CN111915079B - Hybrid KNN wind power prediction method and system - Google Patents
Hybrid KNN wind power prediction method and system Download PDFInfo
- Publication number
- CN111915079B CN111915079B CN202010748717.XA CN202010748717A CN111915079B CN 111915079 B CN111915079 B CN 111915079B CN 202010748717 A CN202010748717 A CN 202010748717A CN 111915079 B CN111915079 B CN 111915079B
- Authority
- CN
- China
- Prior art keywords
- data
- model
- wind power
- knn
- wind
- 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 38
- 238000012549 training Methods 0.000 claims abstract description 47
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 21
- 238000012545 processing Methods 0.000 claims description 38
- 238000012795 verification Methods 0.000 claims description 25
- 238000013499 data model Methods 0.000 claims description 13
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 238000010248 power generation Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004519 manufacturing process 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
- 238000005457 optimization Methods 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Evolutionary Computation (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)
- Development Economics (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Complex Calculations (AREA)
Abstract
The invention discloses a hybrid KNN wind power prediction method and a hybrid KNN wind power prediction system, wherein input data are obtained, and meteorological data before a certain day period is predicted, wherein the meteorological data mainly comprise temperature, humidity, wind speed, wind direction and power and are used as training input data; inputting input data into a pre-trained mixed KNN wind power prediction model combining a fractal theory and a KNN algorithm; and outputting the prediction data of the wind power. The advantages are that: according to the method, local information of adjacent samples can be effectively stored by utilizing a fractal interpolation idea, so that most characteristics of the samples are effectively reserved through an interpolation curve, and a user-defined KNN algorithm is combined, so that the model is relatively simple, the complexity is low, the prediction performance is improved, an effective way is provided for improving the wind power prediction precision, and the influence of wind power grid connection on the stability of a power grid is reduced.
Description
Technical Field
The invention relates to a hybrid KNN wind power prediction method and a hybrid KNN wind power prediction system, and belongs to the technical field of wind power and solar power generation.
Background
In recent years, the country has increased support for clean energy, the wind power generation industry of China has been rapidly developed, fossil energy on the earth is gradually depleted, and research, development and utilization of new energy are imperative. Among them, the wind energy resource has many advantages, such as wide range, almost no pollution, and recycling, and is one of the most potential new energy resources. However, natural wind has the characteristics of strong randomness and intermittence, and can bring certain threat to the stable operation of a power grid when large-scale centralized grid connection is carried out. The method has the advantages that the wind power in a future period can be accurately predicted, and the method has important significance for power dispatching and safe operation.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a hybrid KNN wind power prediction method and a hybrid KNN wind power prediction system.
In order to solve the technical problem, the invention provides a hybrid KNN wind power prediction method, which comprises the steps of obtaining input data, and taking meteorological data before predicting a certain day period, which mainly comprises temperature, humidity, wind speed, wind direction and power, as training input data;
inputting input data into a pre-trained mixed KNN wind power prediction model combining a fractal theory and a KNN algorithm;
and outputting the prediction data of the wind power.
Further, the construction process of the hybrid KNN wind power prediction model comprises the following steps:
acquiring an original data set of wind power data of a wind power plant at a certain historical time, and dividing the original data set into a front part and a rear part according to a time sequence, wherein the front part is used as a training set, and the other part is used as a verification set;
performing fractal interpolation processing on the original data set to obtain a fractal dimension result set;
and (4) performing mixed KNN wind power prediction model training according to the fractal dimension result set, the training set and the verification set, and determining an optimal model.
Further, the process of performing fractal interpolation processing on the original data set to obtain a fractal dimension result set includes:
acquiring data characteristics from an original data set, wherein the data characteristics comprise five-dimensional data of time series, meteorological temperature, wind speed, wind direction and available output;
discretizing available output data, meteorological temperature, wind speed and wind direction data by using an entropy maximum discretization method to obtain discretization data;
and calculating the fractal dimension of available output data, meteorological temperature, wind speed and wind direction according to the discretization data, and determining a fractal dimension result set.
