CN110516844A - Multivariable based on EMD-PCA-LSTM inputs photovoltaic power forecasting method - Google Patents
Multivariable based on EMD-PCA-LSTM inputs photovoltaic power forecasting method Download PDFInfo
- Publication number
- CN110516844A CN110516844A CN201910675922.5A CN201910675922A CN110516844A CN 110516844 A CN110516844 A CN 110516844A CN 201910675922 A CN201910675922 A CN 201910675922A CN 110516844 A CN110516844 A CN 110516844A
- Authority
- CN
- China
- Prior art keywords
- data
- emd
- photovoltaic
- photovoltaic power
- prediction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013277 forecasting method Methods 0.000 title claims abstract description 13
- 238000000034 method Methods 0.000 claims abstract description 16
- 238000004458 analytical method Methods 0.000 claims abstract description 8
- 238000013528 artificial neural network Methods 0.000 claims abstract description 6
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 6
- 238000012549 training Methods 0.000 claims description 32
- 238000012360 testing method Methods 0.000 claims description 11
- 230000007613 environmental effect Effects 0.000 claims description 9
- 238000005259 measurement Methods 0.000 claims description 7
- 238000004519 manufacturing process Methods 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims description 3
- 238000004321 preservation Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 2
- 230000000007 visual effect Effects 0.000 abstract 1
- 238000002474 experimental method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000002354 daily effect Effects 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
Classifications
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—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
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computing Systems (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Primary Health Care (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (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)
Abstract
The invention discloses a kind of, and the multivariable based on EMD-PCA-LSTM inputs photovoltaic power forecasting method, 5 kinds of environment sequences are decomposed using empirical mode decomposition method, intrinsic mode decomposition and the residual components under different time scales are obtained, environment sequence is decomposed into a variety of different volatility series；The key factor for influencing photovoltaic output power is filtered out using principal component analytical method, reduces the dimension of mode input parameter, eliminates redundancy and correlation by the EMD different volatility series decomposed.Finally, completing to model the dynamic time between Multivariate Time Series and photovoltaic power sequence by LSTM neural network, prediction model is constructed, realizes the prediction to photovoltaic output power.This method demonstrates LSTM model in the practicability in photovoltaic prediction field, extend the application category of depth learning technology, economical operation and scheduling for further investigated photovoltaic parallel in system provide a kind of new visual angle, have a good application prospect in reality and engineering application value.
Description
Technical field
The present invention relates to technical field of data prediction, input more particularly, to a kind of multivariable based on EMD-PCA-LSTM
Photovoltaic power prediction technique.
Background technique
Energy crisis and environmental pollution are two hang-ups that the world today faces, and photovoltaic power generation becomes at present meets mankind's use
One of important channel of electricity demanding.Due to being influenced by natural environment and climate condition, the fluctuation and randomness pair of photovoltaic power
Large-scale photovoltaic, which generates electricity by way of merging two or more grid systems, will cause certain influence.In order to guarantee the normal operation of photovoltaic plant and the safety tune of power grid
Degree accurately and timely carries out photovoltaic power prediction and has very important significance.
Summary of the invention
In order to solve the above technical problems, the present invention provides a kind of, the multivariable based on EMD-PCA-LSTM inputs photovoltaic function
Rate prediction technique, comprising steps of
Step 1: obtaining the measured power time series data of the photovoltaic power in photovoltaic plant actual production under inverter, with
And solar irradiance, the relative humidity, air themperature, component temperature, atmospheric pressure of the corresponding environment detector acquisition of photovoltaic array area
5 kinds of environment sequence data of power, the actual measurement sample data set of composition photovoltaic power prediction；
Step 2: data cleansing and down-sampled processing are carried out to the actual measurement sample data set of photovoltaic power prediction；
Step 3: empirical mode decomposition method is utilized, 5 kinds of environment sequence data are decomposed；
Step 4: utilizing principal component analytical method, carries out dimensionality reduction to the characteristic sequence set that environment sequence data are decomposed；
Step 5: the data of characteristic sequence set and photovoltaic output power construction suitable for LSTM network training after dimensionality reduction are utilized
Collection, is divided into training set and test set according to the ratio of 7:3, and training set is inputted in LSMT network and is trained；
Step 6: after model training, training pattern is saved, test set is input in training pattern and is predicted, is passed through
Truthful data and prediction data are compared, the evaluation index of prediction: RMSE, MAE, R is found out2, carry out experimental result and summarize and analyze.
