CN116151545A - Multi-wind motor group power control optimization system - Google Patents
Multi-wind motor group power control optimization system Download PDFInfo
- Publication number
- CN116151545A CN116151545A CN202211543203.6A CN202211543203A CN116151545A CN 116151545 A CN116151545 A CN 116151545A CN 202211543203 A CN202211543203 A CN 202211543203A CN 116151545 A CN116151545 A CN 116151545A
- Authority
- CN
- China
- Prior art keywords
- matrix
- wind turbine
- feature
- characteristic
- values
- 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
- 238000005457 optimization Methods 0.000 title claims abstract description 40
- 238000009826 distribution Methods 0.000 claims abstract description 33
- 238000000034 method Methods 0.000 claims abstract description 28
- 238000000605 extraction Methods 0.000 claims abstract description 20
- 239000011159 matrix material Substances 0.000 claims description 214
- 239000013598 vector Substances 0.000 claims description 48
- 238000013527 convolutional neural network Methods 0.000 claims description 45
- 230000005611 electricity Effects 0.000 claims description 37
- 238000013528 artificial neural network Methods 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 16
- 239000011800 void material Substances 0.000 claims description 16
- 230000004927 fusion Effects 0.000 claims description 14
- 238000012937 correction Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 9
- 238000001994 activation Methods 0.000 claims description 8
- 238000011176 pooling Methods 0.000 claims description 8
- 238000012546 transfer Methods 0.000 claims description 8
- 238000010248 power generation Methods 0.000 claims description 5
- 238000012163 sequencing technique Methods 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 4
- 238000012544 monitoring process Methods 0.000 claims description 4
- 208000011580 syndromic disease Diseases 0.000 claims 1
- 230000008901 benefit Effects 0.000 abstract description 12
- 238000013135 deep learning Methods 0.000 abstract description 9
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 12
- 238000003860 storage Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 8
- 238000004590 computer program Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 5
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 239000000243 solution Substances 0.000 description 3
- 238000011217 control strategy Methods 0.000 description 2
- 238000010790 dilution Methods 0.000 description 2
- 239000012895 dilution Substances 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 239000003550 marker Substances 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000006798 recombination Effects 0.000 description 1
- 238000005215 recombination Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 230000017105 transposition Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/46—Controlling of the sharing of output between the generators, converters, or transformers
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
-
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2300/00—Systems for supplying or distributing electric power characterised by decentralized, dispersed, or local generation
- H02J2300/20—The dispersed energy generation being of renewable origin
- H02J2300/28—The renewable source being wind energy
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Power Engineering (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Wind Motors (AREA)
Abstract
The utility model relates to an intelligent control technical field of wind turbine generator system, it specifically discloses a multiscale motor group power control optimizing system, and it obtains the output power value and the network access price of each wind turbine generator system at a plurality of predetermined time points in predetermined time quantum at first, then adopts artificial intelligence control technique based on deep learning to carry out the feature extraction through the multiscale relativity characteristic distribution information to the output power and the network access price of each wind turbine generator system in high dimension space, so decoding to obtain the operation economic index, and then carry out the priority order to each wind turbine generator system based on the operation economic index of each wind turbine generator system, and carry out power control to wind turbine generator system according to the priority, through this kind of mode, can be in compromise wind turbine generator system operation stability and economic benefits and carry out accurate control to wind turbine generator system work to realize intelligent optimization dispatch and economic operation of new energy.
Description
Technical Field
The application relates to the technical field of intelligent control of wind turbines, and more particularly relates to a multi-wind turbine power control optimization system.
Background
The energy industry is in a transformation period, new energy scale and traditional energy cleaning are two key points of transformation, and the problem of difficult internet surfing in new energy scale is mainly caused by the fact that the new energy cannot realize intelligent optimal scheduling and economic operation as the traditional energy. With the continuous increase of the scale of the units participating in grid connection, the running instability of the wind turbine units, especially the fluctuation of active power, causes a certain impact on the power grid. The existing method is that the wind farm directly receives the AGC regulation of the power grid, but the active control strategy of the wind farm taking the stability of the grid side as the starting point can cause a large number of situations such as wind abandoning, frequent start and stop of a unit, fatigue operation of the unit, serious grid loss and the like, so that the economic operation of the wind farm is influenced, and the long-term development of the wind farm can be greatly influenced.
Therefore, an optimized wind turbine power control scheme for economic operation is desired.
Disclosure of Invention
The present application has been made in order to solve the above technical problems. The embodiment of the application provides a multi-wind turbine generator system power control optimizing system, firstly, output power values and network access electricity prices of all wind turbine generators at a plurality of preset time points in a preset time period are obtained, then, an artificial intelligent control technology based on deep learning is adopted, so that characteristic extraction is carried out on multi-scale relevance characteristic distribution information of the output power and the network access electricity prices of all wind turbine generators in a high-dimensional space, an operation economy index is obtained through decoding, priority ranking is carried out on all wind turbine generators based on the operation economy index of all wind turbine generators, and power control is carried out on the wind turbine generators according to the priority.
