CN110929953A - Photovoltaic power station ultra-short term output prediction method based on cluster analysis - Google Patents
Photovoltaic power station ultra-short term output prediction method based on cluster analysis Download PDFInfo
- Publication number
- CN110929953A CN110929953A CN201911232972.2A CN201911232972A CN110929953A CN 110929953 A CN110929953 A CN 110929953A CN 201911232972 A CN201911232972 A CN 201911232972A CN 110929953 A CN110929953 A CN 110929953A
- Authority
- CN
- China
- Prior art keywords
- data
- prediction
- photovoltaic
- photovoltaic power
- historical
- 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 52
- 238000007621 cluster analysis Methods 0.000 title claims abstract description 20
- 238000010248 power generation Methods 0.000 claims abstract description 53
- 238000013507 mapping Methods 0.000 claims abstract description 4
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000005855 radiation Effects 0.000 abstract description 16
- 230000005611 electricity Effects 0.000 abstract description 3
- 230000007613 environmental effect Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 8
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 239000013598 vector Substances 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000007667 floating Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 230000035515 penetration Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
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
-
- 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
-
- 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)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Primary Health Care (AREA)
- General Health & Medical Sciences (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a photovoltaic power station ultra-short term output prediction method based on cluster analysis, which comprises the following steps of: classifying historical meteorological data by adopting a fuzzy clustering method, and classifying photovoltaic historical power generation power according to the mapping relation between the historical meteorological data and the photovoltaic historical power generation power data; establishing a corresponding photovoltaic power generation power prediction model according to historical meteorological data and photovoltaic historical power generation power data under different classifications; and ultra-short term output prediction of the photovoltaic power station is carried out. The invention provides a new idea for improving the prediction accuracy by respectively modeling aiming at different weather conditions, the generated electricity quantity, the environmental temperature, the radiation intensity, the local weather and the wind speed and the wind direction are considered during prediction, and the generated power of different weather types is predicted by adopting corresponding prediction models according to different weather types, and the models have obvious superiority in calculating speed, prediction accuracy and model complexity.
Description
Technical Field
The invention relates to a photovoltaic power station ultra-short term output prediction method based on cluster analysis, and belongs to the technical field of photovoltaic power generation.
Background
Due to randomness, fluctuation and uncontrollable property of solar energy, when large-scale photovoltaic is connected into a power grid, adverse effects are caused on safety, stability and economical operation of a power system. The photovoltaic output power prediction has important significance on power grid management, scheduling, operation, system optimization, energy full utilization and safe and stable operation of a power grid.
Because the output of the photovoltaic grid connection has randomness, the photovoltaic grid connection system is an uncontrollable power supply relative to a large power grid, and the instability of the output of the photovoltaic grid connection system has influence on the safe and stable operation of the large power grid. Therefore, penetration access of large-scale photovoltaic power generation certainly brings a series of influences to a power grid, the photovoltaic power generation is an intermittent energy source and is influenced by solar radiation intensity, environmental temperature and the like, output power is uncertain, short-term load prediction accuracy of the large power grid is reduced after the photovoltaic power generation is incorporated into the power grid, voltage and frequency fluctuation of the whole system is inevitably caused by large-scale change of photovoltaic output, the power system has the problems of frequency and voltage stability, the difficulty of traditional power generation, control and operation plans is increased, and the conventional power supply and the coordinated scheduling of the conventional power supply are not facilitated for power grid scheduling personnel.
The photovoltaic output prediction is an important support for a large-scale photovoltaic grid-connected technology and is one of effective methods for ensuring safe and stable operation of a power grid, but a single prediction model adopted by the existing photovoltaic output prediction cannot adapt to changeable weather conditions, so that the prediction precision is low.
Disclosure of Invention
Aiming at the defects of the method, the invention provides a photovoltaic power station ultra-short term output prediction method based on cluster analysis, which can improve the accuracy of photovoltaic power station ultra-short term output prediction.