Wherein, the step of data cleansing being carried out to sampled data are as follows: as unit of day, function present in Rejection of samples data
The data that rate is 0 and environmental data is 0；Data acquisition time period is set as early 6:00- evening 19:00, and the sampling interval is
10min, the sampled point of every day data are 79；The variable for including in every group of data has the solar irradiance of 6:00-19:00, sky
5 kinds of temperature degree, component temperature, relative humidity, atmospheric pressure environment sequence data.
Wherein, primal environment sequence data is decomposed using EMD algorithm, obtains IMF points of every kind of environmental factor number
Amount and residual components, to obtain the local feature of beginning environment sequence；IMF points that each environment sequence progress EMD is decomposed
Amount and residual components, are summarized, and obtain the characteristic sequence of total 68 dimension as new characteristic sequence set.
Wherein, principal component analytical method is that initial data is transformed into new feature space by linear transformation, with this
It extracts the main linear component of data, removes noise present in characteristic sequence data, reduce the superfluous of characteristic sequence data
Remaining property and correlation choose the principal component that contribution rate of accumulative total is greater than 95%, as new input to obtained characteristic sequence data
Variable.
It wherein, further include determining to establish for light before training set is inputted in LSMT network the step of being trained
Lie prostrate output power prediction LSTM neural network forecast it needs to be determined that model parameter the step of:
Using the environmental characteristic sequence and photovoltaic power historical data at t-1 moment, the photovoltaic power historical data of t moment is carried out
Prediction；Wherein, mode input layer time step number is 1, and input layer dimension is 9, and hiding number of layers is 1, and hidden layer unit number is 50
A, output layer variable is that number is 1, and training batch is 10, and training round is 100 times；
After model training, preservation model file, and test set data are tested, by experimental result to model
It is verified and is optimized repeatedly, improve the precision of prediction of model.
The invention proposes a kind of, and the multivariable based on EMD-PCA-LSTM inputs photovoltaic power forecasting method, for photovoltaic
Generated output has the characteristics that unstability and apparent interval fluctuate, first with empirical mode decomposition method by 5 kinds of environment
Sequence is decomposed, and is obtained intrinsic modal components and the residual components under different time scales, environment sequence is decomposed into various
Different characteristic fluctuation sequences；The key factor for influencing photovoltaic output power is filtered out using principal component analytical method, reduces mould
Type inputs the dimension of parameter, eliminates redundancy and correlation by the EMD different volatility series decomposed；Pass through LSTM nerve
Network is completed to model the dynamic time between Multivariate Time Series and photovoltaic power sequence, constructs prediction model, final real
Now to the prediction of photovoltaic output power.By means of the invention it is possible to guarantee the normal operation of photovoltaic plant and the sacurity dispatching of power grid,
Accurately and timely carry out photovoltaic power prediction.
Detailed description of the invention
Fig. 1 is the stream that a kind of multivariable based on EMD-PCA-LSTM provided by the invention inputs photovoltaic power forecasting method
Journey schematic diagram.
Fig. 2 is that a kind of multivariable based on EMD-PCA-LSTM provided by the invention inputs patrolling for photovoltaic power forecasting method
Collect schematic diagram.
Specific embodiment
Further more detailed description is made to technical solution of the present invention With reference to embodiment.Obviously, it is retouched
The embodiment stated is only a part of the embodiments of the present invention, instead of all the embodiments.Based on the embodiments of the present invention,
Those of ordinary skill in the art's every other embodiment obtained without making creative work, all should belong to
The scope of protection of the invention.