Further, the mixed KNN wind power prediction model training is carried out according to the fractal dimension result set, the training set and the verification set, and the process of determining the optimal model comprises the following steps:
training the model continuously by iteration under the condition that the obtained fractal dimension result set and the training set meet the circulation to obtain an initial model, optimizing a K value and a weight coefficient in a KNN algorithm by adopting a full-grid search method, taking a decision coefficient r2_ score value as an evaluation index in a verification set, wherein the better the r2_ score value is close to 1, the better the model is, the closer the r2_ score value is to 0, the worse the model is, and the optimal model parameters are stored through continuous iteration to obtain the optimal model.
Further, setting the model expiration time to be 15 days, training wind power data 15 days before the forecast day, loading the optimal model parameters obtained in the previous step, and storing the trained data model to the local;
and inputting the wind speed, wind direction, temperature and humidity meteorological information of the day before the forecast date, and loading the trained data model in the previous step to obtain the forecast power value.
A hybrid KNN wind power prediction system,
the system comprises an acquisition module, a data processing module and a data processing module, wherein the acquisition module is used for acquiring input data, and the meteorological data before a certain day period is predicted to mainly comprise temperature, humidity, wind speed, wind direction and power as training input data;
the processing module is used for inputting input data into a pre-trained mixed KNN wind power prediction model combined with a fractal theory and a KNN algorithm;
and the output module is used for outputting the prediction data of the wind power.
Further, the processing module comprises:
the system comprises a diversity module, a verification module and a data processing module, wherein the diversity module is used for acquiring an original data set of wind power data of a wind power plant at a certain historical time period, and dividing the original data set into a front part and a rear part according to a time sequence, wherein the front part is used as a training set, and the other part is used as a verification set;
the fractal interpolation processing module is used for carrying out fractal interpolation processing on the original data set to obtain a fractal dimension result set;
and the determining module is used for carrying out mixed KNN wind power prediction model training according to the fractal dimension result set, the training set and the verification set so as to determine an optimal model.
Further, the fractal interpolation processing module includes:
the characteristic extraction module is used for extracting data characteristics from the original data set, wherein the data characteristics comprise five-dimensional data of time sequence, meteorological temperature, wind speed, wind direction and available output;
the discretization processing module is used for performing discretization processing on the available output data, the meteorological temperature, the wind speed and the wind direction data by using an entropy maximum discretization method to obtain discretization data;
and the calculation module is used for calculating the fractal dimension of available output data, meteorological temperature, wind speed and wind direction according to the discretization data and determining a fractal dimension result set.
Further, the determining module is further configured to continuously iterate to train the model according to the obtained fractal dimension result set and the training set under the condition that a loop is satisfied, to obtain an initial model, optimize a K value and a weight coefficient in the KNN algorithm by using a full-grid search method, and use a decision coefficient r2_ score value as an evaluation index in the verification set, where the r2_ score value is closer to 1, indicating that the model is better, the r2_ score value is closer to 0, indicating that the model is worse, and continuously iterating to store optimal model parameters, so as to obtain an optimal model.
Further, the determining module is further configured to set the model expiration time to 15 days, train and predict wind power data 15 days before the day, load the optimal model parameters obtained in the previous step, and store the trained data model to the local;
and inputting the wind speed, wind direction, temperature and humidity meteorological information of the day before the forecast date, and loading the trained data model in the previous step to obtain the forecast power value.
The invention has the following beneficial effects:
according to the method, local information of adjacent samples can be effectively stored by utilizing a fractal interpolation idea, so that most characteristics of the samples are effectively reserved through an interpolation curve, and a user-defined KNN algorithm is combined, so that the model is relatively simple, the complexity is low, the prediction performance is improved, an effective way is provided for improving the wind power prediction precision, and the influence of wind power grid connection on the stability of a power grid is reduced.
Drawings
FIG. 1 is a flow chart of a mixed KNN wind power prediction model based on a fractal theory;
FIG. 2 is a comparison of the effect of the prediction model of the present invention and other conventional models.
Detailed Description
In order to make the objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is obvious that the embodiments described below are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The technical scheme of the invention is further explained by the specific implementation mode in combination with the attached drawings.
