CN115660898A - SVR-based classification type wind power short-term power prediction precision improving method and device - Google Patents
SVR-based classification type wind power short-term power prediction precision improving method and device Download PDFInfo
- Publication number
- CN115660898A CN115660898A CN202211559956.6A CN202211559956A CN115660898A CN 115660898 A CN115660898 A CN 115660898A CN 202211559956 A CN202211559956 A CN 202211559956A CN 115660898 A CN115660898 A CN 115660898A
- Authority
- CN
- China
- Prior art keywords
- power
- term
- short
- wind power
- prediction
- 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.)
- Granted
Links
Images
Abstract
The invention provides a classification type wind power short-term power prediction precision improving method and equipment based on SVR, the method combines weather data such as wind speed, wind direction, temperature, humidity, air pressure and the like, uses a K-nearest neighbor algorithm to divide short-term power prediction errors into three different error sections, respectively establishes a power error prediction model based on SVR for the three error sections, adds original prediction power and error prediction values to obtain a new power prediction result, and finally uses a deviation correction algorithm to process the prediction result, thereby obtaining a final short-term power prediction value. By the method, the optimization algorithm model can be independently learned according to different fan characteristics, the influence of the installed position and the fan characteristic change is avoided, and the robustness and the generalization capability are strong; the algorithm model provided by the invention analyzes and processes the errors of power prediction in a finer dimension, and combines a deviation correction algorithm, thereby ensuring the comprehensive improvement of the prediction accuracy of short-term power on different error classifications.
Description
Technical Field
The invention relates to the technical field of power prediction, in particular to a classification type wind power short-term power prediction precision improving method, device, equipment and storage medium based on SVR.
Background
The short-term wind power prediction plays a significant role in adjusting the combination scheme of the generator set, optimizing the power generation strategy of the conventional generator set and improving the new energy consumption capability. Research shows that the output of the wind turbine generator is in a direct relation with the wind speed, and for a fixed type fan, when the wind speed is too high or too low, the fan is in a standby state, the output of the fan is 0, and the normal operation and the output of the fan can be ensured only if the wind speed meets certain conditions. At present, the prediction of the short-term power of the fan is basically realized based on the establishment of an algorithm relation between output and wind speed, but the algorithm model is limited along with the change of seasons, the service life of the fan is prolonged, and the abrasion of hardware functions causes the precision of the short-term power prediction to be worse and worse, which brings the influence of potential acquiescence to the safe operation of the power market.
The method for predicting the short-term power of the wind power plant in the prior art is mainly divided into two main categories, namely a first category which is used for predicting the short-term power by establishing a corresponding sectional type wind power algorithm model according to the type of a fan and key data such as rated power, rated wind speed, cut-in wind speed, cut-out wind speed and the like of the fan; and in the second category, the neural network algorithm model is used for directly predicting the short-term power or the first category of algorithm is used for predicting the wind power, and the neural network algorithm is combined to perform error correction processing on the prediction result.
The method for predicting the short-term power of the wind power plant in the prior art is mainly divided into two main categories described above, wherein the first category of method needs to establish a corresponding algorithm model by combining actual installed positions and fan characteristics, and the model needs to be optimized and adjusted for a long time according to different seasonal characteristics, and in addition, as the service life is prolonged, the model is more difficult to optimize due to the change of the fan characteristics, so that the prediction result of the short-term power is poorer and poorer; the second method comprehensively analyzes and corrects the short-term power integrity morning error, and cannot comprehensively and finely extract the error characteristics between the short-term actual power and the predicted power under different weather conditions, so that the prediction effect of individual days can be obviously improved, and the comprehensive improvement of the short-term power prediction accuracy cannot be realized.
Disclosure of Invention
The invention provides a classification type wind power short-term power prediction precision improving method, device, equipment and storage medium based on SVR (singular value representation, random value representation), aiming at analyzing and processing power prediction errors in a finer dimension and ensuring comprehensive improvement of prediction accuracy of short-term power in different error classifications.