Refering to fig. 1, the present invention provides a kind of, and the multivariable based on EMD-PCA-LSTM inputs photovoltaic power forecasting method,
Comprising steps of
Step 1: obtaining the measured power time series data of the photovoltaic power in photovoltaic plant actual production under inverter, with
And solar irradiance, the relative humidity, air themperature, component temperature, atmospheric pressure of the corresponding environment detector acquisition of photovoltaic array area
5 kinds of environment sequence data of power, the actual measurement sample data set of composition photovoltaic power prediction；
Step 2: data cleansing and down-sampled processing are carried out to the actual measurement sample data set of photovoltaic power prediction；
Step 3: empirical mode decomposition method is utilized, 5 kinds of environment sequence data are decomposed；
Step 4: utilizing principal component analytical method, carries out dimensionality reduction to the characteristic sequence set that environment sequence data are decomposed；
Step 5: the data of characteristic sequence set and photovoltaic output power construction suitable for LSTM network training after dimensionality reduction are utilized
Collection, is divided into training set and test set according to the ratio of 7:3, and training set is inputted in LSMT network and is trained；
Step 6: after model training, training pattern is saved, test set is input in training pattern and is predicted, is passed through
Truthful data and prediction data are compared, the evaluation index of prediction: RMSE, MAE, R is found out2, carry out experimental result and summarize and analyze.
In the actual measurement sample data set making step of photovoltaic power prediction, measured data includes photovoltaic plant actual production
The measured power time series data and the corresponding environment detector of photovoltaic array area of photovoltaic power under middle inverter obtain
5 kinds of solar irradiance, relative humidity, air themperature, component temperature, the atmospheric pressure environment sequence data arrived.
Fig. 2 is that the detailed process of photovoltaic power prediction technique is illustrated, and explains the process of entire method.Due to photovoltaic plant
There are the reasons such as communication appliance fault in actual moving process, carry out data cleansing to sampled data, as unit of day, reject
" bad data " that power present in sample data is 0 and environmental data is 0.Due to effective output period of photovoltaic power
Based on daytime, thus our primary study period be 6:00- in morning evening 19:00, sampling interval 10min, daily
The sampled point of data is 79.The variable for including in every group of data has the solar irradiance, air themperature, component of 6:00-19:00
5 kinds of temperature, relative humidity, atmospheric pressure environment sequence data.
Environment sequence data in experiment sample are non-stationary signal, and are influenced by Changes in weather, are had centainly
Randomness and mutability decompose primal environment sequence data using EMD algorithm, obtain the IMF of every kind of environmental factor number
Component and residual components highlight the local feature of primal environment sequence with this.Each environment sequence progress EMD is decomposed
The IMF component and residual components arrived, is summarized, and the characteristic sequence of available total 68 dimension is as new characteristic sequence collection
It closes.
Principal component analytical method is a kind of Method of Data with Adding Windows of classics, is transformed into initial data newly by linear transformation
Feature space in, the main linear component of data is extracted with this.In order to remove noise present in characteristic sequence data,
The redundancy and correlation for reducing characteristic sequence data carry out principal component analysis to obtained characteristic sequence data.It chooses accumulative
Contribution rate is greater than 95% principal component, as new input variable.
Determine establish for photovoltaic output power prediction LSTM neural network forecast it needs to be determined that model parameter.Utilize t-1
The environmental characteristic sequence and photovoltaic power historical data at moment, predict the photovoltaic power historical data of t moment.It is wherein defeated
Entering layer time step number is 1, and input layer dimension is 9, and hiding number of layers is 1, and hidden layer unit number is 50, and output layer variable is number
It is 1, training batch is 10, and training round is 100 times.After model training, preservation model file, and to test set number
According to being tested, model is verified and optimized repeatedly by experimental result, improves the precision of prediction of model.