The technical scheme adopted for solving the technical problems is as follows:
the embodiment of the invention provides a photovoltaic power station ultra-short term output prediction method based on cluster analysis, which comprises the following steps:
classifying historical meteorological data by adopting a fuzzy clustering method, and classifying photovoltaic historical power generation power according to the mapping relation between the historical meteorological data and the photovoltaic historical power generation power data;
establishing a corresponding photovoltaic power generation power prediction model according to historical meteorological data and photovoltaic historical power generation power data under different classifications;
and predicting the ultra-short-term output of the photovoltaic power station.
As a possible implementation manner of this embodiment, the input data of the photovoltaic power generation power prediction model includes power generation electric quantity, ambient temperature, irradiation intensity, local weather, wind speed and wind direction, and the output data is a power prediction value.
As a possible implementation manner of this embodiment, the weather types at least include sunny weather, cloudy weather, and rainy weather.
As a possible implementation manner of this embodiment, the weather type further includes a fog weather condition.
As a possible implementation manner of this embodiment, the photovoltaic power generation power prediction model is established by using a radial basis function neural network regression method.
As a possible implementation manner of this embodiment, before classifying the photovoltaic generation power historical data, the following steps are further included:
and establishing uncertain description of the sample data to the category based on fuzzy clustering.
As a possible implementation manner of this embodiment, the process of establishing the uncertain description of the sample data on the category based on the fuzzy clustering specifically includes:
let set M have a samples to be classified:
M＝{m1,m2,...,ma} (1)
each sample has b indices describing its characteristics, i.e.
Mi＝{mi1,mi2,...,mib} (2)
The raw data matrix is composed of:
wherein, a is the number of sample types, and b is the number of sample data of each type.
As a possible implementation manner of the present embodiment, a is 5.
As a possible implementation manner of this embodiment, the process of classifying the photovoltaic generation power history data includes:
carrying out dimensionless processing on the original sample data:
wherein i is less than or equal to a, and j is less than or equal to b;
constructing vectors by sample data to be classified in historical data
Then
The correlation coefficient of the photovoltaic power factor is as follows:
in the formula: j is 1,2, …, n is the number of the historical data to be classified, p is the resolution factor;
definition of
And
the degree of association is:
selecting the degree of association gamma with the day to be predicted by comparing and calculating the degree of association with the day to be predictedjLarger historical data as prediction sample data mi,yi}。
As a possible implementation of this embodiment, p is 0.5.
As a possible implementation manner of this embodiment, the photovoltaic power generation power prediction model is:
wherein y is output data, namely a power predicted value of photovoltaic power generation; y isiInput data, namely prediction sample data; y is*Is a non-dimensionalized value of the input data.
As a possible implementation manner of this embodiment, the process of predicting the ultra-short term output of the photovoltaic power station includes:
and inputting prediction sample data, and obtaining a prediction result corresponding to the input prediction sample data, namely a power prediction value according to the photovoltaic power generation power prediction model.
The technical scheme of the embodiment of the invention has the following beneficial effects:
the technical scheme of the embodiment of the invention provides a new idea for improving the prediction accuracy by respectively modeling aiming at different weather conditions, the generated electricity quantity, the environmental temperature, the radiation intensity, the local weather and the wind speed and the wind direction are considered in the prediction process, the corresponding prediction model is adopted to predict the generated power of different weather types according to the different weather types, and the model has obvious superiority in calculating speed, predicting accuracy and model complexity.
Description of the drawings:
FIG. 1 is a flow diagram illustrating a method for cluster analysis based ultra-short term power output prediction for a photovoltaic power plant in accordance with an exemplary embodiment;
FIG. 2 is a schematic diagram of the variation of the power generation of a photovoltaic power plant;
FIG. 3 is a graph of the difference in output values of a photovoltaic power plant under different weather conditions (i.e., under different solar radiation intensities);
FIG. 4 is a schematic diagram of a process for ultra-short term power output prediction for a photovoltaic power plant;
FIG. 5 is a graphical illustration of the predicted outcome of a prior art single prediction model;
fig. 6 is a diagram illustrating a prediction result based on weather classification according to the present invention, fig. 6(a) is a diagram illustrating a prediction result based on sunny weather, fig. 6(b) is a diagram illustrating a prediction result based on cloudy weather, and fig. 6(c) is a diagram illustrating a prediction result based on rainy weather.
