CN112668806B - Photovoltaic power ultra-short-term prediction method based on improved random forest - Google Patents
Photovoltaic power ultra-short-term prediction method based on improved random forest Download PDFInfo
- Publication number
- CN112668806B CN112668806B CN202110058821.0A CN202110058821A CN112668806B CN 112668806 B CN112668806 B CN 112668806B CN 202110058821 A CN202110058821 A CN 202110058821A CN 112668806 B CN112668806 B CN 112668806B
- Authority
- CN
- China
- Prior art keywords
- prediction
- formula
- photovoltaic power
- random forest
- value
- 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
Images
Classifications
-
- 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
Abstract
The invention discloses a photovoltaic power ultra-short term prediction calculation method based on an improved random forest, which is characterized by comprising the following steps of: the method comprises the steps of preliminary prediction of a random forest model, correction of preliminary prediction results in the afternoon period, correction of preliminary prediction results in the morning period, simulation calculation and error analysis, wherein photovoltaic power daily periodicity analysis is applied to double-period sequence prediction obtained by the random forest model, and prediction is carried out by comparing with a basic random forest prediction method, so that the dynamic characteristics of a system can be reflected, and the future power trend can be tracked; the improved random forest ultra-short term prediction model is a modified prediction model, and is simple in calculation, high in prediction performance, clear in physical significance, effective in prediction result and strong in practicability.
Description
Technical Field
The invention relates to the field of photovoltaic power prediction, in particular to a photovoltaic power ultra-short-term prediction method based on an improved random forest.
Background
Photovoltaic power generation has become a new point of growth for renewable energy power generation following wind power generation. Photovoltaic power generation is to convert solar energy resources into electric energy required by people by utilizing equipment. The sunlight has day and night periodicity, and is easily influenced by weather and meteorology, so the photovoltaic power has the characteristics of intermittence, fluctuation and randomness. The accurate prediction of the photovoltaic power directly influences the safe and economic operation of the power grid.
The photovoltaic power ultra-short-term prediction refers to prediction from a prediction moment to the future of 15 minutes to 4 hours, and the time resolution is 15 minutes. The significance of the ultra-short term prediction lies in that a plan curve is corrected in a rolling mode, and active output is adjusted in time.
The existing ultra-short term prediction generally establishes a mapping relation between historical input data and future power output, and can directly predict a future power value according to the historical data so as to obtain higher prediction accuracy. For the artificial intelligence method, the method has great advantages for processing the nonlinear time series, but has the defects that the dynamic characteristic of the system cannot be reflected and the future power trend cannot be tracked.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a photovoltaic power ultra-short-term prediction method based on an improved random forest, which is scientific and reasonable, clear in physical significance, rapid, efficient, simple, practical, higher in precision, capable of reflecting the dynamic characteristics of a system and tracking the trend of future power.
The technical scheme adopted for realizing the aim of the invention is as follows: a photovoltaic power ultra-short term prediction method based on improved random forests is characterized by comprising the following steps: the method comprises the following steps:
1) preliminary prediction of random forest models
The photovoltaic power historical power is divided without pruning by utilizing a random forest model to generate a CART decision tree, the CART decision tree is a binary recursive division technology, a current sample set is divided into two subsets at each node except leaf nodes,
the attribute selection measure adopted by the CART decision tree is a Gini index, and if a data set D comprises m categories, the Gini index G is used D Is defined as formula (1):
in the formula: p is a radical of j Is the frequency of occurrence of the class j element;
the Gini index requires consideration of a binary partition of each attribute, assuming that the binary partition of attribute A divides data set D into D 1 And D 2 ，
The kini index of the sample set D divided by some attribute a at the child node this time is formula (2):
through k rounds of training, a classification model sequence { h }is obtained 1 (X),h 2 (X),…,h k (X), and then forming a multi-classification model system by using the same, wherein the final classification result of the system adopts a simple majority voting method,
the final classification decision result is as follows (3):
wherein, h (x) represents a combined classification model, hi is a single decision tree classification model, Y represents an output variable or a target variable, I (-) is an exemplary function, and equation (1) illustrates the way of using majority voting decision to determine the final classification;
