CN112927097A - Photovoltaic power generation short-term prediction method based on GRA-ABC-Elman model - Google Patents
Photovoltaic power generation short-term prediction method based on GRA-ABC-Elman model Download PDFInfo
- Publication number
- CN112927097A CN112927097A CN202110124002.1A CN202110124002A CN112927097A CN 112927097 A CN112927097 A CN 112927097A CN 202110124002 A CN202110124002 A CN 202110124002A CN 112927097 A CN112927097 A CN 112927097A
- Authority
- CN
- China
- Prior art keywords
- meteorological
- power generation
- elman
- neural network
- photovoltaic power
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000010248 power generation Methods 0.000 title claims abstract description 38
- 238000013528 artificial neural network Methods 0.000 claims abstract description 38
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 14
- 238000012549 training Methods 0.000 claims abstract description 10
- 230000002159 abnormal effect Effects 0.000 claims abstract description 4
- 238000004458 analytical method Methods 0.000 claims abstract description 4
- 235000012907 honey Nutrition 0.000 claims description 28
- 239000013598 vector Substances 0.000 claims description 27
- 241000257303 Hymenoptera Species 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000005284 excitation Effects 0.000 claims description 6
- 210000002569 neuron Anatomy 0.000 claims description 6
- 238000010219 correlation analysis Methods 0.000 claims description 4
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 claims description 3
- 238000005286 illumination Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 2
- 238000011161 development Methods 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000000053 physical method Methods 0.000 description 2
- 230000035945 sensitivity Effects 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- 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"
-
- 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
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Marketing (AREA)
- Biophysics (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Tourism & Hospitality (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Development Economics (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a short-term photovoltaic power generation power prediction method based on a GRA-ABC-Elman model, which mainly comprises the following steps: collecting historical power generation power data of a photovoltaic power station and meteorological information of a meteorological station in a corresponding time period as meteorological and power sample data; removing abnormal and repeated data in meteorological data, performing grey correlation degree analysis on all features, and extracting important features as a training input set of a neural network; establishing an Elman neural network, determining a network topology structure, and initializing various parameters of the network topology structure; continuously optimizing the weight and the threshold of the Elman neural network by using an artificial bee colony algorithm; establishing a photovoltaic power generation short-term prediction model by using the optimized network and inputting meteorological-power sample data for training; and inputting the meteorological information of the day to be predicted into the trained prediction model, and predicting the generated power of the day to be predicted. The method can accurately predict the short-term power generation power of the photovoltaic power station, and is convenient for the dispatching and operation of the power grid.
Description
Technical Field
The invention relates to the field of photovoltaic power generation, in particular to a short-term photovoltaic power generation power prediction method based on a GRA-ABC-Elman model.
Background
In recent years, the problem of environmental pollution is becoming more serious, the requirements of countries in the world on energy conservation and emission reduction are continuously improved, and the global demand for electric energy is increasing year by year, so that the development and utilization of renewable energy are important ways for solving the energy crisis. The photovoltaic power generation rapidly becomes a hot point of world attention and research with unique charm of cleanness, no pollution and inexhaustibility, however, the output of the photovoltaic power generation has strong fluctuation and intermittency due to meteorological environment factors, the stability of the photovoltaic power generation after grid connection is influenced, and the photovoltaic power generation is difficult to be applied in a large scale. Therefore, the generated power of the photovoltaic power station is predicted, so that the adverse effect of photovoltaic grid connection on a power system can be reduced; and the photovoltaic power generation is combined with power grid dispatching, load distribution and the like, the whole power system can be planned, the stability and the utilization rate of the system are greatly improved, and the method has important significance for photovoltaic power stations and power systems.
