CN116845869A - Clustering method, device and equipment for wind-light output typical scene set - Google Patents
Clustering method, device and equipment for wind-light output typical scene set Download PDFInfo
- Publication number
- CN116845869A CN116845869A CN202310798680.5A CN202310798680A CN116845869A CN 116845869 A CN116845869 A CN 116845869A CN 202310798680 A CN202310798680 A CN 202310798680A CN 116845869 A CN116845869 A CN 116845869A
- Authority
- CN
- China
- Prior art keywords
- day
- wind
- similar
- data
- distance
- 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 46
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 22
- 238000001556 precipitation Methods 0.000 claims description 32
- 238000004364 calculation method Methods 0.000 claims description 19
- 238000000926 separation method Methods 0.000 claims description 11
- 230000002123 temporal effect Effects 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 5
- 230000009191 jumping Effects 0.000 claims description 4
- 238000011160 research Methods 0.000 description 8
- 230000006872 improvement Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 4
- 229910052799 carbon Inorganic materials 0.000 description 3
- 238000010248 power generation Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 230000033228 biological regulation Effects 0.000 description 2
- 238000005265 energy consumption Methods 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
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 description 1
- 238000000418 atomic force spectrum Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007621 cluster analysis Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000010792 warming Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/004—Generation forecast, e.g. methods or systems for forecasting future energy generation
-
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/04—Circuit arrangements for ac mains or ac distribution networks for connecting networks of the same frequency but supplied from different sources
- H02J3/06—Controlling transfer of power between connected networks; Controlling sharing of load between connected networks
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/22—The renewable source being solar energy
- H02J2300/24—The renewable source being solar energy of photovoltaic origin
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/28—The renewable source being wind energy
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Photometry And Measurement Of Optical Pulse Characteristics (AREA)
Abstract
The invention discloses a clustering method, a clustering device and clustering equipment for a wind-light output typical scene set, which are used for determining a historical day set of a wind-light power station and acquiring meteorological data of each historical day in the historical day set; acquiring historical days, which reach preset similar conditions with meteorological data of days to be matched, in the historical day set as similar days to obtain a similar day set; calculating a wind-light output characteristic value of the wind-light power station on each similar day; and clustering the similar days according to the wind-light output characteristic values by using a K-Means algorithm to obtain a typical scene set of the wind-light output of the days to be matched. By adopting the method, the decisive influence of meteorological conditions on wind and light output is considered, and the generation efficiency and accuracy of a wind and light output typical scene set are improved.
Description
Technical Field
The invention relates to the technical field of power systems, in particular to a clustering method, device and equipment for a wind-light output typical scene set.
Background
With the continuous warming of global climate and the continuous deterioration of natural environment, the problem of climate control is very important for each country, and the development of the energy industry structure to the characteristics of cleanness, low carbon and sustainability is positively promoted. Under the background, renewable energy sources such as wind, light and the like become important participants on the source side of the novel power system, and the high-proportion renewable energy source system has wide application in operation scenes.
The existing research mainly adopts a scene analysis method to eliminate uncertainty of wind and light output, discretizes and samples a prediction curve and a prediction error or a probability distribution model to generate an original scene set, and then uses a scene reduction technology to merge and reduce the original scene set so as to obtain a typical scene set of wind and light output.
However, wind power and photovoltaic output are extremely easy to be influenced by meteorological conditions, and the wind power and photovoltaic power generation device has the characteristics of randomness, volatility, intermittence and the like, various technologies related to the existing research all depend on a probability distribution model of wind and light, regional characteristics and meteorological characteristics of different areas are not considered, and the problems of large calculation amount of an algorithm, long calculation time, strong coupling of calculation effect and probability distribution model accuracy, poor mobility and the like exist. Therefore, under the new situation that renewable energy sources rapidly develop, research aiming at the operation characteristics of renewable energy sources such as wind, light and the like is needed to be developed, the influence of uncertainty of the renewable energy sources is fully considered when a power system is scheduled, and the optimal scheduling strategy of the system is perfected, so that efficient and reliable access of clean energy sources is realized.
Disclosure of Invention
The embodiment of the invention aims to provide a clustering method, a clustering device and clustering equipment for a wind-light output typical scene set, which consider the decisive influence of meteorological conditions on wind-light output and improve the generation efficiency and accuracy of the wind-light output typical scene set.
In order to achieve the above objective, the embodiment of the present invention provides a clustering method for a typical scene set of wind-light output, including:
determining a historical day set of a wind-solar power station, and acquiring meteorological data of each historical day in the historical day set;
acquiring historical days, which reach preset similar conditions with meteorological data of days to be matched, in the historical day set as similar days to obtain a similar day set;
calculating a wind-light output characteristic value of the wind-light power station on each similar day;
and clustering the similar days according to the wind-light output characteristic values by using a K-Means algorithm to obtain a typical scene set of the wind-light output of the days to be matched.
As an improvement of the above solution, the obtaining, as the similar day, the historical day of the historical day set, which reaches a preset similar condition with the weather data of the day to be matched, to obtain the similar day set specifically includes:
calculating the original time distance and the original space distance of each history day and the day to be matched under different meteorological data;
performing standardization processing on the original space distance of each history day and the day to be matched under different meteorological data by adopting a range method to obtain the standard space distance of each history day and the day to be matched under different meteorological data;
According to the original time distance and the standard space distance of each history day and the day to be matched under different meteorological data, calculating the time distance and the space distance of each history day and the day to be matched;
and acquiring a historical day which meets the requirement that the time distance between the historical day set and the day to be matched is smaller than or equal to a preset time distance threshold value, and the space distance between the historical day set and the day to be matched is smaller than or equal to a preset space distance threshold value, and taking the historical day set as a similar day which reaches the preset similar condition to obtain a similar day set.
