CN112270439B - Ultra-short-term wind power prediction method and device, electronic equipment and storage medium - Google Patents
Ultra-short-term wind power prediction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN112270439B CN112270439B CN202011172181.8A CN202011172181A CN112270439B CN 112270439 B CN112270439 B CN 112270439B CN 202011172181 A CN202011172181 A CN 202011172181A CN 112270439 B CN112270439 B CN 112270439B
- Authority
- CN
- China
- Prior art keywords
- climbing
- matrix
- prediction
- power
- short
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 75
- 238000003860 storage Methods 0.000 title claims abstract description 9
- 230000009194 climbing Effects 0.000 claims abstract description 335
- 239000011159 matrix material Substances 0.000 claims abstract description 214
- 238000004422 calculation algorithm Methods 0.000 claims description 14
- 238000010801 machine learning Methods 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 10
- 238000004590 computer program Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 7
- 238000004140 cleaning Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 7
- 238000003064 k means clustering Methods 0.000 claims description 7
- 238000012937 correction Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 5
- OIGNJSKKLXVSLS-VWUMJDOOSA-N prednisolone Chemical compound O=C1C=C[C@]2(C)[C@H]3[C@@H](O)C[C@](C)([C@@](CC4)(O)C(=O)CO)[C@@H]4[C@@H]3CCC2=C1 OIGNJSKKLXVSLS-VWUMJDOOSA-N 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 230000002159 abnormal effect Effects 0.000 claims description 4
- 238000010248 power generation Methods 0.000 claims description 4
- 230000006870 function Effects 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 238000004891 communication Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Development Economics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The embodiment of the invention provides an ultra-short-term wind power prediction method, an ultra-short-term wind power prediction device, electronic equipment and a storage medium, wherein the method comprises the following steps: determining ultra-short-term prediction series P of t moments after the current moment k according to the real-time power, the real-time wind speed and the short-term prediction power of the current moment k of the wind power plant o The method comprises the steps of carrying out a first treatment on the surface of the Determining a climbing matrix corresponding to short-term preset power at t times after the current time k, inputting climbing characteristics of the climbing matrix and the climbing matrix corresponding to the climbing characteristics into a climbing prediction probability model to obtain a climbing state class, and determining a future climbing state matrix corresponding to the climbing state class; ultra-short-term prediction array P at t times after current time k is corrected based on future climbing state matrix o And determining an ultra-short-term wind power prediction result. According to the embodiment of the invention, the climbing type is determined according to the climbing state type at the current moment, and the ultra-short-term power of the wind power plant is corrected according to the climbing type, so that the ultra-short-term predicted time delay phenomenon can be reduced.
Description
Technical Field
The invention relates to the technical field of electric power, in particular to an ultra-short-term wind power prediction method and device, electronic equipment and a storage medium.
Background
Wind energy is a clean and pollution-free renewable energy source, but wind has randomness and uncontrollability, and unstable wind energy can generate great impact on a power grid after grid connection, so that the safe and stable operation of a power grid system is affected. Wind power prediction is an important means for helping to realize stable running of wind power grid connection. The wind power prediction method can be divided into two types according to different input data: a power prediction method based on numerical weather forecast and a power prediction method based on historical data. The current ultra-short period wind power prediction generally predicts the future short time power by taking the current instantaneous power as the basis, and the method has obvious time delay phenomenon during prediction.
Disclosure of Invention
Aiming at the problems existing in the prior art, the embodiment of the invention provides an ultra-short-term wind power prediction method and device.
In a first aspect, an embodiment of the present invention provides an ultrashort-period wind power prediction method, including:
determining ultra-short-term prediction series P of t moments after the current moment k according to the real-time power, the real-time wind speed and the short-term prediction power of the current moment k of the wind power plant o ；
Determining a climbing matrix corresponding to short-term preset power at t times after the current time k, inputting climbing characteristics corresponding to the climbing matrix and the climbing matrix into a climbing prediction probability model to obtain a climbing state class, and determining a future climbing state matrix corresponding to the climbing state class;
Correcting ultra-short-term prediction number series P of t moments after current moment k based on future climbing state matrix o Determining an ultra-short-term wind power prediction result;
the climbing prediction probability model is obtained by training based on a machine learning algorithm by adopting climbing matrix sample data of historical prediction power of a wind farm and climbing characteristics corresponding to a climbing matrix of the historical prediction power as input data and adopting climbing state categories corresponding to the climbing matrix of the historical prediction power as output data.
Further, the method further comprises the following steps:
acquiring wind farm historical data; wherein the wind farm history data comprises wind farm historic actual wind speed and wind farm historic actual power generation;
based on the wind farm historical data and a preset time window length n, calculating and determining a corresponding actual climbing matrix window by window according to a time sequence;
dividing the actual climbing matrix into j classes according to the corresponding characteristics of the actual climbing matrix to obtain the center point of the j classes of actual climbing matrix;
acquiring historical short-term prediction power, and determining a corresponding prediction climbing matrix and climbing characteristics of the prediction climbing matrix by adopting the window length of the preset time window according to window-by-window calculation of a time sequence;
And taking the central point of the j-class actual climbing matrix as an output value, taking the climbing characteristics of the prediction climbing matrix and the prediction climbing matrix as input values, and establishing the climbing prediction probability model.
Further, determining the climbing matrix specifically includes:
according to the first relation model, performing time difference on the power of the wind power station; wherein the first relationship model is as follows:
P delta ＝P-P sheft(1)
wherein P is delta Is the time sequence difference of the power, P is the grid-connected power, P sheft(1) Is a grid-connected power offset value;
calculating the climbing time length according to the second relation model; wherein the second relationship model is as follows:
t clam ＝t max +t min
wherein t is clam For climbing time length, t max For the corresponding time point of the maximum value of the actual power, t min A corresponding time point for the actual power minimum value;
determining a climbing matrix according to the third relation model; wherein the third relationship model is as follows:
C(k)＝F(k)·E(k)·P delta ·F(k) T
wherein C (k) is a climbing matrix, k represents the current moment, F (k) is n t of the kth point clam Filtering matrix, E (k) is identity matrix, P delta F (k) is the time series difference of the power T N t representing the kth point clam The transpose of the filter matrix, T represents the transpose, and n is the preset time window length.
