CN107563561A - A kind of method and system of photovoltaic prediction - Google Patents
A kind of method and system of photovoltaic prediction Download PDFInfo
- Publication number
- CN107563561A CN107563561A CN201710806475.3A CN201710806475A CN107563561A CN 107563561 A CN107563561 A CN 107563561A CN 201710806475 A CN201710806475 A CN 201710806475A CN 107563561 A CN107563561 A CN 107563561A
- Authority
- CN
- China
- Prior art keywords
- kernel function
- supporting vector
- component
- photovoltaic
- machine model
- 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
Abstract
Photovoltaic sequence is decomposed the invention discloses a kind of method of photovoltaic prediction, including according to integrated Empirical mode decomposition, obtains intrinsic mode functions component and trend component；The first kernel function is selected according to preset rules, then trend component is predicted by supporting vector machine model, obtains the predicted value of trend component, wherein, supporting vector machine model includes the first kernel function.The present invention is predicted for the feature of trend component, improves precision of prediction, and then be ensure that the safe operation of photovoltaic generating system and facilitated staff and carry out rational dispatching of power netwoks.The invention also discloses a kind of system of photovoltaic prediction, there is above-mentioned beneficial effect.
Description
Technical field
The present invention relates to field of photovoltaic power generation, more particularly to a kind of method and system of photovoltaic prediction.
Background technology
Solar energy power generating is as a kind of high conversion efficiency, service life is long, the generation technology without running gear by
The extensive attention and popularization of countries in the world, because the power output of photovoltaic sequence changes closely with environmental factor, have very big
Randomness and fluctuation, so, in order that photovoltaic generating system safe operation and facilitating staff to carry out rational power network
Scheduling is, it is necessary to which the power output following to photovoltaic sequence is accurately predicted, but existing Forecasting Methodology can not accomplish root
It can not be also directed to according to the feature of each decomposed component according to effective decompose is carried out the characteristics of photovoltaic sequence power output
Property predict, cause precision of prediction to be difficult to meet engine request, prediction error is larger.
Therefore, how to provide a kind of scheme for solving above-mentioned technical problem is that those skilled in the art need to solve at present
Problem.
The content of the invention
It is an object of the invention to provide a kind of method of photovoltaic prediction, the present invention carries out pre- for the feature of trend component
Survey, improve precision of prediction, and then ensure that the safe operation of photovoltaic generating system and facilitate staff and carry out rationally
Dispatching of power netwoks；It is a further object of the present invention to provide a kind of system of photovoltaic prediction.
In order to solve the above technical problems, the invention provides a kind of method of photovoltaic prediction, including：
Photovoltaic sequence is decomposed according to integrated Empirical mode decomposition, obtains intrinsic mode functions component and trend point
Amount；
The first kernel function is selected according to preset rules, then the trend component is carried out by supporting vector machine model pre-
Survey, obtain the predicted value of the trend component, wherein, the supporting vector machine model includes the first kernel function.
Preferably, it is described obtain intrinsic mode functions component and trend component after, according to preset rules select kernel function it
Before, this method also includes：
Reconstruct is overlapped to the intrinsic mode functions component and obtains new photovoltaic sequence；
The new photovoltaic sequence is decomposed to obtain details coefficients again according to wavelet decomposition method；
After being predicted then and then to the trend component by supporting vector machine model, this method also includes：
The second kernel function is selected according to preset rules, then the details coefficients are carried out by supporting vector machine model pre-
Survey, obtain the predicted value of the details coefficients, wherein, the SVMs includes the second kernel function, second kernel function
It is different with first kernel function；
The predicted value of the predicted value of the trend component and the details coefficients is overlapped reconstruct, obtains the photovoltaic
The final predicted value of sequence.
Preferably, first kernel function is linear kernel function, wherein, the linear kernel function is K (x, xi)=x × xi,
X is the vector that extracts in space, xiFor supporting vector.
Preferably, second kernel function is gaussian kernel function, wherein, gaussian kernel function isx
For the vector extracted in space, xiFor supporting vector, δ is the width parameter of the gaussian kernel function.
Preferably, the supporting vector machine model also includes multiple parameters, described to obtain intrinsic mode functions component and trend
After component, before selecting the first kernel function according to preset rules, this method also includes：
Each parameter of the supporting vector machine model is optimized respectively according to interior extrapolation method in length and breadth.