Embodiment 1:
This experimental data comes from Shanxi province Taiyuan city photovoltaic energy company.The experimental data essential information is as follows:
It is tested using energy company's subordinate's photovoltaic plant 2018 3 data to the photovoltaic under inverter on October 14
Verifying.The data set suitable for LSTM network training is converted by 6952 time profile datas that sample data is concentrated, and is pressed
It is divided into training set and test set according to the ratio of 7:3, training process updates weight, model using the optimization of adaptability momentum algorithm for estimating
Arameter optimization use experience tuning is combined with grid search.To compare experiment, the present invention uses BP model, LSTM respectively
Model, XGboost model, EMD-LSTM and EMD-PCA-LSTM model compare experiment, and experiment uses identical experimental ring
Border and the number of iterations.Table one is experimental results.
1 test result of table
| Model name | RMSE | MAE | R2 |
| EMD-PCA-LSTM | 35.18 | 15.51 | 0.9446 |
| EMD-LSTM | 38.10 | 19.74 | 0.9418 |
| LSTM | 40.95 | 16.16 | 0.9328 |
| BP | 40.78 | 18.12 | 0.9334 |
| XGBoost | 44.68 | 22.65 | 0.9123 |
Photovoltaic power forecasting method, EMD-PCA- are inputted based on the multivariable based on EMD-PCA-LSTM the invention proposes a kind of
The precision of prediction and prediction effect of LSTM model are relatively high, and EMD-LSTM model increases with input variable, RMSE and MAE
It is improved to some extent, the precision of prediction of model has certain decline, still, when will be original defeated using principal component analysis
When entering variable and being reduced to 9 input variables, compared with single LSTM model, RMSE and MAE are respectively increased, R2Also have and centainly mention
It is high.It is realized using PCA and EMD is decomposed to obtain the dimension-reduction treatment of data, eliminated the redundancy and correlation between variable, mention
The precision of prediction of prediction model is risen, it was demonstrated that the necessity of PCA dimension-reduction treatment.In addition to this, EMD-PCA-LSTM model
Precision of prediction is also significantly better than single LSTM neural network, traditional BP neural network and machine learning regression algorithm
XGBoost model.
The above is only embodiments of the present invention, are not intended to limit the scope of the invention, all to utilize the present invention
Equivalent structure or equivalent flow shift made by specification and accompanying drawing content is applied directly or indirectly in other relevant technologies
Field is included within the scope of the present invention.
Claims (5)
1. a kind of multivariable based on EMD-PCA-LSTM inputs photovoltaic power forecasting method, which is characterized in that
Step 1: the measured power time series number of the array photovoltaic power in photovoltaic plant actual production under inverter is obtained
According to and it is the solar irradiance of photovoltaic array area corresponding environment detector acquisition, relative humidity, air themperature, component temperature, big
5 kinds of environment sequence data of atmospheric pressure, the actual measurement sample data set of composition photovoltaic power prediction；
Step 2: to photovoltaic power prediction actual measurement sample data set carry out data cleansing and
Down-sampled processing；
Step 3: empirical mode decomposition method is utilized, 5 kinds of environment sequence data are decomposed；
Step 4: utilizing principal component analytical method, carries out dimensionality reduction to the characteristic sequence set that environment sequence data are decomposed；
Step 5: the data of characteristic sequence set and photovoltaic output power construction suitable for LSTM network training after dimensionality reduction are utilized
Collection, is divided into training set and test set according to the ratio of 7:3, and training set is inputted in LSMT network and is trained；
Step 6: after model training, training pattern is saved, test set is input in training pattern and is predicted, is passed through
Truthful data and prediction data are compared, the evaluation index of prediction: RMSE, MAE, R is found out2, carry out experimental result and summarize and analyze.
2. the multivariable according to claim 1 based on EMD-PCA-LSTM inputs photovoltaic power forecasting method, feature
Be, to sampled data carry out data cleansing the step of are as follows: as unit of day, power present in Rejection of samples data be 0 with
And the data that environmental data is 0；Data acquisition time period is set as early 6:00- evening 19:00, sampling interval 10min, every number of days
According to sampled point be 79；The variable for including in every group of data has the solar irradiance, air themperature, component temperature of 6:00-19:00
5 kinds of degree, relative humidity, atmospheric pressure environment sequence data.
