CN110738253A - short-term wind power prediction method based on FCM and AFSA-Elman - Google Patents

short-term wind power prediction method based on FCM and AFSA-Elman Download PDF

Info

Publication number
CN110738253A
CN110738253A CN201910971600.5A CN201910971600A CN110738253A CN 110738253 A CN110738253 A CN 110738253A CN 201910971600 A CN201910971600 A CN 201910971600A CN 110738253 A CN110738253 A CN 110738253A
Authority
CN
China
Prior art keywords
elman
afsa
wind
wind power
clustering
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201910971600.5A
Other languages
Chinese (zh)
Inventor
邓华
张颖超
宗阳
章璇
成金杰
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nanjing University of Information Science and Technology
Original Assignee
Nanjing University of Information Science and Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Nanjing University of Information Science and Technology filed Critical Nanjing University of Information Science and Technology
Priority to CN201910971600.5A priority Critical patent/CN110738253A/en
Publication of CN110738253A publication Critical patent/CN110738253A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • 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
    • 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply

Abstract

The invention discloses short-term wind power prediction methods based on FCM and AFSA-Elman, which relate to the field of wind power prediction and comprise the following steps of cleaning and standardizing historical data of a wind field, primarily selecting an input vector of a clustering model, training a fan sample by adopting a fuzzy C clustering algorithm, selecting a proper clustering index, constructing a new input set according to the clustering index, training the input set by using FCM again to obtain the partitioning results of different clusters, obtaining the parameter values of equivalent machine sets of each cluster by adopting a fan capacity weighted aggregation method for each cluster, wherein the parameter values can represent the corresponding cluster, establishing AFSA-Elman prediction models of different clusters according to the equivalent parameters to obtain the prediction results of different clusters, and carrying out capacity weighting on the predicted power of each cluster to obtain the total predicted power of the whole wind power plant.

