CN107947738A - A kind of Forecasting Methodology of solar energy unmanned plane cell plate voltage - Google Patents
A kind of Forecasting Methodology of solar energy unmanned plane cell plate voltage Download PDFInfo
- Publication number
- CN107947738A CN107947738A CN201711304116.4A CN201711304116A CN107947738A CN 107947738 A CN107947738 A CN 107947738A CN 201711304116 A CN201711304116 A CN 201711304116A CN 107947738 A CN107947738 A CN 107947738A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- neural network
- population
- factor
- 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 20
- 238000013528 artificial neural network Methods 0.000 claims abstract description 53
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 27
- 238000012549 training Methods 0.000 claims abstract description 25
- 230000003044 adaptive effect Effects 0.000 claims abstract description 19
- 238000005457 optimization Methods 0.000 claims abstract description 12
- 238000010606 normalization Methods 0.000 claims abstract description 4
- 210000004027 cell Anatomy 0.000 claims description 18
- 231100000219 mutagenic Toxicity 0.000 claims description 12
- 230000003505 mutagenic effect Effects 0.000 claims description 12
- 108090000623 proteins and genes Proteins 0.000 claims description 11
- 210000000349 chromosome Anatomy 0.000 claims description 9
- 210000002569 neuron Anatomy 0.000 claims description 8
- 238000005286 illumination Methods 0.000 claims description 7
- 230000006978 adaptation Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 230000035772 mutation Effects 0.000 claims description 3
- 238000005259 measurement Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000010429 evolutionary process Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000000052 comparative effect Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 210000004218 nerve net Anatomy 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000003321 amplification Effects 0.000 description 1
- 238000011157 data evaluation Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000002803 fossil fuel Substances 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000001172 regenerating effect Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000002945 steepest descent method Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02S—GENERATION OF ELECTRIC POWER BY CONVERSION OF INFRARED RADIATION, VISIBLE LIGHT OR ULTRAVIOLET LIGHT, e.g. USING PHOTOVOLTAIC [PV] MODULES
- H02S50/00—Monitoring or testing of PV systems, e.g. load balancing or fault identification
- H02S50/10—Testing of PV devices, e.g. of PV modules or single PV cells
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- 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
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B10/00—Integration of renewable energy sources in buildings
- Y02B10/10—Photovoltaic [PV]
-
- 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
- Y02E10/00—Energy generation through renewable energy sources
- Y02E10/50—Photovoltaic [PV] energy
-
- 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
Abstract
The present invention provides a kind of Forecasting Methodology of solar energy unmanned plane cell plate voltage.The Forecasting Methodology of the solar energy unmanned plane cell plate voltage includes the following steps:First, the operating parameter of solar panel is obtained, training set and forecast set are obtained according to the operating parameter, and the operational parameter data of solar panel is normalized during to fine day;2nd, BP neural network is built with reference to the training set after normalization;3rd, the initial weight and initial threshold of the BP neural network are obtained according to adaptive differential algorithm optimization;4th, obtained weights and threshold value initial weight and initial threshold as the BP neural network will be optimized, and the BP neural network is trained using the training set;5th, the data to be predicted in the forecast set are inputted in the BP neural network, is contrasted according to the predicted value that the BP neural network exports with actual value, and the accuracy of assessment prediction result.
Description
Technical field
The present invention relates in solar energy unmanned plane solar array voltage predict field, more particularly to a kind of solar energy without
The Forecasting Methodology of man-machine cell plate voltage.
Background technology
With social development, the problem of fossil fuel faces exhaustion, environment, also becomes increasingly conspicuous.To solve this problem, people
Class begins to focus on regenerative resource, and wherein the solar energy of green non-pollution just becomes researcher's focus of interest.As one
Kind efficiently uses the mode of solar energy, and solar energy unmanned plane has had a degree of development.