Detailed Description
The invention is further illustrated by the following examples in conjunction with the accompanying drawings:
in order to clearly explain the technical features of the present invention, the following detailed description of the present invention is provided with reference to the accompanying drawings. The following disclosure provides many different embodiments, or examples, for implementing different features of the invention. To simplify the disclosure of the present invention, the components and arrangements of specific examples are described below. Furthermore, the present invention may repeat reference numerals and/or letters in the various examples. This repetition is for the purpose of simplicity and clarity and does not in itself dictate a relationship between the various embodiments and/or configurations discussed. It should be noted that the components illustrated in the figures are not necessarily drawn to scale. Descriptions of well-known components and processing techniques and procedures are omitted so as to not unnecessarily limit the invention.
FIG. 1 is a flow diagram illustrating a method for cluster analysis based ultra-short term power output prediction for a photovoltaic power plant in accordance with an exemplary embodiment. As shown in fig. 1, the method for predicting the ultra-short term output of the photovoltaic power station based on the cluster analysis provided by the embodiment of the present invention includes the following steps:
classifying historical meteorological data by adopting a fuzzy clustering method, and classifying photovoltaic historical power generation power according to the mapping relation between the historical meteorological data and the photovoltaic historical power generation power data;
establishing a corresponding photovoltaic power generation power prediction model according to historical meteorological data and photovoltaic historical power generation power data under different classifications;
and predicting the ultra-short-term output of the photovoltaic power station.
As a possible implementation manner of this embodiment, the input data of the photovoltaic power generation power prediction model includes power generation electric quantity, ambient temperature, irradiation intensity, local weather, wind speed and wind direction, and the output data is a power prediction value.
As a possible implementation manner of this embodiment, the weather types at least include sunny weather, cloudy weather, and rainy weather.
As a possible implementation manner of this embodiment, the weather type further includes a fog weather condition.
As a possible implementation manner of this embodiment, the photovoltaic power generation power prediction model is established by using a radial basis function neural network regression method.
As a possible implementation manner of this embodiment, before classifying the photovoltaic generation power historical data, the following steps are further included:
and establishing uncertain description of the sample data to the category based on fuzzy clustering.
As a possible implementation manner of this embodiment, the process of establishing the uncertain description of the sample data on the category based on the fuzzy clustering specifically includes:
let set M have a samples to be classified:
M＝{m1,m2,...,ma} (1)
each sample has b indices describing its characteristics, i.e.
Mi＝{mi1,mi2,...,mib} (2)
The raw data matrix is composed of:
where a is the number of sample classes, and a is 5. b is the number of sample data of each type.
The cluster analysis is one of the multivariate statistical analysis and an important branch of unsupervised pattern recognition, and is most widely applied to a plurality of fields such as pattern classification image processing, fuzzy rule processing and the like. It divides a sample without class labels into several subsets according to some criterion, so that similar samples fall into one class as much as possible, and dissimilar samples are divided into different classes. The fuzzy clustering algorithm is a clustering algorithm based on a function optimization method, a calculus calculation technology is used for solving an optimal cost function, a probability density function is used in the clustering method based on the probability algorithm, and therefore a proper model is supposed. And fuzzy clustering establishes an uncertain description of a sample to a class.
As a possible implementation manner of this embodiment, the process of classifying the photovoltaic generation power history data includes:
carrying out non-dimensionalization treatment on the original sample data (adopting an averaging mode):
wherein i is less than or equal to a, and j is less than or equal to b;
5 influencing factors of the power generation quantity, the ambient temperature, the radiation intensity, the local weather, the wind speed and the wind direction in the historical data form a vector
Then
The correlation coefficient of the photovoltaic power factor is as follows:
in the formula: j is 1,2, …, n, n is the number of the historical data to be classified, p is the resolution coefficient, and p is 0.5;
definition of
And
the degree of association is:
by treatingComparing and calculating the association degree of the measuring days, selecting the association degree gamma of the measuring days and the association degree gamma of the measuring days to be predictedjLarger historical data as prediction sample data mi,yiAnd providing data support for the photovoltaic power generation power prediction model.