As shown in fig. 1, a hybrid KNN wind power prediction method,
acquiring input data, and taking meteorological data which is predicted before a certain day and date and mainly comprises temperature, humidity, wind speed, wind direction and power as training input data;
inputting input data into a pre-trained mixed KNN wind power prediction model combining a fractal theory and a KNN algorithm;
and outputting the prediction data of the wind power.
The construction process of the hybrid KNN wind power prediction model comprises the following steps:
acquiring an original data set of wind power data of a wind power plant at a certain historical time, and dividing the original data set into a front part and a rear part according to a time sequence, wherein the front part is used as a training set, and the other part is used as a verification set;
performing fractal interpolation processing on the original data set to obtain a fractal dimension result set;
and (4) performing mixed KNN wind power prediction model training according to the fractal dimension result set, the training set and the verification set, and determining an optimal model.
The process of performing fractal interpolation processing on the original data set to obtain a fractal dimension result set comprises the following steps:
acquiring data characteristics from an original data set, wherein the data characteristics comprise five-dimensional data of time series, meteorological temperature, wind speed, wind direction and available output;
discretizing available output data, meteorological temperature, wind speed and wind direction data by using an entropy maximum discretization method to obtain discretization data;
and calculating the fractal dimension of available output data, meteorological temperature, wind speed and wind direction according to the discretization data, and determining a fractal dimension result set.
The mixed KNN wind power prediction model training is carried out according to the fractal dimension result set, the training set and the verification set, and the process of determining the optimal model comprises the following steps:
training the model by continuous iteration according to the obtained fractal dimension result set and the training set under the condition of meeting the circulation to obtain an initial model, optimizing a K value and a weight coefficient in a KNN algorithm by adopting a full-grid search method, taking a decision coefficient r2_ score value as an evaluation index in a verification set, indicating that the better the model is, indicating that the closer the r2_ score value is to 1, the closer the r2_ score value is to 0, indicating that the worse the model is, and storing optimal model parameters by continuous iteration to obtain the optimal model.
(2) In order to ensure the continuous updating of the data model so as to better predict, setting the model expiration time as 15 days, training wind power data 15 days before the prediction day, loading the optimal model parameters obtained in the previous step, and storing the trained data model to the local; (here, the data of the last 15 days are trained, in order to ensure that the generalization degree of the model is better, the latest data is transmitted to the model, and after parameter adjustment, a set of model parameters has the optimal effect, that is, the set of parameters is used in the model training of the last 15 days, for example, K =10, and the weight coefficient also has a set of optimal values).
And inputting the wind speed, wind direction, temperature and humidity meteorological information of the day before the forecast date, and loading the trained data model in the previous step to obtain the forecast power value.
Correspondingly, the invention also provides a hybrid KNN wind power prediction system which comprises an acquisition module, a comparison module and a prediction module, wherein the acquisition module is used for acquiring input data and taking meteorological data which are predicted before a certain day and mainly comprise temperature, humidity, wind speed, wind direction and power as training input data;
the processing module is used for inputting input data into a pre-trained mixed KNN wind power prediction model combined with a fractal theory and a KNN algorithm;
and the output module is used for outputting the prediction data of the wind power.
The processing module comprises:
the system comprises a diversity module, a verification module and a data processing module, wherein the diversity module is used for acquiring an original data set of wind power data of a wind power plant at a certain historical time period, and dividing the original data set into a front part and a rear part according to a time sequence, wherein the front part is used as a training set, and the other part is used as a verification set;
the fractal interpolation processing module is used for carrying out fractal interpolation processing on the original data set to obtain a fractal dimension result set;
and the determining module is used for carrying out mixed KNN wind power prediction model training according to the fractal dimension result set, the training set and the verification set so as to determine an optimal model.