It is contemplated that the cell power generation of unmanned plane upper surface is had factor to be influenced be subject to very, voltage (power) by it is extraneous because
Element influences to change greatly, and can for a long time continue a journey and be affected to unmanned plane, therefore defeated to the solar panel of upper surface of the airfoil
It is highly important to go out voltage prediction.Solar cell output voltage and weather condition are closely related, in advance to output voltage into
Row prediction, and precision of prediction is improved, tool is of great significance.
The content of the invention
The defects of it is an object of the invention to for the prior art, there is provided a kind of solar energy unmanned plane cell plate voltage it is pre-
Survey method, by adaptive differential evolution algorithm (referred to as:SADE) Optimized BP Neural Network algorithm, so as to substantially improve BP god
Performance through network, improves the precision of predicted value.
Technical scheme is as follows:A kind of Forecasting Methodology of solar energy unmanned plane cell plate voltage includes following step
Suddenly:First, the operating parameter of solar panel is obtained, training set and forecast set are obtained according to the operating parameter, and to fine day
When solar panel operational parameter data be normalized;2nd, BP nerves are built with reference to the training set after normalization
Network;3rd, the initial weight and initial threshold of the BP neural network are obtained according to adaptive differential algorithm optimization;4th, will be excellent
Change the initial weight and initial threshold of obtained weights and threshold value as the BP neural network, and using the training set to institute
BP neural network is stated to be trained;5th, the data to be predicted in the forecast set are inputted in the BP neural network, according to institute
The predicted value for stating BP neural network output is contrasted with actual value, and the accuracy of assessment prediction result.
Preferably, in step 1, the operating parameter of solar panel is measured using solar energy light intensity meter and multimeter,
The operating parameter includes intensity of illumination, ambient temperature and voltage etc.;Moreover, intensity of illumination and ambient temperature are become as input
Amount, voltage is as output variable, operational parameter data when thus obtaining training set and forecast set and choosing fine day, and is returned
One change is handled.
Preferably, in step 2, according to the dimension of input data and output data, the hidden layer of BP neural network is determined
Neuron number and output layer neuron number, its calculation formula is as follows:L1=2m+1;L2=log2n;Wherein, L1 is implicit
Node layer number, L2 are output layer number of nodes, and m is input layer number, and n is output layer number of nodes.
Preferably, in step 3, specifically comprise the following steps:
Set the initial population of adaptive differential evolution algorithm:Based on above-mentioned training set and forecast set, using entity coding
Method, setting initial population size Np, gene dimension D, mutagenic factor F, the excursion for intersecting factor CR and each gene,
And using the error of BP neural network as ideal adaptation angle value in population, in solution space random initializtion population:
And generate at random:
Wherein, xi(0) i-th article of chromosome in the 0th generation in population, x are representedj,i(0) i-th article of chromosome in the 0th generation is represented
The gene of jth, rand (0,1) are represented in the equally distributed random number in (0,1) section;
Mutation operation:Individual difference is realized using difference strategy, randomly selects two different individuals in population, by its to
With treating that variation individual is synthesized into row vector after amount difference scaling, intermediate is produced:
υi(g+1)=xr1(g)+F·(xr2(g)-xr3(g))i≠r1≠r2≠r3
Wherein, F is mutagenic factor, xi(g) represent g for i-th of individual in population;
Crossover operation:To g for population xi(g) and its variation intermediate υi(g+1) crossover operation between individual is carried out:
Wherein, CR is to intersect the factor, jrandFor the random integers of [1,2 ... D];
Selection operation:The individual for entering population of future generation is selected using greedy algorithm, i.e. optimum results are better than previous generation
The individual of chromosome enters population of future generation:
Judge whether the precision of optimum results meets the requirements or whether number of iterations reaches maximum, become if it is not, then returning
ETTHER-OR operation step;If it is, output optimization obtains the initial weight and initial threshold of the BP neural network.
Preferably, set the TSP question factor as:
Wherein, FmaxIt is expressed as mutagenic factor maximum, FminMutagenic factor minimum value is expressed as, Gm represents maximum iteration
Number, G is current iteration number.
Preferably, set it is adaptive intersect the factor as:
Wherein, CRmaxRepresent maximum and intersect the factor, CRminRepresent minimum and intersect the factor.