According to one aspect of the present application, there is provided a multi-wind turbine power control optimization system comprising:
the data monitoring and collecting module is used for obtaining output power values and network access electricity prices of each wind turbine generator at a plurality of preset time points in a preset time period;
the operation index association module is used for respectively arranging the output power values and the network access power prices of the wind turbines at a plurality of preset time points in a preset time period into a power input vector and a network access power price input vector according to a time dimension, and respectively calculating an association matrix between the power input vector and the network access power price input vector to obtain a plurality of operation index association matrices;
the operation index feature extraction module is used for enabling the operation index correlation matrixes to respectively pass through a double-flow network model comprising a first convolution neural network and a second convolution neural network to obtain a plurality of multi-scale index correlation feature matrixes, wherein the first convolution neural network uses a first convolution kernel with a first void rate, and the second convolution neural network uses a second convolution kernel with a second void rate;
the decoding module is used for enabling the multi-scale index association characteristic matrixes to pass through a decoder to obtain a plurality of decoding values used for representing the running economy index; and
And the control result generation module is used for sequencing the priorities of the wind power generation sets based on the decoding values and controlling the power of the wind power generation set according to the priorities.
In the multi-wind turbine generator system power control optimization system, the operation index association module is further configured to calculate products between the transposed vector of the power input vector and the network-entry power price input vector to obtain the plurality of operation index association matrices.
In the above multi-wind turbine generator system power control optimization system, the operation index feature extraction module includes:
a first scale feature extraction unit, configured to perform, in forward transfer of layers, a hole convolution process based on the first convolution kernel, a pooling process along a channel dimension, and a nonlinear activation process on input data, respectively, using layers of the first convolutional neural network model, so as to output a first feature matrix from a last layer of the first convolutional neural network model;
a second scale feature extraction unit, configured to perform, in forward transfer of layers, hole convolution processing based on the second convolution kernel, pooling processing along a channel dimension, and nonlinear activation processing on input data, respectively, using layers of the second convolutional neural network model, so as to output a second feature matrix by a last layer of the second convolutional neural network model;
And the fusion unit is used for fusing the first feature matrix and the second feature matrix to obtain the multi-scale index association feature matrix.
In the above multi-wind turbine generator system power control optimization system, the fusion unit includes:
the first characteristic distribution correction subunit is used for correcting the characteristic values of all positions in the first characteristic matrix based on the second characteristic matrix to obtain a corrected first characteristic matrix;
the second characteristic distribution correction subunit is used for correcting the characteristic values of all positions in the second characteristic matrix based on the first characteristic matrix to obtain a corrected second characteristic matrix;
and the position-based fusion subunit is used for calculating a position-based weighted sum between the corrected first feature matrix and the corrected second feature matrix to obtain the multi-scale index associated feature matrix.
In the multi-wind turbine generator system power control optimization system, the first characteristic distribution corrector subunit is further configured to: correcting the characteristic values of each position in the first characteristic matrix according to the following formula based on the second characteristic matrix to obtain a corrected first characteristic matrix;
Wherein, the formula is:
wherein M is 1 And M 2 The first feature matrix and the second feature matrix,
and->
The feature values of the (i, j) th positions of the first feature matrix and the second feature matrix, respectively, and +.>
And->
The mean value of all eigenvalues of the first eigenvalue matrix and the second eigenvalue matrix, log representing a logarithmic function value based on 2,
and (c) a eigenvalue representing the (i, j) th position of the corrected first eigenvector.
In the multi-wind turbine generator system power control optimization system, the second characteristic distribution correction subunit is further configured to: correcting the characteristic values of each position in the second characteristic matrix according to the following formula based on the first characteristic matrix to obtain a corrected second characteristic matrix;
wherein, the formula is:
wherein M is 1 And M 2 The first feature matrix and the second feature matrix,
and->
The feature values of the (i, j) th positions of the first feature matrix and the second feature matrix, respectively, and +.>
And->
The mean value of all eigenvalues of the first eigenvalue matrix and the second eigenvalue matrix, log representing a logarithmic function value based on 2, 
And (c) a eigenvalue representing the (i, j) th position of the corrected second eigenvector.
In the multi-wind turbine generator system power control optimization system, the decoding module is further configured to: performing decoding regression on the plurality of multi-scale index associated feature matrices using a decoder to obtain the plurality of decoded values according to a formula:
wherein X is each of the plurality of multi-scale index associated feature matrices, Y is each of the plurality of decoded values, and W is a weight matrix.
Compared with the prior art, the multi-wind turbine generator system power control optimizing system provided by the application firstly obtains output power values and network access electricity prices of all wind turbine generators at a plurality of preset time points in a preset time period, then adopts an artificial intelligence control technology based on deep learning to extract characteristics of multi-scale relevance characteristic distribution information of the output power and the network access electricity prices of all wind turbine generators in a high-dimensional space, decodes the characteristic extraction to obtain operation economy indexes, further ranks priority of all wind turbine generators based on the operation economy indexes of all wind turbine generators, and controls power of wind turbine generators according to the priority.
Drawings
The foregoing and other objects, features and advantages of the present application will become more apparent from the following more particular description of embodiments of the present application, as illustrated in the accompanying drawings. The accompanying drawings are included to provide a further understanding of embodiments of the application and are incorporated in and constitute a part of this specification, illustrate the application and not constitute a limitation to the application. In the drawings, like reference numerals generally refer to like parts or steps.
FIG. 1 is a block diagram schematic of a multi-wind turbine power control optimization system according to an embodiment of the present application.
FIG. 2 is a block diagram of an operation index feature extraction module in a multi-wind turbine power control optimization system according to an embodiment of the present application.
FIG. 3 is a block diagram of a fusion unit in a multi-wind turbine power control optimization system according to an embodiment of the application.
FIG. 4 is a flow chart of a multi-wind turbine power control optimization method according to an embodiment of the present application.
Fig. 5 is a schematic diagram of a system architecture of a multi-wind turbine power control optimization method according to an embodiment of the present application.
Fig. 6 is a block diagram of an electronic device according to an embodiment of the present application.
