CN109190786A - A kind of combination forecasting method of photovoltaic efficiency - Google Patents
A kind of combination forecasting method of photovoltaic efficiency Download PDFInfo
- Publication number
- CN109190786A CN109190786A CN201810737933.7A CN201810737933A CN109190786A CN 109190786 A CN109190786 A CN 109190786A CN 201810737933 A CN201810737933 A CN 201810737933A CN 109190786 A CN109190786 A CN 109190786A
- Authority
- CN
- China
- Prior art keywords
- photovoltaic
- combination forecasting
- prediction
- model
- neural network
- 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 12
- 238000003062 neural network model Methods 0.000 claims abstract description 17
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000005286 illumination Methods 0.000 claims description 3
- 238000009434 installation Methods 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 2
- 238000010835 comparative analysis Methods 0.000 claims 1
- 238000000034 method Methods 0.000 abstract description 25
- 238000012512 characterization method Methods 0.000 abstract description 4
- 238000013528 artificial neural network Methods 0.000 description 5
- 238000006467 substitution reaction Methods 0.000 description 3
- 238000012935 Averaging Methods 0.000 description 1
- 101001095088 Homo sapiens Melanoma antigen preferentially expressed in tumors Proteins 0.000 description 1
- 102100037020 Melanoma antigen preferentially expressed in tumors Human genes 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000000843 powder Substances 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
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
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A30/00—Adapting or protecting infrastructure or their operation
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
- Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
-
- 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
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- General Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Tourism & Hospitality (AREA)
- Public Health (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Water Supply & Treatment (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 combination forecasting methods of photovoltaic efficiency, using the actual historical power output time series of photovoltaic plant and corresponding weather forecast weather information parameter as the input of gray model and BP neural network model in the combination forecasting, subsequent time photovoltaic plant performance number is respectively as output;The value of the output is carried out power prediction in conjunction with the space-time characterisation of photovoltaic plant, obtain power prediction value, then assess the power prediction value respectively multiplied by the weighted value of the gray model and the BP neural network model again;The prediction result of the above method is accurate, can greatly improve the precision of prediction of photovoltaic generation power.
Description
Technical field
The invention belongs to photovoltaic power generation electric powder predictions, and in particular to a kind of combined prediction side of photovoltaic efficiency
Method.
Background technique
As modern prediction theory technology continues to develop, BP neural network Predicting Technique, wavelet analysis Predicting Technique, grey
The new Predicting Technique such as predicted method, support vector machines analytical prediction method is gradually applied in photovoltaic power prediction field;Wherein,
Gray model is good at the system for handling few data sample, and BP neural network can be made by constantly adjusting the weight and threshold value of network
It obtains network constantly to be trained, be optimal.During prediction, the data of every kind of prediction technique application are roughly the same, but
It is the angle difference for improving useful information, either qualitative forecasting or quantitative forecast have its advantage and disadvantage.In actual prediction,
Exist not to be mutually exclusive between these methods or model, between them and connect each other, or even is mutually complementary.
In addition, different prediction techniques can extract a variety of different useful informations from given data, but Individual forecast
The shortcomings that method precision is not high, is easily trapped into local optimum such as neural network prediction, combination forecasting method is based on can be abundant
The useful information provided using each individual event prediction technique, comprehensive each method or model advantage, are learnt from other's strong points to offset one's weaknesses, to improve prediction
Precision.Relative to Individual forecast method or model, combination forecasting method can be avoided certain useful informations and lose, more scientific,
But it is related to the problem how weight of single model determines in the combination forecasting that combines of multiple models, it is different
The method of determination will affect precision of prediction.
Summary of the invention
For the drawbacks described above for overcoming the prior art, the purpose of the present invention is to provide a kind of combination of photovoltaic efficiency is pre-
Survey method is determined in combination forecasting on the basis of the weight of each model by being introduced into maximum information entropy theory, carries out photovoltaic
The power prediction of component improves precision of prediction.
Above-mentioned purpose of the invention is achieved through the following technical solutions:
A kind of combination forecasting of photovoltaic efficiency, including gray model and BP neural network model, and the ash
The weight ratio of color model and the BP neural network model is determined by maximum informational entropy.