2) correction of preliminary prediction results during afternoon hours
Afternoon descent area peak: and (3) generating a current maximum force output point, and correcting by adopting the current maximum force output point, wherein the correction method is as shown in formula (4):
wherein, P i To an initial predicted value, P imax To the initial predicted peak value, P i Is a correction value, P tmax The peak value of the actual value of the last moment is obtained by the current moment;
3) correction of the morning hours of preliminary prediction results
When the power is corrected in the morning, different from the afternoon, in the morning, the photovoltaic power is in a gradually rising trend due to the solar radiation intensity, the actual output peak value of the day cannot be obtained at the moment to be predicted, so that the initial prediction result of the random forest cannot be adjusted according to the peak value, the DGM (1,1) algorithm is adopted to predict the peak value for correction,
DGM (1,1) model, the original sequence is set as formula (5):
X (0) ＝(x (0 (1),x (0 (2),...x (0 (n)) (5)
α (1) as a first accumulation subtraction operator, as in equation (6):
α (1) x (1) ＝x (1) (k)-x (1) (k-1),k＝2,...,n (6)
wherein x (1) (k) Is (7)
Parameter value is formula (8)
The DGM (1,1) parameter
Is the least square estimate of (9)
The recurrence function is then the equation (10)
x (1) (k+1)＝γx (1) (k)+ρ (10)
The mathematical model that can ultimately yield the original sequence is as in equation (11)
DGM (1,1) prediction is performed on the power time series from the moment of contribution: inputting the time sequence into DGM (1,1) model for prediction to obtain trend prediction value,
the correction formula of the predicted value obtained by the random forest preliminary prediction in the morning is shown as formula (12):
in the formula
P i To an initial predicted value, P i Is a correction value, T i-1 Is the actual value of the time before the time to be predicted, Y t To measure power, t Pmax Predicting the time of the peak value for the trend line of the predicted value;
4) simulation calculation
Simulation input quantity: analyzing the measured data of the electric field to determine the total installed capacity of the electric field; inputting data: historical power of the photovoltaic power station; the data sampling interval is 15 min; obtaining a photovoltaic power ultra-short term prediction result of the daily prediction time period according to the steps 1) to 3);
5) error analysis
The accuracy of the prediction result is defined as formula (13):
in the formula, P M To predict photovoltaic power; p is P Actual photovoltaic power; n is the number of the predicted points; the Cap is the starting capacity of the photovoltaic power station,
the yield is defined by the formula (14):
in the formula, if
Then B is k 1; if it is
Then B is k When the root mean square error of the prediction result is 0, the root mean square error of the prediction result is formula (15):
the mean absolute error is formula (16):
inputting simulation input quantity according to the step 4), carrying out error calculation on the predicted power calculated by the model and the actual measured power through the error evaluation standard formula (13) -formula (16) in the step 5), and obtaining the predicted root mean square error, the average absolute error, the qualification rate and the accuracy rate.
According to the photovoltaic power ultra-short term prediction calculation method based on the improved random forest, due to the fact that preliminary prediction based on a random forest model is adopted; correcting the output of the preliminary predicted value in the afternoon time period by using the current maximum output point generated in the afternoon; and correcting the output of the preliminary predicted value in the morning time period by predicting the rising trend of the DGM (1,1) model at the occurred moment in the morning, and the like. In the double-period sequence prediction obtained by applying the photovoltaic power daily periodicity analysis to the random forest model, the prediction is carried out by comparing with a basic random forest prediction method, the dynamic characteristics of the system can be reflected, and the future power trend can be tracked; the improved random forest ultra-short term prediction model is a modified prediction model, and is simple in calculation and high in prediction performance. Clear physical significance, effective prediction result and strong practicability.
Drawings
FIG. 1 is a schematic diagram of a random forest model;
FIG. 2 is a photovoltaic power ultra-short term prediction frame diagram based on an improved random forest model;
fig. 3 is a diagram showing the comparison between the original predicted result of the random forest model and the predicted result of the improved random forest model and the real value (afternoon).
Fig. 4 is a diagram showing comparison between the original prediction result of the random forest model and the prediction result of the improved random forest model and the real value (in the morning).
Detailed Description
The following further describes a photovoltaic power ultra-short term prediction calculation method based on an improved random forest model, with reference to the accompanying drawings and specific embodiments.