Further, the classifying the actual climbing matrix into j classes according to the corresponding features of the actual climbing matrix specifically includes:
The actual climbing matrix is classified into j classes according to the corresponding characteristics of the actual climbing matrix by adopting a k-means clustering method; wherein, the actual climbing matrix correspondingly features include: length of climbing t clam Climbing peak P max Climbing start point P min 。
Further, determining an ultra-short-term prediction array P of t moments after the current moment k according to the real-time power, the real-time wind speed and the short-term prediction power of the current moment k of the wind power plant o The method specifically comprises the following steps:
a regression model is adopted, historical actual wind speed of the wind power plant and historical short-term prediction power of the wind power plant are input, real-time power at the moment to be predicted is output, and an ultra-short-term power regression model is established;
bringing the real-time power, the real-time wind speed and the short-term predicted power of the current moment k of the wind power plant into the ultra-short-term power regression model, and determining ultra-short-term predicted number series P of t moments after the current moment k o 。
Further, the ultra-short-term prediction sequence P of t times after the current time k is corrected based on the future climbing state matrix o Determining an ultra-short-term wind power prediction result specifically comprises the following steps:
calculating prediction confidence according to a fourth relation model based on the future climbing state matrix; wherein the fourth relationship model is:
λ＝E|C(k-1) R ·C(k-1)′ P |-E|C(k-1) R |·E|C(k-1)′ P |
Where λ is the confidence of the prediction, E is the desired function, C (k-1) R For the actual climbing matrix at time k-1, C (k-1)' P A future climbing state matrix corresponding to the climbing state of the target class corresponding to the actual climbing matrix at the moment k-1, wherein k represents the current moment;
according to a fifth relation model, ultra-short-term prediction sequence P of t times after the current time k is based on prediction confidence o Correcting; wherein the fifth relational model is：
P f ＝P o +λ·F(k) -1 ·C(k)′ P ·E(k) -1 F(k) -T
Wherein P is f Representing ultra-short-term wind power prediction results P of t times after current time k o An ultrashort-term prediction sequence of t times after the current time k is represented, lambda is prediction confidence, F (k) represents a filter matrix of k points, and C (k)' P Expressed as a future climbing state matrix corresponding to the climbing state type at the time k, E (k) is an identity matrix, E (k) -1 Represents the inversion of E (k), F (k) -T The transpose of F (k) is inverted.
Further, after the wind farm history data is obtained, the method further includes:
performing data cleaning and feature construction on the wind farm historical data;
wherein the data cleaning comprises removing dead values, and/or removing abnormal values, and/or limiting power judgment, and/or limiting power reduction;
The feature construction comprises data expansion, wherein data containing effective features but with the data quantity smaller than a preset value are copied and filled into an original data sequence according to a time sequence.
In a second aspect, an embodiment of the present invention provides an ultrashort-term wind power prediction apparatus, including:
a first determining module, configured to determine an ultrashort-term prediction array P at t times after a current time k according to a real-time power, a real-time wind speed and a short-term prediction power of the current time k of a wind farm o ；
The second determining module is used for determining a climbing matrix corresponding to short-term preset power at t times after the current time k, inputting climbing characteristics of the climbing matrix and the climbing matrix corresponding to the climbing matrix into a climbing prediction probability model, obtaining a climbing state category, and determining a future climbing state matrix corresponding to the climbing state category;
the correction module is used for correcting the ultra-short-term prediction number of t moments after the current moment k based on the future climbing state matrixColumn P o Determining an ultra-short-term wind power prediction result;
the climbing prediction probability model is obtained by training based on a machine learning algorithm by adopting climbing matrix sample data of historical prediction power of a wind farm and climbing characteristics corresponding to a climbing matrix of the historical prediction power as input data and adopting climbing state categories corresponding to the climbing matrix of the historical prediction power as output data.
In a third aspect, an embodiment of the present invention further provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the program to implement the steps of the ultra-short term wind power prediction method according to the first aspect.
In a fourth aspect, embodiments of the present invention also provide a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the ultra-short term wind power prediction method according to the first aspect above.
According to the technical scheme, the ultra-short-term wind power prediction method, the device, the electronic equipment and the storage medium provided by the embodiment of the invention, aiming at the time delay phenomenon of the current ultra-short-term wind power prediction result, the ultra-short-term prediction number series P of t moments after the current moment k is determined according to the real-time power, the real-time wind speed and the short-term prediction power of the current moment k of the wind power plant o Then, a climbing matrix corresponding to short-term preset power at t times after the current time k is input into a climbing prediction probability model, a climbing state class is obtained, and a future climbing state matrix corresponding to the climbing state class is determined; the climbing prediction probability model is obtained by training based on a machine learning algorithm by adopting climbing matrix sample data of historical prediction power of a wind farm and climbing characteristics corresponding to a climbing matrix of the historical prediction power as input data and adopting climbing state categories corresponding to the climbing matrix of the historical prediction power as output data; general purpose medicine Correcting the ultra-short-term prediction number series P of t moments after the current moment k by using a future climbing state matrix o The method can accurately predict the climbing state category at the current moment to determine the climbing type, and the ultra-short-term prediction array P is based on the climbing type o And correcting is carried out, so that an ultra-short-period wind power prediction result is determined, the time delay phenomenon of ultra-short-period prediction can be effectively reduced, the ultra-short-period prediction power prediction precision of a wind power plant is improved, and the prediction accuracy is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of an ultra-short-term wind power prediction method according to an embodiment of the present invention;
FIG. 2 is a schematic diagram illustrating an application effect comparison of an ultra-short-term wind power prediction method according to an embodiment of the present invention;
FIG. 3 is a schematic structural diagram of an ultra-short-term wind power prediction apparatus according to an embodiment of the present invention;
fig. 4 is a schematic physical structure of an electronic device according to an embodiment of the invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments of the present invention. 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. The ultra-short-term wind power prediction method provided by the invention is explained and illustrated in detail by a specific embodiment.
FIG. 1 is a flowchart of an ultra-short-term wind power prediction method according to an embodiment of the present invention; as shown in fig. 1, the method includes:
step 101: determining ultra-short-term prediction series P of t moments after the current moment k according to the real-time power, the real-time wind speed and the short-term prediction power of the current moment k of the wind power plant o 。
In the step, it is required to be explained that the real-time power, the real-time wind speed and the short-term predicted power of the current moment of the wind power plant are obtained; the short-term predicted power is directly downloaded data, can be obtained through purchased meteorological data, can also be obtained through observing cloud layer movement through satellites, can calculate future movement tracks through a model to obtain future predicted wind speeds, and then can obtain the predicted power through a mathematical model.