As a possible implementation manner of this embodiment, the photovoltaic power generation power prediction model is:
for input data m*After the non-dimensionalization of the input items, outputting data (predicted power value of photovoltaic power generation) y*Will also become dimensionless; the output data actual value y is:
wherein y is output data, namely a power predicted value of photovoltaic power generation; y isiInput data, namely prediction sample data; y is*Is a non-dimensionalized value of the input data.
For b indexes contained in the sample, the dimension and the order of magnitude of the b indexes are different, and the original data is directly used for calculation, so that indexes with certain order of magnitude are possibly highlighted, and further, the classification result is influenced.
As a possible implementation manner of this embodiment, as shown in fig. 4, the process of performing ultra-short term output prediction of the photovoltaic power station includes:
selecting corresponding photovoltaic power generation power prediction models according to different weather types, inputting prediction sample data, and obtaining a prediction result corresponding to the input prediction sample data, namely a power prediction value, according to the photovoltaic power generation power prediction models.
Taking a distributed solar photovoltaic power station as an example, the radiation angle, the placement position and the like of the distributed solar photovoltaic power station are fixed, the influence degree can be contained in historical generated energy, and the obvious characteristic of the distributed solar photovoltaic power station is the self high autocorrelation of the time sequence of the generated energy of a photovoltaic array, so that a direct prediction method for predicting solar radiation is avoided. Therefore, the characteristics of the distributed energy output data are researched through a big data analysis method, and the method for predicting the future power generation data by taking the historical power generation data as model input has obvious superiority in calculation speed, prediction accuracy and model complexity compared with an indirect prediction method.
To model the annual weather condition classification, the annual weather condition needs to be classified. The weather is complex and changeable, if prediction modeling is carried out under each weather condition, time and labor are consumed, obtained data samples are few, and the statistical rule is not available. Therefore, the weather conditions with less occurrence times are classified according to certain characteristic parameters and classified into four types of weather conditions, namely sunny weather, cloudy weather, overcast and rainy weather and fog weather. The characteristic parameters mainly select three numerical values which are most closely related to the final generated energy and can reflect the influence of weather on power generation: actual generated power, variance of actual generated power, and generated time period (time during which actual generated power is positive).
The specific parameters are explained as follows:
a. actual generated power: the average value of the generated power from the first positive value in the morning to the last positive value in the evening is taken during the day, namely the actual generated power is averaged every fifteen minutes.
b. Variance of actual generated power: the variance of the actual generated power every fifteen minutes for the generated power during the day from the first positive value in the morning to the end of the last positive value in the evening (reflecting the floating condition of the generated power).
c. The power generation time is as follows: the total power generation time is the power generation power in the day from the first positive value in the morning to the last positive value in the evening.
Examples of the design
A certain photovoltaic power station is taken as a research object, the time period from 2017-01-01 to 2017-03-31 is selected, and the daily sunshine period is 6:00-19: 00. After the radiation angle placement position is determined, the solar radiation intensity is a main factor influencing the photovoltaic output change, and the solar radiation intensity is directly influenced by the cloud layer amount. Fig. 2 is a graph showing a generated energy change of a photovoltaic power station, because all selected weeks are sunny days, the solar radiation intensities are basically the same, the output conditions of the power station are relatively consistent, the change of the photovoltaic generated energy every day has high correlation, and the output value at the next moment is predicted by using the output historical values of the previous three time periods. Fig. 3 shows the difference of the output values of the subject photovoltaic plant under different weather conditions, i.e. different solar radiation intensities. While the trend of the output force value is similar under the same weather condition. After determining the influence factors such as the placing position of the radiation angle, the solar radiation intensity is the main factor influencing the photovoltaic output change. Therefore, under the condition that the weather conditions are the same, the model input data are the output historical values of the previous time period, and a new idea is provided for improving the prediction accuracy by respectively modeling aiming at different weather conditions.
Selecting the data of the whole year from 2017-01-01 to 2017-12-30 during sampling, taking the days with the same proportion at certain intervals in four seasons of spring, summer, autumn and winter, and simultaneously leaving the days with the generated power being zero in the whole day caused by overhauling or other factors. The analysis of the example shows that the data sample in the foggy day is only one day, so the example only selects sunny days, cloudy days and cloudy and rainy days for classification modeling.