In order to solve the above-mentioned technical problem, present invention also offers a kind of system of photovoltaic prediction, including：
Decomposing module, for being decomposed according to integrated Empirical mode decomposition to photovoltaic sequence, obtain intrinsic mode functions
Component and trend component；
Prediction module, for selecting the first kernel function according to preset rules, then by supporting vector machine model to described
Trend component is predicted, and obtains the predicted value of the trend component, wherein, the supporting vector machine model includes the first core letter
Number.
Preferably, it is described obtain intrinsic mode functions component and trend component after, according to preset rules select kernel function it
Before, the decomposing module is additionally operable to that the intrinsic mode functions component is overlapped to reconstruct to obtain new photovoltaic sequence；
The new photovoltaic sequence is decomposed to obtain details coefficients again according to wavelet decomposition method；
After being predicted then and then to the trend component by supporting vector machine model, the prediction module is additionally operable to
The second kernel function is selected according to preset rules, then the details coefficients are predicted by supporting vector machine model, obtained
The predicted value of the details coefficients, wherein, the SVMs includes the second kernel function, second kernel function and described
One kernel function is different；
The predicted value of the predicted value of the trend component and the details coefficients is overlapped reconstruct, obtains the photovoltaic
The final predicted value of sequence.
Preferably, first kernel function is linear kernel function, wherein, the linear kernel function is K (x, xi)=x × xi,
X is the vector that extracts in space, xiFor supporting vector.
Preferably, second kernel function is gaussian kernel function, wherein, gaussian kernel function isx
For the vector extracted in space, xiFor supporting vector, δ is the width parameter of the gaussian kernel function.
Preferably, the supporting vector machine model also includes multiple parameters, described to obtain intrinsic mode functions component and trend
After component, before selecting the first kernel function according to preset rules, the prediction module is additionally operable to be distinguished according to interior extrapolation method in length and breadth
Each parameter of the supporting vector machine model is optimized.
The present invention proposes a kind of method of photovoltaic prediction, including photovoltaic sequence is entered according to integrated Empirical mode decomposition
Row decomposes, and obtains intrinsic mode functions component and trend component；According to preset rules select the first kernel function, then by support to
Amount machine model is predicted to trend component, obtains the predicted value of trend component, wherein, supporting vector machine model includes the first core
Function.
It can be seen that in actual applications, it is logical to integrating Empirical mode decomposition photovoltaic sequence first using the solution of the present invention
Decomposed, obtain reacting the trend component of photovoltaic sequence power output trend, the first core letter is chosen according to preset rules
Number, is predicted for the feature of trend component, improves precision of prediction, and then ensure that the safe operation of photovoltaic generating system
And facilitate staff and carry out rational dispatching of power netwoks.
Present invention also offers a kind of system of photovoltaic prediction, has and above-mentioned photovoltaic Forecasting Methodology identical is beneficial to effect
Fruit.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, below will be to institute in prior art and embodiment
The accompanying drawing needed to use is briefly described, it should be apparent that, drawings in the following description are only some implementations of the present invention
Example, for those of ordinary skill in the art, on the premise of not paying creative work, can also be obtained according to these accompanying drawings
Obtain other accompanying drawings.
Fig. 1 is a kind of flow chart of photovoltaic Forecasting Methodology provided by the present invention；
Fig. 2 is a kind of structural representation of photovoltaic forecasting system provided by the present invention.
Embodiment
The core of the present invention is to provide a kind of method of photovoltaic prediction, and the present invention carries out pre- for the feature of trend component
Survey, improve precision of prediction, and then ensure that the safe operation of photovoltaic generating system and facilitate staff and carry out rationally
Dispatching of power netwoks；Another core of the present invention is to provide a kind of system of photovoltaic prediction.
To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
Part of the embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
Fig. 1 is refer to, Fig. 1 is a kind of flow chart of photovoltaic Forecasting Methodology provided by the present invention, including：
Step 1：Photovoltaic sequence is decomposed according to integrated Empirical mode decomposition, intrinsic mode functions component is obtained and becomes
Gesture component；
Specifically, photovoltaic prediction is exactly the inherent law implied by photovoltaic power historical data, photovoltaic sequence is inferred to
Following power output is arranged, because the historical data of photovoltaic power is in non-linear, and there is higher randomness and saltant type, institute
Regarding the historical data of photovoltaic power as a photovoltaic sequence pair with the present invention, it is pre-processed, and prediction process is had more pin
To property.