Preferably, in step 5, training time, iterative steps evaluation index are assessed using root-mean-square error,
Its formula is as follows:
Wherein, PiFor solar array voltage real output value;PfOutput valve is predicted for solar array voltage;N is data
Sum.
Technical solution provided by the invention has the advantages that:
The Forecasting Methodology of the solar energy unmanned plane cell plate voltage is by adaptive differential evolution algorithm to BP nerve nets
The initial value and threshold value of network optimize, and choose the inverse of mean square error of training data as each population and the score of individual
Function, by continuous convergent, alienation, iteration, exports optimum individual, and BP nerve nets are obtained after being decoded it according to coding rule
The initial weight and initial threshold of network, so that the defects of effectively overcoming its convergence rate slowly and being easily trapped into local optimum, is improved
The precision of prediction, realizes global optimization, can significantly increase prediction effect.
Brief description of the drawings
Fig. 1 is the flow diagram of the Forecasting Methodology of solar energy unmanned plane cell plate voltage provided in an embodiment of the present invention;
Fig. 2 is the flow diagram of adaptive differential evolution algorithm;
Fig. 3 is BP neural network topological structure schematic diagram;
Fig. 4 is the schematic diagram of voltage prediction error amount before and after the optimization of difference algorithm (DE) Optimized BP Neural Network;
Fig. 5 is the signal of voltage prediction error amount before and after the optimization of adaptive differential algorithm (SADE) Optimized BP Neural Network
Figure;
Fig. 6 is the schematic diagram of voltage prediction value before and after the optimization of difference algorithm (DE) Optimized BP Neural Network;
Fig. 7 is the schematic diagram of adaptive differential algorithm (SADE) Optimized BP Neural Network voltage prediction.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, with reference to the accompanying drawings and embodiments, it is right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
The description of specific distinct unless the context otherwise, the present invention in element and component, the shape that quantity both can be single
Formula exists, and form that can also be multiple exists, and the present invention is defined not to this.Although step in the present invention with label into
Arrangement is gone, but is not used to limit the precedence of step, unless expressly stated the order of step or holding for certain step
Based on row needs other steps, otherwise the relative rank of step is adjustable.It is it is appreciated that used herein
Term "and/or" is related to and covers one of associated Listed Items or one or more of any and all possible group
Close.
As depicted in figs. 1 and 2, the Forecasting Methodology of solar energy unmanned plane cell plate voltage provided in an embodiment of the present invention includes
Following steps:
First, the operating parameter of solar panel is obtained, training set and forecast set are obtained according to the operating parameter, and it is right
The operational parameter data of solar panel is normalized during fine day.
Specifically, in step 1, using the operating parameter of light intensity meter and multimeter measurement solar panel, the fortune
Row parameter includes intensity of illumination, ambient temperature and voltage;Moreover, using intensity of illumination and ambient temperature as input variable, voltage
As output variable, operational parameter data when thus obtaining training set and forecast set and choosing fine day, and place is normalized
Reason.
For example, data include am6:00-pm6:The test parameter of 00 solar cell, Intensity of the sunlight and corresponding
Weather temperature.
2nd, BP neural network is built with reference to the training set after normalization.
As shown in figure 3, BP neural network is a kind of multilayer feedforward neural network of error back propagation, and include three-layer network
Network:Data input layer, hidden layer and output layer, wherein hidden layer can be made of one or more layers.Input signal is through input layer
Hidden layer is successively transferred to, finally reaches output layer.Hidden layer and output layer are completed according to the weights and threshold value of corresponding neuron
Data evaluation work.If output result and expected result be when having deviation, error signal is reversely successively transmitted to hidden layer and defeated
Enter layer, using gradient steepest descent method, adjust the weights and threshold value between each neuron.
Specifically, in step 2, according to the dimension of input data and output data, the hidden layer of BP neural network is determined
Neuron number and output layer neuron number, its calculation formula is as follows:
L1=2m+1;L2=log2n;
Wherein, L1 is node in hidden layer, and L2 is output layer number of nodes, and m is input layer number, and n is output node layer
Number.