As an improvement of the scheme, the meteorological data comprise temperature data, wind speed data, precipitation data and cloud amount data;
the standard deviation method is adopted to carry out standard processing on the original space distance of each history day and the day to be matched under different meteorological data to obtain the standard space distance of each history day and the day to be matched under different meteorological data, and the standard space distance comprises the following steps:
according to the original space distance between each history day and the day to be matched under different meteorological data and the maximum space distance and the minimum space distance between the history days and the corresponding meteorological data, calculating to obtain the standard space distance between each history day and the day to be matched under different meteorological data by using the calculation formula of the following extreme difference method:
；
；
；
；
wherein ,、/>、/> and />Respectively represent history daysd i Day to be matched withd 0 Standard spatial distance of temperature data, wind speed data, precipitation data and cloud data, +.>、/>、/> and />Respectively represent history daysd i Day to be matched withd 0 Raw spatial distance of temperature data, wind speed data, precipitation data and cloud amount data; />、/> and />Maximum spatial distance in all history days in temperature data, wind speed data, precipitation data and cloud data, respectively,/->、/> and />The minimum spatial distances in the temperature data, the wind speed data, the precipitation data and the cloud cover data in all the history days are respectively represented.
As an improvement of the above solution, the calculating the time distance and the space distance between each history day and the day to be matched according to the original time distance and the standard space distance between each history day and the day to be matched under different meteorological data specifically includes:
according to the original time distance and the standard space distance of each history day and the day to be matched under different meteorological data, the time distance and the space distance of each history day and the day to be matched are calculated by adopting the following calculation formula:
；
；
wherein , and />Respectively represent history days d i Day to be matched withd 0 The temporal distance between them and the spatial distance between them, and />Respectively represent history daysd i Day to be matched withd 0 Original time distance and standard space distance of temperature data of (a); /> and />Respectively represent history daysd i Day to be matched withd 0 The original time distance and the standard space distance of the wind speed data of (1); /> and />Respectively represent history daysd i Day to be matched withd 0 Original time distance and standard space distance of precipitation data of (a); /> and />Respectively represent history daysd i Day to be matched withd 0 Original temporal distance and standard spatial distance of cloud data of +.>，/>，/>And->Respectively representing the weights of the temperature data, the wind speed data, the precipitation data and the cloud amount data in the distance.
As an improvement of the scheme, the wind-light output characteristic value comprises a mean value of photovoltaic output, a range of photovoltaic output, a mean value of wind power output and a range of wind power output.
As an improvement of the above scheme, the calculating the wind-light output characteristic value of the wind-light power station on each similar day specifically includes:
calculating the average value of the photovoltaic output, the range of the photovoltaic output, the average value of the wind-light output and the range of the wind-light output of the wind-light power station on each similar day according to the following calculation formula:
；
；
；
；
wherein ,is of similar dayd i Mean value of internal photovoltaic output, +.>Is of similar dayd i Mean value of internal wind power output,/->Is of similar dayd i Extremely poor in internal photovoltaic output +.>Is of similar dayd i Extremely poor internal wind power output; />Is of similar dayd i Inner time of daytPhotovoltaic output at time,/->Is of similar dayd i Inner time of daytAnd the wind power output is generated.
As an improvement of the above scheme, the clustering of the similar days according to the wind-light output characteristic value by using a K-Means algorithm is performed to obtain a typical scene set of wind-light output of the day to be matched, which specifically includes:
selecting values according to preset K values, and sequentially selecting different K values;
for the currently selected K value, randomly selecting K similar day scenes from all similar day scenes as centroid scenes, and taking the rest similar day scenes as rest scenes;
calculating the distance between each residual scene and each centroid scene;
classifying the rest scenes into centroid scenes closest to the rest scenes according to a distance matrix formed by the distances, and obtaining a clustering set corresponding to each centroid scene;
in the same cluster set, calculating the sum of the distances between each similar daily scene and other similar daily scenes, and selecting the similar daily scene with the smallest sum of the distances as a new centroid scene;
Judging whether a preset optimal condition or a preset iteration number is reached, if so, obtaining data and probability values of each typical scene; if not, jumping to execute the steps: calculating the distance between each residual scene and each centroid scene;
and calculating a CHI index under each K value, selecting the K value with the largest CHI index as a final clustering number, and acquiring typical scene data and probability values corresponding to the final clustering number as the typical scene set of the solar wind power output to be matched.
As an improvement of the above scheme, the calculation of the CHI index at each K value is specifically:
for the cluster set corresponding to the currently selected K value, calculating the separation degree S among different cluster sets and the closeness M in the same cluster set:
and calculating to obtain the CHI index corresponding to the currently selected K value according to the ratio of the separation degree S among different clustering sets to the compactness M in the same clustering set.
The embodiment of the invention also provides a clustering device of the wind-light output typical scene set, which comprises the following steps:
the historical day collection acquisition module is used for determining a historical day collection of the wind-solar power station and acquiring meteorological data of each historical day in the historical day collection;
the similar day set acquisition module is used for acquiring a historical day, which reaches a preset similar condition with meteorological data of a day to be matched, in the historical day set as a similar day so as to obtain a similar day set;
The wind-light output characteristic value calculation module is used for calculating wind-light output characteristic values of the wind-light power station on each similar day;
and the typical scene clustering module is used for clustering the similar days according to the wind-light output characteristic values by using a K-Means algorithm to obtain a typical scene set of the wind-light output of the days to be matched.
The embodiment of the invention also provides clustering equipment of the wind and light output typical scene set, which comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, wherein the clustering method of the wind and light output typical scene set is realized when the processor executes the computer program.