Specifically, set empirical mode decomposition EEMD is a kind of noise assistance data analytic approach, by the way that white noise is added
Into photovoltaic sequence to be decomposed, the intermittency of photovoltaic sequence is eliminated, suppresses to influence caused by noise in decomposition result,
EEMD adaptively resolves into time series data several intrinsic mode functions components and residual components, can more clearly see
Go out the inherent changing rule of the historical data of photovoltaic power, improve precision of prediction of knowing clearly.
Step 2：The first kernel function is selected according to preset rules, then trend component is carried out by supporting vector machine model
Prediction, obtains the predicted value of trend component, wherein, supporting vector machine model includes the first kernel function.
Specifically, supporting vector machine model is used for pattern classification and regression forecasting, the trend component after EEMD is decomposed
Reflect the general trend of photovoltaic power historical data, choose the first kernel function, specific aim prediction is carried out to trend component, can be with
Know the future trend of photovoltaic power, using the method for the present invention, effectively decomposed according to the characteristics of photovoltaic power, and root
Targetedly predicted according to the feature of its trend component, effectively reduce prediction error, further ensuring precision of prediction expires
The corresponding engine request of foot, and the efficiency high of the present invention program is used, it is simple in construction, it is easily operated.
Specifically, the linear regression function of supporting vector machine model isαiWith
For Lagrange Multiplier and it is not 0, k (x, xi) it is kernel function, x is the vector extracted in space, xiFor supporting vector, b is inclined
Shifting amount, specifically, the selection of parameter and kernel function determines that the present invention does not limit herein according to the actual conditions of prediction object.
Specifically, preset rules are to select corresponding first core letter according to the fluctuation situation of each decomposed component of prediction pair
Number.
The present invention proposes a kind of method of photovoltaic prediction, including photovoltaic sequence is entered according to integrated Empirical mode decomposition
Row decomposes, and obtains intrinsic mode functions component and trend component；According to preset rules select the first kernel function, then by support to
Amount machine model is predicted to trend component, obtains the predicted value of trend component, wherein, supporting vector machine model includes the first core
Function.
It can be seen that in actual applications, it is logical to integrating Empirical mode decomposition photovoltaic sequence first using the solution of the present invention
Decomposed, obtain reacting the trend component of photovoltaic sequence power output trend, the first core letter is chosen according to preset rules
Number, is predicted for the feature of trend component, improves precision of prediction, and then ensure that the safe operation of photovoltaic generating system
And facilitate staff and carry out rational dispatching of power netwoks.
On the basis of above-described embodiment：
As a kind of preferred embodiment, after obtaining intrinsic mode functions component and trend component, selected according to preset rules
Before selecting kernel function, this method also includes：
Reconstruct is overlapped to intrinsic mode functions component and obtains new photovoltaic sequence；
New photovoltaic sequence is decomposed to obtain details coefficients again according to wavelet decomposition method；
After being predicted then and then to trend component by supporting vector machine model, this method also includes：
The second kernel function is selected according to preset rules, then details coefficients are predicted by supporting vector machine model,
The predicted value of details coefficients is obtained, wherein, SVMs includes the second kernel function, and the second kernel function and the first kernel function are not
Together；
The predicted value of the predicted value of trend component and details coefficients is overlapped reconstruct, obtains the final pre- of photovoltaic sequence
Measured value.
Specifically, in order to improve precision of prediction, the present invention has carried out double decomposition processing to photovoltaic sequence, i.e., to photovoltaic sequence
The intrinsic mode functions component obtained after EEMD decomposition is decomposed again, first obtains the superposition reconstruct of intrinsic mode functions component
New photovoltaic sequence, wavelet decomposition is carried out to new photovoltaic sequence, obtains details coefficients, wherein, details coefficients can be well
Reflect the local feature information of photovoltaic power initial data, then details coefficients are carried out targetedly by supporting vector machine model
Prediction, the predicted value for predicting to obtain twice is reconstructed, obtains final prediction result, reduce further prediction error.
Specifically, wavelet transformation is a kind of multiple dimensioned signal decomposition method, by flexible, the translation of wavelet basis, and with
Analyzed signal is compared, and realizes the video analysis localization of signal, and continuous wavelet transformation relational expression isψ (t) is basic function, ψa,b(t) it is wavelet basis, x (t)
∈L2(R) it is a square-integrable signal, a is to characterize yardstick, and b is displacement parameter.
As a kind of preferred embodiment, the first kernel function is linear kernel function, wherein, linear kernel function is K (x, xi)
=x × xi, x is the vector that extracts in space, xiFor supporting vector.