Description

short-term wind power prediction method based on FCM and AFSA-Elman
Technical Field
The invention relates to the field of wind power prediction, in particular to short-term wind power prediction methods based on FCM and AFSA-Elman.
Background
With the rapid development of economy and the tension of energy supply, the world energy structure is converted from a fossil energy system to a sustainable new energy system based on renewable energy, so that the development of sustainable new energy is very important for all countries in the world, wind energy is used as renewable energy, has the characteristics of -wide distribution, inexhaustibility, large storage amount and the like, and plays a prominent role in reducing the emission of greenhouse gases and relieving global warming.
China vigorously develops the wind power generation industry under the promotion of the policy of developing clean energy. In order to promote the development of renewable energy sources and provide guidance for the operation management of a power system, a wind power prediction with higher accuracy is very necessary. The wind power prediction is to predict the output wind speed of the wind power plant in a short term by utilizing a physical simulation calculation and scientific statistical method according to the related data of the meteorological information of the wind power plant, so as to predict the generating power of the wind power plant.
The wind power random fluctuation can cause the wind energy to have stronger uncertainty, and the research significance of wind power prediction is increasingly remarkable under the background of increasing wind field scale and grid scheduling difficulty, for the power department , the extreme value of the power prediction can help timely modify and make a grid scheduling strategy, so that the consumption of wind energy resources is effectively reduced, and the safe and stable operation of a national power system is efficiently ensured, for the power market, the efficient power prediction can improve the evaluation index of the wind power in the power market, and ensure that more wind energy resources are utilized, for the wind farm, a maintainer can selectively maintain a fan according to a power curve simulated by prediction, the fan is prevented from being damaged by severe weather, and the loss of the wind resources can be reduced to the minimum, so that the economic benefit of the wind farm is further improved in the step .
Therefore, for the characteristics of complex terrain of a target wind power plant, numerous units in the plant and the like, wind power prediction methods are urgently needed to predict the wind power.
Disclosure of Invention
The invention aims to provide short-term wind power prediction methods based on FCM and AFSA-Elman, which can better avoid the occurrence of 'dimension disaster' of a power system, have important significance for evaluating the mutual influence between a large-capacity wind power plant and the power system, and provide accurate and effective methods for short-term wind power prediction.
The technical purpose of the invention is realized by the following technical scheme:
short-term wind power prediction method based on FCM and AFSA-Elman comprises the following steps:
step 1: cleaning and standardizing historical data of the wind farm, and establishing observation data of each wind turbine generator in the target wind farm;
step 2: firstly selecting an input vector of a clustering model, training a fan sample by adopting a fuzzy C clustering algorithm, and selecting a proper clustering index;
and step 3: constructing a new input set according to the clustering indexes, and training the new input set by using the FCM again to obtain the partitioning results of different clusters;
and 4, step 4: adopting a fan capacity weighted aggregation method to obtain the parameter value of each cluster equivalent unit, wherein the parameter value can represent the corresponding cluster;
and 5: establishing AFSA-Elman prediction models of different clusters according to the equivalent parameters to obtain prediction results of different clusters;
step 6: and carrying out capacity weighting on the power predicted by each cluster, so as to achieve the total predicted power of the whole wind power plant.
Further , in step 1, the observation data includes the time series formed by the wind speed, wind direction, active power and temperature data of each fan.
Further , in step 2, the distance based on the space measurement feature is used as the similarity measurement of the clustering algorithm, the correlation coefficient is properly converted by introducing the distance idea, and the formula is adopted
Figure BDA0002232282100000031
Computing distances based on spatially measured features, where pxyIs a correlation coefficient, dxyIs a distance based on a spatial metric feature.
Further , in step 3, the wind speed, wind direction and active power are integrated as the grouping index.
And , in step 5, an AFSA-Elman algorithm is adopted as a wind power prediction model, and the weight threshold of the Elman neural network is optimized by utilizing the characteristics of parallel processing and automatic global optimization realization of the artificial fish swarm algorithm.
In conclusion, the invention has the following beneficial effects:
1. the close relation between the correlation and the distance is fully embodied through the distance based on the spatial measurement characteristics, the distance of the spatial measurement characteristics capable of reflecting the correlation between variables can be obtained through conversion of correlation coefficients, and the method has application value for the research of the fan correlation.
Compared with the unoptimized Elman algorithm and the commonly used BP algorithm, the AFSA-Elman algorithm has the advantages that the relative root mean square error (rRMSE) and the relative average absolute error (rMAE) are obviously smaller, the error value is relatively stable, and the prediction curve is closer to the actual power curve. Therefore, the effect of the proposed FCM-AFSA-Elman short-term wind power model is ideal.
Drawings
FIG. 1 is a graph of wind turbine distribution locations for a wind farm in an embodiment of the present invention;
FIG. 2 is a comparison of the prediction curves of the AFSA-Elman algorithm and the BP algorithm in an embodiment of the present invention;
FIG. 3 is a comparison of the prediction curves of the AFSA-Elman algorithm and the Elman algorithm in an embodiment of the present invention;
FIG. 4 is a graph of the predicted absolute error of the BP algorithm in an embodiment of the present invention;
FIG. 5 is a graph of predicted absolute error for the Elman algorithm in an embodiment of the present invention;
FIG. 6 is a graph of the predicted absolute error of the AFSA-Elman algorithm in an embodiment of the present invention.
Detailed Description
The following is a further description of an embodiment of the present invention, which is not intended to be limiting.