Compared with the prior art, the clustering method, the clustering device and the clustering equipment for the wind-light output typical scene set are characterized in that the historical day set of the wind-light power station is determined, and meteorological data of each historical day in the historical day set are obtained; acquiring historical days, which reach preset similar conditions with meteorological data of days to be matched, in the historical day set as similar days to obtain a similar day set; calculating a wind-light output characteristic value of the wind-light power station on each similar day; and clustering the similar days according to the wind-light output characteristic values by using a K-Means algorithm to obtain a typical scene set of the wind-light output of the days to be matched. By adopting the technical means of the embodiment of the invention, the decisive influence of meteorological conditions on wind and light output is studied and considered, similar days are searched in historical big data based on space-time similarity of meteorological data, and a typical scene set of wind and light is obtained by clustering in the similar days, so that the problems that wind and light output has high uncertainty, frequent fluctuation, prominent anti-peak regulation characteristics and the like in the actual dispatching work or dispatching strategy research of the traditional power system can be effectively solved, the wind and light output is difficult to accurately describe, the problems of poor running adjustability, low economical efficiency and low renewable resource consumption rate are caused, the calculated amount of an algorithm can be effectively reduced, the generation efficiency and accuracy of the typical wind and light scene set are improved by using clustering on the basis of fully utilizing meteorological forecast and historical big data, the follow-up dispatching optimization research is facilitated, the running cost of a new energy source is further promoted, the economic and clean power generation advantage is realized, and the new thought is realized for the effective energy consumption of the power system and the economic clean low-carbon power supply.
Drawings
FIG. 1 is a schematic flow chart of a clustering method of a wind-light output typical scene set provided by an embodiment of the invention;
FIG. 2 is a schematic flow chart of a K-Means algorithm in an embodiment of the invention;
FIG. 3 is a view of a wind-solar joint clustering result taking time similarity and space similarity of meteorological data into consideration in an embodiment of the invention; in fig. 3, (a) is a photovoltaic clustering result, and (b) is a wind power clustering result;
FIG. 4 is a graph of a result of wind-solar joint clustering taking into account only temporal similarity in an embodiment of the invention; in fig. 4, (a) is a photovoltaic clustering result, and (b) is a wind power clustering result;
FIG. 5 is a graph of a wind-solar joint clustering result considering only spatial similarity in an embodiment of the invention; in fig. 5, (a) is a photovoltaic clustering result, and (b) is a wind power clustering result;
FIG. 6 is a schematic structural diagram of a clustering device for a typical scene set of wind-solar power output according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of a clustering device for a typical scene set of wind-light output according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a flow chart of a clustering method of a wind-light output typical scene set according to an embodiment of the present invention is shown. The embodiment of the invention provides a clustering method of a wind-light output typical scene set, which specifically comprises the following steps of S11 to S14:
s11, determining a historical day set of a wind-solar power station, and acquiring meteorological data of each historical day in the historical day set;
s12, acquiring a historical day, which reaches a preset similar condition with meteorological data of a day to be matched, in the historical day set as a similar day to obtain a similar day set;
s13, calculating a wind-light output characteristic value of the wind-light power station on each similar day;
s14, clustering the similar days according to the wind and light output characteristic values by using a K-Means algorithm to obtain a typical scene set of wind and light output of the days to be matched.
In the embodiment of the invention, the decisive influence of meteorological conditions on wind and light output is studied and considered, similar days are searched in historical big data based on the space-time similarity of meteorological data, and clustering is carried out in the similar days to obtain a typical scene set of wind and light. By adopting the embodiment of the invention, the problems of high uncertainty, frequent fluctuation, outstanding anti-peak regulation characteristics and the like of wind-light output in the actual dispatching work or dispatching strategy research of the existing power system are effectively solved, the problems of poor running adjustability, low economical efficiency and low renewable resource consumption rate caused by the problems are difficult to describe accurately, the calculated amount of an algorithm can be effectively reduced, the scene set is simplified by using clustering on the basis of fully utilizing weather forecast and historical big data, the generation efficiency and accuracy of a typical output scene set are improved, the follow-up dispatching optimization research of a power grid is facilitated, the new energy consumption is promoted, the running cost of the power system is reduced, the advantages of economy and clean power generation are achieved, and the new idea of effectively consuming new energy and realizing economic clean low-carbon power supply is developed for the power system.
As a preferred embodiment, based on the foregoing embodiment, step S12, that is, the step of obtaining, as the similar day, the historical day that the weather data of the date to be matched reaches the preset similar condition in the historical day set, so as to obtain the similar day set, specifically includes steps S121 to S124:
s121, calculating the original time distance and the original space distance of each history day and the day to be matched under different meteorological data.
In the embodiment of the invention, the characteristic matching method for searching weather similarity days in the history days is provided by considering the similarity of different weather data in time and space.
In particular, for the time similarity index of meteorological data, since shape distance can be used to measure the progression of a sequence over time, the invention is used to measure the time similarity of meteorological data. For sequences，TCalculating the slope of each time period for the length of time gives a slope sequence describing the slope of the time period:
；
wherein ,。
further, a sequence is proposedXCorresponding shape sequence，SThe elements of (a) are given by:
；
wherein ,threpresenting a threshold at which the curve can be considered to be stationary.
On the basis, defineX，YShape distance of two sequences Then the distance may measure the sequenceX，YWalk over timePotential similarity, i.e., the time distance between two sequences, is the smaller the time distance, the more similar the two sequences are in time.
For the spatial similarity index of meteorological data, the invention adopts Euclidean distance to measure the spatial similarity between two sequences. For two sequencesX，YTheir Euclidean distancel X,Y Can be given by:
；
this distance characterizes the spatial similarity between two sequences, i.e. the smaller the spatial distance between two sequences, the more similar the two sequences are in space.
And calculating the original time distance and the original space distance of each historical day and the day to be matched under different meteorological data based on the two indexes.
Preferably, the meteorological data comprises temperature data, wind speed data, precipitation data and cloud data, respectivelyT，W，P，CStep S121 specifically includes:
calculating an original time distance and an original space distance of temperature data of each historical day and the day to be matched;
calculating the original time distance and the original space distance of wind speed data of each historical day and the day to be matched;
calculating the original time distance and the original space distance of precipitation data of each history day and the day to be matched;
And calculating the original time distance and the original space distance of cloud amount data of each history day and the day to be matched.