Therefore, the invention aims to provide a classification type wind power short-term power prediction precision improving method based on SVR, which comprises the following steps:
constructing a wind power short-term power prediction precision lifting model; the wind power short-term power prediction precision improving model comprises an adaptive correlation analysis unit, a classification algorithm unit, an SVR power error prediction unit and a power correction unit;
acquiring wind power short-term actual power, historical data corresponding to wind power short-term predicted power and weather factors, and inputting a wind power short-term power prediction precision promotion model for training;
acquiring the short-term actual power of the electricity of the forecast day, the corresponding short-term forecast power of the wind power and weather factors, and inputting the short-term forecast power of the wind power and the weather factors into the trained wind power short-term power forecast precision promotion model to obtain a forecast result of the short-term power error of the forecast day;
and combining the short-term wind power predicted power of the predicted day with the short-term wind power error predicted result of the predicted day, and obtaining a wind power short-term power correction result through a deviation correction algorithm.
Wherein, in the step of training the wind power short-term power prediction precision lifting model, the method comprises the following steps:
determining any type of correlation between the wind power short-term actual power and weather factors through a self-adaptive correlation analysis unit to obtain strong correlation weather factor characteristic data of the wind power short-term actual power;
inputting feature data of strongly related weather factors into a classification algorithm unit, classifying according to the difference between the short-term actual power of the wind power and the short-term predicted power of the wind power at the same time, and obtaining classification data sets which correspond to different types of strongly related weather factors and are classified according to the difference between the short-term actual power of the wind power and the short-term predicted power of the wind power at the same time, wherein the classification data sets comprise input data and output data;
and inputting the classification data sets corresponding to each error category into an SVR power error prediction unit for training to obtain an SVR power error prediction model corresponding to each error category.
In the step of calculating the strong correlation weather factor characteristic data of the wind power short-term actual power, the correlation between the characteristics of different weather factors and the wind power short-term power is calculated through a formula (1), wherein the formula (1) is expressed as:
wherein the content of the first and second substances,
representing short-term power of wind power
And any weather factor characteristic data
The value of the correlation coefficient between the feature vectors of (a);
n represents the number of test samples;
the larger the calculated value of formula (1) is, the stronger the correlation is; and determining the weather factor characteristic data with the correlation coefficient larger than a preset threshold value as strongly correlated weather factor characteristic data by setting the threshold value.
The classification algorithm unit adopts a K-nearest neighbor classification model for classification.
In the step of classifying according to the wind power short-term actual power and the wind power short-term predicted power at the same time, strongly relevant weather factor characteristic data are input into a K-nearest neighbor classification model, and three classification processing is performed on the basis of the difference between the wind power short-term actual power and the wind power short-term predicted power of corresponding weather factor characteristics, so that three classification results of the strongly relevant weather factor characteristics and a classification data set corresponding to each classification result are obtained.
In the step of respectively inputting the strongly correlated weather factor characteristic data sets corresponding to each error category into an SVR power error prediction unit for training to obtain an SVR power error prediction model corresponding to each error category, three classification data sets obtained by classification are respectively input into an SVR power error prediction unit for training to obtain three SVR power error prediction models.
In the step of combining the short-term wind power predicted power of the predicted day with the short-term wind power error predicted result of the predicted day and obtaining the short-term wind power correction result through the deviation correction algorithm, adding the short-term wind power predicted power of the predicted day with the short-term wind power predicted error of the predicted day to obtain a primary power correction result; processing the preliminary power correction result by using a deviation correction algorithm to obtain a final short-term power correction result; wherein the content of the first and second substances,
the deviation correction algorithm is shown in equation (2):
wherein, the first and the second end of the pipe are connected with each other,
a corrected value representing the short-term wind power at the time t;
The invention provides a classification type wind power short-term power prediction precision improving device based on SVR, which comprises:
the model construction module is used for constructing a wind power short-term power prediction precision promotion model; the wind power short-term power prediction precision improving model comprises an adaptive correlation analysis unit, a classification algorithm unit, an SVR power error prediction unit and a power correction unit;
the model training module is used for acquiring the wind power short-term actual power, historical data corresponding to the wind power short-term predicted power and weather factors, and inputting a wind power short-term power prediction precision improvement model for training;
the error prediction module is used for acquiring the short-term actual power of the electricity on the prediction day, the corresponding short-term predicted power of the wind power and weather factors, and inputting the short-term actual power of the wind power, the corresponding short-term predicted power of the wind power and the weather factors into the trained wind power short-term power prediction precision promotion model to obtain a prediction result of the short-term power of the wind power on the prediction day;
and the power correction module is used for combining the wind power short-term predicted power of the predicted day with the wind power short-term power error prediction result of the predicted day and obtaining a wind power short-term power correction result through a deviation correction algorithm.