The invention discloses short-term wind power prediction methods based on FCM and AFSA-Elman, which comprise the steps of clustering wind turbines by FCM, respectively establishing AFSA-Elman models for each cluster after clustering, and superposing prediction results of each cluster to obtain a final short-term wind power prediction result, and specifically comprise the following steps:
step 1: cleaning and standardizing historical data of a wind field;
the observed data comprises time sequences formed by wind speed, wind direction, active power and temperature data of all the fans.
Step 2: firstly selecting input vectors (all characteristic quantities) of a clustering model, training a fan sample by adopting a fuzzy C clustering algorithm based on the distance of spatial measurement characteristics, and selecting a proper clustering index according to the influence condition of different clustering indexes on a clustering result;
the distance based on spatial measurement characteristics is adopted as similarity measurement of a clustering algorithm, correlation coefficients are properly converted by introducing a distance idea, and a distance formula reference formula is
Figure BDA0002232282100000041
The above equation mainly calculates the distance based on the spatial metric feature, whichMiddle rhoxyIs a correlation coefficient, dxyIs a distance based on a spatial metric feature. Will dxyThe method is applied to the FCM, the problem of Euclidean distance is well solved, and the problem that the FCM clustering algorithm is only suitable for processing data which are compact in class and good in class-to-class separation and spherical data but cannot process non-convex data is solved.
And step 3: constructing a new input set according to the obtained clustering indexes, and training the new input set by using the FCM again to obtain the partitioning results of different clusters;
in the embodiment, because the influence of the temperature element on the clustering result is small, the temperature element is omitted, and three elements of wind speed, wind direction and active power are integrated into a clustering index;
and 4, step 4: obtaining equivalent parameters of equivalent units of all cluster by adopting a fan capacity weighted aggregation method for all clustered clusters, wherein the parameter values can represent the corresponding clusters;
the clustering result judgment criterion defines measurement intra-class distance and inter-class distance measure by using a variance idea, the larger the inter-class distance is, the better the intra-class distance is, the smaller the intra-class distance is, the better the intra-class distance is, and the internal evaluation index reference formula of the clustering result is as follows:
wherein STDI is the ratio of the distance between classes to the distance within the class, ckIs the centroid of cluster k, xtIs the centroid, x, of all samplesiIs the ith sample, n, of the class cluster kkIs the number of samples of class K, which is the number of class clusters of the data set.
And 5: establishing AFSA-Elman prediction models of different clusters according to equivalent parameters to obtain prediction results of different clusters, wherein an AFSA-Elman algorithm is adopted as a wind power prediction model, and the weight threshold of an Elman neural network is optimized by utilizing the characteristics of parallel processing and automatic global optimization realization of an artificial fish swarm algorithm;
in order to improve the prediction accuracy of the Elman algorithm, an Artificial Fish School Algorithm (AFSA) is introduced to optimize the weight threshold of the Elman algorithm, and the optimization process satisfies the following two formulas:
Figure BDA0002232282100000052
Figure BDA0002232282100000061
wherein M ═ M1,m2,…,mn) In order to simulate the current state of the artificial fish,
Figure BDA0002232282100000062
for the position state of the viewpoint at a certain time, the Rand function generates a random number between 0 and 1, Step is the Step length, and Visual is the view field range.
Step 6: and (5) obtaining the prediction results of different clusters based on the AFSA-Elman prediction models of different clusters established in the step (5), and carrying out capacity weighting on the power predicted by each cluster to obtain the total predicted power of the whole wind power plant.
As shown in fig. 1, take a Yunnan grinding bean mountain wind farm as an example: the ground bean mountain wind power station is located in a low-latitude high-altitude terrain, 24 fans with the same model are shared in the ground bean mountain wind power station, the installed capacity of each fan is 2MW, and the total capacity of the wind power station is 48 MW. The cut-in wind speed of the fan is 3m/s, the rated wind speed is 12m/s, the cut-out wind speed is 25m/s, and the rated power is 2 MW.
Because the wind power plant is located in the special mountainous terrain, the influence of the position distribution, the wind speed spatial distribution and the like of the wind generation set on the wind power output is large, and therefore, detailed modeling and analysis of the wind generation set in the wind power plant are necessary.
The geographical distribution of the wind turbine generator is shown in fig. 1, wherein, the position 0 is the position of the wind measuring tower, the positions 1-24 represent the positions of the fans 1 to 24, and the fan 7 has a fault and cannot accurately obtain effective data, so the fan is not taken as a research object, and the invention is further described in :
1. after the invalid observation data are removed, the abnormal data values are removed and the data are classified from the wind turbine data,establishing observation data of each wind turbine generator set in the target wind power plant, and expressing the observation data as Di＝[Di1,Di2,Di3,Di4]Where i ∈ [1,24 ]]Denotes a blower number, Di1,Di2,Di3,Di4And respectively representing time sequences formed by wind speed, wind direction, active power and temperature data of the fan.
2. The wind power station is located on mountainous terrain, and the layout of wind power sets in the wind power station is irregular. Due to the factors such as the influence of terrain, altitude and other units, the operating data captured by each fan in the field has larger difference, so that the wind speed has stronger volatility and intermittency, and a clustering algorithm is introduced to analyze the fans in different operating states in the field. The key of the cluster analysis lies in the clustering index and similarity measurement.
(1) In the aspect of fan characteristic quantity selection, the selection that characteristic indexes can be easily obtained before power generation can be ensured, and the power generation process of a fan can be obviously influenced.
In order to fully represent the close relation between the correlation and the distance, the distance of the space measurement characteristic capable of reflecting the correlation between the variables can be obtained by converting the correlation coefficient
Figure BDA0002232282100000071