S122, carrying out standardization processing on the original space distance of each history day and the day to be matched under different meteorological data by adopting a range method to obtain the standard space distance of each history day and the day to be matched under different meteorological data.
Specifically, all historical day sets are noted asDThenWhere n is the total number of history days. The day to be matched, i.e. the dispatch diary isd 0 。
In consideration of the fact that units and dimensions of different meteorological data are different, the units and dimensions of the different meteorological data are required to be standardized first, and the method adopts a very poor method to perform standardization treatment.
Preferably, step S122 specifically includes:
according to the original space distance between each history day and the day to be matched under different meteorological data and the maximum space distance and the minimum space distance between the history days and the corresponding meteorological data, calculating to obtain the standard space distance between each history day and the day to be matched under different meteorological data by using the calculation formula of the following extreme difference method:
；
；
；
；
wherein ,、/>、/> and />Respectively represent history daysd i Day to be matched with d 0 Temperature data of (2)Standard spatial distance of wind speed data, precipitation data and cloud data, +.>、/>、/> and />Respectively represent history daysd i Day to be matched withd 0 Raw spatial distance of temperature data, wind speed data, precipitation data and cloud amount data; />、/> and />Maximum spatial distance in all history days in temperature data, wind speed data, precipitation data and cloud data, respectively,/->、/> and />The minimum spatial distances in the temperature data, the wind speed data, the precipitation data and the cloud cover data in all the history days are respectively represented.
S123, calculating the time distance and the space distance of each historical day and the day to be matched according to the original time distance and the standard space distance of each historical day and the day to be matched under different meteorological data.
Specifically, step S123 includes:
according to the original time distance and the standard space distance of each history day and the day to be matched under different meteorological data, the time distance and the space distance of each history day and the day to be matched are calculated by adopting the following calculation formula:
；
；
wherein , and />Respectively represent history daysd i Day to be matched withd 0 The temporal distance between them and the spatial distance between them, and />Respectively represent history daysd i Day to be matched withd 0 Original time distance and standard space distance of temperature data of (a); /> and />Respectively represent history daysd i Day to be matched withd 0 The original time distance and the standard space distance of the wind speed data of (1); /> and />Respectively represent history daysd i Day to be matched withd 0 Original time distance and standard space distance of precipitation data of (a); /> and />Respectively represent history daysd i Day to be matched withd 0 Original temporal distance and standard spatial distance of cloud data of +.>，/>，/>And->Respectively representing the weights of the temperature data, the wind speed data, the precipitation data and the cloud amount data in the distance.
S124, obtaining a history day which meets the condition that the time distance between the history day set and the day to be matched is smaller than or equal to a preset time distance threshold value, and the space distance between the history day set and the day to be matched is smaller than or equal to a preset space distance threshold value, and obtaining a similarity day set as a similarity day which meets the preset similarity condition.
From historical day collectionDFind the day to be matched withd 0 Similar day set with sufficient similarity under four meteorological conditions of temperature, wind speed, precipitation and cloud coverFDAfter obtaining the time distances and the space distances between all the historical days and the days to be matched, obtaining elements in the similar day set FD according to the following formula:
；
wherein ,ft h and (3) withfs h The time-space distance threshold and the space distance threshold are preset respectively, alternatively, the space-time distance of the similar days ranked as the first 5% is taken as the threshold, and when the distance is smaller than the threshold, the two are considered to be sufficiently similar in time or space.
If it isFDAnd if the proportion of the number of the similar days to the total number of the historical days is less than 0.5%, the days to be matched are regarded as extreme weather, the subsequent cluster analysis operation is not performed, and separate analysis is needed.
As a preferred implementation manner, the embodiment of the present invention is further implemented on the basis of the foregoing embodiment, and in step S13, the wind-light output characteristic value includes a mean value of the photovoltaic output, a range of the photovoltaic output, a mean value of the wind power output, and a range of the wind power output. The steps are made to obtain a similar day setAnd M is the total number of similar days, calculating the wind-light output characteristic value of the wind-light power station on each similar day, wherein the method specifically comprises the following steps:
calculating the average value of the photovoltaic output, the range of the photovoltaic output, the average value of the wind-light output and the range of the wind-light output of the wind-light power station on each similar day according to the following calculation formula:
；
；
；
；
wherein ,Is of similar dayd i Mean value of internal photovoltaic output, +.>Is of similar dayd i Mean value of internal wind power output,/->Is of similar dayd i Extremely poor in internal photovoltaic output +.>Is of similar dayd i Extremely poor internal wind power output; />Is of similar dayd i Inner time of daytPhotovoltaic output at time,/->Is of similar dayd i Inner time of daytAnd the wind power output is generated.
As a preferred embodiment, step S14, the clustering the similar days according to the wind-light output characteristic value using a K-Means algorithm to obtain a typical scene set of wind-light output of the day to be matched, specifically includes steps S141 to S147:
s141, selecting values according to preset K values, and sequentially selecting different K values;
s142, randomly selecting K similar day scenes from all similar day scenes as centroid scenes for the currently selected K value, and taking the rest similar day scenes as rest scenes;
s143, calculating the distance between each residual scene and each centroid scene;
s144, classifying the rest scenes into centroid scenes closest to the rest scenes according to a distance matrix formed by the distances, and obtaining a clustering set corresponding to each centroid scene;
s145, respectively calculating the sum of the distances between each similar daily scene and other similar daily scenes in the same cluster set, and selecting the similar daily scene with the smallest sum of the distances as a new centroid scene;
S146, judging whether a preset optimal condition or a preset iteration number is reached, and if so, obtaining typical scene data and probability values; if not, jumping to execute the steps: calculating the distance between each residual scene and each centroid scene;
s147, calculating CHI indexes under each K value, selecting the K value with the largest CHI index as a final cluster number, and acquiring typical scene data and probability values corresponding to the final cluster number as the typical scene set of the solar, wind and light output to be matched.