The fractal interpolation processing module comprises:
the characteristic extraction module is used for extracting data characteristics from the original data set, wherein the data characteristics comprise five-dimensional data of time sequence, meteorological temperature, wind speed, wind direction and available output;
the discretization processing module is used for performing discretization processing on available output data, meteorological temperature, wind speed and wind direction data by using an entropy maximum discretization method to obtain discretization data;
and the calculation module is used for calculating the fractal dimension of available output data, meteorological temperature, wind speed and wind direction according to the discretization data and determining a fractal dimension result set.
The determining module is further used for training the model through continuous iteration under the condition that the obtained fractal dimension result set and the training set meet the circulation to obtain an initial model, optimizing the K value and the weight coefficient in the KNN algorithm by adopting a full-grid search method, taking the value of a decision coefficient r2_ score as an evaluation index in the verification set, wherein the value of r2_ score is closer to 1 to indicate that the model is better, the value of r2_ score is closer to 0 to indicate that the model is worse, and the optimal model parameters are stored through continuous iteration to obtain the optimal model.
The determining module is also used for setting the model expiration time to be 15 days, training wind power data 15 days before the predicted day, loading the optimal model parameters obtained in the previous step, and storing the trained data model to the local;
and inputting the wind speed, wind direction, temperature and humidity meteorological information of the day before the forecast date, and loading the trained data model in the previous step to obtain the forecast power value.
FIG. 2 is a comparison graph of the effect of the prediction model in the invention and other common models. Firstly, an original wind power plant data set is obtained, the data is stored as raw _ data _ df after being preprocessed, then the data is converted and stored into a slicer data _ df attribute, and a fractal dimension is calculated according to the raw _ data _ df data and is output to fractional _ dim.
Taking a certain cycle of model training as an example, setting start _ t = '2019-01-0100: 00', vali _ t = '2019-10-1800: 00', current _ date = '2019-10-1800: 15' can be known by current _ row at the current line, and obtaining df _ data of the cycle from a fractional _ dim.csv file according to the start _ t and the current _ date; obtaining latest 8 time point data last _8_ df from the slicer.data _ df by current _ date, calculating fractal dimension of last _8_ df data, storing the fractal dimension of last _8_ df data to tmp _ vector, taking df _ data, tmp _ vector, preset weight1 and K1 as first input of a KNN algorithm, and finally outputting K1 nearest fractal dimension cache ID values; finding out corresponding meteorological temperature, wind speed and wind direction values from the slicer.data _ df according to the output ID value, storing the meteorological temperature, wind speed and wind direction values as observation _ df, obtaining meteorological temperature, wind speed and wind direction values in a verification set according to the current _ row, storing a temporary variable current _ vector, then taking the calculated observation _ df, current _ vector, preset weight weights2 and K2 values as the input of a self-defined KNN algorithm, finding K2 nearest power values by calculating Euclidean distance, continuously iterating under the condition of meeting a circulation condition, simultaneously adopting a full-grid search optimization algorithm, continuously optimizing input parameters of the KNN algorithm to carry out parameter adjustment, finding out an optimal parameter set, and storing the optimal parameter set.
And after the model parameters are optimized, calculating the predicted power value of each predicted time point within 30 days in sequence through the prediction model, and comparing the predicted power value with the actual power value. Meanwhile, the method is compared with a random forest model (RFR), a support vector machine model (SVM) and a gradient lifting regression model (GBDT), the Root Mean Square Error (RMSE) and the prediction accuracy of each algorithm are calculated, and the fact that the mixed KNN algorithm prediction model based on the fractal theory has higher accuracy and effectiveness is proved.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (4)
1. A hybrid KNN wind power prediction method is characterized in that,
acquiring input data, and taking meteorological data including temperature, humidity, wind speed, wind direction and power before a certain day period as training input data;
inputting input data into a pre-trained mixed KNN wind power prediction model combining a fractal theory and a KNN algorithm;
outputting prediction data of wind power;
the construction process of the hybrid KNN wind power prediction model comprises the following steps:
acquiring an original data set of wind power data of a wind power plant at a certain historical time, and dividing the original data set into a front part and a rear part according to a time sequence, wherein the front part is used as a training set, and the other part is used as a verification set;
performing fractal interpolation processing on the original data set to obtain a fractal dimension result set;
performing mixed KNN wind power prediction model training according to the fractal dimension result set, the training set and the verification set to determine an optimal model;
the process of performing fractal interpolation processing on the original data set to obtain a fractal dimension result set comprises the following steps:
acquiring data characteristics from an original data set, wherein the data characteristics comprise five-dimensional data of time series, meteorological temperature, wind speed, wind direction and available output;
discretizing available output data, meteorological temperature, wind speed and wind direction data by using an entropy maximum discretization method to obtain discretization data;
calculating the fractal dimension of available output data, meteorological temperature, wind speed and wind direction according to the discretization data, and determining a fractal dimension result set;
the mixed KNN wind power prediction model training is carried out according to the fractal dimension result set, the training set and the verification set, and the process of determining the optimal model comprises the following steps:
training the model continuously by iteration under the condition that the obtained fractal dimension result set and the training set meet the circulation to obtain an initial model, optimizing a K value and a weight coefficient in a KNN algorithm by adopting a full-grid search method, taking a decision coefficient r2_ score value as an evaluation index in a verification set, wherein the better the r2_ score value is close to 1, the better the model is, the closer the r2_ score value is to 0, the worse the model is, and the optimal model parameters are stored through continuous iteration to obtain the optimal model.