Specifically, the trend component after EEMD decomposition reflects the general trend of photovoltaic power historical data, relatively
Steadily, so the first Selection of kernel function linear kernel function, targetedly to be predicted trend component, prediction result is more accurate
Really, actual conditions are more conformed to.
Certainly, except linear kernel function can be selected to be also an option that other kernel functions as the first kernel function, the present invention
Do not limit herein.
As a kind of preferred embodiment, the second kernel function is gaussian kernel function, wherein, gaussian kernel function isX is the vector that extracts in space, xiFor supporting vector, δ is the width parameter of gaussian kernel function.
Specifically, the details coefficients after wavelet decomposition reflect the local feature of photovoltaic power historical data, fluctuation phase
To relatively strong, so the second Selection of kernel function gaussian kernel function, targetedly to be predicted details coefficients, prediction result is more
Accurately, actual conditions are more conformed to.
Certainly, except gaussian kernel function can be selected to be also an option that other kernel functions as the second kernel function, the present invention
Do not limit herein.
In summary, can be according to each decomposed component of photovoltaic power historical data spy of itself using the solution of the present invention
Point, suitable kernel function is adaptive selected to carry out specific aim prediction, reduce further prediction error.
As a kind of preferred embodiment, supporting vector machine model also includes multiple parameters, obtains intrinsic mode functions component
After trend component, before selecting the first kernel function according to preset rules, this method also includes：
The parameters of supporting vector machine model are optimized respectively according to interior extrapolation method in length and breadth.
Specifically, supporting vector machine model includes multiple parameters, typically can be to branch in order to obtain more accurate prediction result
The parameters for holding vector model are optimized, and the present invention is optimized using interior extrapolation method in length and breadth to parameters, make optimization fast
Faster, convergence precision is higher, further ensures the accuracy of prediction result for degree.
Except of course that can be optimized using interior extrapolation method in length and breadth, other method can also be used, the present invention is not done herein
Limit.
Fig. 2 is refer to, Fig. 2 is a kind of structural representation of photovoltaic forecasting system provided by the present invention, including：
Decomposing module 1, for being decomposed according to integrated Empirical mode decomposition to photovoltaic sequence, obtain intrinsic mode functions
Component and trend component；
Prediction module 2, for selecting the first kernel function according to preset rules, then by supporting vector machine model to trend
Component is predicted, and obtains the predicted value of trend component, wherein, supporting vector machine model includes the first kernel function.
As a kind of preferred embodiment, after obtaining intrinsic mode functions component and trend component, selected according to preset rules
Before selecting kernel function, decomposing module 1 is additionally operable to that intrinsic mode functions component is overlapped to reconstruct to obtain new photovoltaic sequence；
New photovoltaic sequence is decomposed to obtain details coefficients again according to wavelet decomposition method；
After being predicted then and then to trend component by supporting vector machine model, prediction module 2 is additionally operable to according to pre-
Then if details coefficients are predicted, obtain details coefficients by rule the second kernel function of selection by supporting vector machine model
Predicted value, wherein, SVMs includes the second kernel function, and the second kernel function is different with the first kernel function；
The predicted value of the predicted value of trend component and details coefficients is overlapped reconstruct, obtains the final pre- of photovoltaic sequence
Measured value.
As a kind of preferred embodiment, the first kernel function is linear kernel function, wherein, linear kernel function is K (x, xi)
=x × xi, x is the vector that extracts in space, xiFor supporting vector.
As a kind of preferred embodiment, the second kernel function is gaussian kernel function, wherein, gaussian kernel function isX is the vector that extracts in space, xiFor supporting vector, δ is the width parameter of gaussian kernel function.
As a kind of preferred embodiment, supporting vector machine model also includes multiple parameters, obtains intrinsic mode functions component
After trend component, before selecting the first kernel function according to preset rules, prediction module 2 is additionally operable to according to interior extrapolation method point in length and breadth
The other parameters to supporting vector machine model optimize.
Above-described embodiment is refer to for a kind of introduction of the system of photovoltaic prediction provided by the present invention, the present invention is herein
Repeat no more.
Each embodiment is described by the way of progressive in this specification, what each embodiment stressed be and other
The difference of embodiment, between each embodiment identical similar portion mutually referring to.For device disclosed in embodiment
For, because it is corresponded to the method disclosed in Example, so description is fairly simple, related part is said referring to method part
It is bright.