The second aspect of the present invention, a kind of combination forecasting method of photovoltaic efficiency, specifically, comprising the following steps:
The actual historical power output time series of photovoltaic plant and corresponding weather forecast weather information parameter are distinguished
As the input of gray model and BP neural network model in the combination forecasting, subsequent time photovoltaic plant performance number point
It Zuo Wei not export;Again by the value of the output respectively multiplied by the weighted value of the gray model and the BP neural network model,
Power prediction is carried out in conjunction with the space-time characterisation of photovoltaic plant, obtains power prediction value, then assess the power prediction value.
Further, the weather forecast weather information parameter includes intensity of illumination, temperature, humidity, wind speed, photovoltaic array
Installation, solar incident angle degree and transfer efficiency.
It should be noted that the gray model (Grey Model) is referred to as GM model, it is suitable for time-variant nonlinear system
System, is the most frequently used, more succinct gray model, BP neural network model is also known as error backpropagation algorithm, reversed using error
The multilayer feedforward neural network of propagation algorithm;The present invention is by actual historical power output time series of photovoltaic plant and corresponding
Weather forecast weather information parameter is separately input in Individual forecast Model B P and GM (1,1), then exports subsequent time light respectively
Overhead utility performance number obtains power prediction value by output valve respectively multiplied by respective weighted value after superposition.Wherein, each single model
Weighted value determined by maximum information entropy theory, regard combined prediction process as an informix process, i.e., from various single
The statistical nature that the amount of being predicted is extracted in the prediction result of prediction model, the letter of combination forecasting is supplied to as the model
Breath, makes objective prediction to forecasted future value.
Compared with prior art, the beneficial effects of the present invention are:
One, the present invention is effectively reduced by the way that two Individual forecast models are combined prediction to Individual forecast method essence
The dependence of degree makes up the shortcomings that single BP neural network is easily trapped into local optimum, gives full play to two prediction model advantages,
Improve precision of prediction;And in the weight of single model, conventional averaging method is not used, but is carried out by maximum information entropy theory
Weight optimization keeps combined prediction result more accurate, and verifying also further demonstrates that combination forecasting method of the invention can improve photovoltaic
The precision of prediction of generated output.
Two, the present invention utilizes combination forecasting method Accurate Prediction photovoltaic generation power, improves the utilization rate of photovoltaic electric energy, has
Effect mitigates influence of the system access to power grid, to dispatching of power netwoks department reasonable arrangement operation plan, balance of electric power and ener cooperation with
And it is of great significance to entire safe and stable operation of power system.
Detailed description of the invention
Fig. 1 is combination forecasting flow chart;
Fig. 2 is combination forecasting;
Fig. 3 is historical power curve and prediction power curve comparison figure.
Specific embodiment
The technical solution that the invention will now be described in detail with reference to the accompanying drawings, but protection scope of the present invention is not limited to following realities
Apply example.