For example, using environment temperature and intensity of illumination as input layer, voltage, therefore can be true as output layer, i.e. L=5
The structure for determining BP neural network is:Two input layers, five hidden layers and an output layer.
3rd, the initial weight and initial threshold of the BP neural network are obtained according to adaptive differential algorithm optimization.
Specifically, step 3 includes the following steps:
Set the initial population of adaptive differential evolution algorithm:Based on above-mentioned training set and forecast set, using entity coding
Method, setting initial population size Np, gene dimension D, mutagenic factor F, the excursion for intersecting factor CR and each gene,
And using the error of BP neural network as ideal adaptation angle value in population, in solution space random initializtion population:
And generate at random:
Wherein, xi(0) i-th article of chromosome in the 0th generation in population, x are representedj,i(0) i-th article of chromosome in the 0th generation is represented
The gene of jth, rand (0,1) are represented in the equally distributed random number in (0,1) section;
Mutation operation:Individual difference is realized using difference strategy, randomly selects two different individuals in population, by its to
With treating that variation individual is synthesized into row vector after amount difference scaling, intermediate is produced:
υi(g+1)=xr1(g)+F·(xr2(g)-xr3(g))i≠r1≠r2≠r3
Wherein, F is mutagenic factor, xi(g) represent g for i-th of individual in population;
Crossover operation:To g for population xi(g) and its variation intermediate υi(g+1) crossover operation between individual is carried out:
Wherein, CR is to intersect the factor, jrandFor the random integers of [1,2 ... D];
Selection operation:The individual for entering population of future generation is selected using greedy algorithm, i.e. optimum results are better than previous generation
The individual of chromosome enters population of future generation:
Judge whether the precision of optimum results meets the requirements or whether number of iterations reaches maximum, become if it is not, then returning
ETTHER-OR operation step;If it is, output optimization obtains the initial weight and initial threshold of the BP neural network.
It should be noted that in the evolutionary process of adaptive differential algorithm, in order to ensure the validity of solution, it is necessary to judge
Whether each gene meets boundary condition in chromosome, and if any the condition that is unsatisfactory for, then gene is regenerated with random device.
Moreover, in the evolutionary process of adaptive differential algorithm, since mutagenic factor determines the amplification ratio of bias vector
Example, if mutagenic factor is too big, algorithm search efficiency can be lower, precision is not high;If mutagenic factor is too small, it is impossible to meets population
Diversity, it is easy to occur precocity phenomenon, therefore, in the present embodiment, set the TSP question factor as:
Wherein, FmaxIt is expressed as mutagenic factor maximum, FminMutagenic factor minimum value is expressed as, Gm represents maximum iteration
Number, G is current iteration number.
In addition, in the evolutionary process of adaptive differential algorithm, due to intersecting the factor with the increase of iterations, intersect
Rate also dynamic change, initial stage, the larger intersection factor ensured the variation situation of global scope, and the later stage, less crossing-over rate was more paid close attention to
Local convergent, therefore, in the present embodiment, set the adaptive intersection factor as:
Wherein, CRmaxRepresent maximum and intersect the factor, CRminRepresent minimum and intersect the factor.
4th, obtained weights and threshold value initial weight and initial threshold as the BP neural network, and profit will be optimized
The BP neural network is trained with the training set.
Specifically, in step 4, obtained weights and threshold value initial value and threshold value as BP neural network will be optimized,
And BP neural network is trained using training set sample, is learnt.
For example, the learning rate of BP neural network is arranged to 0.1, the biography of iterations 800, hidden layer and output layer is trained
Defeated function selected as S types tangent function ' tansig ', network training function are ' trainlm '.
5th, the data to be predicted in the forecast set are inputted in the BP neural network, according to the BP neural network
The predicted value of output is contrasted with actual value, and the accuracy of assessment prediction result.