3. the multivariable according to claim 1 based on EMD-PCA-LSTM inputs photovoltaic power forecasting method, feature
It is, primal environment sequence data is decomposed using EMD algorithm, obtains the IMF component and residue of every kind of environmental factor number
Component, to obtain the local feature of beginning environment sequence；By each environment sequence progress EMD IMF component decomposed and residue
Component is summarized, and obtains the characteristic sequence of total 68 dimension as new characteristic sequence set.
4. the multivariable according to claim 1 based on EMD-PCA-LSTM inputs photovoltaic power forecasting method, feature
It is, principal component analytical method is that initial data is transformed into new feature space by linear transformation, extracts number with this
According to main linear component, remove characteristic sequence data present in noise, reduce characteristic sequence data redundancy and phase
Guan Xing chooses the principal component that contribution rate of accumulative total is greater than 95%, as new input variable to obtained characteristic sequence data.
5. the multivariable according to claim 1 based on EMD-PCA-LSTM inputs photovoltaic power forecasting method, feature
It is, further includes determining to establish for photovoltaic output work before training set is inputted in LSMT network the step of being trained
Rate prediction LSTM neural network forecast it needs to be determined that model parameter the step of:
Using the environmental characteristic sequence and photovoltaic power historical data at t-1 moment, the photovoltaic power historical data of t moment is carried out
Prediction；Wherein, mode input layer time step number is 1, and input layer dimension is 9, and hiding number of layers is 1, and hidden layer unit number is 50
A, output layer variable is that number is 1, and training batch is 10, and training round is 100 times；
After model training, preservation model file, and test set data are tested, by experimental result to model
It is verified and is optimized repeatedly, improve the precision of prediction of model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910675922.5A CN110516844A (en) | 2019-07-25 | 2019-07-25 | Multivariable based on EMD-PCA-LSTM inputs photovoltaic power forecasting method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910675922.5A CN110516844A (en) | 2019-07-25 | 2019-07-25 | Multivariable based on EMD-PCA-LSTM inputs photovoltaic power forecasting method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110516844A true CN110516844A (en) | 2019-11-29 |
Family
ID=68623147
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910675922.5A Pending CN110516844A (en) | 2019-07-25 | 2019-07-25 | Multivariable based on EMD-PCA-LSTM inputs photovoltaic power forecasting method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110516844A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111144286A (en) * | 2019-12-25 | 2020-05-12 | 北京工业大学 | Urban PM2.5 concentration prediction method fusing EMD and LSTM |
| CN111369070A (en) * | 2020-03-13 | 2020-07-03 | 西安理工大学 | Envelope clustering-based multimode fusion photovoltaic power prediction method |
| 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 |
| CN112561181A (en) * | 2020-12-21 | 2021-03-26 | 国网甘肃省电力公司电力科学研究院 | Photovoltaic power generation prediction system based on Unet network and foundation cloud picture |
| CN112884249A (en) * | 2021-03-25 | 2021-06-01 | 国家海洋信息中心 | Sea area surface water temperature prediction method and device |
| CN113487064A (en) * | 2021-06-10 | 2021-10-08 | 淮阴工学院 | Photovoltaic power prediction method and system based on principal component analysis and improved LSTM |
| CN113673788A (en) * | 2021-09-23 | 2021-11-19 | 国网天津市电力公司 | Photovoltaic power generation power prediction method based on decomposition error correction and deep learning |
| CN113762642A (en) * | 2021-09-23 | 2021-12-07 | 大连理工大学人工智能大连研究院 | Classroom air quality prediction method based on BO-EMD-LSTM deep learning algorithm |
| CN113837434A (en) * | 2021-08-16 | 2021-12-24 | 同盾科技有限公司 | Solar photovoltaic power generation prediction method and device, electronic equipment and storage medium |