2. The hybrid KNN wind power prediction method according to claim 1, characterized in that the model expiration time is set to 15 days, wind power data 15 days before the prediction day are trained, the optimal model parameters obtained in the previous step are loaded, and the trained data model is stored locally;
and inputting the wind speed, wind direction, temperature and humidity meteorological information of the day before the forecast date, and loading the trained data model in the previous step to obtain the forecast power value.
3. A hybrid KNN wind power prediction system is characterized in that,
the system comprises an acquisition module, a data processing module and a data processing module, wherein the acquisition module is used for acquiring input data and taking meteorological data including temperature, humidity, wind speed, wind direction and power before a certain day period as training input data;
the processing module is used for inputting input data into a pre-trained mixed KNN wind power prediction model combined with a fractal theory and a KNN algorithm;
the output module is used for outputting the prediction data of the wind power;
the processing module comprises:
the system comprises a diversity module, a verification module and a data processing module, wherein the diversity module is used for acquiring an original data set of wind power data of a wind power plant at a certain historical time period, and dividing the original data set into a front part and a rear part according to a time sequence, wherein the front part is used as a training set, and the other part is used as a verification set;
the fractal interpolation processing module is used for carrying out fractal interpolation processing on the original data set to obtain a fractal dimension result set;
the determining module is used for carrying out mixed KNN wind power prediction model training according to the fractal dimension result set, the training set and the verification set to determine an optimal model;
the fractal interpolation processing module comprises:
the characteristic extraction module is used for extracting data characteristics from the original data set, wherein the data characteristics comprise five-dimensional data of time sequence, meteorological temperature, wind speed, wind direction and available output;
the discretization processing module is used for performing discretization processing on available output data, meteorological temperature, wind speed and wind direction data by using an entropy maximum discretization method to obtain discretization data;
the calculation module is used for calculating the fractal dimension of available output data, meteorological temperature, wind speed and wind direction according to the discretization data and determining a fractal dimension result set;
the determination module is further used for training the model through continuous iteration according to the obtained fractal dimension result set and the training set under the condition that the circulation is met, so that an initial model is obtained, then a K value and a weight coefficient in a KNN algorithm are optimized through a full-grid search method, a decision coefficient r2_ score value is used as an evaluation index in the verification set, the closer the r2_ score value is to 1, the better the model is, the closer the r2_ score value is to 0, the worse the model is, and the optimal model parameters are stored through continuous iteration to obtain the optimal model.