Detailed Description
Hereinafter, example embodiments according to the present application will be described in detail with reference to the accompanying drawings. It should be apparent that the described embodiments are only some of the embodiments of the present application and not all of the embodiments of the present application, and it should be understood that the present application is not limited by the example embodiments described herein.
Scene overview
As described above, as the scale of the units participating in grid connection becomes larger, the instability of the operation of the wind turbine units, especially the fluctuation of active power, causes a certain impact on the power grid. The existing method is that the wind farm directly receives the AGC regulation of the power grid, but the active control strategy of the wind farm taking the stability of the grid side as the starting point can cause a large number of situations such as wind abandoning, frequent start and stop of a unit, fatigue operation of the unit, serious grid loss and the like, so that the economic operation of the wind farm is influenced, and the long-term development of the wind farm can be greatly influenced. Therefore, an optimized wind turbine power control scheme for economic operation is desired.
At present, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, speech signal processing, and the like. In addition, deep learning and neural networks have also shown levels approaching and even exceeding humans in the fields of image classification, object detection, semantic segmentation, text translation, and the like.
In recent years, the development of deep learning and neural networks provides a new solution idea and scheme for intelligent power control of wind turbines.
Accordingly, considering that the active power of the wind turbine generator has fluctuation to cause unstable operation of the wind turbine generator when the power control of the wind turbine generator is performed, in order to explore a stable operation control scheme for the purpose of economic operation, the correlation relationship between the active power and the network access electricity price in time sequence needs to be mined. Specifically, in the technical scheme of the application, an artificial intelligent control technology based on deep learning is adopted to extract characteristics of multi-scale relevance characteristic distribution information of output power and network-access electricity price of each wind turbine in a high-dimensional space, so that an operation economy index is obtained, priority ranking is carried out on each wind turbine based on the operation economy index of each wind turbine, and power control is carried out on wind farm units according to the priority. Therefore, the operation of the wind turbine can be accurately controlled on the premise of considering the stable operation and the economic benefit of the wind turbine, so that intelligent optimal scheduling and economic operation of new energy sources are realized.
Specifically, in the technical scheme of the application, firstly, output power values and network access electricity prices of each wind turbine generator set at a plurality of preset time points in a preset time period are obtained. And then, in order to dig out the time sequence relevance characteristic distribution representation between the output power and the network access price of each wind turbine, further arranging the output power values and the network access price of each wind turbine at a plurality of preset time points in a preset time period into a power input vector and a network access price input vector according to a time dimension, and then respectively calculating the product between the transposed vector of the power input vector and the network access price input vector to construct an operation index relevance matrix of each wind turbine with time sequence relevance information between the power and the electricity price so as to obtain a plurality of operation index relevance matrices.
Further, feature mining of the operation index correlation matrix is performed using a convolutional neural network model having excellent performance in terms of local implicit correlation feature extraction, particularly, considering that in operation control for the wind turbines, output power and network access electricity prices of the respective wind turbines have different correlation feature representations at different predetermined time points due to fluctuation of the output power. Therefore, in the technical scheme of the application, the operation index correlation matrix is further processed through a double-flow network model comprising a first convolution neural network and a second convolution neural network, so as to extract multi-scale correlation characteristic distribution information among characteristic values of each position in the operation index correlation matrix, namely, extract index multi-scale correlation characteristics in time dimension between output power and network access electricity price of each wind turbine generator respectively, and therefore a plurality of multi-scale index correlation characteristic matrixes are obtained. Accordingly, in one specific example of the present application, the first convolutional neural network uses a first convolutional kernel having a first void fraction and the second convolutional neural network uses a second convolutional kernel having a second void fraction. In this way, the multiscale relevance characteristic distribution information of the operation index relevance matrix between the specific positions on time sequence of the output power and the network-access electricity price of each wind turbine generator can be respectively extracted.
And then decoding the multi-scale index association characteristic matrixes in a decoder to obtain a plurality of decoding values used for representing the running economy index of each wind turbine. And further, based on the plurality of decoding values, sequencing the priorities of the wind turbines, and performing power control on the wind turbines according to the priorities. Therefore, the operation of the wind turbine can be controlled on the premise of considering the stable operation and the economic benefit of the wind turbine, so that intelligent optimal scheduling and economic operation of new energy sources are realized.
Particularly, in the technical solution of the present application, when the operation index correlation matrix is obtained by using a dual-flow network model including a first convolutional neural network and a second convolutional neural network, the operation index correlation matrix needs to be fused with a first feature matrix and a second feature matrix obtained by using the first convolutional neural network and the second convolutional neural network respectively to obtain the multi-scale index correlation feature matrix. And, because the first convolutional neural network and the second convolutional neural network use convolution kernels with different void rates, spatial position errors exist in the high-dimensional feature space of the feature distribution of the first feature matrix and the second feature matrix, so that the fusion effect of the first feature matrix and the second feature matrix is affected.
In the application, the first feature matrix and the second feature matrix are obtained from the operation index association matrix, so that the first feature matrix and the second feature matrix are used as homologous feature matrices and have certain correspondence in feature distribution, and therefore, relative angle probability information representation correction can be respectively carried out on the first feature matrix and the second feature matrix, and the relative angle probability information representation correction is expressed as follows:
wherein the method comprises the steps of
And->
Respectively the first feature matrix M 1 And the second feature matrix M 2 Characteristic value of the (i, j) th position of (c), and +.>
And->
Respectively the first feature matrix M 1 And the second feature matrix M 2 Log represents the base 2 logarithm of the mean of all eigenvalues.