Referring to attached Fig. 1 and 2, two kinds of Single models of gray model and BP neural network model are first passed through, according to maximum information
The weighted value that entropy determines, which is weighted, combines to obtain combination forecasting;Wherein, the weight of gray model and BP neural network model
Value is determined by maximum informational entropy.The method for carrying out photovoltaic efficiency prediction by said combination prediction model, comprising: first
Using the actual historical power output time series of photovoltaic plant and corresponding weather forecast weather information parameter as described
The input of gray model and BP neural network model in combination forecasting, subsequent time photovoltaic plant performance number is respectively as defeated
Out;Again by the value of the output respectively multiplied by the weighted value of the gray model and the BP neural network model, in conjunction with photovoltaic
The space-time characterisation in power station carries out power prediction, obtains power prediction value, then assess the power prediction value;Wherein, weather
Forecast that weather information parameter includes intensity of illumination, temperature, humidity, wind speed, the installation of photovoltaic array, solar incident angle degree and conversion
Efficiency.Detailed process are as follows:
S1: gray model described in Individual forecast model and the BP neural network model respectively carry out photovoltaic plant power
Prediction, the mean value of prediction result are denoted as
S2: the gray model and the BP neural network model are calculated to the information contribution degree of combination forecasting, tool
Body method are as follows: respectively to respectively since t moment point start the total n moment photovoltaic generation power carry out simulation and forecast, as a result record
ForThen the numerical characteristic for calculating prediction photovoltaic generation power must predict that photovoltaic is sent out
The center of electrical power away from forWherein,The information contribution degree
Factor includes: the mean value of predictionWith second-order central away fromWherein, the second-order central takes away from characterization stochastic variable
The degree of scatter of value, value is bigger, and it is bigger that deviation possibility occurs in prediction;
S3: combination forecasting is established based on maximum informational entropy
It regards photovoltaic generation power as stochastic variable, is indicated with X, establish combination forecasting with maximum informational entropy:
Wherein, N is prediction model number, N=2 in the combination forecasting;
piFor weight coefficient of the prediction result in combination forecasting of Individual forecast model;
For predict photovoltaic generation power each rank center away from;
K be each rank center away from order, K=2;
As available from the above equation:
So
And then it is available:
λ is solved by above formula equationk(k=1,2 ..., K), then λkSubstitution acquires λ0;λ0, λ1..., λkSubstitution solves
pi, then by piSubstitution obtains H (X);
S4: it chooses average absolute value percentage error MAPE and assesses power prediction value, verify the combination forecasting
Precision of prediction;Wherein,
PiFor photovoltaic generation power actual value;
PfFor photovoltaic power predicted value;
N is data count, then carries out analysis comparison to the error of the power prediction value, verifies the combination forecasting
Validity and practicability.
Referring to attached drawing 3, historical power curve and prediction power curve comparison can be seen that be obtained by the method for the invention
Power prediction value and power actual value difference very little further prove that combination forecasting method accuracy of the invention is high, compared to
Individual forecast model substantially increases the precision of prediction of photovoltaic generation power.
The above is presently preferred embodiments of the present invention, but the present invention should not be limited to disclosed in the embodiment
Content.So all do not depart from the lower equivalent or modification completed of spirit disclosed in this invention, the model that the present invention protects is both fallen within
It encloses.
Claims (4)
1. a kind of combination forecasting of photovoltaic efficiency, which is characterized in that including gray model and BP neural network model,
And the weight ratio of the gray model and the BP neural network model is determined by maximum informational entropy.
2. a kind of combination forecasting method of photovoltaic efficiency, which comprises the following steps:
S1: the actual historical power output time series of photovoltaic plant and corresponding weather forecast weather information parameter are made respectively
For the input of gray model in combination forecasting described in claim 1 and BP neural network model, subsequent time photovoltaic plant
Performance number is respectively as output;
S2: by the value of output described in step S1 respectively multiplied by the weighted value of the gray model and the BP neural network model,
Obtain power prediction value;
S3: the power prediction value described in step S2 is assessed, is verified.
3. the combination forecasting method of photovoltaic efficiency as claimed in claim 2, which is characterized in that in step S1, the weather
Forecast that weather information parameter includes intensity of illumination, temperature, humidity, wind speed, the installation of photovoltaic array, solar incident angle degree and conversion
Efficiency.