With reference to fig. 1 to 4, the method for ultra-short-term prediction of photovoltaic power based on improved random forest according to the present invention includes the following steps:
1) preliminary prediction of random forest models
The photovoltaic power historical power is divided without pruning by utilizing a random forest model to generate a CART decision tree, the CART decision tree is a binary recursive division technology, a current sample set is divided into two subsets at each node except leaf nodes,
the attribute selection measure adopted by the CART decision tree is a Gini index, and if a data set D comprises m categories, the Gini index G is set D Is defined as formula (1):
in the formula: p is a radical of j Is the frequency of occurrence of the j-type element;
the Gini index requires consideration of a binary partition of each attribute, assuming that a binary partition of attribute A divides data set D into D 1 And D 2 ，
The kini index of the sample set D divided by some attribute a at the child node this time is formula (2):
through k rounds of training, a classification model sequence { h }is obtained 1 (X),h 2 (X),…,h k (X), then using them to form a multi-classification model system, and adopting simple majority voting method to make the final classification result of said system,
the final classification decision result is as follows (3):
where h (x) represents a combined classification model, hi is a single decision tree classification model, Y represents an output variable or target variable, I (-) is an exemplary function, and equation (1) illustrates the way that a majority voting decision is used to determine the final classification.
2) Correction of the afternoon period of the preliminary prediction result
Peak in the afternoon descent region: and (3) generating a current maximum force output point, and correcting by adopting the current maximum force output point, wherein the correction method is as shown in formula (4):
wherein, P i To an initial predicted value, P imax To the initial predicted peak value, P i Is a correction value, P tmax The peak value of the actual value of the last moment is obtained by the current moment.
3) Correction of preliminary prediction results during the morning hours
When the power is corrected in the morning, different from the afternoon, in the morning, the photovoltaic power is in a gradually rising trend due to the solar radiation intensity, the actual output peak value of the day cannot be obtained at the moment to be predicted, so that the initial prediction result of the random forest cannot be adjusted according to the peak value, the DGM (1,1) algorithm is adopted to predict the peak value for correction,
DGM (1,1) model, the original sequence is set as formula (5):
X (0) ＝(x (0 (1),x (0 (2),...x (0 (n)) (5)
α (1) the first accumulation subtraction operator is as follows:
α (1) x (1) ＝x (1) (k)-x (1) (k-1),k＝2,...,n (6)
wherein x is (1) (k) Is (7)
Parameter value is formula (8)
The DGM (1,1) parameter
Is the least square estimate of (9)
The recurrence function is formula (10)
x (1) (k+1)＝γx (1) (k)+ρ (10)
The mathematical model that can ultimately yield the original sequence is as in equation (11)
DGM (1,1) prediction is performed on the power time series from the moment of contribution: inputting the time sequence into DGM (1,1) model for prediction to obtain trend prediction value,
the correction formula of the predicted value obtained by the random forest preliminary prediction in the morning is shown as formula (12):
in the formula
P i To an initial predicted value, P i Is a correction value, T i-1 Is the actual value of the time before the time to be predicted, Y t To measure power, t Pmax And predicting the time of the peak value for the trend line of the predicted value.
4) Simulation calculation
Simulation input quantity: analyzing the measured data of the electric field to determine the total installed capacity of the electric field; inputting data: historical power of the photovoltaic power station; the data sampling interval is 15 min; and obtaining the photovoltaic power ultra-short term prediction result in the daily prediction time period according to the steps 1) to 3).
5) Error analysis
The accuracy of the prediction result is defined as formula (13):
in the formula, P M To predict photovoltaic power; p P Actual photovoltaic power; n is the number of the predicted points; the Cap is the starting capacity of the photovoltaic power station,
the yield is defined by the formula (14):
in the formula, if
Then B is k 1 is ═ 1; if it is
Then B is k When the root mean square error of the prediction result is 0, the root mean square error is expressed by formula (15):
the mean absolute error is formula (16):
inputting simulation input quantity according to the step 4), carrying out error calculation on the predicted power calculated by the model and the actual measured power through the error evaluation standard formula (13) -formula (16) in the step 5), and obtaining the predicted root mean square error, the average absolute error, the qualification rate and the accuracy rate.