Here, the relative class angle probability information indicates that the correction is by the first feature matrix M 1 And the second feature matrix M 2 The relative angle probability information between the two is represented to perform the first characteristic matrix M 1 And the second feature matrix M 2 Geometric dilution of spatial position errors of feature distribution in high-dimensional feature space, thereby generating a first feature matrix M 1 And the second feature matrix M 2 With a certain correspondence between them, based on the first feature matrix M 1 And the second feature matrix M 2 The feature value distribution of each position is compared with the distribution constraint of the whole each other to carry out the implicit context correspondence correction of the feature by the point-by-point regression of the position, thereby improving the first feature matrix M 1 And the second feature matrix M 2 Is a fusion effect of (a). Therefore, the work of the wind turbine can be accurately controlled on the basis of considering the running stability and economic benefit of the wind turbine, so that intelligent optimal scheduling and economic running of new energy sources are realized.
Having described the basic principles of the present application, various non-limiting embodiments of the present application will now be described in detail with reference to the accompanying drawings.
Exemplary System
FIG. 1 is a block diagram schematic of a multi-wind turbine power control optimization system according to an embodiment of the present application. As shown in fig. 1, the multi-wind turbine power control optimization system 100 according to the embodiment of the present application includes: the data monitoring and collecting module 110 is configured to obtain output power values and network access electricity prices of each wind turbine generator at a plurality of preset time points in a preset time period; the operation index association module 120 is configured to arrange output power values and network access power prices of the wind turbines at a plurality of predetermined time points in a predetermined time period into a power input vector and a network access power price input vector according to a time dimension, and calculate association matrices between the power input vector and the network access power price input vector to obtain a plurality of operation index association matrices; an operation index feature extraction module 130, configured to pass the plurality of operation index correlation matrices through a dual-flow network model including a first convolutional neural network and a second convolutional neural network, respectively, to obtain a plurality of multi-scale index correlation feature matrices, where the first convolutional neural network uses a first convolutional kernel with a first void rate, and the second convolutional neural network uses a second convolutional kernel with a second void rate; a decoding module 140, configured to pass the multiple multi-scale index association feature matrices through a decoder to obtain multiple decoded values for representing the running economy index; and a control result generating module 150, configured to prioritize the wind turbines based on the multiple decoding values, and perform power control on the wind farm turbines according to the priorities.
In this embodiment of the present application, the data monitoring and collecting module 110 is configured to obtain output power values and network access electricity prices of each wind turbine generator at a plurality of predetermined time points within a predetermined time period. As described above, considering that the active power of the wind turbine generator has fluctuation to cause unstable operation of the wind turbine generator when performing power control of the wind turbine generator, in order to explore a stable operation control scheme for the purpose of economic operation, it is necessary to dig out the correlation relationship between the active power and the network entry electricity price in time sequence. Specifically, in the technical scheme of the application, an artificial intelligent control technology based on deep learning is adopted to extract characteristics of multi-scale relevance characteristic distribution information of output power and network-access electricity price of each wind turbine in a high-dimensional space, so that an operation economy index is obtained, priority ranking is carried out on each wind turbine based on the operation economy index of each wind turbine, and power control is carried out on wind farm units according to the priority. Therefore, the operation of the wind turbine can be accurately controlled on the premise of considering the stable operation and the economic benefit of the wind turbine, so that intelligent optimal scheduling and economic operation of new energy sources are realized.
In this embodiment of the present application, the operation index association module 120 is configured to arrange, in a predetermined period of time, output power values and network access prices of the wind turbines at a plurality of predetermined time points into a power input vector and a network access price input vector according to a time dimension, and then calculate association matrices between the power input vector and the network access price input vector, respectively, so as to obtain a plurality of operation index association matrices. It should be understood that, in order to dig out the time-series correlation characteristic distribution representation between the output power and the network-access electricity price of each wind turbine generator, the output power values and the network-access electricity price of each wind turbine generator at a plurality of preset time points in a preset time period are respectively arranged into a power input vector and a network-access electricity price input vector according to a time dimension, and then products between transposed vectors of the power input vector and the network-access electricity price input vector are respectively calculated to construct operation index correlation matrices of each wind turbine generator, which have time-series correlation information between power and electricity price, so as to obtain a plurality of operation index correlation matrices.
In a specific embodiment of the present application, the operation index association module is further configured to calculate products between the transposed vector of the power input vector and the input vector of the network entry price to obtain the plurality of operation index association matrices. The method comprises the steps of carrying out transposition multiplication on output power values and network access electricity prices of a plurality of preset time points of the same wind turbine generator within a preset time period, wherein values of positions at non-diagonal positions in an operation index correlation matrix are used for representing the correlation between the output power values and the network access electricity prices of different time points of the wind turbine generator, and the diagonal positions in the operation index correlation matrix are used for representing the correlation between the output power values and the network access electricity prices of the same time point.