In this step, the real-time power, the real-time wind speed and the short-term predicted power at the current time k of the wind farm are brought into a prediction algorithm to obtain an ultrashort-term predicted sequence of t times after the current time (i.e., an ultrashort-term predicted sequence of t times after the current time k) o ) The method comprises the steps of carrying out a first treatment on the surface of the The prediction algorithm can be realized through any regression model, historic actual power, wind speed and predicted power of the wind power plant are input, and the actual power at a time point needing prediction is output to obtain an ultra-short-term power regression model; the real-time power, the wind speed and the short-term predicted power are brought into a regression model to obtain an original ultra-short-term predicted sequence P of n points in the future o 。
Step 102: and determining a climbing matrix corresponding to short-term preset power at t times after the current time k, inputting climbing characteristics corresponding to the climbing matrix and the climbing matrix into a climbing prediction probability model to obtain a climbing state class, and determining a future climbing state matrix corresponding to the climbing state class.
In this step, it should be noted that, the climbing prediction probability model is obtained by using climbing matrix sample data of wind farm historical prediction power and climbing characteristics corresponding to a climbing matrix of the historical prediction power as input data, and using a climbing state class corresponding to a climbing matrix of the historical prediction power as output data, and training based on a machine learning algorithm.
In this step, for example, a short-term preset power corresponding climbing matrix C (k-1) at the current time k and k-1 is calculated P And C (k) P Actual power climbing matrix C (k-1) R And C (k) R Will C (k-1) P And C (k) P The model is brought into a climbing prediction probability model to obtain the category of the climbing state, and a corresponding future climbing matrix C (k-1) is obtained through the category of the climbing state' P And C (k)' P 。
Step 103: correcting ultra-short-term prediction number series P of t moments after current moment k based on future climbing state matrix o And determining an ultra-short-term wind power prediction result.
In this step, it should be noted that the ultrashort-term prediction sequence P at t times after the current time k is corrected by the future climbing state matrix o Obtaining an ultra-short-term predicted value at the final time of 0-t, and determining an ultra-short-term wind power predicted result.
For a better understanding of this step, it may be, for example:
the confidence coefficient of the climbing matrix obtained by prediction at the previous moment is calculated, and the calculation method comprises the following steps:
λ＝E|C(k-1) R ·C(k-1)′ P |-E|C(k-1) R |·E|C(k-1)′ P |
ultra-short-term prediction sequence P for original future n points o The prediction climbing matrix is used for correction, and the calculation method comprises the following steps:
P f ＝P o +λ·F(k) -1 ·C(k)′ P ·E(k) -1 F(k) -T
finally obtained P f And the result is the ultra-short-term wind power prediction result of n points in the future.
Fig. 2 is a comparison diagram of power of a certain wind farm in northwest regions, wherein the ultra-short-term prediction is a prediction value of 2 hours, and it can be seen that the original ultra-short-term prediction has obvious time delay, and the time delay phenomenon is obviously reduced after the climbing prediction is added.
According to the technical scheme, according to the ultra-short-period wind power prediction method provided by the embodiment of the invention, aiming at the time delay phenomenon of the current ultra-short-period wind power prediction result, the ultra-short-period prediction array P of t moments after the current moment k is determined according to the real-time power, the real-time wind speed and the short-period prediction power of the current moment k of the wind power plant o Then, a climbing matrix corresponding to short-term preset power at t times after the current time k is input into a climbing prediction probability model, a climbing state class is obtained, and a future climbing state matrix corresponding to the climbing state class is determined; the climbing prediction probability model is obtained by training based on a machine learning algorithm by adopting climbing matrix sample data of historical prediction power of a wind farm and climbing characteristics corresponding to a climbing matrix of the historical prediction power as input data and adopting climbing state categories corresponding to the climbing matrix of the historical prediction power as output data; correcting ultra-short-term prediction number series P of t moments after current moment k through future climbing state matrix o The method can accurately predict the climbing state category at the current moment to determine the climbing type, and the ultra-short-term prediction array P is based on the climbing type o And correcting is carried out, so that an ultra-short-period wind power prediction result is determined, the time delay phenomenon of ultra-short-period prediction can be effectively reduced, the ultra-short-period prediction power prediction precision of a wind power plant is improved, and the prediction accuracy is improved.
On the basis of the above embodiment, in this embodiment, further includes:
acquiring wind farm historical data; wherein the wind farm history data comprises wind farm historic actual wind speed and wind farm historic actual power generation;
based on the wind farm historical data and a preset time window length n, calculating and determining a corresponding actual climbing matrix window by window according to a time sequence;
dividing the actual climbing matrix into j classes according to the corresponding characteristics of the actual climbing matrix to obtain the center point of the j classes of actual climbing matrix;
acquiring historical short-term prediction power, and determining a corresponding prediction climbing matrix and climbing characteristics of the prediction climbing matrix by adopting the window length of the preset time window according to window-by-window calculation of a time sequence;
And taking the central point of the j-class actual climbing matrix as an output value, taking the climbing characteristics of the prediction climbing matrix and the prediction climbing matrix as input values, and establishing the climbing prediction probability model.
In this embodiment, it should be noted that, the historical actual wind speed of the wind farm and the historical actual wind power of the wind farm are obtained, then a time window with a length of n is taken for the historical data, the interval length can be determined according to the total amount of the historical data and the length of the time window, and the corresponding actual climbing matrix is obtained by calculating window by window according to the time sequence. For example, by taking a time window for the actual data and calculating the climbing time length, t clam ＝t max +t min Wherein t is max T is the time point corresponding to the maximum value of the actual power min The time point corresponding to the minimum value of the actual power; constructing a climbing feature matrix according to the climbing time length: c (k) =f (k) ·e (k) ·p delta ·F(k) T Wherein F (k) is n t of the kth point clam The order filter matrix, E (k), is the identity matrix. The actual climbing matrix can be divided into j classes by a K-means clustering method, and the clustering method can also be realized by adopting other traditional clustering methods. The clustering method comprises the following steps: length of climbing t clam Climbing peak P max Climbing start point P min And a climbing matrix C (k), wherein the clustering output is the center point of the j-class climbing matrix; the main purpose of clustering is to exclude few extreme cases and abnormal cases, the number j of categories can be determined according to requirements, and the larger the j value is, the larger the probability that the extreme cases are taken as one category is, and the smaller the j value is, the fewer the climbing variety is reflected.