Where ρ isxyIs a correlation coefficient, dxyThe method is based on the distance of the space measurement characteristics, and has application value in the research of fan correlation. The distances based on the spatial metric features are taken as similarity metrics and classified into 4 classes, the fuzzy C-means clustering (FCM) algorithm is adopted to analyze the conditions of different clustering indexes, and the analysis results are shown in Table 1.
TABLE 1 FCM clustering result table under different clustering indexes
Figure BDA0002232282100000072
Figure BDA0002232282100000081
From table 1, it can be seen that under different clustering indexes, clustering situations of clustering are greatly different, wherein the element of temperature has little influence on clustering results, so we omit the element and only study the influence of three major elements of wind speed, wind direction and active power.
The clustering effect was analyzed using STDI as an index for evaluating the clustering result, as shown in table 2. The STDI value of the comprehensive grouping index combining the wind speed, the wind direction and the active power is 0.7325, and is larger than the STDI value of the grouping index combining the wind speed, the wind speed and the wind direction, so that the clustering effect of the comprehensive grouping index is proved to be better, and the wind speed, the wind direction and the active power are determined to be integrated into the classification index.
TABLE 2 STDI values under different criteria
Grouping index STDI value
Wind speed 0.4329
Wind speed + wind direction 0.5663
Wind speed + wind direction + active power 0.7325
(2) Similarity measurement is also an important link in cluster analysis. Euclidean distance (Euclidean distance), Pearson coefficient (Pearson) and Space distance (Space distance) based on spatial measurement characteristics are respectively selected as similarity measurement for comparison, fuzzy C mean value clustering is used as a clustering algorithm to cluster the fans in the wind power plant, and clustering results and STDI values are shown in table 3.
TABLE 3 FCM clustering result table under different similarity measures
Figure BDA0002232282100000082
As can be seen from table 3, although the clustering methods are the same, different similarity measures result in different cluster groups, which illustrates that the similarity measures have a great influence on the partition between the clusters. From the STDI value, clustering under distance based on spatial metric features is better than the other two classes.
3. In order to further evaluate the effect of different similarity measures, an AFSA-Elman prediction model is adopted to test data of different clusters, the dynamic information processing capability of the Elman neural network enables the Elman neural network to be widely applied to the prediction problem of time series , in order to improve the prediction accuracy of the Elman algorithm, an Artificial Fish Swarm Algorithm (AFSA) is introduced to optimize the weight threshold of the Elman algorithm, and the optimization process is as follows:
Figure BDA0002232282100000091
Figure BDA0002232282100000092
wherein M ═ M1,m2,…,mn) In order to simulate the current state of the artificial fish,
Figure BDA0002232282100000093
for the position state of the viewpoint at a certain time, the Rand function generates a random number between 0 and 1, Step is the Step length, and Visual is the view field range.
And selecting data of 27 days before 1 month of the wind power plant, namely 2592 groups of data as training samples to train the AFSA-Elman prediction model, predicting the last three days of the trained model, namely 288 groups of data, comparing the predicted data with the actually measured data of the wind power plant, and solving error values. To evaluate the clustering results for different similarity measures, the AFSA-Elman model was used for training and testing.
The method selects four error indexes of Root Mean Square Error (RMSE), Mean Absolute Error (MAE), relative root mean square error (rRMSE) and relative mean absolute error (rMAE) to evaluate the short-term wind speed prediction result of the wind power plant, and the calculation formula is as follows:
(1) root Mean Square Error (RMSE)
Figure BDA0002232282100000094
(2) Mean relative error (MAE)
(3) Relative root mean square error (rRMSE)
Figure BDA0002232282100000102
(4) Relative mean absolute error (rMAE)
Wherein N represents the number of samples,
Figure BDA0002232282100000104
denoted is the ith predictor, xk(i) The ith actual value is shown.
The corresponding prediction error results for the different similarity measures are shown in table 4.
TABLE 4 comparison of prediction accuracy for different similarity measures
Figure BDA0002232282100000105
As can be seen from table 4, the model error values rmse and rMAE under the similarity measurement using the distance based on the spatial metric features are 22.25% and 16.53%, respectively, the model error values under the similarity measurement using the euclidean distance are 25.41% and 19.04%, and the model error values under the similarity measurement using the pearson coefficient are 23.48% and 17.85%, respectively. The feasibility of the proposed similarity measure method, i.e. the distance based spatial metric feature, is verified.
To further evaluate the predicted effect of the AFSA-Elman algorithm, the AFSA-Elman algorithm was compared with the unoptimized Elman algorithm and the conventional BP neural network algorithm, respectively, and the predicted curve comparison is shown in FIGS. 2 and 3, respectively.
Due to the large amount of data, 48 data are extracted from 288 data in the test set for specific analysis. FIG. 2 respectively compares the prediction curve of the AFSA-Elman algorithm, the prediction curve of the BP algorithm and the actual power curve, and the prediction curve of the AFSA-Elman algorithm, the prediction curve of the Elman algorithm and the actual power curve are compared in FIG. 3. Fig. 4, fig. 5 and fig. 6 are absolute error comparison graphs of each predicted power value and an actually measured value in corresponding time by the BP algorithm, the Elman algorithm and the AFSA-Elman algorithm model respectively. As can be seen by comparing the three graphs, the absolute error of the AFSA-Elman model is obviously smaller than that of BP and Elman, and the error value is relatively stable. In order to more intuitively understand the prediction effects of the three algorithms, table 5 lists various prediction error indexes under different algorithms.
TABLE 5 comparison of prediction accuracy for several prediction algorithms
The method has the advantages that the prediction accuracy of the AFSA-Elman algorithm is obviously higher than that of a BP neural network algorithm and an Elman algorithm, the algorithm also has a good prediction effect in summer and autumn with the largest error index, the predicted power sequence is closer to the actual power sequence, the effectiveness of the algorithm in power prediction is demonstrated, and compared with the current algorithm which uses more , the method also has high accuracy.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the spirit and scope of the invention.