In this embodiment of the present application, the operation index feature extraction module 130 is configured to pass the plurality of operation index correlation matrices through a dual-flow network model including a first convolutional neural network and a second convolutional neural network, so as to obtain a plurality of multi-scale index correlation feature matrices, where the first convolutional neural network uses a first convolution kernel with a first void ratio, and the second convolutional neural network uses a second convolution kernel with a second void ratio. It should be understood that, in consideration of that, although the operation index correlation matrix includes time series correlation information between the power and electricity prices of the respective wind turbines, such information is distributed in the operation index correlation matrix, not obvious, but the convolutional neural network model has significant advantages in terms of feature extraction, feature mining of the operation index correlation matrix is performed using the convolutional neural network model, in particular, in consideration of that, in operation control of the wind turbines, the output power and the network entry electricity prices of the respective wind turbines have different correlation feature representations at different predetermined time points due to fluctuation of the output power. Therefore, in the technical scheme of the application, the operation index correlation matrix is further processed through a double-flow network model comprising a first convolution neural network and a second convolution neural network, so as to extract multi-scale correlation characteristic distribution information among characteristic values of each position in the operation index correlation matrix, namely, extract index multi-scale correlation characteristics in time dimension between output power and network access electricity price of each wind turbine generator respectively, and therefore a plurality of multi-scale index correlation characteristic matrixes are obtained. Accordingly, in one specific example of the present application, the first convolutional neural network uses a first convolutional kernel having a first void fraction, and the second convolutional neural network uses a second convolutional kernel having a second void fraction, the first and second convolutional kernels being the same size. In this way, the multiscale relevance characteristic distribution information of the operation index relevance matrix between the specific positions on time sequence of the output power and the network-access electricity price of each wind turbine generator can be respectively extracted.
FIG. 2 is a block diagram of an operation index feature extraction module in a multi-wind turbine power control optimization system according to an embodiment of the present application. As shown in fig. 2, in a specific embodiment of the present application, the operation index feature extraction module 130 includes: a first scale feature extraction unit 131, configured to perform, in forward transfer of layers, a hole convolution process based on the first convolution kernel, a pooling process along a channel dimension, and a nonlinear activation process on input data, respectively, using layers of the first convolutional neural network model, so as to output a first feature matrix from a last layer of the first convolutional neural network model; a second scale feature extraction unit 132 for performing a hole convolution process based on the second convolution kernel, a pooling process along a channel dimension, and a nonlinear activation process on input data in forward transfer of layers using each layer of the second convolutional neural network model, respectively, to output a second feature matrix by a last layer of the second convolutional neural network model; and a fusion unit 133, configured to fuse the first feature matrix and the second feature matrix to obtain the multi-scale index associated feature matrix.
Particularly, in the technical solution of the present application, when the operation index correlation matrix is obtained by using a dual-flow network model including a first convolutional neural network and a second convolutional neural network, the operation index correlation matrix needs to be fused with a first feature matrix and a second feature matrix obtained by using the first convolutional neural network and the second convolutional neural network respectively to obtain the multi-scale index correlation feature matrix. And, because the first convolutional neural network and the second convolutional neural network use convolution kernels with different void rates, spatial position errors exist in the high-dimensional feature space of the feature distribution of the first feature matrix and the second feature matrix, so that the fusion effect of the first feature matrix and the second feature matrix is affected.
In the method, the first feature matrix and the second feature matrix are obtained from the operation index association matrix, so that the first feature matrix and the second feature matrix are used as homologous feature matrices and have certain correspondence in feature distribution, and therefore relative angle probability information representation correction can be carried out on the first feature matrix and the second feature matrix respectively.
FIG. 3 is a block diagram of a fusion unit in a multi-wind turbine power control optimization system according to an embodiment of the application. As shown in fig. 3, in a specific embodiment of the present application, the fusing unit 133 includes: a first feature distribution corrector unit 1331, configured to correct feature values of each position in the first feature matrix based on the second feature matrix to obtain a corrected first feature matrix; a second feature distribution corrector unit 1332, configured to correct feature values of each position in the second feature matrix based on the first feature matrix to obtain a corrected second feature matrix; and a per-position fusion subunit 1333 configured to calculate a per-position weighted sum between the corrected first feature matrix and the corrected second feature matrix to obtain the multi-scale-index-associated feature matrix.
In a specific embodiment of the present application, the first feature distribution corrector subunit is further configured to: correcting the characteristic values of each position in the first characteristic matrix according to the following formula based on the second characteristic matrix to obtain a corrected first characteristic matrix;
wherein, the formula is:
Wherein M is 1 And M 2 The first feature matrix and the second feature matrix,
and m 2i,j The feature values of the (i, j) th positions of the first feature matrix and the second feature matrix, respectively, and +.>
And->
The mean value of all eigenvalues of the first eigenvalue matrix and the second eigenvalue matrix, log representing a logarithmic function value based on 2, +.>
And (c) a eigenvalue representing the (i, j) th position of the corrected first eigenvector.
In a specific embodiment of the present application, the second feature distribution corrector unit is further configured to: correcting the characteristic values of each position in the second characteristic matrix according to the following formula based on the first characteristic matrix to obtain a corrected second characteristic matrix;
wherein, the formula is:
wherein M is 1 And M 2 The first feature matrix and the second feature matrix,
and->
The feature values of the (i, j) th positions of the first feature matrix and the second feature matrix, respectively, and +.>
And->
The mean value of all eigenvalues of the first eigenvalue matrix and the second eigenvalue matrix, log representing a logarithmic function value based on 2,
and (c) a eigenvalue representing the (i, j) th position of the corrected second eigenvector.
Here, the relative class angle probability information indicates that the correction is by the first feature matrix M 1 And the second feature matrix M 2 The relative angle probability information between the two is represented to perform the first characteristic matrix M 1 And the second feature matrix M 2 Geometric dilution of spatial position errors of feature distribution in high-dimensional feature space, thereby generating a first feature matrix M 1 And the second feature matrix M 2 With a certain correspondence between them, based on the first feature matrix M 1 And the second feature matrix M 2 The feature value distribution of each position is compared with the distribution constraint of the whole each other to carry out the implicit context correspondence correction of the feature by the point-by-point regression of the position, thereby improving the first feature matrix M 1 And the second feature matrix M 2 Is a fusion effect of (a). Therefore, the work of the wind turbine can be accurately controlled on the basis of considering the running stability and economic benefit of the wind turbine, so that intelligent optimal scheduling and economic running of new energy sources are realized.