In this embodiment, it should be noted that, the historical short-term prediction power is obtained, and the corresponding prediction climbing matrix and the climbing characteristic of the prediction climbing matrix are calculated by using the same time window, where the calculation mode of the prediction climbing matrix is the same as that of the actual climbing matrix. Wherein, climbing characteristics include: and (5) climbing time length, climbing peak, and climbing starting point.
In this embodiment, the center point of the j-class actual climbing matrix is taken as a target value, the predicted climbing matrix and the characteristics thereof are taken as characteristic values, and a climbing prediction probability model is constructed, that is, the predicted climbing matrix and the characteristics thereof are taken as inputs, the class m (m= … j) corresponding to the actual climbing matrix at the moment corresponding to the predicted climbing matrix is taken as outputs, a probability model for predicting the climbing state through the predicted climbing matrix is manufactured by using a random forest method, and the probability model can also be realized by adopting other traditional probability models.
According to the technical scheme, the method for predicting the ultra-short-term wind power provided by the embodiment of the invention takes the center point of the j-class actual climbing matrix as a target value, takes the predicted climbing matrix and the characteristics thereof as characteristic values, constructs a climbing prediction probability model, the climbing state has N classes, calculates the probability of each class in the N classes, and can determine the climbing type by taking the class with the highest probability corresponding to the climbing prediction probability, so that the follow-up determination of the ultra-short-term wind power prediction result is carried out according to the predicted climbing type, and a more accurate prediction result is obtained.
On the basis of the foregoing embodiment, in this embodiment, the determining a climbing matrix specifically includes:
according to the first relation model, performing time difference on the power of the wind power station; wherein the first relationship model is as follows:
P delta ＝P-P sheft(1)
wherein P is delta Is the time sequence difference of the power, P is the grid-connected power, P sheft(1) Is a grid-connected power offset value;
calculating the climbing time length according to the second relation model; wherein the second relationship model is as follows:
t clam ＝t max +t min
wherein t is clam For climbing time length, t max Is the actual powerMaximum value corresponds to time point, t min A corresponding time point for the actual power minimum value;
determining a climbing matrix according to the third relation model; wherein the third relationship model is as follows:
C(k)＝F(k)·E(k)·P delta ·F(k) T
wherein C (k) is a climbing matrix, k represents the current moment, F (k) is n t of the kth point clam Filtering matrix, E (k) is identity matrix, P delta F (k) is the time series difference of the power T N t representing the kth point clam The transpose of the filter matrix, T represents the transpose, and n is the preset time window length.
In this embodiment, for example, the method for calculating the climbing matrix is:
firstly, performing time difference on actual grid-connected power of a power station:
P delta ＝P-P sheft(1)
wherein P is delta Is the time sequence difference of the power, P is the grid-connected power, P sheft(1) Is a grid-tied power offset value.
Then, by taking a time window for the actual data and calculating the climbing time length:
t clam ＝t max +t min
wherein t is clam For climbing time length, t max For the corresponding time point of the maximum value of the actual power, t min Corresponding to the actual power minimum.
Constructing a climbing matrix according to the climbing time length:
C(k)＝F(k)·E(k)·P delta ·F(k) T
wherein C (k) is a climbing matrix, k represents the current time, k=1, …, L-n, L is the total length of the data, and F (k) is n×t of the kth point clam Filtering matrix, E (k) is identity matrix, P delta F (k) is the time series difference of the power T N t representing the kth point clam The transpose of the filter matrix, T represents the transpose, and n is the preset time window length.
Based on the foregoing embodiment, in this embodiment, the classifying the actual climbing matrix into j classes according to the corresponding features of the actual climbing matrix specifically includes:
the actual climbing matrix is classified into j classes according to the corresponding characteristics of the actual climbing matrix by adopting a k-means clustering method; wherein, the actual climbing matrix correspondingly features include: length of climbing t clam Climbing peak P max Climbing start point P min 。
In this embodiment, it should be noted that, the actual climbing matrix is divided into j classes by using a k-means clustering method, and the k-means clustering method is an unsupervised clustering algorithm, so that the implementation is relatively simple, the clustering effect is excellent, and the convergence speed is high. The principle of the algorithm is that for a given sample set, the sample set is divided into j clusters according to the size between sample distances.
According to the technical scheme, the ultra-short-term wind power prediction method provided by the embodiment of the invention adopts the k-means clustering method, so that the implementation is simple, the clustering effect is excellent, and the convergence speed is high.
Based on the above embodiment, in this embodiment, the ultra-short-term prediction number series P of t times after the current time k is determined according to the real-time power, the real-time wind speed and the short-term prediction power of the current time k of the wind farm o The method specifically comprises the following steps:
a regression model is adopted, historical actual wind speed of the wind power plant and historical short-term prediction power of the wind power plant are input, real-time power at the moment to be predicted is output, and an ultra-short-term power regression model is established;
bringing the real-time power, the real-time wind speed and the short-term predicted power of the current moment k of the wind power plant into the ultra-short-term power regression model, and determining ultra-short-term predicted number series P of t moments after the current moment k o 。
In this embodiment, the ultrashort-period prediction sequence P at t times after the current time k is determined based on the ultrashort-period power regression model o . It should be noted that the regression model can clearly show the significant relationship between the independent and dependent variables, indicating that multiple independent variables are one-to-one The degree of influence of the individual dependent variables is thus beneficial to the data analyst to exclude and estimate a set of optimal variables for use in constructing the predictive model.