Claims (5)

1, short-term wind power prediction method based on FCM and AFSA-Elman, which is characterized in that:
the method comprises the following steps:
step 1: cleaning and standardizing historical data of the wind farm, and establishing observation data of each wind turbine generator in the target wind farm;
step 2: firstly selecting an input vector of a clustering model, training a fan sample by adopting a fuzzy C clustering algorithm, and selecting a proper clustering index;
and step 3: constructing a new input set according to the clustering indexes, and training the new input set by using the FCM again to obtain the partitioning results of different clusters;
and 4, step 4: adopting a fan capacity weighted aggregation method to obtain the parameter value of each cluster equivalent unit, wherein the parameter value can represent the corresponding cluster;
and 5: establishing AFSA-Elman prediction models of different clusters according to the equivalent parameters to obtain prediction results of different clusters;
step 6: and carrying out capacity weighting on the power predicted by each cluster, so as to achieve the total predicted power of the whole wind power plant.
2. The FCM and AFSA-Elman-based short-term wind power prediction methods of claim 1, wherein:
in the step 1, the observation data includes time sequences formed by the wind speed, wind direction, active power and temperature data of each fan.
3. The short-term wind power prediction method based on FCM and AFSA-Elman according to claim 1, wherein in step 2, distance based on spatial metric features is used as similarity measurement of clustering algorithm, correlation coefficient is properly converted by introducing distance idea, and the result is expressed according to formula
Figure FDA0002232282090000011
Computing distances based on spatially measured features, where pxyIs a correlation coefficient, dxyIs a distance based on a spatial metric feature.
4. The short-term wind power prediction method based on FCM and AFSA-Elman according to claim 1, wherein in step 3, three factors of wind speed, wind direction and active power are integrated as a clustering index.
5. The FCM and AFSA-Elman-based short-term wind power prediction method according to claim 1, wherein step 5 adopts AFSA-Elman algorithm as wind power prediction model, and optimizes the weight threshold of Elman neural network by using the characteristics of parallel processing and automatic global optimization of artificial fish swarm algorithm.
CN201910971600.5A 2019-10-14 2019-10-14 short-term wind power prediction method based on FCM and AFSA-Elman Pending CN110738253A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910971600.5A CN110738253A (en) 2019-10-14 2019-10-14 short-term wind power prediction method based on FCM and AFSA-Elman

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910971600.5A CN110738253A (en) 2019-10-14 2019-10-14 short-term wind power prediction method based on FCM and AFSA-Elman

Publications (1)

Publication Number Publication Date
CN110738253A true CN110738253A (en) 2020-01-31

Family

ID=69270012

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910971600.5A Pending CN110738253A (en) 2019-10-14 2019-10-14 short-term wind power prediction method based on FCM and AFSA-Elman

Country Status (1)

Country Link
CN (1) CN110738253A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111476293A (en) * 2020-04-03 2020-07-31 河海大学 Wind power plant grouping method considering wind speed and wind direction
CN112016684A (en) * 2020-08-06 2020-12-01 广西大学 Electric power terminal fingerprint identification method of deep parallel flexible transmission network
CN112580900A (en) * 2021-02-23 2021-03-30 国能日新科技股份有限公司 Short-term power prediction method and system based on single fan modeling
CN113449920A (en) * 2021-06-30 2021-09-28 上海电机学院 Wind power prediction method, system and computer readable medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105703396A (en) * 2016-03-21 2016-06-22 广西电网有限责任公司电力科学研究院 Multi-turbine characterization equivalent modeling method of double-fed wind power plant
CN109165770A (en) * 2018-07-09 2019-01-08 江苏大学 Novel photovoltaic power forecasting method based on AFSA-Elman