4. the combination forecasting method of photovoltaic efficiency as claimed in claim 2, which is characterized in that in step S3, choose average
Absolute value percentage error carries out power prediction value described in analysis comparative evaluation, verifies the prediction essence of the combination forecasting
Degree.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810737933.7A CN109190786A (en) | 2018-07-06 | 2018-07-06 | A kind of combination forecasting method of photovoltaic efficiency |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810737933.7A CN109190786A (en) | 2018-07-06 | 2018-07-06 | A kind of combination forecasting method of photovoltaic efficiency |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109190786A true CN109190786A (en) | 2019-01-11 |
Family
ID=64936123
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810737933.7A Pending CN109190786A (en) | 2018-07-06 | 2018-07-06 | A kind of combination forecasting method of photovoltaic efficiency |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109190786A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110190810A (en) * | 2019-06-04 | 2019-08-30 | 西安工程大学 | The measurement method and application configuration modification method that filth causes photo-voltaic power supply power to lose |
CN110675278A (en) * | 2019-09-18 | 2020-01-10 | 上海电机学院 | Photovoltaic power short-term prediction method based on RBF neural network |
CN110751326A (en) * | 2019-10-17 | 2020-02-04 | 江苏远致能源科技有限公司 | Photovoltaic day-ahead power prediction method and device and storage medium |
CN111339157A (en) * | 2020-02-20 | 2020-06-26 | 南方电网科学研究院有限责任公司 | Method, system and equipment for calculating and predicting daily operation efficiency of power distribution network |
-
2018
- 2018-07-06 CN CN201810737933.7A patent/CN109190786A/en active Pending
Non-Patent Citations (2)
Title |
---|
夏冬 等: "一种新型的风电功率预测综合模型", 《电工技术学报》 * |
涂雨曦: "基于改进神经网络的光伏出力预测", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110190810A (en) * | 2019-06-04 | 2019-08-30 | 西安工程大学 | The measurement method and application configuration modification method that filth causes photo-voltaic power supply power to lose |
CN110675278A (en) * | 2019-09-18 | 2020-01-10 | 上海电机学院 | Photovoltaic power short-term prediction method based on RBF neural network |
CN110751326A (en) * | 2019-10-17 | 2020-02-04 | 江苏远致能源科技有限公司 | Photovoltaic day-ahead power prediction method and device and storage medium |
CN111339157A (en) * | 2020-02-20 | 2020-06-26 | 南方电网科学研究院有限责任公司 | Method, system and equipment for calculating and predicting daily operation efficiency of power distribution network |
CN111339157B (en) * | 2020-02-20 | 2023-05-16 | 南方电网科学研究院有限责任公司 | Method, system and equipment for calculating and predicting daily operation efficiency of power distribution network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Liu et al. | Forecasting power output of photovoltaic system using a BP network method | |
CN109190786A (en) | A kind of combination forecasting method of photovoltaic efficiency | |
CN108921339B (en) | Quantile regression-based photovoltaic power interval prediction method for genetic support vector machine | |
CN110880789B (en) | Economic dispatching method for wind power and photovoltaic combined power generation system | |
CN104978611A (en) | Neural network photovoltaic power generation output prediction method based on grey correlation analysis | |
Kani et al. | A new ANN-based methodology for very short-term wind speed prediction using Markov chain approach | |
Kolhe et al. | GA-ANN for short-term wind energy prediction | |
CN109636054A (en) | Solar energy power generating amount prediction technique based on classification and error combination prediction | |
CN110212551B (en) | Micro-grid reactive power automatic control method based on convolutional neural network | |
CN110163444A (en) | A kind of water demand prediction method based on GASA-SVR | |
CN111340305A (en) | Building operation energy consumption prediction method | |
CN106611243A (en) | Residual correction method for wind speed prediction based on GARCH (Generalized ARCH) model | |
Luo et al. | Short-term photovoltaic generation forecasting based on similar day selection and extreme learning machine | |
CN104915727A (en) | Multi-dimensional isomorphic heterogeneous BP neural network optical power ultrashort-term prediction method | |
CN110956304A (en) | Distributed photovoltaic power generation capacity short-term prediction method based on GA-RBM | |
CN112836876B (en) | Power distribution network line load prediction method based on deep learning | |
CN114037209A (en) | Comprehensive benefit analysis method and device for distributed photovoltaic access direct-current power distribution system | |
CN116485139A (en) | Short-term photovoltaic power generation amount prediction method based on multi-feature fusion | |
CN115034608A (en) | Distribution network risk assessment method based on distribution network element and neural network | |
Zhang et al. | A hybrid load forecasting method based on neural network in smart grid | |
CN111932401A (en) | Multi-party trust service interaction method based on block chain | |
CN116523148B (en) | Distribution network distribution transformer overload early warning method, device and equipment | |
Di | Power system short term load forecasting based on weather factors | |
CN116596279B (en) | Intelligent park energy consumption scheduling system | |
Mohammed et al. | Ultra-short-term wind power prediction using a hybrid model |
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: 20190111 |