The factors affecting photovoltaic output are many and difficult to accurately describe with a certain deterministic model. Based on the characteristics of discontinuity, uncertainty and periodicity of the output of the photovoltaic power generation system and the nonlinear relation between the output of the photovoltaic system and the influence factors of the output of the photovoltaic system, a Radial Basis Function Neural Network (RBFNN) regression method is used for establishing an output prediction model of the photovoltaic system, and the output of the photovoltaic system is classified according to the corresponding weather type.
Because the model input is a historical force output value which changes from 0 to a peak value, the neural network is a nonlinear model, the nonlinear input value is mapped into a high-dimensional space so as to realize linearization, and when the historical force output value is applied to the prediction model as the input, the too large change range can increase the fitting strength of the model, thereby reducing the fitting precision. Therefore, the input data of the model is normalized, namely the data range is limited to (0,1), so that the interval between the data is reduced, the fitting difficulty is reduced, and the relevance and the difference of the data are kept. Therefore, under the condition that the weather conditions are the same, the model input data are the output historical values of the previous time period, and a new idea is provided for improving the prediction accuracy by respectively modeling aiming at different weather conditions.
In order to cope with the complexity and the variability of weather conditions, a cluster analysis method is adopted to classify weather characteristics into four main categories of sunny days, cloudy days, overcast and rainy days and fog days, and then historical data of a plurality of photovoltaic outputs with the same photovoltaic output type are combined into a data sequence with highly similar photovoltaic output characteristics. When the output of the photovoltaic system is predicted, the weather type and the meteorological characteristics of the day to be predicted are known according to weather forecast, and then a prediction model corresponding to the weather type of the predicted day is selected.
Based on historical output data of the distributed energy, researching output characteristics and influence factors of the distributed energy by adopting a big data method, and constructing a big data analysis model of the output of the distributed energy;
based on historical contribution data and coarse-grained meteorological information and model data relationships and characteristics that need to be established by meteorological forecasting techniques, the following is data information used in the prediction process.
a. Inputting data: electricity generation quantity, environment temperature, irradiation intensity, local weather, wind speed and wind direction (meteorological station monitoring data, every 15 min);
b. outputting data: predicted power value (15 min ahead);
c. the data interval is 15 min.
The input data of each prediction model comprises a historical photovoltaic output value (15-minute average value) of the day before the prediction day, and the maximum value, the minimum value and the average value of the air temperature shown by the weather forecast of the prediction day; the model output data are photovoltaic output values for the predicted day (15 min mean). The power prediction results and prediction errors using the single model and the weather classification model, respectively, during the test period are shown in fig. 5 and 6.
The prediction effect of the prediction model based on the weather classification (as shown in fig. 6(a), 6(b) and 6 (c)) is improved compared with that of a single prediction model (as shown in fig. 5). According to different weather conditions, the law of photovoltaic output has different characteristics: the photovoltaic output changes along with the solar radiation intensity at the sunny moment, and can reach the maximum value near the noon; the photovoltaic output at a multi-cloud moment is shielded by a cloud layer to present regular diversity; photovoltaic output is generally low in rainy weather, which is due to the fact that solar radiation is reduced more due to long-term shielding of large cloud layers.
The foregoing is only a preferred embodiment of the present invention, and it will be apparent to those skilled in the art that various modifications and improvements can be made without departing from the principle of the present invention, and these modifications and improvements are also considered to be within the scope of the present invention.
Claims (10)
1. A photovoltaic power station ultra-short term output prediction method based on cluster analysis is characterized by comprising the following steps:
classifying historical meteorological data by adopting a fuzzy clustering method, and classifying photovoltaic historical power generation power according to the mapping relation between the historical meteorological data and the photovoltaic historical power generation power data;
establishing a corresponding photovoltaic power generation power prediction model according to historical meteorological data and photovoltaic historical power generation power data under different classifications;
and predicting the ultra-short-term output of the photovoltaic power station.
2. The method of claim 1, wherein the input data of the photovoltaic power generation power prediction model comprises power generation capacity, ambient temperature, irradiation intensity, local weather, wind speed and wind direction, and the output data is a power prediction value.
3. The cluster analysis based prediction method of very short term output of photovoltaic power plants of claim 1 wherein the weather types include at least sunny, cloudy and rainy weather conditions.