Preferably, the calculating the CHI index at each K value is specifically:
for the cluster set corresponding to the currently selected K value, calculating the separation degree S among different cluster sets and the closeness M in the same cluster set:
and calculating to obtain the CHI index corresponding to the currently selected K value according to the ratio of the separation degree S among different clustering sets to the compactness M in the same clustering set.
In the embodiment of the invention, after the similar day set is obtained, a wind-solar combined clustering method improved by a coarse clustering algorithm is further constructed, and because weather data obtained by weather forecast is not absolutely accurate and has a certain error, the number of similar days contained in the similar day set is often more, and the characteristics are repeated, so that the similar day set is inconvenient to use, the method simplifies the similar day set by adopting the clustering algorithm, thereby obtaining a typical scene set only containing a plurality of specific conditions, and facilitating the subsequent dispatching or planning research of a ground power system.
The number of clusters and the initial cluster center are required to be predetermined by a common K-Means clustering algorithm, and the superior clustering result often comprises two characteristics: i.e. the similarity between clusters is high and the similarity between clusters is low. Based on the above, the CHI (Calinski-Harabasz index, CHI)) index can better measure the separation degree S between different clusters and the compactness M in the same cluster, and the ratio of S to M is the size of the CHI index.
The degree of separation S between different cluster sets and the closeness M within the same cluster set are expressed as:
；
；
according to the separation degree S among different clustering sets and the closeness M in the same clustering set, calculating CHI indexes corresponding to the currently selected K values, wherein the CHI indexes specifically comprise:
；
wherein ,Krepresenting the number of cluster sets;C i andc i respectively represent the firstiA cluster set and a cluster center thereof;representing the center of the sample set;n i represent the firstiThe number of samples contained by the cluster set;Nrepresenting the total number of samples of the sample set.
And clustering the characteristic values of the wind and light output according to the similar days by using a K-Means algorithm. The core idea of K-Means mean clustering is: k initial scenes are selected as centroids, distances between the rest scenes and all centroids are calculated and included in clusters where the centroids with minimum distances are located, new cluster centers are formed through iterative updating until iteration conditions are met, the final centroid of the obtained cluster is a typical scene, and the probability is the sum of probabilities of all scenes in the cluster. Referring to fig. 2, a flow chart of a K-Means algorithm in an embodiment of the present invention is shown, which specifically includes the steps of:
Randomly selecting M scenes as centroids, and collecting centroids scenesThe method comprises the steps of carrying out a first treatment on the surface of the Calculating the remaining scene set +.>Distance from centroid scene:the method comprises the steps of carrying out a first treatment on the surface of the According to->The distance matrix is composed, the rest scenes are classified to the mass centers closest to the rest scenes, and a cluster set is obtained>，/>Representing a set of scenes in a family. Calculating a centroid: let a certain distance +.>In Ls scenes, calculating the sum CTs of the distances between each scene and other scenes, selectingScene->And updating the mass centers for each cluster for the new cluster center, determining a new mass center set, judging whether an optimal condition or the iteration number is reached, and if so, obtaining each typical data and probability value.
And clustering all similar days in the similar day set by using a K-Means algorithm to obtain a typical scene set capable of effectively representing the solar wind light output condition to be matched. And selecting different k values, clustering the same similar day set, calculating CHI indexes under different k values, and selecting the k value with the largest CHI index as a final clustering number, wherein a corresponding clustering result is a typical scene set of the solar energy and wind energy output of the day to be matched.
As an example, historical wind power output and photovoltaic power output between 2000-2020 twenty-one years of a certain region generated by GSEE (global solar simulator) model and VWF (virtual wind farm) model, and temperature, wind speed, precipitation, cloud amount data in the same period of time of a certain region in MERRA-2 dataset are used as historical day data. The photovoltaic distance weight is selected to be 0.3, the wind power distance weight is selected to be 0.5, the precipitation distance weight is selected to be 0.1, and the cloud distance weight is selected to be 0.1. And selecting a random one of 7665 historical days as a day to be matched, and verifying the validity of the method. The space-time distance selection threshold was set to the first 5%, and finally the matching resulted in 94 similar days, of which the time and space distances for 20 similar days were listed, as shown in table 1.
TABLE 1 spatiotemporal distance of partially similar days
And clustering the obtained similar day set by adopting a k-means algorithm, respectively carrying out five clustering experiments on each value of the clustering number from 2 to 9, and selecting a primary experiment result with the maximum CHI index as a final clustering result, wherein the corresponding clustering number is the optimal clustering number. Table 2 shows the CHI index of each experiment at different cluster numbers, and the first experiment result with cluster number 5 was selected as the final cluster result according to the table information.
TABLE 2 CHI index at different cluster numbers
Referring to fig. 3, fig. 3 is a wind-solar combined clustering result considering time similarity and space similarity of meteorological data in the embodiment of the invention, wherein a black dotted line represents a real output curve, and as can be seen from the graph, a clustering center better reflects the shape of the real curve, namely the similarity in time, whether wind power or photovoltaic; in terms of spatial similarity, the clustering result of the photovoltaic comprises two extreme cases with larger distances from a real curve, the occurrence probability of the two extreme cases is 5.32% and 13.88%, the volatility of the photovoltaic can be represented to a certain extent, and compared with the spatial volatility of the wind power clustering center, the spatial volatility of the wind power clustering center is larger, because wind power output is more random than that of the photovoltaic, and is almost influenced by wind speed only, and the similarity judgment of wind power output can be influenced by the consideration of other meteorological data such as temperature when calculating the comprehensive space-time distance.
Referring to fig. 4 and 5, fig. 4 is a wind-solar combined clustering result considering only time similarity in the embodiment of the present invention; FIG. 5 is a graph of the wind-solar joint clustering result considering only spatial similarity in an embodiment of the invention. As can be seen from fig. 4, when only the temporal similarity is considered, the spatial offset is larger, although the cluster center is closer to the shape of the true force curve; in contrast to fig. 5, the spatial distance between the cluster center and the true curve decreases, but the temporal distance increases. In combination, it is necessary to consider both temporal and spatial similarities.