The photovoltaic power ultra-short term prediction calculation method based on the improved random forest model is characterized in that field station actual measurement data of a certain photovoltaic set are taken as an example for analysis, and the sampling interval is 15 min. The installed capacity of the power station is 650 MW; the evaluation indexes of the prediction results are shown in table 1:
TABLE 1 prediction accuracy statistics
Tab.1 prediction accuracy statistics
The description of the present invention is not intended to be exhaustive or to limit the scope of the claims, and those skilled in the art will be able to conceive other substantially equivalent alternatives without inventive step in light of the teachings of the embodiments of the present invention.
Claims (1)
1. A photovoltaic power ultra-short term prediction method based on improved random forests is characterized by comprising the following steps: the method comprises the following steps: 1) preliminary prediction of random forest models
The photovoltaic power historical power is divided without pruning by utilizing a random forest model to generate a CART decision tree, the CART decision tree is a binary recursive division technology, a current sample set is divided into two subsets at each node except leaf nodes,
the attribute selection measure adopted by the CART decision tree is a Gini index, and if a data set D comprises m categories, the Gini index G is set D Is defined as formula (1):
in the formula: p is a radical of formula j Is the frequency of occurrence of the class j element;
the Gini index requires consideration of a binary partition of each attribute, assuming that the binary partition of attribute A divides data set D into D 1 And D 2 ，
The kini index of the sample set D divided by some attribute a at the child node this time is formula (2):
through k rounds of training, a classification model sequence { h }is obtained 1 (X),h 2 (X),…,h k (X), and then forming a multi-classification model system by using the same, wherein the final classification result of the system adopts a simple majority voting method,
the final classification decision result is as follows (3):
where h (x) represents a combined classification model, hi is a single decision tree classification model, Y represents an output variable, I (-) is an exemplary function, and equation (1) illustrates the way that a majority voting decision is used to determine the final classification;
2) correction of preliminary prediction results during afternoon hours
Afternoon descent area peak: and (3) generating a current maximum force output point, and correcting by adopting the current maximum force output point, wherein the correction method is as shown in formula (4):
wherein, P i To an initial predicted value, P imax To the initial predicted peak value, P i Is a correction value, P tmax The peak value of the actual value of the last moment is obtained by the current moment;
3) correction of preliminary prediction results during the morning hours
When the power is corrected in the morning, different from the afternoon, the photovoltaic power is in a gradually rising trend in the morning due to the solar radiation intensity, the actual output peak value of the day cannot be obtained at the moment to be predicted, so that the initial prediction result of the random forest cannot be adjusted according to the peak value, and the DGM algorithm is adopted to predict the peak value for correction;
DGM prediction is carried out on the power time sequence from the moment of output: inputting the time sequence into a DGM model for prediction to obtain a trend prediction value,
the correction formula of the predicted value obtained by the random forest preliminary prediction in the morning is shown as formula (5):
in the formula
P i To an initial predicted value, P i Is a correction value, T i-1 Is an actual value of a time before the time to be predicted, Y t To measure power, t Pmax Predicting the time of the peak value for the trend line of the predicted value;
4) simulation calculation
Simulation input quantity: analyzing the measured data of the electric field to determine the total installed capacity of the electric field; inputting data: historical power of the photovoltaic power station; the data sampling interval is 15 min; obtaining a photovoltaic power ultra-short term prediction result of the daily prediction time period according to the steps 1) to 3);
5) error analysis
The accuracy of the prediction result is defined as formula (6):
in the formula, P M To predict photovoltaic power; p is P Actual photovoltaic power; n is the number of the predicted points; the Cap is the starting capacity of the photovoltaic power station,