The photovoltaic power generation output power can adopt various prediction methods. According to the source of the prediction information, a physical method, a statistical method and an artificial intelligence method can be divided. The physical prediction method is to predict by means of a sunlight irradiation transfer equation, a photovoltaic module operation equation and the like according to geographic information and meteorological information of the position of the photovoltaic power station. The method does not depend on historical data, can predict data acquired on site according to geographic information and parameters of the photovoltaic module, but is difficult to model, and the parameters of the photovoltaic module can change slowly, so that the method is limited in anti-interference capability and not strong in robustness. A representative physical prediction method includes: ASHRAE model on sunny days, HOTTEL model, REST model, Nielsen model with cloud weather, etc. Different from a physical method, the statistical prediction method depends on historical data, and searches a statistical rule according to input historical meteorological data, power data and operation conditions to predict the output power. A representative statistical prediction method includes: time series methods, regression analysis methods, grey theory methods, fuzzy theory methods, space-time correlation methods, and the like.
In recent years, an artificial intelligence prediction method mainly based on an artificial neural network has attracted much attention, is an operational model formed by interconnection of a large number of nodes, has high fitting and generalization capabilities, and has the characteristic of short training time. However, the conventional neural network only has feedforward and no feedback, has poor sensitivity to historical data and weak capability of processing dynamic information, and does not have the capability of adapting to time-varying characteristics. Therefore, the Elman neural network is adopted, and the network is characterized in that a carrying layer is added to an implicit layer on the basis of the traditional BP network and serves as a one-step delay operator to achieve the purpose of memorizing, so that various defects and problems of the traditional neural network can be effectively solved.
The photovoltaic power station short-term power prediction method is various, a hybrid improved Kmeans-GRA-Elman algorithm is a model with a novel prediction method and an accurate prediction result, but compared with the method for performing parameter optimization on the weight and the threshold of the Elman neural network by using the artificial bee colony algorithm, the method effectively solves the problems that the network generalization capability is not strong and the Elman neural network is easy to fall into a minimum value and the like due to random initialization of the artificial neural network weight and the threshold, the development capability of the bee colony algorithm is enhanced, the optimal solution of the Elman neural network weight and the threshold can be quickly found, and the predicted photovoltaic power station short-term power is more accurate and effective. At present, no study for applying the GRA-ABC-Elman model to photovoltaic power station short-term power prediction is found in published documents and patents.
Disclosure of Invention
In view of the above, the invention aims to provide a short-term photovoltaic power generation power prediction method based on a GRA-ABC-Elman model, which can predict the short-term photovoltaic power generation power more quickly and accurately.
In order to solve the problems in the prior art, the technical scheme adopted by the invention is as follows:
the photovoltaic power generation short-term prediction method based on the GRA-ABC-Elman model comprises the following steps:
collecting historical power generation power data of a photovoltaic power station and meteorological information of a meteorological station in a corresponding time period as meteorological and power sample data;
eliminating abnormal and repeated data in the meteorological data, performing grey correlation degree analysis on all features in the meteorological data, and extracting important features in the meteorological data to be used as a training input set of an Elman neural network;
establishing an Elman neural network, determining the network topology structure and the number of nodes of an input layer, an output layer and a hidden layer, and initializing all parameters of the Elman neural network;
fourthly, continuously optimizing the weight and the threshold of the Elman neural network by utilizing an artificial bee colony algorithm;
establishing a photovoltaic power generation short-term prediction model by using the optimized neural network, and inputting meteorological and power sample data to train the prediction model;
sixthly, inputting the weather information of the day to be predicted, which is obtained from the numerical weather forecast NWP, into the trained prediction model, and predicting the power generation power of the day to be predicted.
Further, the meteorological information provided by the meteorological station in the step (i) mainly comprises illumination radiance, temperature, humidity, wind speed and atmospheric pressure.
Further, the grey correlation analysis is performed on all the characteristics in the meteorological data in the step two, and the method mainly comprises the following steps:
a. data normalization:
in the formula (1), xA(i) Representing a value obtained after the meteorological sample data is normalized; x is the number ofmax、xminRespectively representing the maximum value and the minimum value in the meteorological sample data, wherein x represents the meteorological sample data value at a certain moment;
b. and (3) calculating a correlation coefficient:
in the formula (2), ζA(i) Representing a correlation coefficient; rho represents an adjustable coefficient for controlling the discrimination of the correlation coefficient;
c. calculating the relevance:
d. and extracting the important features with the relevance close to 1 as an Elman neural network training input set.