Based on the above embodiment, in this embodiment, the ultra-short term prediction sequence P at t times after the current time k is corrected based on the future climbing state matrix o Determining an ultra-short-term wind power prediction result specifically comprises the following steps:
calculating prediction confidence according to a fourth relation model based on the future climbing state matrix; wherein the fourth relationship model is:
λ＝E|C(k-1) R ·C(k-1)′ P |-EC(k-1) R |·E|C(k-1)′ P |
where λ is the confidence of the prediction, E is the desired function, C (k-1) R For the actual climbing matrix at time k-1, C (k-1)' P A future climbing state matrix corresponding to the climbing state of the target class corresponding to the actual climbing matrix at the moment k-1, wherein k represents the current moment;
according to a fifth relation model, ultra-short-term prediction sequence P of t times after the current time k is based on prediction confidence o Correcting; wherein the fifth relationship model is:
P f ＝P o +λ·F(k) -1 ·C(k)′ P ·E(k) -1 F(k) -T
wherein P is f Representing ultra-short-term wind power prediction results P of t times after current time k o An ultrashort-term prediction sequence of t times after the current time k is represented, lambda is prediction confidence, F (k) represents a filter matrix of k points, and C (k)' P Expressed as a future climbing state matrix corresponding to the climbing state type at the time k, E (k) is an identity matrix, E (k) -1 Represents the inversion of E (k), F (k) -T The transpose of F (k) is inverted, and is an operator.
According to the technical scheme, the ultra-short-term wind power prediction method provided by the embodiment of the invention is based on the future climbing state matrix, and the prediction confidence is calculated according to the fourth relation model, so that the corresponding probability is obtainedHow large the rate is, the larger the confidence interval is, the higher the confidence level is; ultra-short-term prediction sequence P for original future n points on the basis o (i.e., ultra-short-term prediction sequence P at t times after the current time k) o ) And correcting is carried out, so that an ultra-short-period wind power prediction result is determined, the time delay phenomenon of ultra-short-period prediction can be effectively reduced, the ultra-short-period prediction power prediction precision of a wind power plant is improved, and the prediction accuracy is improved.
On the basis of the foregoing embodiment, in this embodiment, after obtaining the wind farm history data, the method further includes:
performing data cleaning and feature construction on the wind farm historical data;
wherein the data cleaning comprises removing dead values, and/or removing abnormal values, and/or limiting power judgment, and/or limiting power reduction;
The feature construction comprises data expansion, wherein data containing effective features but with the data quantity smaller than a preset value are copied and filled into an original data sequence according to a time sequence.
According to the technical scheme, the ultra-short-term wind power prediction method provided by the embodiment of the invention obtains objective and accurate historical data by carrying out data cleaning and feature construction on the historical data of the wind power plant, and the historical data contains more effective features, so that the follow-up prediction work is facilitated.
Fig. 3 is a schematic structural diagram of an ultra-short-term wind power prediction apparatus according to an embodiment of the present invention, as shown in fig. 3, the apparatus includes: a first determination module 201, a second determination module 202, and a correction module 203, wherein:
wherein, the first determining module 201 is configured to determine an ultrashort-period prediction array P at t times after the current time k according to the real-time power, the real-time wind speed and the short-period prediction power of the current time k of the wind farm o ；
A second determining module 202, configured to determine a climbing matrix corresponding to short-term preset power at t times after the current time k, input climbing features of the climbing matrix and the climbing matrix corresponding to the climbing matrix to a climbing prediction probability model, obtain a climbing state class, and determine a future climbing state matrix corresponding to the climbing state class;
A correction module 203 for correcting the ultra-short-term prediction sequence P at t times after the current time k based on the future climbing state matrix o Determining an ultra-short-term wind power prediction result;
the climbing prediction probability model is obtained by training based on a machine learning algorithm by adopting climbing matrix sample data of historical prediction power of a wind farm and climbing characteristics corresponding to a climbing matrix of the historical prediction power as input data and adopting climbing state categories corresponding to the climbing matrix of the historical prediction power as output data.
The ultra-short-term wind power prediction device provided by the embodiment of the invention can be particularly used for executing the ultra-short-term wind power prediction method of the embodiment, and the technical principle and the beneficial effects of the ultra-short-term wind power prediction device are similar, and the ultra-short-term wind power prediction device can be particularly referred to the embodiment and is not repeated here.
Based on the same inventive concept, an embodiment of the present invention provides an electronic device, referring to fig. 4, including the following details: a processor 301, a communication interface 303, a memory 302 and a communication bus 304;
wherein, the processor 301, the communication interface 303 and the memory 302 complete the communication with each other through the communication bus 304; the communication interface 303 is used for realizing information transmission between the modeling software and related devices such as an intelligent manufacturing equipment module library; the processor 301 is configured to invoke a computer program in the memory 302, and when the processor executes the computer program, the method provided by the above method embodiments is implemented, for example, when the processor executes the computer program, the following steps are implemented: determining ultra-short-term prediction series P of t moments after the current moment k according to the real-time power, the real-time wind speed and the short-term prediction power of the current moment k of the wind power plant o The method comprises the steps of carrying out a first treatment on the surface of the Determining a climbing matrix corresponding to short-term preset power at t times after the current time k, inputting climbing characteristics corresponding to the climbing matrix and the climbing matrix into a climbing prediction probability model to obtain a climbing state category, and determining the climbing state category with the climbing matrixThe state category corresponds to a future climbing state matrix; correcting ultra-short-term prediction number series P of t moments after current moment k based on future climbing state matrix o Determining an ultra-short-term wind power prediction result; the climbing prediction probability model is obtained by training based on a machine learning algorithm by adopting climbing matrix sample data of historical prediction power of a wind farm and climbing characteristics corresponding to a climbing matrix of the historical prediction power as input data and adopting climbing state categories corresponding to the climbing matrix of the historical prediction power as output data.
Based on the same inventive concept, a further embodiment of the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the methods provided by the above method embodiments, e.g. determining an ultrashort-predicted sequence P of t moments after a current moment k of a wind farm, based on a real-time power, a real-time wind speed and a short-term predicted power of said current moment k o The method comprises the steps of carrying out a first treatment on the surface of the Determining a climbing matrix corresponding to short-term preset power at t times after the current time k, inputting climbing characteristics corresponding to the climbing matrix and the climbing matrix into a climbing prediction probability model to obtain a climbing state class, and determining a future climbing state matrix corresponding to the climbing state class; correcting ultra-short-term prediction number series P of t moments after current moment k based on future climbing state matrix o Determining an ultra-short-term wind power prediction result; the climbing prediction probability model is obtained by training based on a machine learning algorithm by adopting climbing matrix sample data of historical prediction power of a wind farm and climbing characteristics corresponding to a climbing matrix of the historical prediction power as input data and adopting climbing state categories corresponding to the climbing matrix of the historical prediction power as output data.
