns
Move matrix P：
For the random number that the randomizer in step 5) generates, an initial mark is obtained using initial random number
Label obtain the corresponding section of corresponding two tuple of the random number using follow-up random number, and state is shifted by the first item of two tuples
To Section 2, successive ignition generates the stochastic simulation time series being made of label；By the time series, according to generating random number
Device randomly selects data on the one as analogue data in irradiation intensity database, and all moment points of traversal time sequence obtain
The sunlight irradiation intensity data simulated.
Compared with prior art, the invention has the advantages that：
Sunlight irradiation intensity initial data is carried out feature extraction by the present invention, is inputted, is reduced using feature as cluster
Input quantity dimension promotes computational efficiency；A characteristic quantity present invention introduces cymomotive force as initial data is shown daily
The fluctuation of data enables input quantity more fully express the essential attribute of data；The present invention can be to sunlight irradiation intensity number
Data with existing library is enriched, while ensure that analogue data in the case where data with existing is relatively fewer according to stochastic simulation is carried out
Still there are the features of initial data, and data basis is provided for the mass data analysis of photovoltaic generation.
Description of the drawings
Fig. 1 is the data Method of Stochastic block diagram of sunlight irradiation intensity of the present invention.
Specific implementation mode
The present invention is described in further detail below in conjunction with the accompanying drawings：
Referring to Fig. 1, the data Method of Stochastic of sunlight irradiation intensity of the present invention includes the following steps：
1) the irradiation intensity data more than 1 year are obtained by the sunlight irradiation intensity collection to website；
2) the irradiation intensity data obtained to step 1) pre-process, and the excessive abnormal point removal of the amount of will deviate from is irradiated
Intensity data library；Selected characteristic includes：For daily data xN(N is the quantity of daily sampled point), mean value：
Standard deviation：Kurtosis：The degree of bias：
Maximum value：H=(Max (x (i)) |I=1~N)；Cymomotive force：MFIN=AFMN×RFCN；
Cymomotive force MFINFor same day average fluctuation margin AFMNWith reversed fluctuation number RFCNProduct.Average fluctuation margin
AFMNFor the sum of the absolute value of amplitude of variation at same day whole consecutive number strong point.Reversed fluctuation number RFCNIt is all adjacent for the same day
There is reversed total degree in fluctuation.
3) feature extraction is carried out to step 2) the data obtained library, using daily feature vector as input, is calculated using cluster
Method carries out clustering, obtains four classes cluster as a result, respectively fine day, cloudy day, rainy day, cloudy four kinds of weather patterns, it is corresponding not
Same photovoltaic generation sunlight irradiation weather condition；
4) daily cluster result obtained by step 3) is constituted into time series, obtain it is all kinds of between probability transfer matrix；
By daily cluster result (i.e. weather pattern label) makeup time sequence, adjacent label on the two constitutes one in time series
Two tuples
(i, j) i, j ∈ { fine day, cloudy day, the rainy day, cloudy }
The frequency for counting each two tuple, is calculated the probability P (i, j) to each weather pattern of next day, and generating probability turns
Matrix P is moved, it is cloudy
5) probability transfer matrix obtained by step 2) the data obtained library and step 4) is simulated using randomizer
Sunlight irradiation intensity data.For the random number that randomizer generates, an initial mark is obtained using initial random number
Label obtain the corresponding section of corresponding two tuple of the random number using follow-up random number, and state is shifted by the first item of two tuples
To Section 2, successive ignition generates the stochastic simulation time series being made of label.By the time series, according to generating random number
Device randomly selects data on the one as analogue data in irradiation intensity database, and all moment points of traversal time sequence obtain
The sunlight irradiation intensity data simulated.
The above content is merely illustrative of the invention's technical idea, and protection scope of the present invention cannot be limited with this, every to press
According to technological thought proposed by the present invention, any change done on the basis of technical solution each falls within claims of the present invention
Protection domain within.