A third object of the present invention is to provide an electronic apparatus, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executable by the at least one processor to enable the at least one processor to perform the steps of the method of the preceding claims.
A fourth object of the present invention is to propose a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the steps of the method according to the aforementioned technical solution.
The invention is different from the prior art, and provides a classified short-term power prediction precision improving method based on SVR, which combines weather data such as wind speed, wind direction, temperature, humidity, air pressure and the like, divides short-term power prediction errors into three different error sections by using a K-nearest neighbor algorithm, respectively establishes a power error prediction model based on SVR for the three error sections, adds original prediction power and error prediction values to obtain a new power prediction result, and finally processes the prediction result by using a deviation correction algorithm, thereby obtaining a final short-term power prediction value. By the method, the optimization algorithm model can be independently learned according to different fan characteristics, the influence of the installation position and the fan characteristic change is avoided, and the robustness and the generalization capability are strong; the algorithm model provided by the invention analyzes and processes the errors of power prediction in a finer dimension, and combines a deviation correction algorithm, thereby ensuring the comprehensive improvement of the prediction accuracy of short-term power on different error classifications.
Drawings
The present invention and/or additional aspects and advantages will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a schematic flow diagram of a classified wind power short-term power prediction accuracy improving method based on SVR provided by the invention.
FIG. 2 is a logic schematic diagram of a classified wind power short-term power prediction accuracy improving method based on SVR provided by the invention.
FIG. 3 is a schematic flow diagram of an SVR error regression algorithm model in the SVR-based classification type wind power short-term power prediction precision improving method provided by the invention.
Fig. 4 is a schematic structural diagram of the classified wind power short-term power prediction precision improving device based on the SVR provided by the invention.
FIG. 5 is a block diagram of a non-transitory computer readable storage medium storing computer instructions according to the present invention.
Detailed Description
Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.
As shown in fig. 1, a method for improving accuracy of short-term wind power prediction based on SVR provided in an embodiment of the present invention includes:
s110: constructing a wind power short-term power prediction precision lifting model; the wind power short-term power prediction precision improving model comprises a self-adaptive correlation analysis unit, a classification algorithm unit, an SVR power error prediction unit and a power correction unit.
The power prediction precision improvement logic is shown in FIG. 2, historical data of weather factor characteristics strongly related to wind power short-term power are selected through a correlation analysis algorithm, input into a classification algorithm model for classification processing, an SVR error regression algorithm model is trained respectively based on a weather factor characteristic data set of a classification result, after model training is completed, real-time wind power short-term prediction power data and weather factor characteristic data of a prediction day are classified through the classification algorithm model, and then the classification result is input into the trained SVR error regression algorithm model according to class correspondence; and combining the wind power short-term power error prediction result output by the model with the corresponding wind power short-term prediction power, and correcting by a deviation correction algorithm to obtain a wind power short-term power correction result.
S120: and acquiring the short-term actual power of the wind power, the historical data corresponding to the short-term predicted power of the wind power and weather factors, and inputting a wind power short-term power prediction precision promotion model for training.
The training process comprises the following steps:
s121: and determining the correlation of any type of the wind power short-term actual power and the weather factors through a self-adaptive correlation analysis unit to obtain strong correlation weather factor characteristic data of the wind power short-term actual power.