The apparatus embodiments described above are merely illustrative, wherein elements illustrated as separate elements may or may not be physically separate, and elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on such understanding, the foregoing technical solutions may be embodied essentially or in part in the form of a software product, which may be stored in a computer-readable storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the various embodiments or methods of some parts of the embodiments.
Furthermore, in the present disclosure, such as "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present invention, the meaning of "plurality" means at least two, for example, two, three, etc., unless specifically defined otherwise.
Moreover, in the present invention, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.
Furthermore, in the description herein, reference to the terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (8)
1. An ultra-short term wind power prediction method is characterized by comprising the following steps:
acquiring wind farm historical data; wherein the wind farm history data comprises wind farm historic actual wind speed and wind farm historic actual power generation;
based on the wind farm historical data and a preset time window length n, calculating and determining a corresponding actual climbing matrix window by window according to a time sequence;
dividing the actual climbing matrix into j classes according to the corresponding characteristics of the actual climbing matrix to obtain the center point of the j classes of actual climbing matrix;
acquiring historical short-term prediction power, and determining a corresponding prediction climbing matrix and climbing characteristics of the prediction climbing matrix by adopting the window length of the preset time window according to window-by-window calculation of a time sequence;
Taking the central point of the j-class actual climbing matrix as an output value, taking the climbing characteristics of the prediction climbing matrix and the prediction climbing matrix as input values, and establishing a climbing prediction probability model;
determining ultra-short-term prediction series P of t moments after the current moment k according to the real-time power, the real-time wind speed and the short-term prediction power of the current moment k of the wind power plant o ；
Determining a climbing matrix corresponding to short-term preset power at t times after the current time k, inputting climbing characteristics corresponding to the climbing matrix and the climbing matrix into the climbing prediction probability model to obtain a climbing state class, and determining a future climbing state matrix corresponding to the climbing state class;
correcting ultra-short-term prediction number series P of t moments after current moment k based on future climbing state matrix o Determining an ultra-short-term wind power prediction result;
the climbing prediction probability model is obtained by training based on a machine learning algorithm by adopting climbing matrix sample data of historical prediction power of a wind farm and climbing characteristics corresponding to a climbing matrix of the historical prediction power as input data and adopting climbing state categories corresponding to the climbing matrix of the historical prediction power as output data;
The determination process of the climbing matrix is as follows:
according to the first relation model, performing time difference on the power of the wind power station; wherein the first relationship model is as follows: p (P) delta ＝P-P sheft(1) The method comprises the steps of carrying out a first treatment on the surface of the Wherein P is delta Is the time sequence difference of the power, P is the grid-connected power, P sheft(1) Is a grid-connected power offset value;
calculating the climbing time length according to the second relation model; wherein the second relationship model is as follows: t is t clam ＝t max +t min The method comprises the steps of carrying out a first treatment on the surface of the Wherein t is clam For climbing time length, t max For the corresponding time point of the maximum value of the actual power, t min A corresponding time point for the actual power minimum value;
determining a climbing matrix according to the third relation model; wherein the third relationship model is as follows: c (k) =f (k) ·e (k) ·p delta ·F(k) T The method comprises the steps of carrying out a first treatment on the surface of the Wherein C (k) is a climbing matrix, k represents the current moment, and F (k) is the kth pointFiltering matrix, E (k) is identity matrix, P delta F (k) is the time series difference of the power T Represents +.>The transpose of the filter matrix, T represents the transpose, and n is the preset time window length.
2. The ultra-short term wind power prediction method according to claim 1, wherein the classifying the actual climbing matrix into j classes according to the corresponding features of the actual climbing matrix specifically includes:
the actual climbing matrix is classified into j classes according to the corresponding characteristics of the actual climbing matrix by adopting a k-means clustering method; wherein, the actual climbing matrix correspondingly features include: length of climbing t clam Climbing peak P max Climbing start point P min 。
3. The ultra-short term wind power prediction method according to claim 1, wherein the ultra-short term prediction array P of t times after the current time k is determined according to the real-time power, the real-time wind speed and the short term prediction power of the current time k of the wind farm o The method specifically comprises the following steps:
a regression model is adopted, historical actual wind speed of the wind power plant and historical short-term prediction power of the wind power plant are input, real-time power at the moment to be predicted is output, and an ultra-short-term power regression model is established;
bringing the real-time power, the real-time wind speed and the short-term predicted power of the current moment k of the wind power plant into the ultra-short-term power regression model, and determining ultra-short-term predicted number series P of t moments after the current moment k o 。
4. The ultra-short term wind power prediction method according to claim 1, wherein the ultra-short term prediction sequence P at t times after the current time k is corrected based on a future climbing state matrix o Determining an ultra-short-term wind power prediction result specifically comprises the following steps:
calculating prediction confidence according to a fourth relation model based on the future climbing state matrix; wherein the fourth relationship model is:
λ＝E|C(k-1) R ·C(k-1)' P |-E|C(k-1) R |·E|C(k-1)' P |
Where λ is the confidence of the prediction, E is the desired function, C (k-1) R For the actual climbing matrix at time k-1, C (k-1)' P A future climbing state matrix corresponding to the climbing state of the target class corresponding to the actual climbing matrix at the moment k-1, wherein k represents the current moment;
according to a fifth relation model, ultra-short-term prediction sequence P of t times after the current time k is based on prediction confidence o Correcting; wherein the fifth relationship model is:
P f ＝P o +λ·F(k) -1 ·C(k)' P ·E(k) -1 F(k) -T
wherein P is f Representing ultra-short-term wind power prediction results P of t times after current time k o An ultrashort-term prediction sequence of t times after the current time k is represented, lambda is prediction confidence, F (k) represents a filter matrix of k points, and C (k)' P Expressed as a future climbing state matrix corresponding to the climbing state type at the time k, E (k) is an identity matrix, E (k) -1 Represents the inversion of E (k), F (k) -T The transpose of F (k) is inverted.
5. The ultra-short term wind power prediction method according to claim 1, further comprising, after obtaining the wind farm history data:
performing data cleaning and feature construction on the wind farm historical data;
wherein the data cleaning comprises removing dead values, and/or removing abnormal values, and/or limiting power judgment, and/or limiting power reduction;
The feature construction comprises data expansion, wherein data containing effective features but with the data quantity smaller than a preset value are copied and filled into an original data sequence according to a time sequence.