4. The cluster analysis based photovoltaic power plant ultra-short term contribution prediction method of claim 1 wherein the weather type further comprises fog weather conditions.
5. The cluster analysis based photovoltaic power plant ultra-short term output prediction method of claim 1, wherein the photovoltaic power generation power prediction model is built using a radial basis neural network regression method.
6. The method for predicting the ultra-short term output of the photovoltaic power station based on the cluster analysis as claimed in any one of claims 1 to 5, wherein before classifying the historical photovoltaic power generation power data, the method further comprises the following steps:
and establishing uncertain description of the sample data to the category based on fuzzy clustering.
7. The photovoltaic power plant ultra-short term output prediction method based on cluster analysis as claimed in claim 6, wherein the process of establishing the uncertain description of the sample data to the category based on fuzzy clustering specifically comprises:
let set M have a samples to be classified:
M＝{m1,m2,...,ma}(1)
each sample has b indices describing its characteristics, i.e.
Mi＝{mi1,mi2,...,mib}(2)
The raw data matrix is composed of:
wherein, a is the number of sample types, and b is the number of sample data of each type.
8. The method for predicting the ultra-short term output of the photovoltaic power station based on the cluster analysis as claimed in any one of claims 1 to 5, wherein the step of classifying the historical photovoltaic power generation power data comprises the following steps:
carrying out dimensionless processing on the original sample data:
wherein i is less than or equal to a, and j is less than or equal to b;
sample data to be classified in historical data are formedVector quantity
Then
The correlation coefficient of the photovoltaic power factor is as follows:
in the formula: j is 1,2, …, n is the number of the historical data to be classified, p is the resolution factor;
definition of
And
the degree of association is:
selecting the degree of association gamma with the day to be predicted by comparing and calculating the degree of association with the day to be predictedjLarger historical data as prediction sample data mi,yi}。
9. The method for predicting the ultra-short term output of the photovoltaic power station based on the cluster analysis according to any one of claims 1 to 5, wherein the photovoltaic power generation power prediction model is as follows:
wherein y is output data, namely a power predicted value of photovoltaic power generation; y isiInput data, namely prediction sample data; y is*Is a non-dimensionalized value of the input data.
10. The photovoltaic power plant ultra-short term output prediction method based on cluster analysis according to any one of claims 1-5, characterized in that the process of performing the photovoltaic power plant ultra-short term output prediction is as follows:
and inputting prediction sample data, and obtaining a prediction result corresponding to the input prediction sample data, namely a power prediction value according to the photovoltaic power generation power prediction model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911232972.2A CN110929953A (en) | 2019-12-04 | 2019-12-04 | Photovoltaic power station ultra-short term output prediction method based on cluster analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911232972.2A CN110929953A (en) | 2019-12-04 | 2019-12-04 | Photovoltaic power station ultra-short term output prediction method based on cluster analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110929953A true CN110929953A (en) | 2020-03-27 |
Family
ID=69856897
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911232972.2A Pending CN110929953A (en) | 2019-12-04 | 2019-12-04 | Photovoltaic power station ultra-short term output prediction method based on cluster analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110929953A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738522A (en) * | 2020-06-28 | 2020-10-02 | 电子科技大学中山学院 | Photovoltaic power generation power prediction method, storage medium and terminal equipment |
| CN112052985A (en) * | 2020-08-07 | 2020-12-08 | 南京易司拓电力科技股份有限公司 | Lightgbm-based medium-short term low voltage prediction method |