Specifically, in step 5, import data in Matlab, carry out simulation and prediction, will predict obtained result with
Actual comparison, assesses training time, iterative steps evaluation index using root-mean-square error, its formula is as follows:
Wherein, PiFor solar array voltage real output value;PfOutput valve is predicted for solar array voltage;N is data
Sum.
By BP neural network algorithm, differential evolution algorithm Optimized BP Neural Network (referred to as:DE Optimized BP Neural Networks) and
The adaptive differential evolution algorithm Optimized BP Neural Network that the embodiment of the present invention proposes is (referred to as:SADE Optimized BP Neural Networks),
Comparative result is predicted, it is as shown in the table:
Algorithm title | Root-mean-square error | Iterative steps | Training time (S) |
BP neural network | 0.0103 | 49 | 10.213 |
DE Optimized BP Neural Networks | 0.00784 | 37 | 9.541 |
SADE Optimized BP Neural Networks | 0.00368 | 19 | 10.427 |
As shown in figs. 4-7, according to comparative test result, adaptive differential evolution algorithm proposed by the invention optimization
BP neural network optimizing ability is stronger, better for prediction solar cell panel voltages.
It is obvious to a person skilled in the art that the invention is not restricted to the details of above-mentioned one exemplary embodiment, Er Qie
In the case of without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Profit requires rather than described above limits, it is intended that all in the implication and scope of the equivalency of claim by falling
Change is included in the present invention.Any reference numeral in claim should not be considered as to the involved claim of limitation.
Moreover, it will be appreciated that although the present specification is described in terms of embodiments, not each embodiment is only wrapped
Containing an independent technical solution, this narrating mode of specification is only that those skilled in the art should for clarity
Using specification as an entirety, the technical solutions in the various embodiments may also be suitably combined, forms those skilled in the art
It is appreciated that other embodiment.
Claims (7)
- A kind of 1. Forecasting Methodology of solar energy unmanned plane cell plate voltage, it is characterised in that:Include the following steps:First, the operating parameter of solar panel is obtained, training set and forecast set are obtained according to the operating parameter, and to fine day When solar panel operational parameter data be normalized;2nd, BP neural network is built with reference to the training set after normalization;3rd, the initial weight and initial threshold of the BP neural network are obtained according to adaptive differential algorithm optimization;4th, obtained weights and threshold value initial weight and initial threshold as the BP neural network will be optimized, and utilize institute Training set is stated to be trained the BP neural network;5th, the data to be predicted in the forecast set are inputted in the BP neural network, is exported according to the BP neural network Predicted value contrasted with actual value, and the accuracy of assessment prediction result.
- A kind of 2. Forecasting Methodology of solar energy unmanned plane cell plate voltage according to claim 1, it is characterised in that:In step In rapid one, using light intensity meter and multimeter measurement solar panel operating parameter, the operating parameter include intensity of illumination, Ambient temperature and voltage;Moreover, thus intensity of illumination and ambient temperature are obtained as input variable, voltage as output variable Operational parameter data when to training set and forecast set and choosing fine day, and be normalized.
- A kind of 3. Forecasting Methodology of solar energy unmanned plane cell plate voltage according to claim 1, it is characterised in that:In step In rapid two, according to the dimension of input data and output data, the neuron number and output layer of the hidden layer of BP neural network are determined Neuron number, its calculation formula is as follows:L1=2m+1;L2=log2n;Wherein, L1For node in hidden layer, L2For output layer number of nodes, m is input layer number, and n is output layer number of nodes.