| CN114819382A (en) * | 2022-05-11 | 2022-07-29 | 湘潭大学 | Photovoltaic power prediction method based on LSTM |
| CN115096357A (en) * | 2022-06-07 | 2022-09-23 | 大连理工大学 | Indoor environment quality prediction method based on CEEMDAN-PCA-LSTM |
| CN116404645A (en) * | 2023-06-07 | 2023-07-07 | 山东大学 | Distributed photovoltaic short-term power prediction method and system considering space-time correlation characteristics |
| CN116933152A (en) * | 2023-06-07 | 2023-10-24 | 哈尔滨工业大学（威海） | Wave information prediction method and system based on multidimensional EMD-PSO-LSTM neural network |
| CN117194962A (en) * | 2023-09-13 | 2023-12-08 | 安徽国麒科技有限公司 | Photovoltaic power generation amount prediction method based on deep learning algorithm |
| CN117272851A (en) * | 2023-11-23 | 2023-12-22 | 太原理工大学 | Modeling prediction method for plant light receiving quantity under saline-alkali soil photovoltaic panel |
| CN117852721A (en) * | 2024-01-15 | 2024-04-09 | 国网山东省电力公司潍坊供电公司 | EPL-based park medium-long term terminal energy service demand prediction method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103049798A (en) * | 2012-12-05 | 2013-04-17 | 浙江大学城市学院 | Short-period electric generation power forecasting method applied to photovoltaic electric generation system |
| CN105589998A (en) * | 2015-12-23 | 2016-05-18 | 华北电力大学 | Photovoltaic output power super-short-term prediction method |
| CN108280551A (en) * | 2018-02-02 | 2018-07-13 | 华北电力大学 | A kind of photovoltaic power generation power prediction method using shot and long term memory network |
-
2019
- 2019-07-25 CN CN201910675922.5A patent/CN110516844A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103049798A (en) * | 2012-12-05 | 2013-04-17 | 浙江大学城市学院 | Short-period electric generation power forecasting method applied to photovoltaic electric generation system |
| CN105589998A (en) * | 2015-12-23 | 2016-05-18 | 华北电力大学 | Photovoltaic output power super-short-term prediction method |
| CN108280551A (en) * | 2018-02-02 | 2018-07-13 | 华北电力大学 | A kind of photovoltaic power generation power prediction method using shot and long term memory network |
Non-Patent Citations (1)
| Title |
|---|
| 杨茂等: "基于FA－PCA－LSTM的光伏发电短期功率预测", 《昆明理工大学学报( 自然科学版)》, vol. 44, no. 1, 28 February 2019 (2019-02-28), pages 1 - 5 * |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111144286A (en) * | 2019-12-25 | 2020-05-12 | 北京工业大学 | Urban PM2.5 concentration prediction method fusing EMD and LSTM |
| CN111369070A (en) * | 2020-03-13 | 2020-07-03 | 西安理工大学 | Envelope clustering-based multimode fusion photovoltaic power prediction method |
| CN111369070B (en) * | 2020-03-13 | 2023-06-27 | 西安理工大学 | Multimode fusion photovoltaic power prediction method based on envelope clustering |
| 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 |
| CN112561181A (en) * | 2020-12-21 | 2021-03-26 | 国网甘肃省电力公司电力科学研究院 | Photovoltaic power generation prediction system based on Unet network and foundation cloud picture |
| CN112884249A (en) * | 2021-03-25 | 2021-06-01 | 国家海洋信息中心 | Sea area surface water temperature prediction method and device |
| CN113487064A (en) * | 2021-06-10 | 2021-10-08 | 淮阴工学院 | Photovoltaic power prediction method and system based on principal component analysis and improved LSTM |
| CN113837434A (en) * | 2021-08-16 | 2021-12-24 | 同盾科技有限公司 | Solar photovoltaic power generation prediction method and device, electronic equipment and storage medium |
| CN113673788A (en) * | 2021-09-23 | 2021-11-19 | 国网天津市电力公司 | Photovoltaic power generation power prediction method based on decomposition error correction and deep learning |
| CN113762642A (en) * | 2021-09-23 | 2021-12-07 | 大连理工大学人工智能大连研究院 | Classroom air quality prediction method based on BO-EMD-LSTM deep learning algorithm |
| 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 |
| CN115096357A (en) * | 2022-06-07 | 2022-09-23 | 大连理工大学 | Indoor environment quality prediction method based on CEEMDAN-PCA-LSTM |
| CN116404645B (en) * | 2023-06-07 | 2023-08-25 | 山东大学 | Distributed photovoltaic short-term power prediction method and system considering space-time correlation characteristics |