Referring to fig. 6, a schematic structural diagram of a clustering device for a wind-light output typical scene set according to an embodiment of the present invention is provided, and an embodiment of the present invention provides a clustering device 20 for a wind-light output typical scene set, including:
a historical day set acquisition module 21, configured to determine a historical day set of the wind-solar power station, and acquire meteorological data of each historical day in the historical day set;
a similar day set obtaining module 22, configured to obtain, as a similar day, a historical day that reaches a preset similar condition with weather data of a day to be matched in the historical day set, so as to obtain a similar day set;
The wind-light output characteristic value calculation module 23 is used for calculating wind-light output characteristic values of the wind-light power station on each similar day;
and the typical scene clustering module 24 is used for clustering the similar days according to the wind-light output characteristic values by using a K-Means algorithm to obtain a typical scene set of the wind-light output of the days to be matched.
The similar day set obtaining module 22 is specifically configured to:
calculating the original time distance and the original space distance of each history day and the day to be matched under different meteorological data;
performing standardization processing on the original space distance of each history day and the day to be matched under different meteorological data by adopting a range method to obtain the standard space distance of each history day and the day to be matched under different meteorological data;
according to the original time distance and the standard space distance of each history day and the day to be matched under different meteorological data, calculating the time distance and the space distance of each history day and the day to be matched;
and acquiring a historical day which meets the requirement that the time distance between the historical day set and the day to be matched is smaller than or equal to a preset time distance threshold value, and the space distance between the historical day set and the day to be matched is smaller than or equal to a preset space distance threshold value, and taking the historical day set as a similar day which reaches the preset similar condition to obtain a similar day set.
Preferably, the meteorological data comprises temperature data, wind speed data, precipitation data and cloud cover data;
the standard deviation method is adopted to carry out standard processing on the original space distance of each history day and the day to be matched under different meteorological data to obtain the standard space distance of each history day and the day to be matched under different meteorological data, and the standard space distance comprises the following steps:
according to the original space distance between each history day and the day to be matched under different meteorological data and the maximum space distance and the minimum space distance between the history days and the corresponding meteorological data, calculating to obtain the standard space distance between each history day and the day to be matched under different meteorological data by using the calculation formula of the following extreme difference method:
；
；
；
；
wherein ,、/>、/> and />Respectively represent history daysd i Day to be matched withd 0 Standard spatial distance of temperature data, wind speed data, precipitation data and cloud data, +.>、/>、/> and />Respectively represent history daysd i Day to be matched withd 0 Raw spatial distance of temperature data, wind speed data, precipitation data and cloud amount data; />、/> and />Maximum spatial distance in all history days in temperature data, wind speed data, precipitation data and cloud data, respectively,/- >、/> and />The minimum spatial distances in the temperature data, the wind speed data, the precipitation data and the cloud cover data in all the history days are respectively represented.
Calculating the time distance and the space distance between each history day and the day to be matched according to the original time distance and the standard space distance between each history day and the day to be matched under different meteorological data, wherein the method specifically comprises the following steps:
according to the original time distance and the standard space distance of each history day and the day to be matched under different meteorological data, the time distance and the space distance of each history day and the day to be matched are calculated by adopting the following calculation formula:
；
；
wherein , and />Respectively represent history daysd i Day to be matched withd 0 The temporal distance between them and the spatial distance between them, and />Respectively represent history daysd i Day to be matched withd 0 Raw time distance and scale of temperature data of (a)A quasi-spatial distance; /> and />Respectively represent history daysd i Day to be matched withd 0 The original time distance and the standard space distance of the wind speed data of (1); /> and />Respectively represent history daysd i Day to be matched withd 0 Original time distance and standard space distance of precipitation data of (a); /> and />Respectively represent history daysd i Day to be matched withd 0 Original temporal distance and standard spatial distance of cloud data of +. >，/>，/>And->Respectively representing the weights of the temperature data, the wind speed data, the precipitation data and the cloud amount data in the distance.
Preferably, the wind-light output characteristic value comprises a mean value of photovoltaic output, a range of photovoltaic output, a mean value of wind power output and a range of wind power output.
The wind-light output characteristic value calculating module 23 is specifically configured to:
calculating the average value of the photovoltaic output, the range of the photovoltaic output, the average value of the wind-light output and the range of the wind-light output of the wind-light power station on each similar day according to the following calculation formula:
；
；
；
；/>
wherein ,is of similar dayd i Mean value of internal photovoltaic output, +.>Is of similar dayd i Mean value of internal wind power output,/->Is of similar dayd i Extremely poor in internal photovoltaic output +.>Is of similar dayd i Extremely poor internal wind power output; />Is of similar dayd i Inner time of daytPhotovoltaic output at time,/->Is of similar dayd i Inner time of daytAnd the wind power output is generated.
Preferably, the exemplary scene clustering module 24 is specifically configured to:
selecting values according to preset K values, and sequentially selecting different K values;
for the currently selected K value, randomly selecting K similar day scenes from all similar day scenes as centroid scenes, and taking the rest similar day scenes as rest scenes;
Calculating the distance between each residual scene and each centroid scene;
classifying the rest scenes into centroid scenes closest to the rest scenes according to a distance matrix formed by the distances, and obtaining a clustering set corresponding to each centroid scene;
in the same cluster set, calculating the sum of the distances between each similar daily scene and other similar daily scenes, and selecting the similar daily scene with the smallest sum of the distances as a new centroid scene;
judging whether a preset optimal condition or a preset iteration number is reached, if so, obtaining data and probability values of each typical scene; if not, jumping to execute the steps: calculating the distance between each residual scene and each centroid scene;
and calculating a CHI index under each K value, selecting the K value with the largest CHI index as a final clustering number, and acquiring typical scene data and probability values corresponding to the final clustering number as the typical scene set of the solar wind power output to be matched.