Claims (7)
1. a kind of data Method of Stochastic of sunlight irradiation intensity, which is characterized in that include the following steps：
1) the irradiation intensity data more than 1 year are obtained by the sunlight irradiation intensity collection to website；
2) the irradiation intensity data of acquisition are pre-processed, the excessive abnormal point removal of the amount of will deviate from obtains irradiation intensity number
According to library；
3) feature extraction is carried out to irradiation intensity database, using daily feature vector as input, is carried out using clustering algorithm
Clustering obtains four classes cluster as a result, respectively fine day, cloudy day, rainy day and cloudy four kinds of weather patterns, corresponding different light
Volt power generation sunlight irradiation weather condition；
4) daily cluster result is constituted into time series, obtain it is all kinds of between probability transfer matrix；
5) by irradiation intensity database and probability transfer matrix sunlight irradiation intensity number is simulated using randomizer
According to.
2. the data Method of Stochastic of sunlight irradiation intensity according to claim 1, which is characterized in that in step 2)
Selected characteristic includes：For daily data xN, N is the quantity of daily sampled point, mean value：
Standard deviation：
Kurtosis：
The degree of bias：
Maximum value：
H=(Max (x (i)) |I=1~N)
Cymomotive force：
MFIN=AFMN×RFCN。
3. the data Method of Stochastic of sunlight irradiation intensity according to claim 2, which is characterized in that cymomotive force
MFINFor same day average fluctuation margin AFMNWith reversed fluctuation number RFCNProduct.
4. the data Method of Stochastic of sunlight irradiation intensity according to claim 2, which is characterized in that average fluctuation
Amplitude A FMNFor the sum of the absolute value of amplitude of variation at same day whole consecutive number strong point.
5. the data Method of Stochastic of sunlight irradiation intensity according to claim 2, which is characterized in that reversed fluctuation
Number RFCNThere is reversed total degree for the same day all adjacent fluctuations.
6. the data Method of Stochastic of sunlight irradiation intensity according to claim 1, which is characterized in that in step 4)
By daily cluster result, makeup time sequence, wherein cluster result is weather pattern label；Adjacent two days in time series
Label constitute two tuples：
(i, j) i, j ∈ { fine day, cloudy day, the rainy day, cloudy }
The frequency for counting each two tuple, is calculated the probability P (i, j) to each weather pattern of next day, and generating probability shifts square
Battle array P：
7. the data Method of Stochastic of sunlight irradiation intensity according to claim 1, which is characterized in that for step
5) random number that the randomizer in generates, obtains an initial labels using initial random number, utilizes follow-up random number
The corresponding section of corresponding two tuple of the random number is obtained, state is transferred to Section 2, successive ignition by the first item of two tuples
Generate the stochastic simulation time series being made of label；By the time series, according to random number generator, in irradiation intensity data
Data on the one are randomly selected in library as analogue data, all moment points of traversal time sequence, the sunlight spoke simulated
According to intensity data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810107185.4A CN108364236A (en) | 2018-02-02 | 2018-02-02 | A kind of data Method of Stochastic of sunlight irradiation intensity |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810107185.4A CN108364236A (en) | 2018-02-02 | 2018-02-02 | A kind of data Method of Stochastic of sunlight irradiation intensity |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108364236A true CN108364236A (en) | 2018-08-03 |
Family
ID=63004263
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810107185.4A Pending CN108364236A (en) | 2018-02-02 | 2018-02-02 | A kind of data Method of Stochastic of sunlight irradiation intensity |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108364236A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103337042A (en) * | 2013-03-04 | 2013-10-02 | 中国电力科学研究院 | Ground solar irradiation clearance model construction method based on two-way progress |
| CN103390199A (en) * | 2013-07-18 | 2013-11-13 | 国家电网公司 | Photovoltaic power generation capacity/power prediction device |
| CN104361399A (en) * | 2014-08-14 | 2015-02-18 | 国网宁夏电力公司 | Solar irradiation intensity minute-scale predication method |