| CN112365081A (en) * | 2020-11-24 | 2021-02-12 | 国家电网有限公司 | Photovoltaic power station power generation capacity prediction method and device |
| CN112446554A (en) * | 2020-12-18 | 2021-03-05 | 阳光电源股份有限公司 | Power prediction model establishing method, power prediction method and device |
| CN112561181A (en) * | 2020-12-21 | 2021-03-26 | 国网甘肃省电力公司电力科学研究院 | Photovoltaic power generation prediction system based on Unet network and foundation cloud picture |
| CN112766573A (en) * | 2021-01-20 | 2021-05-07 | 国网安徽省电力有限公司电力科学研究院 | Power generation data prediction method and device and electronic equipment |
| CN112925824A (en) * | 2021-02-25 | 2021-06-08 | 山东大学 | Photovoltaic power prediction method and system for extreme weather type |
| CN113627674A (en) * | 2021-08-12 | 2021-11-09 | 中国华能集团清洁能源技术研究院有限公司 | Distributed photovoltaic power station output prediction method and device and storage medium |
| CN116050666A (en) * | 2023-03-20 | 2023-05-02 | 中国电建集团江西省电力建设有限公司 | Photovoltaic power generation power prediction method for irradiation characteristic clustering |
| CN116316615A (en) * | 2023-05-25 | 2023-06-23 | 国网江西省电力有限公司电力科学研究院 | Data enhancement-based distributed light Fu Qun short-term power prediction method and system |
| CN117791687A (en) * | 2024-02-28 | 2024-03-29 | 长峡数字能源科技(湖北)有限公司 | Energy management method of photovoltaic energy storage system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766990A (en) * | 2017-11-10 | 2018-03-06 | 河海大学 | A kind of Forecasting Methodology of photovoltaic power station power generation power |
| CN108197744A (en) * | 2018-01-02 | 2018-06-22 | 华北电力大学（保定） | A kind of determining method and system of photovoltaic generation power |
| CN108280546A (en) * | 2018-01-22 | 2018-07-13 | 长沙有色冶金设计研究院有限公司 | Photovoltaic power generation power prediction method and forecasting system |
| CN108846527A (en) * | 2018-08-27 | 2018-11-20 | 云南电网有限责任公司电力科学研究院 | A kind of photovoltaic power generation power prediction method |
| CN109685257A (en) * | 2018-12-13 | 2019-04-26 | 国网青海省电力公司 | A kind of photovoltaic power generation power prediction method based on Support vector regression |
-
2019
- 2019-12-04 CN CN201911232972.2A patent/CN110929953A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766990A (en) * | 2017-11-10 | 2018-03-06 | 河海大学 | A kind of Forecasting Methodology of photovoltaic power station power generation power |
| CN108197744A (en) * | 2018-01-02 | 2018-06-22 | 华北电力大学（保定） | A kind of determining method and system of photovoltaic generation power |
| CN108280546A (en) * | 2018-01-22 | 2018-07-13 | 长沙有色冶金设计研究院有限公司 | Photovoltaic power generation power prediction method and forecasting system |
| CN108846527A (en) * | 2018-08-27 | 2018-11-20 | 云南电网有限责任公司电力科学研究院 | A kind of photovoltaic power generation power prediction method |
| CN109685257A (en) * | 2018-12-13 | 2019-04-26 | 国网青海省电力公司 | A kind of photovoltaic power generation power prediction method based on Support vector regression |
Non-Patent Citations (1)
| Title |
|---|
| 于秋玲等: "基于模糊聚类和支持向量机的短期光伏功率预测", 《电力***及其自动化学报》 * |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111738522A (en) * | 2020-06-28 | 2020-10-02 | 电子科技大学中山学院 | Photovoltaic power generation power prediction method, storage medium and terminal equipment |
| CN112052985B (en) * | 2020-08-07 | 2023-11-03 | 南京易司拓电力科技股份有限公司 | Middle-short-term low-voltage prediction method based on lightgbm |
| CN112052985A (en) * | 2020-08-07 | 2020-12-08 | 南京易司拓电力科技股份有限公司 | Lightgbm-based medium-short term low voltage prediction method |
| CN112365081A (en) * | 2020-11-24 | 2021-02-12 | 国家电网有限公司 | Photovoltaic power station power generation capacity prediction method and device |
| CN112446554A (en) * | 2020-12-18 | 2021-03-05 | 阳光电源股份有限公司 | Power prediction model establishing method, power prediction method and device |
| CN112446554B (en) * | 2020-12-18 | 2024-05-14 | 阳光电源股份有限公司 | Power prediction model establishment method, power prediction method and device |
| CN112561181A (en) * | 2020-12-21 | 2021-03-26 | 国网甘肃省电力公司电力科学研究院 | Photovoltaic power generation prediction system based on Unet network and foundation cloud picture |
| CN112766573A (en) * | 2021-01-20 | 2021-05-07 | 国网安徽省电力有限公司电力科学研究院 | Power generation data prediction method and device and electronic equipment |