Preferably, the calculating the CHI index at each K value is specifically:
for the cluster set corresponding to the currently selected K value, calculating the separation degree S among different cluster sets and the closeness M in the same cluster set:
and calculating to obtain the CHI index corresponding to the currently selected K value according to the ratio of the separation degree S among different clustering sets to the compactness M in the same clustering set.
It should be noted that, the clustering device for the wind-light output typical scene set provided by the embodiment of the invention is used for executing all flow steps of the clustering method for the wind-light output typical scene set in the embodiment, and the working principles and beneficial effects of the two are in one-to-one correspondence, so that the description is omitted.
Referring to fig. 7, which is a schematic structural diagram of a clustering device for a wind-light-output typical scene set according to an embodiment of the present invention, the embodiment of the present invention further provides a clustering device 30 for a wind-light-output typical scene set, which includes a processor 31, a memory 32, and a computer program stored in the memory and configured to be executed by the processor, where the processor implements the clustering method for a wind-light-output typical scene set according to any one of the embodiments.
Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored on a computer readable storage medium, which when executed may comprise the steps of the embodiments of the methods described above. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), or the like.
While the foregoing is directed to the preferred embodiments of the present invention, it will be appreciated by those skilled in the art that changes and modifications may be made without departing from the principles of the invention, such changes and modifications are also intended to be within the scope of the invention.
Claims (10)
1. The clustering method of the wind-light output typical scene set is characterized by comprising the following steps of:
determining a historical day set of a wind-solar power station, and acquiring meteorological data of each historical day in the historical day set;
acquiring historical days, which reach preset similar conditions with meteorological data of days to be matched, in the historical day set as similar days to obtain a similar day set;
calculating a wind-light output characteristic value of the wind-light power station on each similar day;
and clustering the similar days according to the wind-light output characteristic values by using a K-Means algorithm to obtain a typical scene set of the wind-light output of the days to be matched.
2. The method for clustering a wind-solar power output typical scene set according to claim 1, wherein the step of obtaining the historical day, which is the historical day of which the weather data of the day to be matched reaches a preset similar condition, in the historical day set is used as a similar day, so as to obtain a similar day set, includes:
Calculating the original time distance and the original space distance of each history day and the day to be matched under different meteorological data;
performing standardization processing on the original space distance of each history day and the day to be matched under different meteorological data by adopting a range method to obtain the standard space distance of each history day and the day to be matched under different meteorological data;
according to the original time distance and the standard space distance of each history day and the day to be matched under different meteorological data, calculating the time distance and the space distance of each history day and the day to be matched;
and acquiring a historical day which meets the requirement that the time distance between the historical day set and the day to be matched is smaller than or equal to a preset time distance threshold value, and the space distance between the historical day set and the day to be matched is smaller than or equal to a preset space distance threshold value, and taking the historical day set as a similar day which reaches the preset similar condition to obtain a similar day set.
3. The method for clustering a wind-solar power output typical scene set according to claim 2, wherein the meteorological data comprises temperature data, wind speed data, precipitation data and cloud cover data;
the standard deviation method is adopted to carry out standard processing on the original space distance of each history day and the day to be matched under different meteorological data to obtain the standard space distance of each history day and the day to be matched under different meteorological data, and the standard space distance comprises the following steps:
According to the original space distance between each history day and the day to be matched under different meteorological data and the maximum space distance and the minimum space distance between the history days and the corresponding meteorological data, calculating to obtain the standard space distance between each history day and the day to be matched under different meteorological data by using the calculation formula of the following extreme difference method:
；
；
；
；
wherein ,、/>、/> and />Respectively represent history daysd i Day to be matched withd 0 Standard spatial distance of temperature data, wind speed data, precipitation data and cloud data, +.>、/>、/> and />Respectively represent history daysd i Day to be matched withd 0 Raw spatial distance of temperature data, wind speed data, precipitation data and cloud amount data;、/> and />Maximum spatial distance in all history days in temperature data, wind speed data, precipitation data and cloud data, respectively,/->、/> and />The minimum spatial distances in the temperature data, the wind speed data, the precipitation data and the cloud cover data in all the history days are respectively represented.
4. The method for clustering a wind-solar power output typical scene set according to claim 2, wherein the calculating the time distance and the space distance between each historical day and the day to be matched according to the original time distance and the standard space distance between each historical day and the day to be matched under different meteorological data specifically comprises:
According to the original time distance and the standard space distance of each history day and the day to be matched under different meteorological data, the time distance and the space distance of each history day and the day to be matched are calculated by adopting the following calculation formula:
；
；
wherein , and />Respectively represent history daysd i Day to be matched withd 0 Distance in time and distance in space, +.> and />Respectively represent history daysd i Day to be matched withd 0 Original time distance and standard space distance of temperature data of (a); and />Respectively represent history daysd i Day to be matched withd 0 The original time distance and the standard space distance of the wind speed data of (1); /> and />Respectively represent history daysd i Day to be matched withd 0 Original time distance and standard space distance of precipitation data of (a); /> and />Respectively represent history daysd i Day to be matched withd 0 Original temporal distance and standard spatial distance of cloud data of +.>，/>，/>And->Respectively representing the weights of the temperature data, the wind speed data, the precipitation data and the cloud amount data in the distance.
5. The method for clustering a wind-light-output typical scene set according to claim 1, wherein the wind-light-output characteristic value comprises a mean value of photovoltaic output, a range of photovoltaic output, a mean value of wind power output and a range of wind power output.