Further, the nonlinear state space expression of the Elman neural network established in the third step is as follows:
y(k)=g(w2h(k)) (4)
h(k)=f(w1u(k-1)+w3xc(k)) (5)
xc(k)=h(k-1) (6)
in the formulas (4), (5) and (6), f (—) is the excitation function of hidden layer neurons; g (—) is the excitation function of the neurons of the output layer; y is an m-dimensional output node vector; h is an n-dimensional hidden layer node unit vector; u is an r-dimensional input node vector; x is the number ofcFeeding back a state vector for the n-dimensional carrying layer; w is a1,w2,w3Respectively representing the weight from the input layer to the hidden layer, the weight from the hidden layer to the output layer, and the weight from the receiving layer to the hidden layer.
Further, the step IV of continuously optimizing the weight and the threshold of the Elman neural network by using the artificial bee colony algorithm comprises the following steps:
a. initializing a honey source vector:
xmi=li+rand(0,1)*(ui-li) (7)
in the formula (7), xmiRepresenting all honey source vectors, wherein m is 1,2.. N represents a total of N solution vectors, and i is 1,2.. N represents that each solution vector contains N variables; u. ofiAnd liRespectively represent xmiMaximum and minimum boundary values of;
calculating the value fit of each initial honey source fitnessm(xm) The calculation formula is as follows:
b. bee hiring stage:
in the formula (9), vmiRepresenting a new honey source vector found by the employed bees; x is the number ofkiRepresenting a randomly selected honey source;a random number representing a degree of control disturbance;
c. following the bee stage:
in the formula (10), PmA probability indicating whether a following bee chooses to follow a employed bee;
d. and (3) a bee scouting stage:
formula (9) is calculated together with formula (9) for hiring bees to find new honey sources.
The invention has the advantages and beneficial effects that:
according to the photovoltaic power generation short-term prediction method based on the GRA-ABC-Elman model, parameter optimization is carried out on the weight and the threshold of the Elman neural network by using the artificial bee colony algorithm, the problems that the network generalization capability is not strong and the Elman neural network is easy to fall into a minimum value and the like due to random initialization of the weight and the threshold of the artificial neural network are effectively solved, the development capability of the bee colony algorithm is enhanced, the optimal solution of the weight and the threshold of the Elman neural network can be quickly found, the photovoltaic power generation short-term power generation power can be quickly and accurately predicted, and the photovoltaic power generation short-term prediction method has the advantages of time-varying characteristic adapting capability and high sensitivity. The method is combined with power grid dispatching, load distribution and the like, and the whole power system can be planned and operated, so that the stability and the utilization rate of the system are greatly improved, and the dispatching operation of the power grid is facilitated.
Drawings
The invention is described in detail below with reference to the following figures and examples:
FIG. 1 is a flow chart of an implementation process of the photovoltaic power generation short-term prediction method based on the GRA-ABC-Elman model;
FIG. 2 is a flow chart of data preprocessing of the photovoltaic power generation power short-term prediction method based on the GRA-ABC-Elman model;
FIG. 3 is a schematic structural diagram of an Elman neural network;
FIG. 4 is a flow chart of an artificial bee colony optimization algorithm;
Detailed Description
As shown in FIG. 1, the photovoltaic power generation power short-term prediction method based on the GRA-ABC-Elman model, wherein GRA refers to grey correlation analysis, ABC refers to artificial bee colony algorithm, comprises the following steps:
collecting historical power generation power data of a photovoltaic power station and meteorological information of a meteorological station in a corresponding time period as meteorological and power sample data. The meteorological information provided by the meteorological station mainly comprises illumination radiance, temperature, humidity, wind speed and atmospheric pressure.
And secondly, eliminating abnormal and repeated data in the meteorological data, and performing grey correlation degree analysis on all the features to extract important features as a training input set of the neural network.
And thirdly, establishing an Elman neural network, determining a network topology structure and initializing various parameters of the network topology structure.