The foregoing description of the disclosed embodiments, professional and technical personnel in the field are enable to realize or using the present invention.
A variety of modifications to these embodiments will be apparent for those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the present invention.Therefore, it is of the invention
The embodiments shown herein is not intended to be limited to, and is to fit to and principles disclosed herein and features of novelty phase one
The most wide scope caused.
Claims (10)
- A kind of 1. method of photovoltaic prediction, it is characterised in that including：Photovoltaic sequence is decomposed according to integrated Empirical mode decomposition, obtains intrinsic mode functions component and trend component；The first kernel function is selected according to preset rules, then the trend component is predicted by supporting vector machine model, The predicted value of the trend component is obtained, wherein, the supporting vector machine model includes the first kernel function.
- 2. according to the method for claim 1, it is characterised in that it is described obtain intrinsic mode functions component and trend component it Afterwards, before selecting kernel function according to preset rules, this method also includes：Reconstruct is overlapped to the intrinsic mode functions component and obtains new photovoltaic sequence；The new photovoltaic sequence is decomposed to obtain details coefficients again according to wavelet decomposition method；After being predicted then and then to the trend component by supporting vector machine model, this method also includes：The second kernel function is selected according to preset rules, then the details coefficients are predicted by supporting vector machine model, The predicted value of the details coefficients is obtained, wherein, the SVMs includes the second kernel function, second kernel function and institute State the first kernel function difference；The predicted value of the predicted value of the trend component and the details coefficients is overlapped reconstruct, obtains the photovoltaic sequence Final predicted value.
- 3. according to the method for claim 1, it is characterised in that first kernel function is linear kernel function, wherein, it is described Linear kernel function is K (x, xi)=x × xi, x is the vector that extracts in space, xiFor supporting vector.
- 4. according to the method for claim 2, it is characterised in that second kernel function is gaussian kernel function, wherein, Gauss Kernel function isX is the vector that extracts in space, xiFor supporting vector, δ is the width of the gaussian kernel function Spend parameter.
- 5. according to the method described in claim 1-4 any one, it is characterised in that the supporting vector machine model also includes more Individual parameter, it is described obtain intrinsic mode functions component and trend component after, according to preset rules select the first kernel function before, should Method also includes：Each parameter of the supporting vector machine model is optimized respectively according to interior extrapolation method in length and breadth.
- A kind of 6. system of photovoltaic prediction, it is characterised in that including：Decomposing module, for being decomposed according to integrated Empirical mode decomposition to photovoltaic sequence, obtain intrinsic mode functions component And trend component；Prediction module, for selecting the first kernel function according to preset rules, then by supporting vector machine model to the trend Component is predicted, and obtains the predicted value of the trend component, wherein, the supporting vector machine model includes the first kernel function.
- 7. system according to claim 1, it is characterised in that it is described obtain intrinsic mode functions component and trend component it Afterwards, before selecting kernel function according to preset rules, the decomposing module is additionally operable to be overlapped the intrinsic mode functions component Reconstruct obtains new photovoltaic sequence；The new photovoltaic sequence is decomposed to obtain details coefficients again according to wavelet decomposition method；After being predicted then and then to the trend component by supporting vector machine model, the prediction module is additionally operable to basis Preset rules select the second kernel function, and then the details coefficients are predicted by supporting vector machine model, obtain described The predicted value of details coefficients, wherein, the SVMs includes the second kernel function, second kernel function and first core Function is different；The predicted value of the predicted value of the trend component and the details coefficients is overlapped reconstruct, obtains the photovoltaic sequence Final predicted value.
- 8. system according to claim 1, it is characterised in that first kernel function is linear kernel function, wherein, it is described Linear kernel function is K (x, xi)=x × xi, x is the vector that extracts in space, xiFor supporting vector.
- 9. system according to claim 7, it is characterised in that second kernel function is gaussian kernel function, wherein, Gauss Kernel function isX is the vector that extracts in space, xiFor supporting vector, δ is the width of the gaussian kernel function Spend parameter.