the yield is defined as formula (7):
in the formula, if
Then B is k 1; if it is
Then B is k ＝0，
The root mean square error of the prediction result is formula (8):
the mean absolute error is formula (9):
inputting simulation input quantity according to the step 4), carrying out error calculation on the predicted power calculated by the model and the actual measured power through the error evaluation standard formula (6) -formula (9) in the step 5), and obtaining the predicted root mean square error, the average absolute error, the qualification rate and the accuracy rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110058821.0A CN112668806B (en) | 2021-01-17 | 2021-01-17 | Photovoltaic power ultra-short-term prediction method based on improved random forest |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110058821.0A CN112668806B (en) | 2021-01-17 | 2021-01-17 | Photovoltaic power ultra-short-term prediction method based on improved random forest |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112668806A CN112668806A (en) | 2021-04-16 |
| CN112668806B true CN112668806B (en) | 2022-09-06 |
Family
ID=75415400
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110058821.0A Active CN112668806B (en) | 2021-01-17 | 2021-01-17 | Photovoltaic power ultra-short-term prediction method based on improved random forest |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112668806B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113537582B (en) * | 2021-07-02 | 2022-05-24 | 东北电力大学 | Photovoltaic power ultra-short-term prediction method based on short-wave radiation correction |
| CN113722375B (en) * | 2021-08-10 | 2023-10-17 | 国网浙江省电力有限公司电力科学研究院 | Double-layer hybrid photovoltaic prediction method based on multi-algorithm optimization and computer equipment |
| CN116307269B (en) * | 2023-05-16 | 2023-10-31 | 华能山东泰丰新能源有限公司 | Photovoltaic power generation power prediction method and device based on artificial intelligence |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018143071A (en) * | 2017-02-28 | 2018-09-13 | 京セラ株式会社 | Information processing device and control method of information processing device |
| CN109150104A (en) * | 2018-08-10 | 2019-01-04 | 江南大学 | A kind of diagnosing failure of photovoltaic array method based on random forests algorithm |
| CN109919353A (en) * | 2019-01-11 | 2019-06-21 | 杭州电子科技大学 | A kind of distributed photovoltaic prediction technique of the ARIMA model based on spatial coherence |
| CN110516840A (en) * | 2019-07-15 | 2019-11-29 | 国网甘肃省电力公司电力科学研究院 | Short term prediction method based on the wind light generation power output for improving random forest method |
| CN110852492A (en) * | 2019-10-25 | 2020-02-28 | 东北电力大学 | Photovoltaic power ultra-short-term prediction method for finding similarity based on Mahalanobis distance |
| CN111598337A (en) * | 2020-05-15 | 2020-08-28 | 国网河北省电力有限公司电力科学研究院 | Method for predicting short-term output of distributed photovoltaic system |
| CN111815027A (en) * | 2020-06-09 | 2020-10-23 | 山东大学 | Photovoltaic station generated power prediction method and system |
| CN111950752A (en) * | 2019-05-14 | 2020-11-17 | 京东方科技集团股份有限公司 | Photovoltaic power station generating capacity prediction method, device and system and storage medium thereof |
| CN112036440A (en) * | 2020-07-31 | 2020-12-04 | 山东科技大学 | Satellite attitude control system fault diagnosis and early warning method based on random forest |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3114905A1 (en) * | 2014-03-06 | 2017-01-11 | Philips Lighting Holding B.V. | Intelligent lighting system with predictive maintenance scheduling and method of operation thereof |
| JP6742894B2 (en) * | 2016-06-09 | 2020-08-19 | 株式会社日立製作所 | Data prediction system and data prediction method |
| US10797511B2 (en) * | 2018-09-28 | 2020-10-06 | Johnson Controls Technology Company | Photovoltaic energy system with stationary energy storage control and power factor correction |
-
2021
- 2021-01-17 CN CN202110058821.0A patent/CN112668806B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2018143071A (en) * | 2017-02-28 | 2018-09-13 | 京セラ株式会社 | Information processing device and control method of information processing device |
| CN109150104A (en) * | 2018-08-10 | 2019-01-04 | 江南大学 | A kind of diagnosing failure of photovoltaic array method based on random forests algorithm |
| CN109919353A (en) * | 2019-01-11 | 2019-06-21 | 杭州电子科技大学 | A kind of distributed photovoltaic prediction technique of the ARIMA model based on spatial coherence |
| CN111950752A (en) * | 2019-05-14 | 2020-11-17 | 京东方科技集团股份有限公司 | Photovoltaic power station generating capacity prediction method, device and system and storage medium thereof |