6. An ultra-short term wind power prediction device, comprising:
the first determining module is used for acquiring wind power plant historical data; wherein the wind farm history data comprises wind farm historic actual wind speed and wind farm historic actual power generation; based on the wind farm historical data and a preset time window length n, calculating and determining a corresponding actual climbing matrix window by window according to a time sequence; dividing the actual climbing matrix into j classes according to the corresponding characteristics of the actual climbing matrix to obtain the center point of the j classes of actual climbing matrix; acquiring historical short-term prediction power, and determining a corresponding prediction climbing matrix and climbing characteristics of the prediction climbing matrix by adopting the window length of the preset time window according to window-by-window calculation of a time sequence; taking the central point of the j-class actual climbing matrix as an output value, taking the climbing characteristics of the prediction climbing matrix and the prediction climbing matrix as input values, and establishing a climbing prediction probability model; determining ultra-short-term prediction series P of t moments after the current moment k according to the real-time power, the real-time wind speed and the short-term prediction power of the current moment k of the wind power plant o ；
The second determining module is used for determining a climbing matrix corresponding to short-term preset power at t times after the current time k, inputting climbing characteristics corresponding to the climbing matrix and the climbing matrix into the climbing prediction probability model to obtain a climbing state class, and determining a future climbing state matrix corresponding to the climbing state class;
the correction module is used for correcting the ultra-short-term prediction array P at t times after the current time k based on the future climbing state matrix o Determining an ultra-short-term wind power prediction result;
the climbing prediction probability model is obtained by training based on a machine learning algorithm by adopting climbing matrix sample data of historical prediction power of a wind farm and climbing characteristics corresponding to a climbing matrix of the historical prediction power as input data and adopting climbing state categories corresponding to the climbing matrix of the historical prediction power as output data;
the determination process of the climbing matrix is as follows:
according to the first relation model, performing time difference on the power of the wind power station; wherein the first relationship model is as follows: p (P) delta ＝P-P sheft(1) The method comprises the steps of carrying out a first treatment on the surface of the Wherein P is delta Is the time sequence difference of the power, P is the grid-connected power, P sheft(1) Is a grid-connected power offset value;
calculating the climbing time length according to the second relation model; wherein the second relationship model is as follows: t is t clam ＝t max +t min The method comprises the steps of carrying out a first treatment on the surface of the Wherein t is clam For climbing time length, t max For the corresponding time point of the maximum value of the actual power, t min A corresponding time point for the actual power minimum value;
determining a climbing matrix according to the third relation model; wherein the third relationship model is as follows: c (k) =f (k) ·e (k) ·p delta ·F(k) T The method comprises the steps of carrying out a first treatment on the surface of the Wherein C (k) is a climbing matrix, k represents the current moment, and F (k) is the kth pointFiltering matrix, E (k) is identity matrix, P delta F (k) is the time series difference of the power T Represents +.>The transpose of the filter matrix, T represents the transpose, and n is the preset time window length.
7. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the ultra-short term wind power prediction method of any one of claims 1 to 5 when the program is executed by the processor.
8. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor implements the ultra-short term wind power prediction method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011172181.8A CN112270439B (en) | 2020-10-28 | 2020-10-28 | Ultra-short-term wind power prediction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011172181.8A CN112270439B (en) | 2020-10-28 | 2020-10-28 | Ultra-short-term wind power prediction method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112270439A CN112270439A (en) | 2021-01-26 |
| CN112270439B true CN112270439B (en) | 2024-03-08 |
Family
ID=74345557
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011172181.8A Active CN112270439B (en) | 2020-10-28 | 2020-10-28 | Ultra-short-term wind power prediction method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112270439B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469467B (en) * | 2021-09-02 | 2021-11-30 | 国能日新科技股份有限公司 | Wind power ultra-short term prediction method and device based on band-pass filtering |
| CN114254805A (en) * | 2021-11-22 | 2022-03-29 | 华北电力大学 | Time window identification method, device, equipment and storage medium for climbing event |
| CN117332901A (en) * | 2023-10-17 | 2024-01-02 | 南方电网数字电网研究院有限公司 | New energy small time scale power prediction method adopting layered time aggregation strategy |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008064081A (en) * | 2006-09-08 | 2008-03-21 | Yoshito Hirata | Areal wind turbine generator system by wind state prediction control corresponding to optional observation point number |
| CN103279804A (en) * | 2013-04-29 | 2013-09-04 | 清华大学 | Super short-period wind power prediction method |
| CN103679282A (en) * | 2013-09-30 | 2014-03-26 | 清华大学 | Prediction method for wind power ramp |
| CN104699936A (en) * | 2014-08-18 | 2015-06-10 | 沈阳工业大学 | Sector management method based on CFD short-term wind speed forecasting wind power plant |
| CN104732290A (en) * | 2015-03-24 | 2015-06-24 | 河海大学 | Predicating method for wind electricity power ramp event |
| CN104794546A (en) * | 2015-04-29 | 2015-07-22 | 武汉大学 | Wind power climbing forecasting method based on deep confidence network classifying method |
| CN105160434A (en) * | 2015-09-15 | 2015-12-16 | 武汉大学 | Wind power ramp event prediction method by adopting SVM to select forecasting model |
| CN106374465A (en) * | 2016-11-10 | 2017-02-01 | 南京信息工程大学 | GSA-LSSVM model-based short period wind electricity generation power prediction method |
| CN106779208A (en) * | 2016-12-08 | 2017-05-31 | 贵州电网有限责任公司电力科学研究院 | A kind of wind-powered electricity generation ultra-short term power forecasting method based on virtual anemometer tower technology |
| CN106933778A (en) * | 2017-01-22 | 2017-07-07 | 中国农业大学 | A kind of wind power combination forecasting method based on climbing affair character identification |
| CN107067099A (en) * | 2017-01-25 | 2017-08-18 | 清华大学 | Wind power probability forecasting method and device |