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105703396A (en) * 2016-03-21 2016-06-22 广西电网有限责任公司电力科学研究院 Multi-turbine characterization equivalent modeling method of double-fed wind power plant
CN109165770A (en) * 2018-07-09 2019-01-08 江苏大学 Novel photovoltaic power forecasting method based on AFSA-Elman

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
汪玉环: ""股票间相关性测量方法的研究及应用"", 《中国优秀博硕士学位论文全文数据库(硕士)经济与管理科学辑》 *
祝牧: ""大型风电场群运行特性与优化控制研究"", 《中国优秀博硕士学位论文全文数据库(博士)工程科技Ⅱ辑》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111476293A (en) * 2020-04-03 2020-07-31 河海大学 Wind power plant grouping method considering wind speed and wind direction
CN111476293B (en) * 2020-04-03 2022-08-30 河海大学 Wind power plant grouping method considering wind speed and wind direction
CN112016684A (en) * 2020-08-06 2020-12-01 广西大学 Electric power terminal fingerprint identification method of deep parallel flexible transmission network
CN112016684B (en) * 2020-08-06 2022-09-02 广西大学 Electric power terminal fingerprint identification method of deep parallel flexible transmission network
CN112580900A (en) * 2021-02-23 2021-03-30 国能日新科技股份有限公司 Short-term power prediction method and system based on single fan modeling
CN113449920A (en) * 2021-06-30 2021-09-28 上海电机学院 Wind power prediction method, system and computer readable medium

Similar Documents

Publication Publication Date Title
Shi et al. Hybrid forecasting model for very-short term wind power forecasting based on grey relational analysis and wind speed distribution features
CN110738253A (en) short-term wind power prediction method based on FCM and AFSA-Elman
Ge et al. A hybrid model for short-term PV output forecasting based on PCA-GWO-GRNN
Mosadeghy et al. A time-dependent approach to evaluate capacity value of wind and solar PV generation
Liu et al. Wind power plant prediction by using neural networks
Yang et al. Day-ahead forecasting of photovoltaic output power with similar cloud space fusion based on incomplete historical data mining
CN111753893A (en) Wind turbine generator power cluster prediction method based on clustering and deep learning
Singh Application of artificial neural networks for predicting generated wind power
CN114021483A (en) Ultra-short-term wind power prediction method based on time domain characteristics and XGboost
CN110909310A (en) Photovoltaic short-term power generation capacity prediction method and system based on model parameter optimization
CN112884601A (en) Electric power system operation risk assessment method based on weather regionalization strategy
Wang et al. Multiscale prediction of wind speed and output power for the wind farm
Lin et al. Clustering wind turbines for a large wind farm using spectral clustering approach based on diffusion mapping theory
CN113610285A (en) Power prediction method for distributed wind power
Zhang et al. Graph Neural Network Based Short-Term Solar Irradiance Forcasting Model Considering Surrounding Meteorological Factors
Ghazanfari-Rad et al. Wind Speed Forecasting and Probability Distribution Analysis Using Measured Data at Weather Stations in Iran
Chen et al. Short-term wind power prediction based on BP neural network improved by t-tent-SSA algorithm
Huang et al. Power Prediction Method of Distributed Photovoltaic Digital Twin System Based on GA-BP
CN115829165B (en) Distributed photovoltaic operation condition analysis method and device based on power generation performance difference
Rao et al. Bright sunshine duration index-based prediction of solar PV power using ANN approach
Petrusev et al. Comparing Time Series Classification And Forecasting To Automatically Detect Distributed Generation
Mundotiya et al. Comprehensive Review on Uncertainty in Wind Power Forecasting-Models and Challenges
Dou et al. Day-Ahead Correction of Numerical Weather Prediction Solar Irradiance Forecasts Based on Similar Day Analysis
Yang et al. Wind-storage combined system based on just-in-time-learning prediction model with dynamic error compensation
CN113283702B (en) Power flow state evaluation method and device integrating safety and stability characteristics of power system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication

Application publication date: 20200131

RJ01 Rejection of invention patent application after publication