And fourthly, continuously optimizing the weight and the threshold of the Elman neural network by utilizing an artificial bee colony algorithm.
And fifthly, establishing a photovoltaic power generation short-term prediction model by using the optimized network and inputting meteorological-power sample data for training.
Sixthly, inputting the weather information of the day to be predicted, which is obtained from the numerical weather forecast NWP, into the trained prediction model, and predicting the power generation power of the day to be predicted.
As shown in fig. 2, the grey correlation analysis of all the features in the weather sample data in the second step mainly includes the following steps:
a. and (4) normalizing the data, reducing the difference of absolute numerical values of the data, and scaling various data to a range of [0, 1 ]. The specific formula is as follows:
in the formula (1), xA(i) And representing the value obtained by normalizing the meteorological sample data at the ith moment of the characteristic A. x is the number ofmaxRepresents the maximum value, x, in the meteorological sample data at the ith timeminAnd x represents the value of the meteorological sample data at the ith moment.
b. And (3) calculating a correlation coefficient:
in the formula (2), ζA(i) Representing the correlation coefficient of the ith moment in the characteristic A sample data; rho represents an adjustable coefficient for controlling the discrimination of the correlation coefficient, and the value is 0.5.
c. Carrying out average calculation on each correlation coefficient to obtain the correlation degree, wherein the calculation formula is as follows;
d. and extracting the important features with the relevance close to 1 as an Elman neural network training input set.
As shown in fig. 3, the nonlinear state space expression of the Elman neural network established in step three is:
y(k)=g(w2h(k)) (4)
h(k)=f(w1u(k-1)+w3xc(k)) (5)
xc(k)=h(k-1) (6)
in the formulas (4), (5) and (6), f (—) is the excitation function of hidden layer neurons; g (—) is the excitation function of the neurons of the output layer; y ═ y1,y2,...,ym]Outputting a node vector for m dimensions; h ═ h1,h2,...,hn]A hidden layer node unit vector is expressed in n dimensions; u ═ u1,u2,...,ur]Inputting a node vector in an r dimension; x is the number ofc=[xc1,xc2,...,xcn]Feeding back a state vector for the n-dimensional carrying layer; w is a1,w2,w3Respectively representing the weight from the input layer to the hidden layer, the weight from the hidden layer to the output layer, and the weight from the receiving layer to the hidden layer.
As shown in fig. 4, the step of optimizing the weight and the threshold of the Elman neural network by the artificial bee colony algorithm in the step (iv) is as follows:
a. initializing all honey source vectors, wherein the initialization parameters comprise: the number and the position of honey sources, the maximum iteration number iter max and a threshold limit;
xmi=li+rand(0,1)*(ui-li) (7)
in the formula (7), xmiRepresenting all honey source vectors, wherein m is 1,2.. N represents a total of N solution vectors, and i is 1,2.. N represents that each solution vector contains N variables; u. ofiAnd liRespectively represent xmiMaximum and minimum boundary values.
Calculating the value fit of each initial honey source fitnessm(xm) The calculation formula is as follows, and the value of the honey source fitness represents the quality of the generated solution;
b. employing bees to search for a new honey source, namely generating a new solution;
in the formula (9), vmiRepresenting a new source of honey found by the employed bees; x is the number ofkiRepresenting a randomly selected honey source;a random number representing a degree of control disturbance, taken as [ -1, 1 [ ]]。
Calculating the value fit of the fitness of the new solutionm(xm) Comparing the solution with the old solution, and if the fitness value of the new solution is better than that of the old solution, hiring bees to remember that the new solution forgets the old solution; otherwise, the old solution will be retained.
c. A bee following stage;
the follower bees in the honeycomb analyze the information brought back by the hired bee, and a roulette strategy is adopted to select the honey source for tracking exploitation, and the calculation formula is as follows:
in the formula (10), PmA probability indicating whether a following bee chooses to follow a employed bee; fitm(xm) Representing the value of the fitness of the honey source.
