CN111695736B - Photovoltaic power generation short-term power prediction method based on multi-model fusion - Google Patents
Photovoltaic power generation short-term power prediction method based on multi-model fusion Download PDFInfo
- Publication number
- CN111695736B CN111695736B CN202010541414.0A CN202010541414A CN111695736B CN 111695736 B CN111695736 B CN 111695736B CN 202010541414 A CN202010541414 A CN 202010541414A CN 111695736 B CN111695736 B CN 111695736B
- Authority
- CN
- China
- Prior art keywords
- power generation
- data
- historical
- fusion
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- 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
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- 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
-
- 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
-
- 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 application discloses a photovoltaic power generation short-term power prediction method based on multi-model fusion, which comprises the following steps: step 1, weather historical data and corresponding photovoltaic power generation historical data of a predicted photovoltaic power generation area are obtained, and a first power predicted value corresponding to the current weather data is calculated according to the collected current weather data, weather historical data and photovoltaic power generation historical data by using a local weighted linear regression model; step 2, calculating a second power predicted value corresponding to the current moment by adopting a recurrent neural network according to the photovoltaic power generation historical data and the corresponding power generation time; and 3, under the condition of minimum fusion error value, calculating a first fusion weight and a second fusion weight corresponding to the local weighted linear regression model and the recurrent neural network, and combining the first power predicted value and the second power predicted value to calculate a third power predicted value at the current moment and under meteorological conditions. Through the technical scheme in this application, the accuracy of photovoltaic power generation short-term power prediction is improved.
Description
Technical Field
The application relates to the technical field of photovoltaic power generation, in particular to a photovoltaic power generation short-term power prediction method based on multi-model fusion.
Background
Climate change and energy crisis drive the use and development of solar power generation. Solar energy is considered one of the most abundant and promising candidate energy sources for large-scale power generation. However, one inherent feature of all renewable energy sources is the high degree of intermittence, the generation of electricity being entirely dependent on weather and weather parameters, and therefore the output of solar energy is not fully controlled or planned in advance.
The photovoltaic system is ensured to be safely integrated into the intelligent power grid, accurate short-term power prediction is achieved, and the photovoltaic system is an important component of a new energy management system. Unexpected fluctuations in solar energy production may have a significant impact on daily operation and management, and the health of the whole grid may have a negative impact on the quality of life of new energy consumers, if an accurate short-term power prediction is lacking.
In the prior art, a single prediction model is generally used for short-term power prediction, so that the adaptability is not strong, and the data predicted by the single prediction model is one of linear and nonlinear, so that unavoidable prediction risks exist in the data predicted by the model, and the stability of the photovoltaic grid connection can be influenced.
Disclosure of Invention
The purpose of the present application is: and a linear and nonlinear prediction fusion mode is adopted, prediction risks existing in the prediction data are dispersed, the accuracy of short-term power prediction of photovoltaic power generation is improved, and the reliability of photovoltaic grid connection is further improved.
The technical scheme of the application is as follows: the invention provides a photovoltaic power generation short-term power prediction method based on multi-model fusion, which comprises the following steps: step 1, weather historical data and corresponding photovoltaic power generation historical data of a predicted photovoltaic power generation area are obtained, and a first power predicted value corresponding to the current weather data is calculated according to the collected current weather data, weather historical data and photovoltaic power generation historical data by using a local weighted linear regression model; step 2, calculating a second power predicted value corresponding to the current moment by adopting a recurrent neural network according to the photovoltaic power generation historical data and the corresponding power generation time; and 3, under the condition of minimum fusion error value, calculating a first fusion weight and a second fusion weight corresponding to the local weighted linear regression model and the recurrent neural network, and combining the first power predicted value and the second power predicted value to calculate a third power predicted value at the current moment and under meteorological conditions.
In any of the above technical solutions, further, in step 1, specifically includes: step 11, dividing meteorological historical data and photovoltaic power generation historical data into a training set and a testing set; step 12, calculating correlation coefficients between the meteorological historical data and the photovoltaic power generation capacity historical data in the training set in sequence, constructing a local weighted linear regression model, and calculating predicted power generation capacity data corresponding to the meteorological historical data in the test set; step 13, correcting the local weighted linear regression model according to the predicted power generation data and the historical data of the photovoltaic power generation in the test set by combining the loss function, and judging that the local weighted linear regression model completes correction when the loss function converges; and 14, carrying out power prediction on the acquired current meteorological data according to the corrected local weighted linear regression model, and recording a prediction result as a first power prediction value.
In any of the above technical solutions, further, in step 13, a calculation formula of the loss function is:
wherein m is the number of data in the test set, and x (i) For the predicted power generation amount data corresponding to the ith weather history data in the test set, y (i) Photovoltaic power generation amount historical data, w, corresponding to the ith weather historical data i (x) For the ith predicted power generation amount data x (i) Is used for the weight value of (a),
for the ith predicted power generation amount data x (i) The j-th adjacent point in the preset range, k is a weight change parameter, theta (i) For the corresponding matrix of correlation coefficients,
wherein, the value of each element in the correlation coefficient matrix is determined by the value of the correlation coefficient, if the value of the correlation coefficient is [ -0.5,0.5], the value of the corresponding element is 0, otherwise, the value is 1.
In any one of the above technical solutions, further, in step 3, under the condition that the fusion error value is minimum, the first fusion weight and the second fusion weight corresponding to the local weighted linear regression model and the recurrent neural network specifically include:
step 31, calculating a first historical power predicted value corresponding to weather historical data according to a local weighted linear regression model, and calculating a second historical power predicted value corresponding to power generation time according to a recurrent neural network;
step 32, calculating a first fusion weight and a second fusion weight by adopting a fusion error calculation formula according to the first historical power predicted value, the second historical power predicted value and the photovoltaic power generation amount historical data, wherein the fusion error calculation formula is as follows:
p c (n) ＝w 1 p 1 (n) +w 2 p 2 (n)
wherein p is t (n) For the nth photovoltaic power generation amount history data, p 1 (n) For the corresponding first historical power predicted value, p 2 (n) For the corresponding second historical power predicted value, w 1 W is the first fusion weight 2 P is the second fusion weight c (n) E is the n-th historical power predicted value after fusion 1 E is the error value corresponding to the first historical power predicted value 2 For the error value corresponding to the second historical power predictor, var (·) is variance and cov (·) is covariance.
In any of the above solutions, further, the weather history data includes: irradiance, visibility, ambient temperature, dew point temperature, ambient humidity, wind speed, barometric pressure, altitude.
The beneficial effects of this application are:
according to the technical scheme, the local weighted linear regression model is adopted, the current meteorological data is utilized to conduct linear prediction on the photovoltaic power generation short-term power, a loss function and a correlation coefficient matrix determined by correlation coefficients of historical data are introduced, the local weighted linear regression model is corrected, and the accuracy of linear model prediction is improved.
According to the method and the device, the fusion error is introduced, the first fusion weight and the second fusion weight are calculated, the fusion of the local weighted linear regression model and the prediction result of the nonlinear recurrent neural network is achieved, the prediction risk existing in the linear and nonlinear prediction data is dispersed as the result of the short-term power prediction, the accuracy of the short-term power prediction of photovoltaic power generation is improved, and the influence of photovoltaic grid-connected fluctuation on a power grid is reduced.
Drawings
The advantages of the foregoing and/or additional aspects of the present application will become apparent and readily appreciated from the description of the embodiments, taken in conjunction with the accompanying drawings, wherein:
FIG. 1 is a schematic flow chart of a photovoltaic power generation short-term power prediction method based on multi-model fusion according to one embodiment of the present application;
fig. 2 is a schematic diagram of a recurrent neural network according to one embodiment of the present application.
Detailed Description
In order that the above-recited objects, features and advantages of the present application will be more clearly understood, a more particular description of the application will be rendered by reference to the appended drawings and appended detailed description. It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present application, however, the present application may be practiced otherwise than as described herein, and thus the scope of the present application is not limited to the specific embodiments disclosed below.
As shown in fig. 1, the present embodiment provides a photovoltaic power generation short-term power prediction method based on multi-model fusion, including: step 1, weather historical data and corresponding photovoltaic power generation amount historical data of a predicted photovoltaic power generation area are obtained, and a local weighted linear regression model is utilized to calculate a first power predicted value corresponding to the current weather data according to the collected current weather data, weather historical data and photovoltaic power generation amount historical data, wherein the weather historical data comprises: irradiance, visibility, ambient temperature, dew point temperature, ambient humidity, wind speed, barometric pressure, altitude.
Specifically, historical data of a predicted photovoltaic power generation area for one year is selected, the historical data comprises meteorological historical data and photovoltaic power generation capacity historical data, the historical data is processed, the historical data is grouped by taking every 15 minutes as a statistical unit, a plurality of groups of data are obtained, 80% of the plurality of groups of data are randomly divided into training sets, and the rest 20% of the plurality of groups of data are divided into test sets.
Further, in step 1, specifically includes:
step 11, dividing meteorological historical data and photovoltaic power generation historical data into a training set and a testing set;
step 12, calculating correlation coefficients between the meteorological historical data and the photovoltaic power generation capacity historical data in the training set in sequence, constructing a local weighted linear regression model, and calculating predicted power generation capacity data corresponding to the meteorological historical data in the test set;
specifically, in this embodiment, a locally weighted linear regression (Locally weighted linear regression, WLR) model based on principal component analysis (Principal Component Analysis, PCA) is used as a linear model, and each set of data is used as each data point in the linear analysis process to perform linear analysis on weather history data and photovoltaic power generation history data.
Through linear operation on the historical data, the correlation coefficient between each meteorological historical data and the photovoltaic power generation capacity historical data can be calculated, in particular, parameters such as irradiance, visibility, ambient temperature, ambient humidity and the like have great influence on the photovoltaic power generation capacity, therefore, a threshold value of minus 0.5,0.5 is set, a correlation coefficient matrix is generated through comparison with the correlation coefficient, when the value of the correlation coefficient is within the range of minus 0.5, the value of the corresponding element in the correlation coefficient matrix is 0, and otherwise, the value of the corresponding element is 1.
And 13, correcting the local weighted linear regression model according to the predicted power generation data and the historical data of the photovoltaic power generation in the test set by combining the loss function, and judging that the local weighted linear regression model completes correction when the loss function converges, wherein the calculation formula of the loss function is as follows:
wherein m is the number of data in the test set, and x (i) For the predicted power generation amount data corresponding to the ith weather history data in the test set, y (i) Photovoltaic power generation amount historical data, w, corresponding to the ith weather historical data i (x) For the ith predicted power generation amount data x (i) Is used for the weight value of (a),
for the ith predicted power generation amount data x (i) The j-th adjacent point in the preset range, k is a weight change parameter, theta (i) For the corresponding matrix of correlation coefficients,
wherein, the value of each element in the correlation coefficient matrix is determined by the value of the correlation coefficient, if the value of the correlation coefficient is [ -0.5,0.5], the value of the corresponding element is 0, otherwise, the value is 1.
Specifically, when the local weighted linear regression model is used for predicting the generated energy, not all data are selected for linear regression prediction, but adjacent data near the predicted point (data point) are selected, so that after the local weighted linear regression model is built by the data in the training set, a loss function is introduced to correct the local weighted linear regression model, so that the larger the weight corresponding to the data point with the closer predicted point distance is, the larger the contribution to the regression coefficient in the local weighted linear regression model is. And the correlation coefficient matrix constructed according to the correlation coefficients is fully utilized to realize the correction of the local weighted linear regression model, so that the accuracy of linear prediction of the local weighted linear regression model is improved.
And 14, carrying out power prediction on the acquired current meteorological data according to the corrected local weighted linear regression model, and recording a prediction result as a first power prediction value.
specifically, as shown in fig. 2, in order to utilize the time-dependent change trend in the historical data, a recurrent neural network is used as the nonlinear prediction model in the present embodiment. When constructing the recurrent neural network, the recurrent neural network comprises three hidden layers of LSTM, tanh and Sigmoid, and then an ideal distribution initial value is selected for the weight value in the recurrent neural network by using an Xavier-He initialization method.
And then training the recurrent neural network by utilizing the historical data of the photovoltaic power generation amount and the corresponding power generation time, and confirming the stability of a predicted result by using 10 times of cross verification, and finally obtaining the stable recurrent neural network so as to realize the calculation of a second power predicted value at the current moment.
step 31, calculating a first historical power predicted value corresponding to weather historical data according to a local weighted linear regression model, and calculating a second historical power predicted value corresponding to power generation time according to a recurrent neural network;
step 32, according to the first historical power predicted value, the second historical power predicted value and the photovoltaic power generation amount historical data, a fusion error calculation formula is adopted, and under the condition that the fusion error value is minimum, a corresponding first fusion weight and second fusion weight are calculated, wherein the fusion error calculation formula is as follows:
p c (n) ＝w 1 p 1 (n) +w 2 p 2 (n)
wherein p is t (n) For the nth photovoltaic power generation amount history data, p 1 (n) For the corresponding first historical power predicted value, p 2 (n) For the corresponding second historical power predicted value, w 1 W is the first fusion weight 2 P is the second fusion weight c (n) E is the n-th historical power predicted value after fusion 1 E is the error value corresponding to the first historical power predicted value 2 For the error value corresponding to the second historical power predictor, var (·) is variance and cov (·) is covariance.
Specifically, a first historical power predicted value obtained by setting a local weighted linear regression model is p 1 (n) The second historical power predicted value obtained by the recurrent neural network is p 2 (n) After the weight is introduced, the corresponding historical power predicted value p c (n) The method comprises the following steps:
p c (n) ＝w 1 p 1 (n) +w 2 p 2 (n)
wherein w is 1 W is the first fusion weight 2 And the second fusion weight value.
Taking linear prediction, nonlinear prediction and photovoltaic power generation amount historical data p into consideration t (n) In this embodiment, a fusion error calculation manner is adopted, and a corresponding first fusion weight w is calculated under the condition that a fusion error value is minimum 1 And is the second fusion weight w 2 At this time, the fusion error calculation formula is:
p c (n) ＝w 1 p 1 (n) +w 2 p 2 (n)
at this time, the historical power predicted value p c (n) And photovoltaic power generation history data p t (n) The variance of the error is:
where cov (·) is covariance.
By calculating the variance Var (e) c ) Can be obtained by:
therefore, by the above-mentioned limitation, it can be countedCalculating a first fusion weight w 1 And a second fusion weight w 2 。
The short-term power prediction method in the embodiment is utilized to predict through statistics of historical data of a certain photovoltaic power station in China, and the calculated first fusion weight w is calculated 1 =0.3, second fusion weight w 2 =0.7, and further compared with the short-term power prediction method of a single prediction model, the prediction accuracy is improved by about 8% compared with the linear model, and the prediction accuracy is improved by about 5% compared with the nonlinear model. In summary, by the method in the embodiment, the reliability and stability of the photovoltaic power generation grid connection can be improved.
The technical scheme of the application is explained in detail above with reference to the accompanying drawings, and the application provides a photovoltaic power generation short-term power prediction method based on multi-model fusion, which comprises the following steps: step 1, weather historical data and corresponding photovoltaic power generation historical data of a predicted photovoltaic power generation area are obtained, and a first power predicted value corresponding to the current weather data is calculated according to the collected current weather data, weather historical data and photovoltaic power generation historical data by using a local weighted linear regression model; step 2, calculating a second power predicted value corresponding to the current moment by adopting a recurrent neural network according to the photovoltaic power generation historical data and the corresponding power generation time; and 3, under the condition of minimum fusion error value, calculating a first fusion weight and a second fusion weight corresponding to the local weighted linear regression model and the recurrent neural network, and combining the first power predicted value and the second power predicted value to calculate a third power predicted value at the current moment and under meteorological conditions. Through the technical scheme in the application, the accuracy of photovoltaic power generation power prediction is improved.
The steps in the present application may be sequentially adjusted, combined, and pruned according to actual requirements.
The units in the device can be combined, divided and pruned according to actual requirements.
Although the present application is disclosed in detail with reference to the accompanying drawings, it is to be understood that such descriptions are merely illustrative and are not intended to limit the application of the present application. The scope of the present application is defined by the appended claims and may include various modifications, alterations, and equivalents to the invention without departing from the scope and spirit of the application.
Claims (2)
1. A photovoltaic power generation short-term power prediction method based on multi-model fusion is characterized by comprising the following steps:
step 1, weather historical data and corresponding photovoltaic power generation historical data of a predicted photovoltaic power generation area are obtained, and a first power predicted value corresponding to the current weather data is calculated according to the collected current weather data, the weather historical data and the photovoltaic power generation historical data by utilizing a local weighted linear regression model; in step 1, specifically, the method includes:
step 11, dividing the meteorological historical data and the photovoltaic power generation capacity historical data into a training set and a testing set;
step 12, calculating correlation coefficients between the meteorological historical data and photovoltaic power generation capacity historical data in the training set in sequence, constructing the local weighted linear regression model, and calculating predicted power generation capacity data corresponding to the meteorological historical data in the test set;
step 13, correcting the local weighted linear regression model according to the predicted power generation amount data and the photovoltaic power generation amount historical data in the test set by combining a loss function, and judging that the local weighted linear regression model is corrected when the loss function converges; in step 13, the calculation formula of the loss function is:
wherein m is the number of data in the test set, and x (i) For the predicted power generation amount data corresponding to the ith weather history data in the test set, y (i) Photovoltaic power generation amount historical data, w, corresponding to the ith weather historical data i (x) For the ith predicted power generation amount data x (i) Is used for the weight value of (a),
for the ith predicted power generation amount data x (i) The j-th adjacent point in the preset range, k is a weight change parameter, theta (i) For the corresponding matrix of correlation coefficients,
wherein, the value of each element in the correlation coefficient matrix is determined by the value of the correlation coefficient, if the value of the correlation coefficient is [ -0.5,0.5], the value of the corresponding element is 0, otherwise, the value is 1;
step 14, carrying out power prediction on the collected current meteorological data according to the corrected local weighted linear regression model, and recording a prediction result as the first power prediction value;
step 2, calculating a second power predicted value corresponding to the current moment by adopting a recurrent neural network according to the photovoltaic power generation historical data and the corresponding power generation time;
step 3, under the condition that the fusion error value is minimum, calculating a third power predicted value under the current moment and weather conditions by combining the first power predicted value and the second power predicted value, wherein under the condition that the fusion error value is minimum, calculating the first fusion weight and the second fusion weight corresponding to the local weighted linear regression model and the recurrent neural network, specifically comprising:
step 31, calculating a first historical power predicted value corresponding to the weather history data according to the local weighted linear regression model, and calculating a second historical power predicted value corresponding to the power generation time according to the recurrent neural network;
step 32, calculating the first fusion weight and the second fusion weight according to the first historical power predicted value, the second historical power predicted value and the photovoltaic power generation amount historical data by adopting a fusion error calculation formula, wherein the fusion error calculation formula is as follows:
p c (n) ＝w 1 p 1 (n) +w 2 p 2 (n)
wherein p is t (n) For the nth photovoltaic power generation amount history data, p 1 (n) For the corresponding first historical power predicted value, p 2 (n) For the corresponding second historical power predicted value, w 1 For the first fusion weight, w 2 For the second fusion weight, p c (n) E is the n-th historical power predicted value after fusion 1 E, for the error value corresponding to the first historical power predicted value 2 For the error value corresponding to the second historical power prediction value, var (·) is variance and cov (·) is covariance.
2. The photovoltaic power generation short-term power prediction method based on multi-model fusion according to claim 1, wherein the weather history data includes: irradiance, visibility, ambient temperature, dew point temperature, ambient humidity, wind speed, barometric pressure, altitude.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010541414.0A CN111695736B (en) | 2020-06-15 | 2020-06-15 | Photovoltaic power generation short-term power prediction method based on multi-model fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010541414.0A CN111695736B (en) | 2020-06-15 | 2020-06-15 | Photovoltaic power generation short-term power prediction method based on multi-model fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111695736A CN111695736A (en) | 2020-09-22 |
| CN111695736B true CN111695736B (en) | 2023-04-21 |
Family
ID=72480925
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010541414.0A Active CN111695736B (en) | 2020-06-15 | 2020-06-15 | Photovoltaic power generation short-term power prediction method based on multi-model fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111695736B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112365040B (en) * | 2020-11-03 | 2022-03-01 | 哈尔滨工业大学 | Short-term wind power prediction method based on multi-channel convolution neural network and time convolution network |
| CN112418559A (en) * | 2020-12-09 | 2021-02-26 | 贵州优策网络科技有限公司 | User selection behavior prediction method and device |
| CN114493050B (en) * | 2022-04-08 | 2022-07-29 | 南方电网数字电网研究院有限公司 | Multi-dimensional fusion new energy power parallel prediction method and device |
| CN116011686B (en) * | 2023-03-28 | 2023-06-20 | 山东创宇环保科技有限公司 | Charging shed photovoltaic power generation reserve prediction method based on multi-data fusion |
| CN116701888B (en) * | 2023-08-09 | 2023-10-17 | 国网浙江省电力有限公司丽水供电公司 | Auxiliary model data processing method and system for clean energy enterprises |
| CN117350897A (en) * | 2023-12-04 | 2024-01-05 | 旗星(天津)科技有限公司 | Method and system for predicting conversion power by considering air humidity |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106529719A (en) * | 2016-11-08 | 2017-03-22 | 中国地质大学（武汉） | Method of predicting wind power of wind speed fusion based on particle swarm optimization algorithm |
| CN106529718A (en) * | 2016-11-04 | 2017-03-22 | 贵州电网有限责任公司电力科学研究院 | Non-runoff small hydropower station short term power weighted prediction method |
| CN108196317A (en) * | 2018-02-06 | 2018-06-22 | 南京邮电大学 | A kind of Predictive meteorological methods for micro-grid system |
| CN110705771A (en) * | 2019-09-26 | 2020-01-17 | 国家电网公司华北分部 | Method and device for predicting and optimizing power generation power of new energy of regional power grid |
| CN111191854A (en) * | 2020-01-10 | 2020-05-22 | 上海积成能源科技有限公司 | Photovoltaic power generation prediction model and method based on linear regression and neural network |
Family Cites Families (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102102626B (en) * | 2011-01-30 | 2012-08-29 | 华北电力大学 | Method for forecasting short-term power in wind power station |
| CN103743401A (en) * | 2013-12-31 | 2014-04-23 | 电子科技大学 | Asynchronous fusion method based on multi-model flight path quality |
| CN104809532A (en) * | 2015-05-25 | 2015-07-29 | 海南汉能薄膜太阳能有限公司 | Method for predicting generating capacity of photovoltaic system |
| US10389136B2 (en) * | 2015-10-08 | 2019-08-20 | Con Edison Battery Storage, Llc | Photovoltaic energy system with value function optimization |
| US10734811B2 (en) * | 2017-11-27 | 2020-08-04 | Ihi Inc. | System and method for optimal control of energy storage system |
| CN109325398B (en) * | 2018-06-30 | 2020-10-09 | 东南大学 | Human face attribute analysis method based on transfer learning |
| CN110766134A (en) * | 2019-09-25 | 2020-02-07 | 福州大学 | Photovoltaic power station short-term power prediction method based on cyclic neural network |
| CN111222674A (en) * | 2019-10-08 | 2020-06-02 | 南昌大学 | Short-term photovoltaic power generation capacity prediction method based on long-term and short-term memory neural network |
| CN111145245B (en) * | 2019-12-31 | 2023-07-21 | 哈尔滨工业大学（深圳） | Short-time adjacent rainfall forecasting method, system and computer-readable storage medium |
| CN111210095B (en) * | 2020-03-12 | 2022-02-18 | 深圳前海微众银行股份有限公司 | Power generation amount prediction method, device, equipment and computer readable storage medium |
-
2020
- 2020-06-15 CN CN202010541414.0A patent/CN111695736B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106529718A (en) * | 2016-11-04 | 2017-03-22 | 贵州电网有限责任公司电力科学研究院 | Non-runoff small hydropower station short term power weighted prediction method |
| CN106529719A (en) * | 2016-11-08 | 2017-03-22 | 中国地质大学（武汉） | Method of predicting wind power of wind speed fusion based on particle swarm optimization algorithm |
| CN108196317A (en) * | 2018-02-06 | 2018-06-22 | 南京邮电大学 | A kind of Predictive meteorological methods for micro-grid system |
| CN110705771A (en) * | 2019-09-26 | 2020-01-17 | 国家电网公司华北分部 | Method and device for predicting and optimizing power generation power of new energy of regional power grid |
| CN111191854A (en) * | 2020-01-10 | 2020-05-22 | 上海积成能源科技有限公司 | Photovoltaic power generation prediction model and method based on linear regression and neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111695736A (en) | 2020-09-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111695736B (en) | Photovoltaic power generation short-term power prediction method based on multi-model fusion | |
| CN111008728B (en) | Prediction method for short-term output of distributed photovoltaic power generation system | |
| CN112257941B (en) | Photovoltaic power station short-term power prediction method based on improved Bi-LSTM | |
| CN110909919A (en) | Photovoltaic power prediction method of depth neural network model with attention mechanism fused | |
| CN111260126B (en) | Short-term photovoltaic power generation prediction method considering correlation degree of weather and meteorological factors | |
| CN111488896B (en) | Distribution line time-varying fault probability calculation method based on multi-source data mining | |
| CN107403015B (en) | Short-term optical power prediction method based on time series similarity | |
| CN109840633B (en) | Photovoltaic output power prediction method, system and storage medium | |
| CN113537582B (en) | Photovoltaic power ultra-short-term prediction method based on short-wave radiation correction | |
| CN114792156A (en) | Photovoltaic output power prediction method and system based on curve characteristic index clustering | |
| CN115796004A (en) | Photovoltaic power station ultra-short term power intelligent prediction method based on SLSTM and MLSTNet models | |
| CN113516271A (en) | Wind power cluster power day-ahead prediction method based on space-time neural network | |
| CN114399081A (en) | Photovoltaic power generation power prediction method based on weather classification | |
| CN115759467A (en) | Time-division integrated learning photovoltaic prediction method for error correction | |
| CN113449847B (en) | Offshore wind power rolling prediction method considering second-level time series wind speed change | |
| CN111815051B (en) | GRNN photovoltaic power generation prediction method considering weather influence factors | |
| CN113991711A (en) | Capacity configuration method for energy storage system of photovoltaic power station | |
| KR20190071174A (en) | Method and system for short-term wind speed prediction based on pressure data | |
| CN115544895B (en) | Photovoltaic power station annual output guarantee rate model optimization method | |
| Sun et al. | Short-term photovoltaic power prediction modeling based on AdaBoost algorithm and Elman | |
| CN116484998A (en) | Distributed photovoltaic power station power prediction method and system based on meteorological similar day | |
| CN109447843A (en) | A kind of photovoltaic power generation output forecasting method and device | |
| CN111967660B (en) | Ultra-short-term photovoltaic prediction residual error correction method based on SVR | |
| CN112949936A (en) | Short-term photovoltaic power prediction method based on similar-day wavelet transform and multilayer perceptron | |
| KR20180023078A (en) | Prediction method of generation quantity in solar energy generation using weather information |
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 |