4. The hybrid KNN wind power prediction system according to claim 3, wherein the determination module is further configured to set a model expiration time to 15 days, train wind power data that predicts 15 days before the day, load the optimal model parameters obtained in the previous step, and store the trained data model locally;
and inputting the wind speed, wind direction, temperature and humidity meteorological information of the day before the forecast date, and loading the trained data model in the previous step to obtain the forecast power value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010748717.XA CN111915079B (en) | 2020-07-30 | 2020-07-30 | Hybrid KNN wind power prediction method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010748717.XA CN111915079B (en) | 2020-07-30 | 2020-07-30 | Hybrid KNN wind power prediction method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111915079A CN111915079A (en) | 2020-11-10 |
| CN111915079B true CN111915079B (en) | 2022-08-26 |
Family
ID=73286753
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010748717.XA Active CN111915079B (en) | 2020-07-30 | 2020-07-30 | Hybrid KNN wind power prediction method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111915079B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113642640A (en) * | 2021-08-13 | 2021-11-12 | 华北电力科学研究院有限责任公司 | Regional power grid minimum output prediction method and device |
| CN115600639B (en) * | 2022-09-30 | 2023-11-14 | 国网四川省电力公司眉山供电公司 | Wind speed sensor, wind speed prediction method of power transmission line and early warning system |
| CN116527206B (en) * | 2023-07-03 | 2023-09-01 | 南京华飞数据技术有限公司 | Digital signal processing method for improving KNN method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719029A (en) * | 2016-03-11 | 2016-06-29 | 中国农业大学 | Combined wind power prediction method based on wind speed fluctuation characteristic extraction |
| CN110232460A (en) * | 2018-03-05 | 2019-09-13 | 中国电力科学研究院有限公司 | A kind of wind-powered electricity generation water power synthetic operation method for visualizing based on wind power prediction |
-
2020
- 2020-07-30 CN CN202010748717.XA patent/CN111915079B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105719029A (en) * | 2016-03-11 | 2016-06-29 | 中国农业大学 | Combined wind power prediction method based on wind speed fluctuation characteristic extraction |
| CN110232460A (en) * | 2018-03-05 | 2019-09-13 | 中国电力科学研究院有限公司 | A kind of wind-powered electricity generation water power synthetic operation method for visualizing based on wind power prediction |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111915079A (en) | 2020-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111915079B (en) | Hybrid KNN wind power prediction method and system | |
| CN103390116B (en) | Use the photovoltaic power station power generation power forecasting method of stepping mode | |
| CN111461463B (en) | Short-term load prediction method, system and equipment based on TCN-BP | |
| CN109800898A (en) | A kind of intelligence short-term load forecasting method and system | |
| CN110598947A (en) | Load prediction method based on improved cuckoo-neural network algorithm | |
| CN111697560A (en) | Method and system for predicting load of power system based on LSTM | |
| CN116760006A (en) | Wind power prediction method, device, equipment and storage medium | |
| CN112967154B (en) | Assessment method and device for Well-rolling of power system | |
| CN115600820A (en) | Training method of power generation prediction model and calculation method of wind power generation power | |
| CN110210657B (en) | Fan power prediction method and system based on single machine model and computer storage medium | |
| CN111178593A (en) | Photovoltaic system output power prediction method and device | |
| CN113723717A (en) | Method, device, equipment and readable storage medium for predicting short-term load before system day | |
| CN114139788A (en) | Method and device for predicting power of offshore wind turbine generator | |
| CN113283638A (en) | Load extreme curve prediction method and system based on fusion model | |
| Huang et al. | Probabilistic prediction intervals of wind speed based on explainable neural network | |
| CN112215383A (en) | Distributed photovoltaic power generation power prediction method and system | |
| CN117117858B (en) | Wind turbine generator power prediction method, device and storage medium | |
| CN109101592A (en) | A kind of photovoltaic big data processing platform of fast and low-cost | |
| CN117313043B (en) | Wind power generation power prediction method | |
| CN117371581A (en) | New energy generated power prediction method, device and storage medium | |
| CN117151303B (en) | Ultra-short-term solar irradiance prediction method and system based on hybrid model | |
| CN117435602A (en) | Power resource updating method, device, computer equipment and storage medium | |
| CN117937587A (en) | Distribution optimization method and system for distributed energy | |
| CN113705067B (en) | Microgrid optimization operation strategy generation method, system, equipment and storage medium | |
| CN114298374A (en) | Demand response capability determination method and system considering user power curve loss |
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 |