- A kind of 4. Forecasting Methodology of solar energy unmanned plane cell plate voltage according to claim 1, it is characterised in that:In step In rapid three, specifically comprise the following steps:Set the initial population of adaptive differential evolution algorithm:Based on above-mentioned training set and forecast set, using entity coding method, Initial population size Np, gene dimension D, mutagenic factor F, the excursion for intersecting factor CR and each gene are set, and will The error of BP neural network is as ideal adaptation angle value in population, in solution space random initializtion population:And generate at random:<mrow> <msub> <mi>x</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>i</mi> </mrow> </msub> <mrow> <mo>(</mo> <mn>0</mn> <mo>)</mo> </mrow> <mo>=</mo> <msubsup> <mi>x</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>i</mi> </mrow> <mi>L</mi> </msubsup> <mo>+</mo> <mi>r</mi> <mi>a</mi> <mi>n</mi> <mi>d</mi> <mrow> <mo>(</mo> <mn>0</mn> <mo>,</mo> <mn>1</mn> <mo>)</mo> </mrow> <mo>&CenterDot;</mo> <mrow> <mo>(</mo> <msubsup> <mi>x</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>i</mi> </mrow> <mi>U</mi> </msubsup> <mo>-</mo> <msubsup> <mi>x</mi> <mrow> <mi>j</mi> <mo>,</mo> <mi>i</mi> </mrow> <mi>L</mi> </msubsup> <mo>)</mo> </mrow> </mrow>Wherein, xi(0) i-th article of chromosome in the 0th generation in population, x are representedj,i(0) jth of i-th article of chromosome in the 0th generation is represented Gene, rand (0,1) are represented in the equally distributed random number in (0,1) section;Mutation operation:Individual difference is realized using difference strategy, randomly selects two different individuals in population, by its vector difference With treating that variation individual is synthesized into row vector after scaling, intermediate is produced:υi(g+1)=xr1(g)+F·(xr2(g)-xr3(g))i≠r1≠r2≠r3Wherein, F is mutagenic factor, xi(g) represent g for i-th of individual in population;Crossover operation:To g for population xi(g) and its variation intermediate υi(g+1) crossover operation between individual is carried out:Wherein, CR is to intersect the factor, jrandFor the random integers of [1,2 ... D];Selection operation:The individual for entering population of future generation is selected using greedy algorithm, i.e. optimum results are dyed better than previous generation The individual of body enters population of future generation:Judge whether the precision of optimum results meets the requirements or whether number of iterations reaches maximum, if it is not, then returning to variation behaviour Make step;If it is, output optimization obtains the initial weight and initial threshold of the BP neural network.
- A kind of 5. Forecasting Methodology of solar energy unmanned plane cell plate voltage according to claim 4, it is characterised in that:Setting The TSP question factor is:<mrow> <mi>F</mi> <mo>=</mo> <msub> <mi>F</mi> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> <mo>-</mo> <mrow> <mo>(</mo> <msub> <mi>F</mi> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>F</mi> <mrow> <mi>m</mi> <mi>i</mi> <mi>n</mi> </mrow> </msub> <mo>)</mo> </mrow> <mo>*</mo> <msup> <mi>e</mi> <mrow> <mn>1</mn> <mo>-</mo> <mfrac> <msub> <mi>G</mi> <mi>m</mi> </msub> <mrow> <mi>G</mi> <mi>m</mi> <mo>-</mo> <mi>G</mi> <mo>+</mo> <mn>1</mn> </mrow> </mfrac> </mrow> </msup> </mrow>Wherein, FmaxIt is expressed as mutagenic factor maximum, FminMutagenic factor minimum value is expressed as, Gm represents maximum number of iterations, G For current iteration number.
- A kind of 6. Forecasting Methodology of solar energy unmanned plane cell plate voltage according to claim 4, it is characterised in that:Setting Adaptively intersecting the factor is:<mrow> <mi>C</mi> <mi>R</mi> <mo>=</mo> <msub> <mi>CR</mi> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> <mo>-</mo> <mfrac> <mrow> <mi>G</mi> <mrow> <mo>(</mo> <msub> <mi>CR</mi> <mrow> <mi>m</mi> <mi>a</mi> <mi>x</mi> </mrow> </msub> <mo>-</mo> <msub> <mi>CR</mi> <mrow> <mi>m</mi> <mi>i</mi> <mi>n</mi> </mrow> </msub> <mo>)</mo> </mrow> </mrow> <msub> <mi>G</mi> <mi>m</mi> </msub> </mfrac> </mrow>Wherein, CRmaxRepresent maximum and intersect the factor, CRminRepresent minimum and intersect the factor.