If the calculated selection probability is larger than the randomly generated numerical value, the following bees select to follow the employed bees, otherwise, the following bees are converted into scout bees, and a new honey source is searched.
d. And (3) a bee scouting stage:
the honey source has a parameter trim, the initial value is 0, and the value of the mined primary trim is added with 1. If a honey source is mined for a plurality of times, namely the value of the real is too high and exceeds the preset threshold limit, the honey source needs to be abandoned, and the scout bees, which are also employed bees, need to find a new food source v againmi. Formula (9) is calculated together with formula (9) for hiring bees to find new honey sources.
Claims (5)
1. The photovoltaic power generation short-term prediction method based on the GRA-ABC-Elman model is characterized by comprising the following steps of:
collecting historical power generation power data of a photovoltaic power station and meteorological information of a meteorological station in a corresponding time period as meteorological and power sample data;
eliminating abnormal and repeated data in the meteorological data, performing grey correlation degree analysis on all features in the meteorological data, and extracting important features in the meteorological data to be used as a training input set of an Elman neural network;
establishing an Elman neural network, determining the network topology structure and the number of nodes of an input layer, an output layer and a hidden layer, and initializing all parameters of the Elman neural network;
fourthly, continuously optimizing the weight and the threshold of the Elman neural network by utilizing an artificial bee colony algorithm;
establishing a photovoltaic power generation short-term prediction model by using the optimized neural network, and inputting meteorological and power sample data to train the prediction model;
sixthly, inputting the weather information of the day to be predicted, which is obtained from the numerical weather forecast NWP, into the trained prediction model, and predicting the power generation power of the day to be predicted.
2. The GRA-ABC-Elman model-based short-term prediction method of photovoltaic power generation power as claimed in claim 1, wherein: the meteorological information provided by the meteorological station in the step I mainly comprises illumination radiance, temperature, humidity, wind speed and atmospheric pressure.
3. The GRA-ABC-Elman model-based short-term prediction method of photovoltaic power generation power as claimed in claim 1, wherein: performing grey correlation analysis on all the characteristics in the meteorological data in the second step, mainly comprising the following steps:
a. data normalization:
in the formula (1), xA(i) Representing a value obtained after the meteorological sample data is normalized; x is the number ofmax、xminRespectively representing the maximum value and the minimum value in the meteorological sample data, wherein x represents the meteorological sample data value at a certain moment;
b. and (3) calculating a correlation coefficient:
in the formula (2), ζA(i) Representing a correlation coefficient; rho represents an adjustable coefficient for controlling the discrimination of the correlation coefficient;
c. calculating the relevance:
d. and extracting the important features with the relevance close to 1 as an Elman neural network training input set.
4. The GRA-ABC-Elman model-based short-term prediction method of photovoltaic power generation power as claimed in claim 1, wherein: the nonlinear state space expression of the Elman neural network established in the step three is as follows:
y(k)=g(w2h(k)) (4)
h(k)=f(w1u(k-1)+w3xc(k)) (5)
xc(k)=h(k-1) (6)
in the formulas (4), (5) and (6), f (—) is the excitation function of hidden layer neurons; g (—) is the excitation function of the neurons of the output layer; y is an m-dimensional output node vector; h is an n-dimensional hidden layer node unit vector; u is an r-dimensional input node vector; x is the number ofcFeeding back a state vector for the n-dimensional carrying layer; w is a1,w2,w3Respectively representing the weight from the input layer to the hidden layer, the weight from the hidden layer to the output layer, and the weight from the receiving layer to the hidden layer.