In this embodiment, the decoding module 140 is configured to pass the multiple multi-scale index association feature matrices through a decoder to obtain multiple decoded values that are used to represent the running economy index. The multi-scale index association characteristic matrixes are decoded and returned in a decoder to obtain a plurality of decoding values used for representing the running economy index of each wind turbine.
In a specific embodiment of the present application, the decoding module is further configured to: performing decoding regression on the plurality of multi-scale index associated feature matrices using a decoder to obtain the plurality of decoded values according to a formula:
wherein X is each of the plurality of multi-scale index associated feature matrices, Y is each of the plurality of decoded values, and M is a weight matrix.
In this embodiment of the present application, the control result generating module 150 is configured to prioritize the wind turbine generator sets based on the multiple decoding values, and perform power control on the wind turbine generator sets according to the priorities. It should be understood that each of the plurality of decoded values represents an operational economy index of the respective wind turbine, and the magnitudes of the plurality of decoded values are ordered so as to prioritize the respective wind turbine, and more specifically, the larger the decoded value is, the higher the operational economy index of the wind turbine is, the higher the priority of the wind turbine is, and the power control is performed on the wind farm according to the priority. Of course, the power of the wind farm unit can be controlled simply according to the priority of the wind farm unit. But in actual cases, the priority of the wind power plant unit is more adopted to control the power by combining with other information.
In a specific embodiment of the present application, the control result generating module includes:
the information acquisition unit is used for acquiring the state information of the wind power plant unit in real time and outputting the power P of the PCC point of the wind power plant in real time wf Power grid active power scheduling instruction P ref Wind power prediction value P pef ；
The unit classification unit is used for classifying the units into five types including a first type of marker post unit, a second type of on-line unit, a third type of fault unit, a fourth type of unit to be started and a fifth type of unit to be started according to the collected state information of the unit.
The priority judging unit is used for judging the priority of the wind power plant unit according to the running economy index of the unit;
the difference value calculation unit is used for scheduling the instruction P according to the active power of the power grid ref Predicted wind power value P pef Real-time output power P of wind farm PCC point wf The difference ΔP is calculated by the following formulaIs that
When Δp >0, using a power-up control unit; when Δp <0, using a power-down control unit;
the power-up control unit is used for combining the types of the units and respectively performing power-up control according to the priority of the units;
the power-down control unit is used for combining the types of the units and respectively performing power-down control according to the priority of the units;
And the circulation unit is used for returning to the information acquisition unit and carrying out the next round of control cycle.
In summary, according to the multi-wind turbine generator system power control optimizing system provided by the embodiment of the application, firstly, output power values and network access electricity prices of each wind turbine generator system at a plurality of preset time points in a preset time period are obtained, then, an artificial intelligent control technology based on deep learning is adopted to extract characteristics of multi-scale relevance characteristic distribution information of the output power and the network access electricity price of each wind turbine generator system in a high-dimensional space, so that an operation economy index is obtained through decoding, priority ranking is carried out on each wind turbine generator system based on the operation economy index of each wind turbine generator system, and power control is carried out on the wind turbine generator system according to the priority.
As described above, the multi-wind turbine power control optimization system 100 according to the embodiment of the present application may be implemented in various terminal devices, for example, a server or the like where a multi-wind turbine power control optimization algorithm is deployed. In one example, the multi-wind turbine power control optimization system 100 may be integrated into the terminal device as a software module and/or hardware module. For example, the multi-wind turbine power control optimization system 100 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the multi-wind turbine power control optimization system 100 can also be one of a number of hardware modules of the terminal device.
Alternatively, in another example, the multi-wind turbine power control optimization system 100 and the terminal device may be separate devices, and the multi-wind turbine power control optimization system 100 may be connected to the terminal device through a wired and/or wireless network and transmit the interaction information in a agreed data format.
Exemplary method
FIG. 4 is a flow chart of a multi-wind turbine power control optimization method according to an embodiment of the present application. As shown in fig. 4, the multi-wind turbine generator system power control optimization method according to the embodiment of the application includes: s110, obtaining output power values and network access electricity prices of each wind turbine generator at a plurality of preset time points in a preset time period; s120, respectively arranging output power values and network access prices of the wind turbines at a plurality of preset time points in a preset time period into a power input vector and a network access price input vector according to a time dimension, and respectively calculating an incidence matrix between the power input vector and the network access price input vector to obtain a plurality of operation index incidence matrices; s130, enabling the operation index correlation matrixes to respectively pass through a double-flow network model comprising a first convolution neural network and a second convolution neural network to obtain a plurality of multi-scale index correlation feature matrixes, wherein the first convolution neural network uses a first convolution kernel with a first void rate, and the second convolution neural network uses a second convolution kernel with a second void rate; s140, the multi-scale index association characteristic matrixes are passed through a decoder to obtain a plurality of decoding values for representing the running economy index; and S150, sequencing the priorities of the wind power generation sets based on the decoding values, and controlling the power of the wind power plant sets according to the priorities.