| CN110516840A (en) * | 2019-07-15 | 2019-11-29 | 国网甘肃省电力公司电力科学研究院 | Short term prediction method based on the wind light generation power output for improving random forest method |
| CN110852492A (en) * | 2019-10-25 | 2020-02-28 | 东北电力大学 | Photovoltaic power ultra-short-term prediction method for finding similarity based on Mahalanobis distance |
| CN111598337A (en) * | 2020-05-15 | 2020-08-28 | 国网河北省电力有限公司电力科学研究院 | Method for predicting short-term output of distributed photovoltaic system |
| CN111815027A (en) * | 2020-06-09 | 2020-10-23 | 山东大学 | Photovoltaic station generated power prediction method and system |
| CN112036440A (en) * | 2020-07-31 | 2020-12-04 | 山东科技大学 | Satellite attitude control system fault diagnosis and early warning method based on random forest |
Non-Patent Citations (5)
| Title |
|---|
| A novel adaptive discrete grey model with time-varying parameters for long-term photovoltaic power generation forecasting;Song Ding 等;《Energy Conversion and Management》;20201120;第1-15页 * |
| A Short-Term Forecasting Method for Photovoltaic Power Based on Ensemble Adaptive Boosting Random Forests;Guanjie Wang 等;《2020 IEEE/IAS 56th Industrial and Commercial Power Systems Technical Conference (I&CPS)》;20200825;第1-8页 * |
| Day-ahead power forecasting in a large-scale photovoltaic plant based on weather classification using LSTM;Mingming Gao 等;《Energy》;20190729;第1-12页 * |
| Fault Diagnosis Method of Photovoltaic Array Based on Random Forest Algorithm;Sizhe Gong 等;《2020 39th Chinese Control Conference (CCC)》;20200909;第4249-4254页 * |
| 基于决策树理论的风电功率实时预测方法;杨茂 等;《电测与仪表》;20180610;第120-124页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112668806A (en) | 2021-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112668806B (en) | Photovoltaic power ultra-short-term prediction method based on improved random forest | |
| CN102930358B (en) | A kind of neural net prediction method of photovoltaic power station power generation power | |
| CN106529814B (en) | Distributed photovoltaic ultra-short term prediction method based on Adaboost clustering and Markov chain | |
| CN110059891B (en) | Photovoltaic power station output power prediction method based on VMD-SVM-WSA-GM combined model | |
| CN111027775A (en) | Step hydropower station generating capacity prediction method based on long-term and short-term memory network | |
| Verma et al. | Markov models based short term forecasting of wind speed for estimating day-ahead wind power | |
| CN110717610B (en) | Wind power prediction method based on data mining | |
| Lu et al. | A hybrid approach for day-ahead forecast of PV power generation | |
| CN114970952B (en) | Photovoltaic output short-term prediction method and system considering environmental factors | |
| CN116341613A (en) | Ultra-short-term photovoltaic power prediction method based on Informar encoder and LSTM | |
| CN110852492A (en) | Photovoltaic power ultra-short-term prediction method for finding similarity based on Mahalanobis distance | |
| Alanazi et al. | Day-ahead solar forecasting using time series stationarization and feed-forward neural network | |
| Li et al. | Short-term prediction of the output power of PV system based on improved grey prediction model | |
| CN110363338B (en) | Cos attenuated atomic decomposition and chaos theory-based wind power ultra-short-term prediction method | |
| CN106503828B (en) | Ultra-short term chaotic prediction method for photovoltaic output power | |
| Omer et al. | Adaptive boosting and bootstrapped aggregation based ensemble machine learning methods for photovoltaic systems output current prediction | |
| Khatib et al. | Assessment of three learning machines for long-term prediction of wind energy in Palestine | |
| CN111967660B (en) | Ultra-short-term photovoltaic prediction residual error correction method based on SVR | |
| Han et al. | Hybrid model based on EEMD, ARMA and elman for photovoltaic power prediction | |
| Zhang et al. | Short-term wind power interval prediction based on gd-lstm and bootstrap techniques | |
| CN113112085A (en) | New energy station power generation load prediction method based on BP neural network | |
| Shen et al. | PV power interval prediction based on EEMD-LSTM method | |
| Zhang et al. | Prediction of PV output based on local mean decomposition under limited information | |
| Wang et al. | A new improved combined model algorithm for the application of photovoltaic power prediction | |
| Zhang et al. | Research on the prediction of solar energy generation based on measured environmental data |
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 |