5. The GRA-ABC-Elman model-based short-term prediction method of photovoltaic power generation power as claimed in claim 1, wherein: the step IV of continuously optimizing the weight and the threshold of the Elman neural network by utilizing the artificial bee colony algorithm comprises the following steps:
a. initializing a honey source vector:
xmi=li+rand(0,1)*(ui-li) (7)
in the formula (7), xmiRepresenting all honey source vectors, wherein m is 1,2.. N represents a total of N solution vectors, and i is 1,2.. N represents that each solution vector contains N variables; u. ofiAnd liRespectively represent xmiMaximum and minimum boundary values of;
calculating the value fit of each initial honey source fitnessm(xm) The calculation formula is as follows:
b. bee hiring stage:
in the formula (9), vmiRepresenting a new honey source vector found by the employed bees; x is the number ofkiRepresenting a randomly selected honey source;a random number representing a degree of control disturbance;
c. following the bee stage:
in the formula (10), PmA probability indicating whether a following bee chooses to follow a employed bee;
d. and (3) a bee scouting stage:
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110124002.1A CN112927097A (en) | 2021-01-29 | 2021-01-29 | Photovoltaic power generation short-term prediction method based on GRA-ABC-Elman model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110124002.1A CN112927097A (en) | 2021-01-29 | 2021-01-29 | Photovoltaic power generation short-term prediction method based on GRA-ABC-Elman model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112927097A true CN112927097A (en) | 2021-06-08 |
Family
ID=76168398
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110124002.1A Pending CN112927097A (en) | 2021-01-29 | 2021-01-29 | Photovoltaic power generation short-term prediction method based on GRA-ABC-Elman model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112927097A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113507128A (en) * | 2021-07-14 | 2021-10-15 | 国网青海省电力公司 | Near-field reactive power optimization configuration method for extra-high voltage direct current converter station |
CN113762646A (en) * | 2021-10-15 | 2021-12-07 | 国网山东省电力公司菏泽供电公司 | Photovoltaic short-term power intelligent prediction method and system |
CN115115124A (en) * | 2022-07-13 | 2022-09-27 | 国网江苏省电力有限公司营销服务中心 | Distributed photovoltaic output prediction method and device based on DTW meteorological offset correction |
CN117117859A (en) * | 2023-10-20 | 2023-11-24 | 华能新能源股份有限公司山西分公司 | Photovoltaic power generation power prediction method and system based on neural network |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108734321A (en) * | 2018-05-17 | 2018-11-02 | 江苏大学 | A kind of short-term load forecasting method based on the Elman neural networks for improving ABC algorithms |
CN108876013A (en) * | 2018-05-29 | 2018-11-23 | 福州大学 | One kind being based on best similar day and Elman neural fusion photovoltaic plant short term power prediction technique |
CN109002915A (en) * | 2018-07-13 | 2018-12-14 | 福州大学 | Photovoltaic plant short term power prediction technique based on Kmeans-GRA-Elman model |
CN109165774A (en) * | 2018-08-07 | 2019-01-08 | 燕山大学 | A kind of short-term photovoltaic power prediction technique |
CN111967079A (en) * | 2020-08-17 | 2020-11-20 | 河海大学 | Foundation pit deformation prediction method based on improved artificial bee colony algorithm and BP neural network |
AU2020104000A4 (en) * | 2020-12-10 | 2021-02-18 | Guangxi University | Short-term Load Forecasting Method Based on TCN and IPSO-LSSVM Combined Model |
-
2021
- 2021-01-29 CN CN202110124002.1A patent/CN112927097A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108734321A (en) * | 2018-05-17 | 2018-11-02 | 江苏大学 | A kind of short-term load forecasting method based on the Elman neural networks for improving ABC algorithms |
CN108876013A (en) * | 2018-05-29 | 2018-11-23 | 福州大学 | One kind being based on best similar day and Elman neural fusion photovoltaic plant short term power prediction technique |
CN109002915A (en) * | 2018-07-13 | 2018-12-14 | 福州大学 | Photovoltaic plant short term power prediction technique based on Kmeans-GRA-Elman model |
CN109165774A (en) * | 2018-08-07 | 2019-01-08 | 燕山大学 | A kind of short-term photovoltaic power prediction technique |