Fig. 5 is a schematic diagram of a system architecture of a multi-wind turbine power control optimization method according to an embodiment of the present application. As shown in fig. 5, in the embodiment of the present application, first, output power values of each wind turbine generator set at a plurality of predetermined time points in a predetermined time period are obtained, and the output power values of each wind turbine generator set at a plurality of predetermined time points in the predetermined time period are arranged as a power input vector according to a time dimension. Meanwhile, acquiring the network access electricity prices of each wind turbine generator set at a plurality of preset time points in a preset time period, and arranging the network access electricity prices of each wind turbine generator set at the plurality of preset time points in the preset time period into network access electricity price input vectors according to the time dimension. And then, respectively calculating the incidence matrixes between the power input vector and the network-access electricity price input vector to obtain a plurality of operation index incidence matrixes. And then, respectively passing the operation index correlation matrixes through a double-flow network model comprising a first convolution neural network and a second convolution neural network to obtain a plurality of multi-scale index correlation feature matrixes, and passing the plurality of multi-scale index correlation feature matrixes through a decoder to obtain a plurality of decoding values for representing the operation economy index. And finally, sequencing the priority of each wind turbine generator based on the plurality of decoding values, and controlling the power of the wind turbine generator according to the priority.
In a specific embodiment of the present application, the calculating the correlation matrix between the power input vector and the network-entry power price input vector to obtain a plurality of operation index correlation matrices includes: and respectively calculating products between the transposed vector of the power input vector and the network-access electricity price input vector to obtain the operation index association matrixes.
In a specific embodiment of the present application, the passing the multiple operation index correlation matrices through a dual-flow network model including a first convolutional neural network and a second convolutional neural network to obtain multiple multi-scale index correlation feature matrices includes: using each layer of the first convolutional neural network model to respectively perform cavity convolution processing based on the first convolutional kernel, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transfer of the layer so as to output a first feature matrix by the last layer of the first convolutional neural network model; using each layer of the second convolutional neural network model to respectively perform hole convolution processing based on the second convolutional kernel, pooling processing along a channel dimension and nonlinear activation processing on input data in forward transfer of the layer so as to output a second feature matrix by the last layer of the second convolutional neural network model; and fusing the first feature matrix and the second feature matrix to obtain the multi-scale index association feature matrix.
In a specific embodiment of the present application, the fusing the first feature matrix and the second feature matrix to obtain the multi-scale index associated feature matrix includes: correcting the characteristic values of each position in the first characteristic matrix based on the second characteristic matrix to obtain a corrected first characteristic matrix; correcting the characteristic values of each position in the second characteristic matrix based on the first characteristic matrix to obtain a corrected second characteristic matrix; and calculating a weighted sum of the corrected first feature matrix and the corrected second feature matrix according to positions to obtain the multi-scale index associated feature matrix.
In a specific embodiment of the present application, the correcting, based on the second feature matrix, the feature values of each position in the first feature matrix to obtain a corrected first feature matrix includes: correcting the characteristic values of each position in the first characteristic matrix according to the following formula based on the second characteristic matrix to obtain a corrected first characteristic matrix;
wherein, the formula is:
wherein M is 1 And M 2 The first feature matrix and the second feature matrix, 
And->
The feature values of the (i, j) th positions of the first feature matrix and the second feature matrix, respectively, and +.>
And->
The mean value of all eigenvalues of the first eigenvalue matrix and the second eigenvalue matrix, log representing a logarithmic function value based on 2,
and (c) a eigenvalue representing the (i, j) th position of the corrected first eigenvector.
In a specific embodiment of the present application, the correcting, based on the first feature matrix, the feature values of each position in the second feature matrix to obtain a corrected second feature matrix includes: correcting the characteristic values of each position in the second characteristic matrix according to the following formula based on the first characteristic matrix to obtain a corrected second characteristic matrix;
wherein, the formula is:
wherein M is 1 And M 2 The first feature matrix and the second feature matrix,
and->
The feature values of the (i, j) th positions of the first feature matrix and the second feature matrix, respectively, and +.>
And->
The mean value of all eigenvalues of the first eigenvalue matrix and the second eigenvalue matrix, log representing a logarithmic function value based on 2,
and (c) a eigenvalue representing the (i, j) th position of the corrected second eigenvector. / >
In a specific embodiment of the present application, the passing the plurality of multi-scale index associated feature matrices through a decoder to obtain a plurality of decoded values for representing the running economy index includes: performing decoding regression on the plurality of multi-scale index associated feature matrices using a decoder to obtain the plurality of decoded values according to a formula:
wherein X is each of the plurality of multi-scale index associated feature matrices, Y is each of the plurality of decoded values, and W is a weight matrix.
In a specific embodiment of the present application, the prioritizing the wind turbine generator based on the plurality of decoding values and performing power control on the wind turbine generator according to the priorities includes: s210, collecting state information of wind power plant units in real time and real-time output power P of PCC points of wind power plant wf Power grid active power scheduling instruction P ref Wind power prediction value P pef The method comprises the steps of carrying out a first treatment on the surface of the S220, a unit classification unit is used for classifying the units into five types including a first type of marker post unit, a second type of on-line unit, a third type of fault unit, a fourth type of unit to be started and a fifth type of unit to be started according to collected state information of the units. S230, judging the priority of the wind farm unit according to the running economy index of the unit; s240, according to the power grid active power scheduling instruction P ref Predicted wind power value P pef Real-time output power P of wind farm PCC point wf The difference ΔP is calculated by the following formula
When DeltaP>0, the following step S250 is performed; when DeltaP<0, the following step S260 is performed; s250, combining the unit types, and respectively performing power increasing control according to the priority of the unit; s260, combining the unit types, and respectively performing power reduction control according to the priority of the unit; s270, the process returns to step S210, and the next control cycle is performed.