| CN112925824A (en) * | 2021-02-25 | 2021-06-08 | 山东大学 | Photovoltaic power prediction method and system for extreme weather type |
| CN113627674A (en) * | 2021-08-12 | 2021-11-09 | 中国华能集团清洁能源技术研究院有限公司 | Distributed photovoltaic power station output prediction method and device and storage medium |
| CN116050666A (en) * | 2023-03-20 | 2023-05-02 | 中国电建集团江西省电力建设有限公司 | Photovoltaic power generation power prediction method for irradiation characteristic clustering |
| CN116316615A (en) * | 2023-05-25 | 2023-06-23 | 国网江西省电力有限公司电力科学研究院 | Data enhancement-based distributed light Fu Qun short-term power prediction method and system |
| CN116316615B (en) * | 2023-05-25 | 2023-09-12 | 国网江西省电力有限公司电力科学研究院 | Data enhancement-based distributed light Fu Qun short-term power prediction method and system |
| CN117791687A (en) * | 2024-02-28 | 2024-03-29 | 长峡数字能源科技(湖北)有限公司 | Energy management method of photovoltaic energy storage system |
| CN117791687B (en) * | 2024-02-28 | 2024-05-14 | 长峡数字能源科技(湖北)有限公司 | Energy management method of photovoltaic energy storage system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110929953A (en) | Photovoltaic power station ultra-short term output prediction method based on cluster analysis | |
| Graditi et al. | Comparison of Photovoltaic plant power production prediction methods using a large measured dataset | |
| Musbah et al. | Energy management of hybrid energy system sources based on machine learning classification algorithms | |
| US20170286838A1 (en) | Predicting solar power generation using semi-supervised learning | |
| Yang et al. | Day-ahead forecasting of photovoltaic output power with similar cloud space fusion based on incomplete historical data mining | |
| Wang et al. | A comprehensive review for wind, solar, and electrical load forecasting methods | |
| Wang et al. | Comparative study of machine learning approaches for predicting short-term photovoltaic power output based on weather type classification | |
| CN114792156B (en) | Photovoltaic output power prediction method and system based on curve characteristic index clustering | |
| Anuradha et al. | Analysis of solar power generation forecasting using machine learning techniques | |
| Park et al. | Multi-layer RNN-based short-term photovoltaic power forecasting using IoT dataset | |
| CN114399081A (en) | Photovoltaic power generation power prediction method based on weather classification | |
| Yang et al. | Photovoltaic power forecasting with a rough set combination method | |
| Mayer et al. | Probabilistic modeling of future electricity systems with high renewable energy penetration using machine learning | |
| Karabiber et al. | Short term PV power estimation by means of extreme learning machine and support vector machine | |
| Jazayeri et al. | Artificial neural network-based all-sky power estimation and fault detection in photovoltaic modules | |
| Putra et al. | Modeling of high uncertainty photovoltaic generation in quasi dynamic power flow on distribution systems: A case study in Java Island, Indonesia | |
| Kader et al. | Active power control strategy for wind farms based on power prediction errors distribution considering regional data | |
| Yan et al. | A high concentrated photovoltaic output power predictive model based on Fuzzy Clustering and RBF neural network | |
| Mdluli et al. | Development of short term solar radiation forecasting using AI techniques | |
| Cao et al. | Short-term forecasting and uncertainty analysis of photovoltaic power based on the FCM-WOA-BILSTM model | |
| Castillo-Rojas et al. | Daily prediction model of photovoltaic power generation using a hybrid architecture of recurrent neural networks and shallow neural networks | |
| Tamal et al. | Forecasting of solar photovoltaic output energy using lstm machine learning algorithm | |
| CN113723670A (en) | Photovoltaic power generation power short-term prediction method with variable time window | |
| Zhang et al. | Multi‐Input Deep Convolutional Neural Network Model for Short‐Term Power Prediction of Photovoltaics | |
| Varughese et al. | Prediction Of Solar Power Generation Based On Machine Learning 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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200327 |