Calculating the correlation between the characteristics of different weather factors and the wind power short-term power through a formula (1), wherein the formula (1) is expressed as follows:
wherein the content of the first and second substances,
representing short-term power of wind power
And any weather factor characteristic data
The value of the correlation coefficient between the feature vectors of (a);
n represents the number of test samples;
determining the correlation between the short-term wind power and weather factors such as wind speed, wind direction, temperature, humidity and air pressure, determining the characteristic type of the algorithm model, further ensuring the self-adaptability and robustness of the algorithm model under different environments, different seasons and different years, obtaining the correlation between different characteristic vectors by using a formula (1), wherein the larger the calculation value of the formula (1) is, the stronger the correlation is; and determining the weather factor characteristic data with the correlation coefficient larger than a preset threshold value as strongly correlated weather factor characteristic data by setting the threshold value.
S122: and inputting the characteristic data of the strongly correlated weather factors into a classification algorithm unit, classifying according to the difference between the short-term actual power of the wind power and the short-term predicted power of the wind power at the same time, and obtaining classification data sets which correspond to different types of strongly correlated weather factors and are classified according to the difference between the short-term actual power of the wind power and the short-term predicted power of the wind power at the same time, wherein the classification data sets comprise input data and output data.
In the invention, a classification algorithm unit adopts a K-nearest neighbor classification model for classification.
And inputting the strongly relevant weather factor characteristic data into a K-nearest neighbor classification model, and performing three-classification treatment on the wind power short-term actual power and the wind power short-term predicted power based on the corresponding weather factor characteristics to obtain three classification results of the strongly relevant weather factor characteristics and a strongly relevant weather factor characteristic data set corresponding to each classification result.
S123: and inputting the classification data sets corresponding to each error category into an SVR power error prediction unit for training respectively to obtain an SVR power error prediction model corresponding to each error category.
S130: and acquiring the short-term actual power of the electricity on the prediction day, the short-term prediction power of the wind power and weather factors, and inputting the short-term actual power of the wind power and the weather factors into the trained wind power short-term power prediction precision promotion model to obtain a prediction result of the short-term power error of the wind power on the prediction day.
And respectively inputting the classification data sets corresponding to each error category into an SVR power error prediction unit for training to obtain an SVR power error prediction model corresponding to each error category, and respectively inputting the three classification data sets obtained by classification into an SVR power error prediction unit for training to obtain three different SVR power error prediction models. The flow of the SVR error regression algorithm model is shown in FIG. 3.
S140: and combining the wind power short-term predicted power of the predicted day with the wind power short-term power error prediction result of the predicted day, and obtaining a wind power short-term power correction result through a deviation correction algorithm.
Adding the wind power short-term predicted power of the predicted day and the wind power short-term predicted error of the predicted day to obtain a preliminary power correction result; processing the preliminary power correction result by using a deviation correction algorithm to obtain a final short-term power correction result; wherein the content of the first and second substances,
the deviation correction algorithm is shown in equation (2):
wherein the content of the first and second substances,
a corrected value representing the short-term wind power at the time t;
Compared with the prior art, the invention provides the self-adaptive correlation analysis method for the historical data, the method can weaken the dependence of the algorithm model on hard conditions such as the type, the installation position, the service life and the like of the fan, and the generalization capability and the self-adaptability of the model are improved; the method combining the K-nearest neighbor algorithm and the SVR regression model can obtain the error characteristics between the short-term actual power and the predicted power on a finer granularity, so that the accuracy and the reliability of error prediction are greatly improved; the invention combines and uses the deviation correction algorithm to ensure the rationality and normalization of the final correction result of the short-term power. The invention avoids the defects of the existing short-term power prediction algorithm, combines the advantages of the prior art, and greatly improves the practicability of the algorithm and the accuracy of the prediction precision from the aspects of the whole design idea and the realization function.