| CN107909227A (en) * | 2017-12-20 | 2018-04-13 | 北京金风慧能技术有限公司 | Ultra-short term predicts the method, apparatus and wind power generating set of wind power |
| CN108074015A (en) * | 2017-12-25 | 2018-05-25 | 中国电力科学研究院有限公司 | A kind of ultrashort-term wind power prediction method and system |
| US10041475B1 (en) * | 2017-02-07 | 2018-08-07 | International Business Machines Corporation | Reducing curtailment of wind power generation |
| CN109033027A (en) * | 2018-08-08 | 2018-12-18 | 长春工程学院 | A kind of high speed fitful wind leads to the prediction technique of climbing event under wind power |
| CN109523060A (en) * | 2018-10-22 | 2019-03-26 | 上海交通大学 | Ratio optimization method of the high proportion renewable energy under transmission and distribution network collaboration access |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110149268A1 (en) * | 2009-12-17 | 2011-06-23 | Marchant Alan B | Dynamic 3d wind mapping system and method |
| US8600572B2 (en) * | 2010-05-27 | 2013-12-03 | International Business Machines Corporation | Smarter-grid: method to forecast electric energy production and utilization subject to uncertain environmental variables |
| US9230219B2 (en) * | 2010-08-23 | 2016-01-05 | Institute Of Nuclear Energy Research Atomic Energy Council, Executive Yuan | Wind energy forecasting method with extreme wind speed prediction function |
| US20130184838A1 (en) * | 2012-01-06 | 2013-07-18 | Michigan Aerospace Corporation | Resource optimization using environmental and condition-based monitoring |
| EP2820735A1 (en) * | 2012-04-11 | 2015-01-07 | Siemens Aktiengesellschaft | Electric device, and method for controlling an electric energy generator |
-
2020
- 2020-10-28 CN CN202011172181.8A patent/CN112270439B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008064081A (en) * | 2006-09-08 | 2008-03-21 | Yoshito Hirata | Areal wind turbine generator system by wind state prediction control corresponding to optional observation point number |
| CN103279804A (en) * | 2013-04-29 | 2013-09-04 | 清华大学 | Super short-period wind power prediction method |
| CN103679282A (en) * | 2013-09-30 | 2014-03-26 | 清华大学 | Prediction method for wind power ramp |
| CN104699936A (en) * | 2014-08-18 | 2015-06-10 | 沈阳工业大学 | Sector management method based on CFD short-term wind speed forecasting wind power plant |
| CN104732290A (en) * | 2015-03-24 | 2015-06-24 | 河海大学 | Predicating method for wind electricity power ramp event |
| CN104794546A (en) * | 2015-04-29 | 2015-07-22 | 武汉大学 | Wind power climbing forecasting method based on deep confidence network classifying method |
| CN105160434A (en) * | 2015-09-15 | 2015-12-16 | 武汉大学 | Wind power ramp event prediction method by adopting SVM to select forecasting model |
| CN106374465A (en) * | 2016-11-10 | 2017-02-01 | 南京信息工程大学 | GSA-LSSVM model-based short period wind electricity generation power prediction method |
| CN106779208A (en) * | 2016-12-08 | 2017-05-31 | 贵州电网有限责任公司电力科学研究院 | A kind of wind-powered electricity generation ultra-short term power forecasting method based on virtual anemometer tower technology |
| CN106933778A (en) * | 2017-01-22 | 2017-07-07 | 中国农业大学 | A kind of wind power combination forecasting method based on climbing affair character identification |
| CN107067099A (en) * | 2017-01-25 | 2017-08-18 | 清华大学 | Wind power probability forecasting method and device |
| US10041475B1 (en) * | 2017-02-07 | 2018-08-07 | International Business Machines Corporation | Reducing curtailment of wind power generation |
| CN107909227A (en) * | 2017-12-20 | 2018-04-13 | 北京金风慧能技术有限公司 | Ultra-short term predicts the method, apparatus and wind power generating set of wind power |
| CN108074015A (en) * | 2017-12-25 | 2018-05-25 | 中国电力科学研究院有限公司 | A kind of ultrashort-term wind power prediction method and system |
| CN109033027A (en) * | 2018-08-08 | 2018-12-18 | 长春工程学院 | A kind of high speed fitful wind leads to the prediction technique of climbing event under wind power |
| CN109523060A (en) * | 2018-10-22 | 2019-03-26 | 上海交通大学 | Ratio optimization method of the high proportion renewable energy under transmission and distribution network collaboration access |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112270439A (en) | 2021-01-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112270439B (en) | Ultra-short-term wind power prediction method and device, electronic equipment and storage medium | |
| Mahmoud et al. | An advanced approach for optimal wind power generation prediction intervals by using self-adaptive evolutionary extreme learning machine | |
| CN110852902A (en) | Photovoltaic power generation power prediction method based on BAS-BP | |
| CN104573876A (en) | Wind power plant short-period wind speed prediction method based on time sequence long memory model | |
| CN109088407B (en) | Power distribution network state estimation method based on deep belief network pseudo-measurement modeling | |
| CN106875033A (en) | A kind of wind-powered electricity generation cluster power forecasting method based on dynamic self-adapting | |
| CN114492675B (en) | Intelligent fault cause diagnosis method for capacitor voltage transformer | |
| CN111523728B (en) | Four-stage hybrid short-term wind direction prediction method | |
| CN111461463A (en) | Short-term load prediction method, system and equipment based on TCN-BP | |
| CN115049024B (en) | Training method and device of wind speed prediction model, electronic equipment and storage medium | |
| CN116306798A (en) | Ultra-short time wind speed prediction method and system | |
| CN116680540A (en) | Wind power prediction method based on deep learning | |
| Sharma et al. | Wind speed forecasting using hybrid ANN-Kalman filter techniques | |
| CN110096730B (en) | Method and system for rapidly evaluating voltage of power grid | |
| CN109146131A (en) | A kind of wind-power electricity generation prediction technique a few days ago | |
| Phan et al. | Application of a new Transformer-based model and XGBoost to improve one-day-ahead solar power forecasts | |
| CN109840308B (en) | Regional wind power probability forecasting method and system | |
| CN114611799B (en) | Time sequence neural network new energy output multi-step prediction method based on supervised learning | |
| CN113991752B (en) | Quasi-real-time intelligent control method and system for power grid | |
| CN115907131A (en) | Method and system for building electric heating load prediction model in northern area | |
| Guoqiang et al. | Study of RBF neural network based on PSO algorithm in nonlinear system identification | |
| CN114638396A (en) | Photovoltaic power prediction method and system based on neural network instantiation | |
| CN114638421A (en) | Method for predicting requirement of generator set spare parts | |
| Zhang et al. | Wind farm wind power prediction method based on CEEMDAN and DE optimized DNN neural network | |
| Niu et al. | Research on power load forecasting based on combined model of Markov and BP neural networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |