CN110909919A - Photovoltaic power prediction method of depth neural network model with attention mechanism fused - Google Patents
Photovoltaic power prediction method of depth neural network model with attention mechanism fused Download PDFInfo
- Publication number
- CN110909919A CN110909919A CN201911079279.6A CN201911079279A CN110909919A CN 110909919 A CN110909919 A CN 110909919A CN 201911079279 A CN201911079279 A CN 201911079279A CN 110909919 A CN110909919 A CN 110909919A
- Authority
- CN
- China
- Prior art keywords
- model
- neural network
- prediction
- data
- attention mechanism
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000003062 neural network model Methods 0.000 title claims abstract description 20
- 238000013528 artificial neural network Methods 0.000 claims abstract description 31
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 24
- 238000004364 calculation method Methods 0.000 claims abstract description 10
- 230000007787 long-term memory Effects 0.000 claims abstract description 8
- 230000006403 short-term memory Effects 0.000 claims abstract description 8
- 238000012549 training Methods 0.000 claims description 16
- 230000015654 memory Effects 0.000 claims description 11
- 238000002474 experimental method Methods 0.000 claims description 7
- 230000008034 disappearance Effects 0.000 claims description 5
- 238000004880 explosion Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 230000000306 recurrent effect Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000012795 verification Methods 0.000 claims description 3
- 210000002569 neuron Anatomy 0.000 claims description 2
- 238000000605 extraction Methods 0.000 abstract description 5
- 238000013136 deep learning model Methods 0.000 abstract description 2
- 230000009977 dual effect Effects 0.000 abstract description 2
- 238000010248 power generation Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000000053 physical method Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000005855 radiation Effects 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000005684 electric field Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield 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
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Software Systems (AREA)
- Marketing (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Development Economics (AREA)
- Game Theory and Decision Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Photovoltaic Devices (AREA)
Abstract
A photovoltaic power prediction method of a deep neural network model integrating an attention mechanism belongs to the technical field of renewable energy photovoltaic power. The method comprises the steps that firstly, a hybrid neural network based on a long-term and short-term memory neural network and a convolutional neural network is selected as a prediction model according to the characteristics of photovoltaic data, and an optimal connection mode is considered; secondly, in order to reduce the calculation time of the model and more accurately extract high-quality feature information which can be used for photovoltaic prediction, an attention mechanism model is added in the aspect of model feature extraction. The advantages of the proposed hybrid deep learning model are proved through comparison of different prediction models; the selection of high quality features is made possible by the application of attention model. The reasonable mixed model mode can realize the dual pursuit of high prediction precision and low calculation cost.
Description
Technical Field
The invention belongs to the technical field of renewable energy photovoltaic power prediction, and particularly relates to a photovoltaic power prediction method of a deep neural network model fused with an attention mechanism.
Background
Photovoltaic power generation is increasing at a relatively rapid rate every year. Photovoltaic power generation has a natural uncontrollable nature due to the randomness of light and day and night periodicity, which are typical of fluctuating and intermittent power sources. The output of the photovoltaic power generation system is greatly influenced by factors such as weather and climate. These features lead to new challenges for the power system after the high-capacity/high-ratio photovoltaic is connected to the grid, such as increasing the difficulty and complexity of grid dispatching. The prediction of the generated power of the photovoltaic is one of the key basic technologies for improving the running quality of a power system and reducing the reserve capacity.
The photovoltaic power can be classified into direct prediction and indirect prediction according to the prediction process; the prediction time scale classification can be divided into ultra-short term prediction, medium term prediction and long term prediction; the prediction mechanism of the prediction method can be divided into a physical method, a statistical method and a machine learning method. The physical method is that a mathematical model is established according to a photovoltaic power generation principle, data such as solar radiation, temperature, humidity, cloud cover, air pressure, wind speed and the like are obtained by means of numerical weather forecast and the like, and parameters such as a photovoltaic system installation angle, photovoltaic array conversion efficiency, battery conditions and the like are combined to directly calculate to obtain photovoltaic power generation power. The prediction accuracy of the physical method strongly depends on the accuracy degree of numerical weather forecast information, but the bottleneck is met in the aspect of improving the numerical weather forecast accuracy at present. The statistical method is to establish an incidence mapping relation between input data and output data through curve fitting, parameter estimation, incidence relation analysis and other means on processed historical data such as solar radiation degree, photovoltaic power generation output and the like, so as to realize prediction on future photovoltaic power generation output. However, the implementation of the statistical prediction method requires a large amount of processed correct historical data, and the implementation process is difficult in terms of data acquisition and processing calculation amount. Machine learning has the ability to efficiently extract high-dimensional complex nonlinear features and to map directly from input to output. However, for the traditional neural network, problems of overfitting, gradient disappearance, explosion and the like of network training are easily caused by excessive input data, hidden layer number and node number of each hidden layer. The deep neural network has higher feature extraction capability than the common neural network, and can remarkably improve the problem of gradient disappearance of the original neural network. To date, some success has been achieved in predicting wind speed, wind power, load, solar irradiance, etc. based on deep learning predictive models. The attention mechanism is similar to the human selective visual attention mechanism in nature, and the core goal is to select the more critical information for the current task goal from a plurality of information.
Therefore, the photovoltaic electric field power is modeled and predicted by using a deep learning algorithm, and the depth features extracted by the neural network model are subjected to weighted summation by using an attention mechanism, so that feature information with higher weight of a prediction result, namely the selection of high-quality features is obtained. The method has important significance for improving the accuracy and stability of the photovoltaic power prediction model, can reduce the interference of useless information on the model, reduces the calculation time, and provides a new idea for predicting the renewable energy power.
Disclosure of Invention
The invention aims to provide a photovoltaic power prediction method of a depth neural network model fused with an attention mechanism, which comprises the steps of firstly selecting a hybrid neural network based on a long-short term memory neural network and a convolutional neural network as a prediction model according to the characteristics of photovoltaic data, and considering the optimal connection mode; secondly, in order to reduce the calculation time of the model and more accurately extract high-quality feature information which can be used for photovoltaic prediction, an attention mechanism model is added in the aspect of model feature extraction.
The purpose of the invention is realized as follows:
the photovoltaic power prediction method of the depth neural network model with the attention mechanism fused comprises the following steps:
the method comprises the following steps: downloading numerical weather forecast and solar irradiance data including wind speed, wind direction, temperature, humidity, global irradiance, scattering irradiance and other data and corresponding historical power data of a photovoltaic field through a website;
step two: processing the acquired historical data, including removing abnormal data, filling missing data, unifying resolution, converting data formats and the like;
step three: selecting a deep neural network model to form a hybrid model and selecting a connection mode of the hybrid model; setting parameters of the deep hybrid neural network, including parameters such as the number of hidden layers, the number of neurons in each layer, the learning rate, the learning step length, the iteration times, the number of convolution kernels, the number of long and short term neural network units and the like;
step four: selecting a proper attention mechanism model and a model placing position on the attention mechanism model;
step five: dividing the preprocessed historical data into a training set, a verification set and a test set according to a certain proportion;
step six: selecting and training a mixed model according to a trial-and-error method to obtain the optimal model parameters for prediction;
step seven: and comparing the prediction result with different models and analyzing the prediction results obtained by the different models to prove the advantages of the proposed models.
Selecting a long-short term memory neural network (LSTM) to extract the time characteristics of the data in the second step; the internal storage unit and the door mechanism of the long-short term memory neural network overcome the problems of gradient disappearance and gradient explosion in the traditional recurrent neural network training; the door mechanism comprises a forgetting door, an input door, an updating door and an output door; and obtaining the prediction results of the long-term and short-term memory neural network model under different historical data lengths through model training and a large number of experiments.
Selecting a convolutional neural network CNN to extract the spatial characteristics of the data in the second step; the convolutional neural network mainly comprises a convolutional layer, a pooling layer, a full-link layer, an output layer and the like; and obtaining the prediction results of the convolutional neural network model under different historical data lengths through model training and a large number of experiments.
In the fourth step, the selection of high-quality features is extracted by means of an attention mechanism model; the most intuitive method for measuring the importance of the features by the power mechanism model is to rely on the weight, that is, the weight of each feature is calculated firstly during each recognition, then the features are weighted and summed, the larger the weight is, the greater the contribution of the feature to the current recognition is, and the formula is as follows:
wherein L isxValue is the element in Source for the length of Source.
The invention has the beneficial effects that:
the advantages of the proposed hybrid deep learning model are proved through comparison of different prediction models; the selection of high quality features is made possible by the application of attention model. The reasonable mixed model mode can realize the dual pursuit of high prediction precision and low calculation cost.
Drawings
FIG. 1 is a system diagram of photovoltaic power prediction;
FIG. 2 is a historical data breakdown diagram;
FIG. 3 is a diagram of a convolutional neural network architecture;
FIG. 4 is a diagram of a long-short term memory neural network model architecture;
fig. 5 is an essential idea of the attention mechanism.
Detailed Description
The invention is further described below with reference to the accompanying drawings. The drawings are only for the purpose of illustrating the principles of the invention and are not to be construed as limiting the scope of the invention.
The invention mainly aims at the field of renewable energy source prediction, and the deep neural network has a very good effect on processing nonlinear data and extracting the depth characteristics of the data. In order to improve the accuracy of a photovoltaic power prediction model, the invention provides a photovoltaic power prediction model of a hybrid deep neural network fused with an attention mechanism. Firstly, selecting a hybrid neural network based on a long-term and short-term memory neural network and a convolutional neural network as a prediction model according to the characteristics of photovoltaic data, and considering an optimal connection mode; secondly, in order to reduce the calculation time of the model and more accurately extract high-quality feature information which can be used for photovoltaic prediction, an attention mechanism model is added in the aspect of model feature extraction. The system architecture for photovoltaic power prediction is shown in fig. 1. The drawings are only for the purpose of illustrating the principles of the invention and are not to be construed as limiting the scope of the invention.
The invention predicts photovoltaic power based on a deep neural network, considers numerical weather forecast data and irradiance data, including wind speed, wind direction, temperature, humidity, global irradiance, scattered irradiance and other factors, relates to model selection and connection of a hybrid model, and attention mechanism characteristic selection, and specifically comprises the following steps:
step 1) downloading complete numerical weather forecast data, irradiance data and corresponding photovoltaic field historical power data suitable for prediction from a website. Preprocessing the downloaded historical data, including rejecting abnormal data; filling missing data by using technologies such as average value and interpolation; normalizing the data by utilizing a normalization technology, and unifying the data resolution by utilizing a linear interpolation technology; conversion of data format, etc.
And 2) dividing the obtained preprocessed historical data into a training set, a verification set and a test set according to a certain proportion.
And 3) selecting a deep neural network model to form a mixed model, and training and verifying the mixed model. The convolutional neural network CNN is selected in the present invention to extract the spatial features of the data. Local connectivity and weight sharing are two main features of convolutional neural networks. The convolutional neural network mainly comprises a convolutional layer, a pooling layer, a full-link layer, an output layer and the like. Fig. 3 is a main structural diagram of a convolutional neural network. And obtaining the prediction results of the convolutional neural network model under different historical data lengths through model training and a large number of experiments.
And 4) extracting the time characteristics of the data mainly by means of a long-term and short-term memory neural network (LSTM). The long-term and short-term memory neural network is a time recursion neural network and is used for learning long-term dependence information; the internal memory unit and the door mechanism thereof overcome the problems of gradient disappearance and gradient explosion in the traditional recurrent neural network training. The door mechanism comprises a forgetting door, an input door, an updating door and an output door. FIG. 4 is a block diagram of a long-short term memory neural network model. And obtaining the prediction results of the long-term and short-term memory neural network model under different historical data lengths through model training and a large number of experiments.
The core calculation formula of the long-short term memory neural network model is as follows:
ft=σ(wf×[ht-1,xt]+bf) (1)
it=σ(wi×[ht-1,xt]+bi) (2)
gt=tanh(wg×[ht-1,xt]+bg) (3)
ct=ft×ct-1+it×gt(4)
ot=σ(wo×[ht-1,xt]+bo) (5)
ht=ot×tanh(ct) (6)
and 5) selecting high-quality features by means of an attention mechanism model to extract. The most intuitive method for the power mechanism model to measure the importance of the features depends on the weight, namely, the weight of each feature is calculated firstly during each recognition, then the features are weighted and summed, and the larger the weight is, the greater the contribution of the feature to the current recognition is. The attention mechanism model is mainly attached under the framework of an Encoder-Decoder, is a very common calculation framework, and the essential idea of the model is shown in fig. 5, and can be written as the following formula:
wherein L isxIs the length of Source. Essentially, the attention mechanism is to perform weighted summation on Value values of elements in Source, and Query and Key are used forA weight coefficient corresponding to Value is calculated.
And 6) the hybrid neural network carries out photovoltaic power prediction through different combination modes of the hybrid convolutional neural network and the long-term and short-term memory neural network. In the invention, the best combination mode and prediction effect of the mixed model are selected by comparing two mixed models (a mixed model of a convolutional neural network + a long-short term memory neural network (CNN + LSTM) for extracting the spatial characteristics of data and then extracting the time characteristics of the data, and a mixed model of a long-short term memory neural network + a convolutional neural network (LSTM + CNN) for extracting the time characteristics of data and then extracting the spatial characteristics of the data). Where the attention mechanism model is intended to consider extraction of high quality depth features placed before the fully connected layer. The optimal network model parameters are obtained through a trial and error method, and a large number of experiments show that the hybrid model (LSTM + CNN) has the optimal prediction effect.
And 7) comparing the prediction results obtained by different models to prove that the selected mixed model has the best prediction effect.
Claims (4)
1. The photovoltaic power prediction method of the depth neural network model fused with the attention mechanism is characterized by comprising the following steps of:
the method comprises the following steps: downloading numerical weather forecast and solar irradiance data including wind speed, wind direction, temperature, humidity, global irradiance, scattering irradiance and other data and corresponding historical power data of a photovoltaic field through a website;
step two: processing the acquired historical data, including removing abnormal data, filling missing data, unifying resolution, converting data formats and the like;
step three: selecting a deep neural network model to form a hybrid model and selecting a connection mode of the hybrid model; setting parameters of the deep hybrid neural network, such as the number of hidden layers, the number of neurons in each layer, the learning rate, the learning step length, the iteration times, the number of convolution kernels, the number of long and short term neural network units and the like;
step four: selecting a proper attention mechanism model and a model placing position on the attention mechanism model;
step five: dividing the preprocessed historical data into a training set, a verification set and a test set according to a certain proportion;
step six: and selecting and training the mixed model according to a trial and error method to obtain the optimal model parameter for prediction, and obtaining a prediction result.
2. The method for predicting photovoltaic power of the deep neural network model with the integrated attention mechanism according to claim 1, wherein: selecting a long-short term memory neural network (LSTM) to extract the time characteristics of the data in the second step; the internal storage unit and the door mechanism of the long-short term memory neural network overcome the problems of gradient disappearance and gradient explosion in the traditional recurrent neural network training; the door mechanism comprises a forgetting door, an input door, an updating door and an output door; and obtaining the prediction results of the long-term and short-term memory neural network model under different historical data lengths through model training and a large number of experiments.
3. The method for predicting photovoltaic power of the deep neural network model with the integrated attention mechanism according to claim 1, wherein: selecting a convolutional neural network CNN to extract the spatial characteristics of the data in the second step; the convolutional neural network mainly comprises a convolutional layer, a pooling layer, a full-link layer, an output layer and the like; and obtaining the prediction results of the convolutional neural network model under different historical data lengths through model training and a large number of experiments.
4. The method for predicting photovoltaic power of the deep neural network model with the integrated attention mechanism according to claim 1, wherein: in the fourth step, the selection of high-quality features is extracted by means of an attention mechanism model; the most intuitive method for measuring the importance of the features by the attention mechanism model is to rely on the weight, namely, the weight of each feature is calculated firstly during each recognition, then the features are weighted and summed, the larger the weight is, the greater the contribution of the feature to the current recognition is, and the calculation formula is
Wherein L isxIs the length of Source. Value is an element in Source.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911079279.6A CN110909919A (en) | 2019-11-07 | 2019-11-07 | Photovoltaic power prediction method of depth neural network model with attention mechanism fused |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911079279.6A CN110909919A (en) | 2019-11-07 | 2019-11-07 | Photovoltaic power prediction method of depth neural network model with attention mechanism fused |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110909919A true CN110909919A (en) | 2020-03-24 |
Family
ID=69816439
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911079279.6A Pending CN110909919A (en) | 2019-11-07 | 2019-11-07 | Photovoltaic power prediction method of depth neural network model with attention mechanism fused |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110909919A (en) |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111461455A (en) * | 2020-04-16 | 2020-07-28 | 武汉大学 | Behavior prediction method based on correlation period attention mechanism |
CN111563611A (en) * | 2020-04-13 | 2020-08-21 | 北京工业大学 | Cloud data center renewable energy space-time prediction method for orientation graph convolutional network |
CN111680454A (en) * | 2020-06-16 | 2020-09-18 | 北京工业大学 | Fan blade icing fault prediction method based on double attention mechanism |
CN112183886A (en) * | 2020-10-22 | 2021-01-05 | 南京信息工程大学 | Short-time approaching rainfall prediction method based on convolutional network and attention mechanism |
CN112215478A (en) * | 2020-09-27 | 2021-01-12 | 珠海博威电气股份有限公司 | Power coordination control method and device for optical storage power station and storage medium |
CN112364477A (en) * | 2020-09-29 | 2021-02-12 | 中国电器科学研究院股份有限公司 | Outdoor empirical prediction model library generation method and system |
CN112434891A (en) * | 2020-12-18 | 2021-03-02 | 中国计量大学上虞高等研究院有限公司 | Method for predicting solar irradiance time sequence based on WCNN-ALSTM |
CN113011674A (en) * | 2021-03-31 | 2021-06-22 | 电子科技大学中山学院 | Photovoltaic power generation prediction method and device, electronic equipment and storage medium |
CN113205223A (en) * | 2021-05-17 | 2021-08-03 | 上海电力大学 | Electric quantity prediction system and prediction method thereof |
CN113240217A (en) * | 2021-07-12 | 2021-08-10 | 国网江西省电力有限公司电力科学研究院 | Photovoltaic power generation prediction method and device based on integrated prediction model |
CN113408785A (en) * | 2021-05-20 | 2021-09-17 | 上海晨翘智能科技有限公司 | Method, device, equipment and storage medium for predicting optical power |
CN113497445A (en) * | 2021-09-08 | 2021-10-12 | 国网江西省电力有限公司电力科学研究院 | Combined prediction method and system for output of regional multi-scale new energy power station |
CN113689045A (en) * | 2021-08-27 | 2021-11-23 | 山东浪潮科学研究院有限公司 | Method, device and medium for predicting grid-connected electric quantity of photovoltaic region |
CN114819382A (en) * | 2022-05-11 | 2022-07-29 | 湘潭大学 | Photovoltaic power prediction method based on LSTM |
CN115330553A (en) * | 2022-06-22 | 2022-11-11 | 四川大学 | Equipment feature based multi-layer optimal selection non-intrusive load decomposition method |
CN115564100A (en) * | 2022-09-22 | 2023-01-03 | 济南大学 | Photovoltaic power prediction method, system and equipment |
CN116384593A (en) * | 2023-06-01 | 2023-07-04 | 深圳市国电科技通信有限公司 | Distributed photovoltaic output prediction method and device, electronic equipment and medium |
CN117557304A (en) * | 2024-01-11 | 2024-02-13 | 国网浙江省电力有限公司 | Electric quantity and electricity price level fusion prediction method based on modal decomposition and neural network |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW201835788A (en) * | 2017-02-24 | 2018-10-01 | 美商高通公司 | Method and apparatus for multi-dimensional sequence prediction |
CN109284870A (en) * | 2018-10-08 | 2019-01-29 | 南昌大学 | Short-term method for forecasting photovoltaic power generation quantity based on shot and long term Memory Neural Networks |
CN110059878A (en) * | 2019-04-15 | 2019-07-26 | 中国计量大学 | Based on CNN LSTM photovoltaic power generation power prediction model and its construction method |
-
2019
- 2019-11-07 CN CN201911079279.6A patent/CN110909919A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TW201835788A (en) * | 2017-02-24 | 2018-10-01 | 美商高通公司 | Method and apparatus for multi-dimensional sequence prediction |
CN109284870A (en) * | 2018-10-08 | 2019-01-29 | 南昌大学 | Short-term method for forecasting photovoltaic power generation quantity based on shot and long term Memory Neural Networks |
CN110059878A (en) * | 2019-04-15 | 2019-07-26 | 中国计量大学 | Based on CNN LSTM photovoltaic power generation power prediction model and its construction method |
Cited By (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111563611A (en) * | 2020-04-13 | 2020-08-21 | 北京工业大学 | Cloud data center renewable energy space-time prediction method for orientation graph convolutional network |
CN111563611B (en) * | 2020-04-13 | 2023-11-24 | 北京工业大学 | Cloud data center renewable energy space-time prediction method for graph rolling network |
CN111461455A (en) * | 2020-04-16 | 2020-07-28 | 武汉大学 | Behavior prediction method based on correlation period attention mechanism |
CN111461455B (en) * | 2020-04-16 | 2022-05-13 | 武汉大学 | Behavior prediction method based on association cycle attention mechanism |
CN111680454A (en) * | 2020-06-16 | 2020-09-18 | 北京工业大学 | Fan blade icing fault prediction method based on double attention mechanism |
CN112215478A (en) * | 2020-09-27 | 2021-01-12 | 珠海博威电气股份有限公司 | Power coordination control method and device for optical storage power station and storage medium |
CN112215478B (en) * | 2020-09-27 | 2024-05-28 | 珠海博威电气股份有限公司 | Power coordination control method and device for optical storage station and storage medium |
CN112364477A (en) * | 2020-09-29 | 2021-02-12 | 中国电器科学研究院股份有限公司 | Outdoor empirical prediction model library generation method and system |
CN112364477B (en) * | 2020-09-29 | 2022-12-06 | 中国电器科学研究院股份有限公司 | Outdoor empirical prediction model library generation method and system |
CN112183886B (en) * | 2020-10-22 | 2024-03-15 | 南京信息工程大学 | Short-time adjacent rainfall prediction method based on convolution network and attention mechanism |
CN112183886A (en) * | 2020-10-22 | 2021-01-05 | 南京信息工程大学 | Short-time approaching rainfall prediction method based on convolutional network and attention mechanism |
CN112434891A (en) * | 2020-12-18 | 2021-03-02 | 中国计量大学上虞高等研究院有限公司 | Method for predicting solar irradiance time sequence based on WCNN-ALSTM |
CN113011674A (en) * | 2021-03-31 | 2021-06-22 | 电子科技大学中山学院 | Photovoltaic power generation prediction method and device, electronic equipment and storage medium |
CN113205223A (en) * | 2021-05-17 | 2021-08-03 | 上海电力大学 | Electric quantity prediction system and prediction method thereof |
CN113408785A (en) * | 2021-05-20 | 2021-09-17 | 上海晨翘智能科技有限公司 | Method, device, equipment and storage medium for predicting optical power |
CN113240217A (en) * | 2021-07-12 | 2021-08-10 | 国网江西省电力有限公司电力科学研究院 | Photovoltaic power generation prediction method and device based on integrated prediction model |
CN113689045A (en) * | 2021-08-27 | 2021-11-23 | 山东浪潮科学研究院有限公司 | Method, device and medium for predicting grid-connected electric quantity of photovoltaic region |
CN113689045B (en) * | 2021-08-27 | 2023-06-27 | 山东浪潮科学研究院有限公司 | Photovoltaic area grid-connected electric quantity prediction method, device and medium |
CN113497445A (en) * | 2021-09-08 | 2021-10-12 | 国网江西省电力有限公司电力科学研究院 | Combined prediction method and system for output of regional multi-scale new energy power station |
CN114819382A (en) * | 2022-05-11 | 2022-07-29 | 湘潭大学 | Photovoltaic power prediction method based on LSTM |