As shown in fig. 4, the invention provides a classified wind power short-term power prediction accuracy improving apparatus 300 based on SVR, comprising:
the model construction module 310 is used for constructing a wind power short-term power prediction precision improvement model; the wind power short-term power prediction precision improving model comprises a self-adaptive correlation analysis unit, a classification algorithm unit and an SVR power error prediction unit;
the model training module 320 is used for acquiring wind power short-term actual power, historical data corresponding to wind power short-term predicted power and weather factors, and inputting a wind power short-term power prediction precision promotion model for training;
the error prediction module 330 is used for acquiring the short-term actual power of the electricity on the predicted day, the corresponding short-term predicted power of the wind power and weather factors, and inputting the power into the trained wind power short-term power prediction precision promotion model to obtain a predicted result of the short-term power error of the wind power on the predicted day;
and the power correction unit 340 is used for combining the wind power short-term predicted power of the predicted day with the wind power short-term power error prediction result of the predicted day, and obtaining a wind power short-term power correction result through a deviation correction algorithm.
In order to implement the embodiment, the present invention further provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute the steps in the method for improving the accuracy of the classified wind power short-term power prediction according to the above technical scheme.
As shown in fig. 5, the non-transitory computer readable storage medium 800 includes a memory 810 of instructions executable by the processor 820 to perform the method according to the categorized wind short term power prediction accuracy boost, interface 830. Alternatively, the storage medium may be a non-transitory computer readable storage medium, for example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.
In order to implement the embodiment, the invention further provides a non-transitory computer-readable storage medium on which a computer program is stored, and the computer program, when executed by a processor, implements the classified wind power short-term power prediction accuracy improvement according to the embodiment of the invention.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean 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, a schematic representation of the terms does not necessarily refer 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, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" 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 defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless explicitly specified otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
The logic and/or steps represented in the flowcharts or otherwise described herein, such as an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the described embodiments, various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
One of ordinary skill in the art will appreciate that all or part of the steps carried by the method implementing the embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and the program, when executed, includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a separate product, may also be stored in a computer-readable storage medium.
The mentioned storage medium may be a read-only memory, a magnetic or optical disk, etc. While embodiments of the present invention have been shown and described above, it will be understood that the embodiments are illustrative and not to be construed as limiting the invention, and that changes, modifications, substitutions and alterations can be made therein by those of ordinary skill in the art without departing from the scope of the present invention.
Claims (10)
1. A classification type wind power short-term power prediction precision improving method based on SVR is characterized by comprising the following steps:
constructing a wind power short-term power prediction precision lifting model; the wind power short-term power prediction precision improving model comprises a self-adaptive correlation analysis unit, a classification algorithm unit, an SVR power error prediction unit and a power correction unit;
acquiring wind power short-term actual power, historical data corresponding to wind power short-term predicted power and weather factors, and inputting the wind power short-term predicted power prediction precision promotion model for training;
acquiring the electric short-term actual power of a prediction day, the corresponding wind power short-term prediction power and weather factors, and inputting the electric short-term actual power, the corresponding wind power short-term prediction power and the weather factors into the trained wind power short-term power prediction precision promotion model to obtain a prediction result of the wind power short-term power error of the prediction day;
and combining the wind power short-term predicted power of the predicted day with the wind power short-term power error prediction result of the predicted day, and obtaining a wind power short-term power correction result through a deviation correction algorithm.
2. The SVR-based classification type wind power short-term power prediction accuracy improving method of claim 1, wherein in the step of training the wind power short-term power prediction accuracy improving model, the method comprises:
determining the correlation of any type of the wind power short-term actual power and the weather factors through the self-adaptive correlation analysis unit to obtain strongly correlated weather factor characteristic data of the wind power short-term actual power;
inputting the strong correlation weather factor characteristic data into a classification algorithm unit, classifying according to the difference between the wind power short-term actual power and the simultaneous wind power short-term predicted power, and obtaining classification data sets which correspond to different types of strong correlation weather factors and are classified according to the difference between the wind power short-term actual power and the simultaneous wind power short-term predicted power, wherein the classification data sets comprise input data and output data;
and inputting the classification data sets corresponding to each error category into the SVR power error prediction unit for training to obtain an SVR power error prediction model corresponding to each error category.