- 10. according to the system described in claim 6-9 any one, it is characterised in that the supporting vector machine model also includes Multiple parameters, it is described obtain intrinsic mode functions component and trend component after, according to preset rules select the first kernel function before, The prediction module is additionally operable to excellent according to each parameter progress of the interior extrapolation method in length and breadth respectively to the supporting vector machine model Change.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710806475.3A CN107563561A (en) | 2017-09-08 | 2017-09-08 | A kind of method and system of photovoltaic prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710806475.3A CN107563561A (en) | 2017-09-08 | 2017-09-08 | A kind of method and system of photovoltaic prediction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107563561A true CN107563561A (en) | 2018-01-09 |
Family
ID=60980280
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710806475.3A Pending CN107563561A (en) | 2017-09-08 | 2017-09-08 | A kind of method and system of photovoltaic prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107563561A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150302313A1 (en) * | 2014-04-22 | 2015-10-22 | State Grid Corporation Of China | Method of predicating ultra-short-term wind power based on self-learning composite data source |
| CN106682782A (en) * | 2016-12-30 | 2017-05-17 | 国网新疆电力公司电力科学研究院 | Short-term photovoltaic power prediction method based on EWT-KMPMR (empirical wavelet transform and kernel minimax probability machine classification) |
-
2017
- 2017-09-08 CN CN201710806475.3A patent/CN107563561A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150302313A1 (en) * | 2014-04-22 | 2015-10-22 | State Grid Corporation Of China | Method of predicating ultra-short-term wind power based on self-learning composite data source |
| CN106682782A (en) * | 2016-12-30 | 2017-05-17 | 国网新疆电力公司电力科学研究院 | Short-term photovoltaic power prediction method based on EWT-KMPMR (empirical wavelet transform and kernel minimax probability machine classification) |
Non-Patent Citations (2)
| Title |
|---|
| 茆美琴等: "基于 EEMD-SVM 方法的光伏电站短期出力预测", 《中国电机工程学报》 * |
| 马留洋等: "基于经验模态分解和小波包分解的组合风电功率预测模型分析", 《内蒙古电力技术》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103136598B (en) | Monthly calculation of power load machine Forecasting Methodology based on wavelet analysis | |
| CN102855385A (en) | Wind power generation short-period load forecasting method | |
| CN106651188A (en) | Electric transmission and transformation device multi-source state assessment data processing method and application thereof | |
| CN107861606A (en) | A kind of heterogeneous polynuclear power cap method by coordinating DVFS and duty mapping | |
| CN109948833A (en) | A kind of Hydropower Unit degradation trend prediction technique based on shot and long term memory network | |
| CN110148068A (en) | One kind being based on meteorological data similarity analysis and LSTM neural fusion photovoltaic plant ultra-short term power forecasting method | |
| Ehsan et al. | A scenario‐based robust investment planning model for multi‐type distributed generation under uncertainties | |
| CN113988477A (en) | Photovoltaic power short-term prediction method and device based on machine learning and storage medium | |
| Li et al. | Eigenvalue‐optimisation‐based optimal power flow with small‐signal stability constraints | |
| CN107330589A (en) | Satellite network coordinates the quantitative evaluation method and system of risk | |
| CN105373856A (en) | Wind electricity power short-term combined prediction method considering run detection method reconstruction | |
| CN107491833A (en) | A kind of method and apparatus of photovoltaic power generation power prediction | |
| CN109840633A (en) | Photovoltaic output power predicting method, system and storage medium | |
| CN104573371A (en) | Optimal wavelet packet transformation and AR-SVM based non-stable wind speed prediction method | |
| CN102930352A (en) | Power grid basic construction project cost prediction method based on multi-core support vector regression | |
| CN105956252B (en) | Based on the multiple dimensioned Forecast model method of ultra-short term wind speed for generating moldeed depth belief network | |
| CN107145968A (en) | Photovoltaic apparatus life cycle cost Forecasting Methodology and system based on BP neural network | |
| Huang et al. | Stochastic economic dispatch with wind using versatile probability distribution and L-BFGS-B based dual decomposition | |
| CN108537359A (en) | Photovoltaic power generation power prediction based on wavelet analysis and BP neural network | |
| CN109614384A (en) | Power-system short-term load forecasting method under Hadoop frame | |
| Bodenbenner et al. | Design science in practice: designing an electricity demand response system | |
| CN107563561A (en) | A kind of method and system of photovoltaic prediction | |
| Risco-Ramos et al. | Use of a business intelligence framework in the management of the quality of electricity supply in small and medium-sized companies | |
| CN106815659A (en) | A kind of ultra-short term forecast of solar irradiance method and its device based on mixed model | |
| Gobu et al. | Two-stage machine learning framework for simultaneous forecasting of price-load in the smart grid |
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: 20180109 |
|
| RJ01 | Rejection of invention patent application after publication |