CN114819382B (en) * | 2022-05-11 | 2024-05-24 | 湘潭大学 | LSTM-based photovoltaic power prediction method |
CN115330553A (en) * | 2022-06-22 | 2022-11-11 | 四川大学 | Equipment feature based multi-layer optimal selection non-intrusive load decomposition method |
CN115564100A (en) * | 2022-09-22 | 2023-01-03 | 济南大学 | Photovoltaic power prediction method, system and equipment |
CN116384593A (en) * | 2023-06-01 | 2023-07-04 | 深圳市国电科技通信有限公司 | Distributed photovoltaic output prediction method and device, electronic equipment and medium |
CN116384593B (en) * | 2023-06-01 | 2023-08-18 | 深圳市国电科技通信有限公司 | Distributed photovoltaic output prediction method and device, electronic equipment and medium |
CN117557304A (en) * | 2024-01-11 | 2024-02-13 | 国网浙江省电力有限公司 | Electric quantity and electricity price level fusion prediction method based on modal decomposition and neural network |
CN117557304B (en) * | 2024-01-11 | 2024-03-29 | 国网浙江省电力有限公司 | Electric quantity and electricity price level fusion prediction method based on modal decomposition and neural network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110909919A (en) | Photovoltaic power prediction method of depth neural network model with attention mechanism fused | |
Li et al. | Photovoltaic power forecasting with a hybrid deep learning approach | |
CN107766990B (en) | Method for predicting power generation power of photovoltaic power station | |
Liu et al. | An improved photovoltaic power forecasting model with the assistance of aerosol index data | |
CN109002915B (en) | Photovoltaic power station short-term power prediction method based on Kmeans-GRA-Elman model | |
CN112561058B (en) | Short-term photovoltaic power prediction method based on Stacking-integrated learning | |
CN111753893A (en) | Wind turbine generator power cluster prediction method based on clustering and deep learning | |
CN114792156B (en) | Photovoltaic output power prediction method and system based on curve characteristic index clustering | |
CN112215428B (en) | Photovoltaic power generation power prediction method and system based on error correction and fuzzy logic | |
CN115688982B (en) | Building photovoltaic data complement method based on WGAN and whale optimization algorithm | |
CN116307291B (en) | Distributed photovoltaic power generation prediction method and prediction terminal based on wavelet decomposition | |
CN114399081A (en) | Photovoltaic power generation power prediction method based on weather classification | |
CN113052469A (en) | Method for calculating wind-solar-water-load complementary characteristic of small hydropower area lacking measurement runoff | |
CN115829126A (en) | Photovoltaic power generation power prediction method based on multi-view self-adaptive feature fusion | |
CN116629416A (en) | Photovoltaic power station power prediction method and device | |
CN115759389A (en) | Day-ahead photovoltaic power prediction method based on weather type similar day combination strategy | |
CN116632820A (en) | Wind power prediction method based on sequence extension VMD and wind speed fluctuation characteristics | |
CN115186923A (en) | Photovoltaic power generation power prediction method and device and electronic equipment | |
CN110852492A (en) | Photovoltaic power ultra-short-term prediction method for finding similarity based on Mahalanobis distance | |
CN113837434A (en) | Solar photovoltaic power generation prediction method and device, electronic equipment and storage medium | |
Alharbi et al. | Short-term wind speed and temperature forecasting model based on gated recurrent unit neural networks | |
CN110555566B (en) | B-spline quantile regression-based photoelectric probability density prediction method | |
CN117060407A (en) | Wind power cluster power prediction method and system based on similar day division | |
CN112734073A (en) | Photovoltaic power generation short-term prediction method based on long and short-term memory network | |
CN116845875A (en) | WOA-BP-based short-term photovoltaic output prediction method and device |
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: 20200324 |
|
RJ01 | Rejection of invention patent application after publication |