6. The method for clustering a wind-solar power output typical scene set according to claim 5, wherein the calculating wind-solar power output characteristic values of the wind-solar power plant on each similar day specifically comprises:
calculating the average value of the photovoltaic output, the range of the photovoltaic output, the average value of the wind-light output and the range of the wind-light output of the wind-light power station on each similar day according to the following calculation formula:
；
；
；
；
wherein ,is of similar dayd i Mean value of internal photovoltaic output, +.>Is of similar dayd i Mean value of internal wind power output,/->Is of similar dayd i Extremely poor in internal photovoltaic output +.>Is of similar dayd i Extremely poor internal wind power output; />Is of similar dayd i Inner time of daytPhotovoltaic output at time,/->Is of similar dayd i Inner time of daytAnd the wind power output is generated.
7. The method for clustering the wind-light output typical scene set according to claim 1, wherein the clustering the similar days according to the wind-light output characteristic value by using a K-Means algorithm to obtain the wind-light output typical scene set of the day to be matched specifically comprises:
selecting values according to preset K values, and sequentially selecting different K values;
for the currently selected K value, randomly selecting K similar day scenes from all similar day scenes as centroid scenes, and taking the rest similar day scenes as rest scenes;
Calculating the distance between each residual scene and each centroid scene;
classifying the rest scenes into centroid scenes closest to the rest scenes according to a distance matrix formed by the distances, and obtaining a clustering set corresponding to each centroid scene;
in the same cluster set, calculating the sum of the distances between each similar daily scene and other similar daily scenes, and selecting the similar daily scene with the smallest sum of the distances as a new centroid scene;
judging whether a preset optimal condition or a preset iteration number is reached, if so, obtaining data and probability values of each typical scene; if not, jumping to execute the steps: calculating the distance between each residual scene and each centroid scene;
and calculating a CHI index under each K value, selecting the K value with the largest CHI index as a final clustering number, and acquiring typical scene data and probability values corresponding to the final clustering number as the typical scene set of the solar wind power output to be matched.
8. The method for clustering a wind-solar power output typical scene set according to claim 7, wherein the calculating the CHI index at each K value specifically comprises:
for the cluster set corresponding to the currently selected K value, calculating the separation degree S among different cluster sets and the closeness M in the same cluster set:
And calculating to obtain the CHI index corresponding to the currently selected K value according to the ratio of the separation degree S among different clustering sets to the compactness M in the same clustering set.
9. A cluster device for a scene set typical of wind and light output, comprising:
the historical day collection acquisition module is used for determining a historical day collection of the wind-solar power station and acquiring meteorological data of each historical day in the historical day collection;
the similar day set acquisition module is used for acquiring a historical day, which reaches a preset similar condition with meteorological data of a day to be matched, in the historical day set as a similar day so as to obtain a similar day set;
the wind-light output characteristic value calculation module is used for calculating wind-light output characteristic values of the wind-light power station on each similar day;
and the typical scene clustering module is used for clustering the similar days according to the wind-light output characteristic values by using a K-Means algorithm to obtain a typical scene set of the wind-light output of the days to be matched.
10. A cluster device for a wind-light-output-typical scene set, comprising a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, the processor implementing the cluster method for a wind-light-output-typical scene set according to any one of claims 1 to 8 when the computer program is executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310798680.5A CN116845869A (en) | 2023-07-03 | 2023-07-03 | Clustering method, device and equipment for wind-light output typical scene set |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310798680.5A CN116845869A (en) | 2023-07-03 | 2023-07-03 | Clustering method, device and equipment for wind-light output typical scene set |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116845869A true CN116845869A (en) | 2023-10-03 |
Family
ID=88168392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310798680.5A Pending CN116845869A (en) | 2023-07-03 | 2023-07-03 | Clustering method, device and equipment for wind-light output typical scene set |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116845869A (en) |
-
2023
- 2023-07-03 CN CN202310798680.5A patent/CN116845869A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107766990B (en) | Method for predicting power generation power of photovoltaic power station | |
| US10103548B2 (en) | Operating a solar power generating system | |
| CN109002915B (en) | Photovoltaic power station short-term power prediction method based on Kmeans-GRA-Elman model | |
| JP6693330B2 (en) | Solar power system operation | |
| CN109165774A (en) | A kind of short-term photovoltaic power prediction technique | |
| CN110619360A (en) | Ultra-short-term wind power prediction method considering historical sample similarity | |
| CN112990558B (en) | Meteorological temperature and illumination prediction method based on deep migration learning | |
| CN105701572B (en) | Photovoltaic short-term output prediction method based on improved Gaussian process regression | |
| CN113065278B (en) | Frequent pattern mining-based prediction method for wind power small-occurrence event statistical characteristic model in rich period | |
| CN112186761B (en) | Wind power scene generation method and system based on probability distribution | |
| CN114511132A (en) | Photovoltaic output short-term prediction method and prediction system | |
| Ghasvarian Jahromi et al. | A novel method for day-ahead solar power prediction based on hidden Markov model and cosine similarity | |
| CN115829105A (en) | Photovoltaic power prediction method based on historical data feature search | |
| CN112561189A (en) | Method suitable for predicting power generation capacity of photovoltaic power station | |
| CN110852492A (en) | Photovoltaic power ultra-short-term prediction method for finding similarity based on Mahalanobis distance | |
| KR20230076695A (en) | Solar power generation forcasting device and method reflecting solar periodic time variable | |
| CN113920349A (en) | Wind and light typical scene construction method containing meteorological data based on density peak value-FCM | |
| CN114882373A (en) | Multi-feature fusion sandstorm prediction method based on deep neural network | |
| CN116845869A (en) | Clustering method, device and equipment for wind-light output typical scene set | |
| Wu et al. | Forecast of short-term electricity price based on data analysis | |
| CN115481788A (en) | Load prediction method and system for phase change energy storage system | |
| CN111291326B (en) | Clustering effectiveness index establishing method combining intra-class similarity and inter-class difference | |
| CN112132344A (en) | Short-term wind power prediction method based on similar day and FRS-SVM | |
| CN115618258B (en) | Method and system for extracting key operation modes of power system planning | |
| Zhang et al. | Load prediction based on depthwise separable convolution model |
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 |