- A kind of 7. Forecasting Methodology of solar energy unmanned plane cell plate voltage according to claim 1, it is characterised in that:In step In rapid five, training time, iterative steps evaluation index are assessed using root-mean-square error, its formula is as follows:<mrow> <mi>R</mi> <mi>M</mi> <mi>S</mi> <mi>E</mi> <mo>=</mo> <msqrt> <mrow> <mfrac> <mn>1</mn> <mi>N</mi> </mfrac> <munderover> <mo>&Sigma;</mo> <mrow> <mi>i</mi> <mo>=</mo> <mn>1</mn> </mrow> <mi>N</mi> </munderover> <msup> <mrow> <mo>(</mo> <msub> <mi>P</mi> <mi>i</mi> </msub> <mo>-</mo> <msub> <mi>P</mi> <mi>f</mi> </msub> <mo>)</mo> </mrow> <mn>2</mn> </msup> </mrow> </msqrt> </mrow>Wherein, PiFor solar array voltage real output value;PfOutput valve is predicted for solar array voltage;N is total for data Number.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711304116.4A CN107947738A (en) | 2017-12-11 | 2017-12-11 | A kind of Forecasting Methodology of solar energy unmanned plane cell plate voltage |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711304116.4A CN107947738A (en) | 2017-12-11 | 2017-12-11 | A kind of Forecasting Methodology of solar energy unmanned plane cell plate voltage |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107947738A true CN107947738A (en) | 2018-04-20 |
Family
ID=61945464
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711304116.4A Pending CN107947738A (en) | 2017-12-11 | 2017-12-11 | A kind of Forecasting Methodology of solar energy unmanned plane cell plate voltage |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107947738A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109635831A (en) * | 2018-10-30 | 2019-04-16 | 成都云材智慧数据科技有限公司 | A kind of anode material for lithium-ion batteries voltage performance prediction technique |
CN111428913A (en) * | 2020-03-06 | 2020-07-17 | 中国科学技术大学 | Performance prediction method and performance prediction system of proton exchange membrane fuel cell |
CN112329997A (en) * | 2020-10-26 | 2021-02-05 | 国网河北省电力有限公司雄安新区供电公司 | Power demand load prediction method and system, electronic device, and storage medium |
CN113128108A (en) * | 2021-04-07 | 2021-07-16 | 汕头大学 | Method for determining diameter of jet grouting pile based on differential evolution artificial intelligence |
WO2023060579A1 (en) * | 2021-10-15 | 2023-04-20 | 厦门大学 | Solar cell structure performance prediction method |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150061387A1 (en) * | 2008-09-13 | 2015-03-05 | Moixa Energy Holdings Limited | Systems, devices and methods for electricity provision, usage monitoring, analysis, and enabling improvements in efficiency |
CN104636823A (en) * | 2015-01-23 | 2015-05-20 | 中国农业大学 | Wind power prediction method |
CN105913151A (en) * | 2016-04-12 | 2016-08-31 | 河海大学常州校区 | Photovoltaic power station power generation amount predication method based on adaptive mutation particle swarm and BP network |
CN106228278A (en) * | 2016-08-19 | 2016-12-14 | 长园深瑞继保自动化有限公司 | Photovoltaic power prognoses system |
CN107194524A (en) * | 2017-07-28 | 2017-09-22 | 合肥工业大学 | A kind of Coal and Gas Outbursts Prediction method based on RBF neural |
-
2017
- 2017-12-11 CN CN201711304116.4A patent/CN107947738A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20150061387A1 (en) * | 2008-09-13 | 2015-03-05 | Moixa Energy Holdings Limited | Systems, devices and methods for electricity provision, usage monitoring, analysis, and enabling improvements in efficiency |
CN104636823A (en) * | 2015-01-23 | 2015-05-20 | 中国农业大学 | Wind power prediction method |
CN105913151A (en) * | 2016-04-12 | 2016-08-31 | 河海大学常州校区 | Photovoltaic power station power generation amount predication method based on adaptive mutation particle swarm and BP network |
CN106228278A (en) * | 2016-08-19 | 2016-12-14 | 长园深瑞继保自动化有限公司 | Photovoltaic power prognoses system |
CN107194524A (en) * | 2017-07-28 | 2017-09-22 | 合肥工业大学 | A kind of Coal and Gas Outbursts Prediction method based on RBF neural |
Non-Patent Citations (1)