| CN116933152A (en) * | 2023-06-07 | 2023-10-24 | 哈尔滨工业大学（威海） | Wave information prediction method and system based on multidimensional EMD-PSO-LSTM neural network |
| CN116933152B (en) * | 2023-06-07 | 2024-05-03 | 哈尔滨工业大学（威海） | Wave information prediction method and system based on multidimensional EMD-PSO-LSTM neural network |
| CN116404645A (en) * | 2023-06-07 | 2023-07-07 | 山东大学 | Distributed photovoltaic short-term power prediction method and system considering space-time correlation characteristics |
| CN117194962A (en) * | 2023-09-13 | 2023-12-08 | 安徽国麒科技有限公司 | Photovoltaic power generation amount prediction method based on deep learning algorithm |
| CN117272851A (en) * | 2023-11-23 | 2023-12-22 | 太原理工大学 | Modeling prediction method for plant light receiving quantity under saline-alkali soil photovoltaic panel |
| CN117272851B (en) * | 2023-11-23 | 2024-02-02 | 太原理工大学 | Modeling prediction method for plant light receiving quantity under saline-alkali soil photovoltaic panel |
| CN117852721A (en) * | 2024-01-15 | 2024-04-09 | 国网山东省电力公司潍坊供电公司 | EPL-based park medium-long term terminal energy service demand prediction method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110516844A (en) | Multivariable based on EMD-PCA-LSTM inputs photovoltaic power forecasting method | |
| CN109376951B (en) | Photovoltaic probability prediction method | |
| CN106779223B (en) | Photovoltaic system power generation real-time prediction method and device | |
| Zhao et al. | Performance evaluation and accuracy enhancement of a day-ahead wind power forecasting system in China | |
| Kardakos et al. | Application of time series and artificial neural network models in short-term forecasting of PV power generation | |
| CN106920007B (en) | PM based on second-order self-organizing fuzzy neural network2.5Intelligent prediction method | |
| Zhang et al. | Short term wind energy prediction model based on data decomposition and optimized LSSVM | |
| CN113496311A (en) | Photovoltaic power station generated power prediction method and system | |
| CN102479339A (en) | Method and system for forecasting short-term wind speed of wind farm based on hybrid neural network | |
| CN104657791B (en) | A kind of wind farm group wind speed profile prediction technique based on correlation analysis | |
| CN108520269A (en) | A kind of wind speed forecasting method and forecasting wind speed system | |
| CN101604356A (en) | A kind of method for building up of uncertain mid-and-long term hydrologic forecast model | |
| CN111695666A (en) | Wind power ultra-short term conditional probability prediction method based on deep learning | |
| CN115860797B (en) | Electric quantity demand prediction method suitable for new electricity price reform situation | |
| CN106600060A (en) | Method for predicting amount of solar radiation based on similar day sunny coefficient correction | |
| CN116128039A (en) | Construction method and prediction method of surface water quality prediction model | |
| CN116151464A (en) | Photovoltaic power generation power prediction method, system and storable medium | |
| Qu et al. | Research on short‐term output power forecast model of wind farm based on neural network combination algorithm | |
| CN103605908A (en) | Wind speed sequence forecasting method based on Kalman filtering | |
| Cheng et al. | Application of clustering analysis in the prediction of photovoltaic power generation based on neural network | |
| Khalyasmaa et al. | Averaged errors as a risk factor for intelligent forecasting systems operation in the power industry | |
| CN118095891A (en) | Active power distribution network payload prediction method and system considering source load meteorological characteristic decoupling | |
| Shiwakoti et al. | Time Series Analysis of Electricity Demand Forecasting Using Seasonal ARIMA and an Exponential Smoothing Model | |
| CN112132344A (en) | Short-term wind power prediction method based on similar day and FRS-SVM | |
| CN116565850A (en) | Wind power ultra-short-term prediction method based on QR-BLSTM |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191129 |