3. The SVR-based classification type wind power short-term power prediction accuracy improving method according to claim 2, characterized in that in the step of calculating the strongly correlated weather factor characteristic data of the wind power short-term actual power, the correlation between the characteristics of different weather factors and the wind power short-term power is calculated by formula (1), wherein formula (1) is expressed as:
wherein the content of the first and second substances,
representing short-term power of wind power
And any weather factor characteristic data
The value of the correlation coefficient between the feature vectors of (a);
n represents the number of test samples;
the larger the calculated value of formula (1) is, the stronger the correlation is; and determining the weather factor characteristic data with the correlation coefficient larger than a preset threshold value as strongly correlated weather factor characteristic data by setting the threshold value.
4. The SVR-based classification-type wind power short-term power prediction accuracy improvement method of claim 1, wherein said classification algorithm unit performs classification using a K-nearest neighbor classification model.
5. The SVR-based classification type wind power short-term power prediction accuracy improving method according to claim 3, wherein in the step of classifying according to the difference between the wind power short-term actual power and the wind power short-term predicted power at the same time, the strongly correlated weather factor feature data is input into a K-nearest neighbor classification model, and three classification processing is performed on the wind power short-term actual power and the wind power short-term predicted power based on the corresponding weather factor features to obtain three classification results of the strongly correlated weather factor features and a classification data set corresponding to each classification result.
6. The SVR-based classified wind power short-term power prediction accuracy improving method according to claim 5, wherein in the step of inputting the classified data sets corresponding to each error category into one SVR power error prediction unit for training to obtain the SVR power error prediction model corresponding to each error category, the three classified data sets are respectively input into one SVR power error prediction unit for training to obtain three SVR power error prediction models.
7. The SVR-based classified wind power short-term power prediction accuracy improving method according to claim 6, wherein in the step of combining the wind power short-term predicted power on the predicted day with the wind power short-term power error predicted result on the predicted day and obtaining the wind power short-term power corrected result by a deviation correction algorithm, the wind power short-term predicted power on the predicted day and the wind power short-term power predicted error on the predicted day are added to obtain a preliminary power corrected result; processing the preliminary power correction result by using a deviation correction algorithm to obtain a final short-term power correction result; wherein the content of the first and second substances,
the deviation correction algorithm is shown in formula (2):
wherein, the first and the second end of the pipe are connected with each other,
a corrected value representing the short-term power of the wind power at the time t;
8. The utility model provides a short-term power prediction accuracy hoisting device of categorised formula wind-powered electricity generation based on SVR which characterized in that includes:
the model construction module is used for constructing a wind power short-term power prediction precision promotion model; the wind power short-term power prediction precision improving model comprises a self-adaptive correlation analysis unit, a classification algorithm unit, an SVR power error prediction unit and a power correction unit;
the model training module is used for acquiring wind power short-term actual power, historical data corresponding to wind power short-term predicted power and weather factors, and inputting the wind power short-term predicted precision improving model for training;
the error prediction module is used for acquiring the short-term actual power of the electricity on the prediction day, the corresponding short-term predicted power of the wind power and weather factors, and inputting the short-term predicted power of the wind power into the trained wind power short-term power prediction precision promotion model to obtain a prediction result of the short-term power error of the wind power on the prediction day;
and the power correction module is used for combining the wind power short-term predicted power of the predicted day with the wind power short-term power error prediction result of the predicted day, and obtaining a wind power short-term power correction result through a deviation correction algorithm.