CN111967079A (en) * | 2020-08-17 | 2020-11-20 | 河海大学 | Foundation pit deformation prediction method based on improved artificial bee colony algorithm and BP neural network |
AU2020104000A4 (en) * | 2020-12-10 | 2021-02-18 | Guangxi University | Short-term Load Forecasting Method Based on TCN and IPSO-LSSVM Combined Model |
Non-Patent Citations (1)
Title |
---|
李刚,刘佳林,王腾飞,丁坤: "基于相似日理论和IPSO-Elman 模型的短期光伏发电功率预测", 测控技术, vol. 39, no. 2, pages 1 - 8 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113507128A (en) * | 2021-07-14 | 2021-10-15 | 国网青海省电力公司 | Near-field reactive power optimization configuration method for extra-high voltage direct current converter station |
CN113507128B (en) * | 2021-07-14 | 2023-07-21 | 国网青海省电力公司 | Near-field reactive power optimal configuration method for extra-high voltage direct current converter station |
CN113762646A (en) * | 2021-10-15 | 2021-12-07 | 国网山东省电力公司菏泽供电公司 | Photovoltaic short-term power intelligent prediction method and system |
CN115115124A (en) * | 2022-07-13 | 2022-09-27 | 国网江苏省电力有限公司营销服务中心 | Distributed photovoltaic output prediction method and device based on DTW meteorological offset correction |
CN115115124B (en) * | 2022-07-13 | 2024-01-30 | 国网江苏省电力有限公司营销服务中心 | Distributed photovoltaic output prediction method and device based on DTW meteorological offset correction |
CN117117859A (en) * | 2023-10-20 | 2023-11-24 | 华能新能源股份有限公司山西分公司 | Photovoltaic power generation power prediction method and system based on neural network |
CN117117859B (en) * | 2023-10-20 | 2024-01-30 | 华能新能源股份有限公司山西分公司 | Photovoltaic power generation power prediction method and system based on neural network |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109002915B (en) | Photovoltaic power station short-term power prediction method based on Kmeans-GRA-Elman model | |
CN112927097A (en) | Photovoltaic power generation short-term prediction method based on GRA-ABC-Elman model | |
CN107766990B (en) | Method for predicting power generation power of photovoltaic power station | |
CN110705789A (en) | Photovoltaic power station short-term power prediction method | |
CN105574615B (en) | wavelet-BP neural network wind power prediction method based on spatial correlation and GA | |
CN111414717A (en) | XGboost-L ightGBM-based unit power prediction method | |
CN110717610B (en) | Wind power prediction method based on data mining | |
CN109086928A (en) | Photovoltaic plant realtime power prediction technique based on SAGA-FCM-LSSVM model | |
CN109523084A (en) | A kind of ultrashort-term wind power prediction method based on pivot analysis and machine learning | |
CN112215428B (en) | Photovoltaic power generation power prediction method and system based on error correction and fuzzy logic | |
Kolhe et al. | GA-ANN for short-term wind energy prediction | |
CN112990558B (en) | Meteorological temperature and illumination prediction method based on deep migration learning | |
CN110009135B (en) | Wind power prediction method based on width learning | |
CN112149905A (en) | Photovoltaic power station short-term power prediction method based on wavelet transformation and wavelet neural network | |
Xiao et al. | Online sequential extreme learning machine algorithm for better predispatch electricity price forecasting grids | |
CN115115127A (en) | Low-voltage transformer area intelligent regulation and control method and system based on deep learning | |
Luo et al. | Short-term photovoltaic generation forecasting based on similar day selection and extreme learning machine | |
Meng et al. | A new PV generation power prediction model based on GA-BP neural network with artificial classification of history day | |
CN116187540A (en) | Wind power station ultra-short-term power prediction method based on space-time deviation correction | |
Joudaki | Long-term wind speed and power forecasting based on lstm: A comprehensive study | |
CN117875752A (en) | Power system flexible operation domain assessment method based on self-organizing map decision tree | |
CN109359671B (en) | Classification intelligent extraction method for hydropower station reservoir dispatching rules | |
CN116499023A (en) | Intelligent control method and system for geothermal coupling solar heating station | |
Xu et al. | One-day ahead forecast of PV output based on deep belief network and weather classification | |
Lv et al. | Research on short-term load forecasting approach for 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 |