| CN106251001A (en) * | 2016-07-18 | 2016-12-21 | 南京工程学院 | A kind of based on the photovoltaic power Forecasting Methodology improving fuzzy clustering algorithm |
| CN106529814A (en) * | 2016-11-21 | 2017-03-22 | 武汉大学 | Distributed photovoltaic ultra-short-term forecasting method based on Adaboost clustering and Markov chain |
-
2018
- 2018-02-02 CN CN201810107185.4A patent/CN108364236A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103337042A (en) * | 2013-03-04 | 2013-10-02 | 中国电力科学研究院 | Ground solar irradiation clearance model construction method based on two-way progress |
| CN103390199A (en) * | 2013-07-18 | 2013-11-13 | 国家电网公司 | Photovoltaic power generation capacity/power prediction device |
| CN104361399A (en) * | 2014-08-14 | 2015-02-18 | 国网宁夏电力公司 | Solar irradiation intensity minute-scale predication method |
| CN106251001A (en) * | 2016-07-18 | 2016-12-21 | 南京工程学院 | A kind of based on the photovoltaic power Forecasting Methodology improving fuzzy clustering algorithm |
| CN106529814A (en) * | 2016-11-21 | 2017-03-22 | 武汉大学 | Distributed photovoltaic ultra-short-term forecasting method based on Adaboost clustering and Markov chain |
Non-Patent Citations (2)
| Title |
|---|
| WEIDONG ZHANG 等: ""Simulation and Analysis of the Power Output Fluctuation of Photovoltaic Modules Based on NREL One-minute Irradiance Data"", 《2013 INTERNATIONAL CONFERENCE ON MATERIALS FOR RENEWABLE ENERGY AND ENVIRONMENT》 * |
| 赵宇等: ""基于改进马尔科夫链的风电功率时间序列模型"", 《电力建设》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108306303B (en) | Voltage stability evaluation method considering load increase and new energy output randomness | |
| CN110070226A (en) | Photovoltaic power prediction technique and system based on convolutional neural networks and meta learning | |
| CN103870923B (en) | The wind energy turbine set group of planes polymerization of coagulation type hierarchical clustering algorithm based on comentropy | |
| CN103106544A (en) | Photovoltaic power generation prediction system based on T-S-type fuzzy neural network | |
| CN105426956A (en) | Ultra-short-period photovoltaic prediction method | |
| CN106447098A (en) | Photovoltaic ultra-short period power predicting method and device | |
| CN103390116A (en) | Method for predicting electricity generation power of photovoltaic power station in step-by-step way | |
| CN104638672A (en) | Determining method of photovoltaic transmission power limit considering variable correlation | |
| CN108335010A (en) | A kind of wind power output time series modeling method and system | |
| CN103400204A (en) | Forecasting method for solar photovoltaic electricity generation amount based on SVM (support vector machine) - Markov combination method | |
| CN104682387A (en) | Probability load flow calculation method based on multi-zone interactive iteration | |
| CN103942736A (en) | Wind power station multi-machine equivalent modeling method | |
| CN105896578A (en) | Random production simulation method used for wind energy-solar photovoltaic energy-stored energy combined power generating system | |
| CN111859283B (en) | Scene generation method considering multi-energy charge time sequence and correlation | |
| CN110909911A (en) | Method for aggregating multidimensional time series data considering space-time correlation | |
| CN112925824A (en) | Photovoltaic power prediction method and system for extreme weather type | |
| Sunny et al. | Design and simulation of maximum power point tracking of photovoltaic system using ANN | |
| CN105956252A (en) | Generative deep belief network-based multi-scale forecast modeling method for ultra-short-term wind speed | |
| CN110909994A (en) | Small hydropower station power generation amount prediction method based on big data drive | |
| CN104268435B (en) | A kind of method for the Probabilistic Load Flow for calculating multimachine balance | |
| CN108364236A (en) | A kind of data Method of Stochastic of sunlight irradiation intensity | |
| CN104700156A (en) | Wind power prediction method for screening variables based on support vector machine | |
| CN110084430A (en) | A method of considering space-time characterisation design distributed photovoltaic power output prediction model | |
| Xu et al. | One-day ahead forecast of PV output based on deep belief network and weather classification | |
| Ghosh et al. | Optimal sizing and placement of DG units in radial distribution system using cuckoo search algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180803 |
|
| RJ01 | Rejection of invention patent application after publication |