9. An electronic device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps of the method of any one of claims 1-7.
10. A non-transitory computer readable storage medium having stored thereon computer instructions for causing a computer to perform the steps of the method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211559956.6A CN115660898B (en) | 2022-12-06 | 2022-12-06 | SVR-based classification type wind power short-term power prediction precision improving method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211559956.6A CN115660898B (en) | 2022-12-06 | 2022-12-06 | SVR-based classification type wind power short-term power prediction precision improving method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115660898A true CN115660898A (en) | 2023-01-31 |
| CN115660898B CN115660898B (en) | 2023-03-28 |
Family
ID=85020227
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211559956.6A Active CN115660898B (en) | 2022-12-06 | 2022-12-06 | SVR-based classification type wind power short-term power prediction precision improving method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115660898B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116664015A (en) * | 2023-07-26 | 2023-08-29 | 深圳市森树强电子科技有限公司 | Intelligent charging pile management system and method thereof |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150154504A1 (en) * | 2012-12-17 | 2015-06-04 | Arizona Board Of Regents On Behalf Of Arizona State University | Support vector machine enhanced models for short-term wind farm generation forecasting |
| CN108667069A (en) * | 2018-04-19 | 2018-10-16 | 河海大学 | A kind of short-term wind power forecast method returned based on Partial Least Squares |
| CN111091236A (en) * | 2019-11-27 | 2020-05-01 | 长春吉电能源科技有限公司 | Multi-classification deep learning short-term wind power prediction method classified according to pitch angles |
| CN113361761A (en) * | 2021-06-01 | 2021-09-07 | 山东大学 | Short-term wind power integration prediction method and system based on error correction |
-
2022
- 2022-12-06 CN CN202211559956.6A patent/CN115660898B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150154504A1 (en) * | 2012-12-17 | 2015-06-04 | Arizona Board Of Regents On Behalf Of Arizona State University | Support vector machine enhanced models for short-term wind farm generation forecasting |
| CN108667069A (en) * | 2018-04-19 | 2018-10-16 | 河海大学 | A kind of short-term wind power forecast method returned based on Partial Least Squares |
| CN111091236A (en) * | 2019-11-27 | 2020-05-01 | 长春吉电能源科技有限公司 | Multi-classification deep learning short-term wind power prediction method classified according to pitch angles |
| CN113361761A (en) * | 2021-06-01 | 2021-09-07 | 山东大学 | Short-term wind power integration prediction method and system based on error correction |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116664015A (en) * | 2023-07-26 | 2023-08-29 | 深圳市森树强电子科技有限公司 | Intelligent charging pile management system and method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115660898B (en) | 2023-03-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11409347B2 (en) | Method, system and storage medium for predicting power load probability density based on deep learning | |
| CN109488528B (en) | Method and device for adjusting fan yaw system | |
| CN115660898B (en) | SVR-based classification type wind power short-term power prediction precision improving method and device | |
| CN117411189B (en) | Monitoring data enhancement method of micro-grid coordination controller | |
| CN110165698B (en) | Wind power plant smooth grid-connection method for realizing prospective error asset conversion | |
| CN110837915A (en) | Low-voltage load point prediction and probability prediction method for power system based on hybrid integrated deep learning | |
| CN113822418A (en) | Wind power plant power prediction method, system, device and storage medium | |
| CN112465239A (en) | Desulfurization system operation optimization method based on improved PSO-FCM algorithm | |
| CN116227637A (en) | Active power distribution network oriented refined load prediction method and system | |
| CN115526258A (en) | Power system transient stability evaluation method based on Spearman correlation coefficient feature extraction | |
| CN115758151A (en) | Combined diagnosis model establishing method and photovoltaic module fault diagnosis method | |
| CN115018200A (en) | Power load prediction method and system based on deep learning and considering multiple influence factors | |
| CN108832623B (en) | Physical-statistical hybrid two-stage wind power prediction method | |
| CN113177355B (en) | Power load prediction method | |
| CN114564513A (en) | Sea fog prediction method, device, equipment and storage medium based on neural network | |
| CN112802011A (en) | Fan blade defect detection method based on VGG-BLS | |
| CN116979505A (en) | Power grid short-term load prediction method and system | |
| CN112598148A (en) | Fan variable pitch motor temperature fault early warning method based on collaborative expression and LightGBM algorithm | |
| CN111060755A (en) | Electromagnetic interference diagnosis method and device | |
| CN115313522A (en) | Spare capacity configuration method and system considering new energy output prediction deviation | |
| CN114169226A (en) | Short-term power load prediction method, computer device, and storage medium | |
| CN113947007A (en) | Energy load prediction method, device and equipment | |
| CN112668764B (en) | Offshore wind farm energy storage system optimization configuration method based on cloud model and FCM algorithm | |
| CN116760026B (en) | Accurate prediction method for unit output considering parameter unrecognizable | |
| CN117791626B (en) | Intelligent comprehensive power box power supply optimization method |
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 |