Title |
---|
李倩,周彬倩,张建成,李嘉俊: "基于自适应差分进化和BP神经网络的光伏功率预测", 《陕西电力》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109635831A (en) * | 2018-10-30 | 2019-04-16 | 成都云材智慧数据科技有限公司 | A kind of anode material for lithium-ion batteries voltage performance prediction technique |
CN111428913A (en) * | 2020-03-06 | 2020-07-17 | 中国科学技术大学 | Performance prediction method and performance prediction system of proton exchange membrane fuel cell |
CN112329997A (en) * | 2020-10-26 | 2021-02-05 | 国网河北省电力有限公司雄安新区供电公司 | Power demand load prediction method and system, electronic device, and storage medium |
CN113128108A (en) * | 2021-04-07 | 2021-07-16 | 汕头大学 | Method for determining diameter of jet grouting pile based on differential evolution artificial intelligence |
WO2023060579A1 (en) * | 2021-10-15 | 2023-04-20 | 厦门大学 | Solar cell structure performance prediction method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107947738A (en) | A kind of Forecasting Methodology of solar energy unmanned plane cell plate voltage | |
CN103413174B (en) | Based on the short-term wind speed multistep forecasting method of degree of depth learning method | |
CN104361393B (en) | Data predication method is used for based on the improved neural network model of particle swarm optimization algorithm | |
CN104636801A (en) | Transmission line audible noise prediction method based on BP neural network optimization | |
CN107316099A (en) | Ammunition Storage Reliability Forecasting Methodology based on particle group optimizing BP neural network | |
CN108205717A (en) | A kind of photovoltaic generation power Multiple Time Scales Forecasting Methodology | |
CN103793887B (en) | Short-term electric load on-line prediction method based on self-adaptive enhancement algorithm | |
CN106910337A (en) | A kind of traffic flow forecasting method based on glowworm swarm algorithm Yu RBF neural | |
Yan et al. | A general method to estimate wind farm power using artificial neural networks | |
CN103105246A (en) | Greenhouse environment forecasting feedback method of back propagation (BP) neural network based on improvement of genetic algorithm | |
Weerasuriya et al. | Performance evaluation of population-based metaheuristic algorithms and decision-making for multi-objective optimization of building design | |
CN104636985A (en) | Method for predicting radio disturbance of electric transmission line by using improved BP (back propagation) neural network | |
CN106529818A (en) | Water quality evaluation prediction method based on fuzzy wavelet neural network | |
CN110942205A (en) | Short-term photovoltaic power generation power prediction method based on HIMVO-SVM | |
Cai et al. | Short‐term load forecasting method based on deep neural network with sample weights | |
CN104134103B (en) | Utilize the method for the BP neural network model prediction hot oil pipeline energy consumption of amendment | |
CN106779177A (en) | Multiresolution wavelet neutral net electricity demand forecasting method based on particle group optimizing | |
CN103927460A (en) | Wind power plant short-term wind speed prediction method based on RBF | |
CN110264012A (en) | Renewable energy power combination prediction technique and system based on empirical mode decomposition | |
CN105608295A (en) | Multi-objective evolutionary algorithm (MOEA) and radial basis function (RBF) neural network optimization modeling method of coking furnace pressure | |
CN110516318A (en) | Airfoil Design method based on radial basis function neural network agent model | |
Zhu et al. | Structural safety monitoring of high arch dam using improved ABC-BP model | |
CN103489037B (en) | A kind of Forecasting Methodology that can power generating wind resource | |
CN104361399A (en) | Solar irradiation intensity minute-scale predication method | |
CN109408896B (en) | Multi-element intelligent real-time monitoring method for anaerobic sewage treatment gas production |
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: 20180420 |