Here, it will be understood by those skilled in the art that the specific operations of the respective steps in the above-described multi-wind turbine generator system power control optimizing system have been described in detail in the above description of the multi-wind turbine generator system power control optimizing system with reference to fig. 1 to 3, and thus, repetitive descriptions thereof will be omitted.
Exemplary electronic device
Next, an electronic device according to an embodiment of the present application is described with reference to fig. 6.
Fig. 6 is a block diagram of an electronic device according to an embodiment of the present application.
As shown in fig. 6, the electronic device 10 includes one or more processors 11 and a memory 12.
The processor 11 may be a Central Processing Unit (CPU) or other form of processing unit having data processing and/or instruction execution capabilities, and may control other components in the electronic device 10 to perform desired functions.
In one example, the electronic device 10 may further include: an input device 13 and an output device 14, which are interconnected by a bus system and/or other forms of connection mechanisms (not shown).
The input means 13 may comprise, for example, a keyboard, a mouse, etc.
The output device 14 may output various information including a decoded value, a control result, and the like to the outside. The output means 14 may include, for example, a display, speakers, a printer, and a communication network and remote output devices connected thereto, etc.
Exemplary computer program product and computer readable storage Medium
In addition to the methods and apparatus described above, embodiments of the present application may also be a computer program product comprising computer program instructions which, when executed by a processor, cause the processor to perform the steps of a multi-wind turbine power control optimization method according to various embodiments of the present application described in the "exemplary methods" section of this specification.
The computer program product may write program code for performing the operations of embodiments of the present application in any combination of one or more programming languages, including an object oriented programming language such as Java, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computing device, partly on the user's device, as a stand-alone software package, partly on the user's computing device, partly on a remote computing device, or entirely on the remote computing device or server.
Furthermore, embodiments of the present application may also be a computer-readable storage medium, having stored thereon computer program instructions, which when executed by a processor, cause the processor to perform the steps of a multi-wind turbine power control optimization method according to various embodiments of the present application described in the above "exemplary methods" section of the present specification.
The computer readable storage medium may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may include, for example, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the foregoing. More specific examples (a non-exhaustive list) of the readable storage medium would include the following: an electrical connection having one or more wires, a portable disk, a hard disk, random Access Memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
The basic principles of the present application have been described above in connection with specific embodiments, however, it should be noted that the advantages, benefits, effects, etc. mentioned in the present application are merely examples and not limiting, and these advantages, benefits, effects, etc. are not to be considered as necessarily possessed by the various embodiments of the present application. Furthermore, the specific details disclosed herein are for purposes of illustration and understanding only, and are not intended to be limiting, as the application is not intended to be limited to the details disclosed herein as such.
The block diagrams of the devices, apparatuses, devices, systems referred to in this application are only illustrative examples and are not intended to require or imply that the connections, arrangements, configurations must be made in the manner shown in the block diagrams. As will be appreciated by one of skill in the art, the devices, apparatuses, devices, systems may be connected, arranged, configured in any manner. Words such as "including," "comprising," "having," and the like are words of openness and mean "including but not limited to," and are used interchangeably therewith. The terms "or" and "as used herein refer to and are used interchangeably with the term" and/or "unless the context clearly indicates otherwise. The term "such as" as used herein refers to, and is used interchangeably with, the phrase "such as, but not limited to.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent to the present application.
The previous description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.
Claims (7)
1. A multi-wind turbine power control optimization system, comprising:
the data monitoring and collecting module is used for obtaining output power values and network access electricity prices of each wind turbine generator at a plurality of preset time points in a preset time period;
the operation index association module is used for respectively arranging the output power values and the network access power prices of the wind turbines at a plurality of preset time points in a preset time period into a power input vector and a network access power price input vector according to a time dimension, and respectively calculating an association matrix between the power input vector and the network access power price input vector to obtain a plurality of operation index association matrices;
the operation index feature extraction module is used for enabling the operation index correlation matrixes to respectively pass through a double-flow network model comprising a first convolution neural network and a second convolution neural network to obtain a plurality of multi-scale index correlation feature matrixes, wherein the first convolution neural network uses a first convolution kernel with a first void rate, and the second convolution neural network uses a second convolution kernel with a second void rate;
The decoding module is used for enabling the multi-scale index association characteristic matrixes to pass through a decoder to obtain a plurality of decoding values used for representing the running economy index; and
and the control result generation module is used for sequencing the priorities of the wind power generation sets based on the decoding values and controlling the power of the wind power generation set according to the priorities.
2. The multi-wind turbine power control optimization system of claim 1, wherein the operation index correlation module is further configured to calculate products between transpose vectors of the power input vectors and the grid-connected power rate input vectors, respectively, to obtain the plurality of operation index correlation matrices.
3. The multi-wind turbine power control optimization system of claim 2, wherein the operation index feature extraction module comprises:
a first scale feature extraction unit, configured to perform, in forward transfer of layers, a hole convolution process based on the first convolution kernel, a pooling process along a channel dimension, and a nonlinear activation process on input data, respectively, using layers of the first convolutional neural network model, so as to output a first feature matrix from a last layer of the first convolutional neural network model;
A second scale feature extraction unit, configured to perform, in forward transfer of layers, hole convolution processing based on the second convolution kernel, pooling processing along a channel dimension, and nonlinear activation processing on input data, respectively, using layers of the second convolutional neural network model, so as to output a second feature matrix by a last layer of the second convolutional neural network model;
and the fusion unit is used for fusing the first feature matrix and the second feature matrix to obtain the multi-scale index association feature matrix.
4. A multi-wind turbine power control optimization system in accordance with claim 3, wherein said fusion unit comprises:
the first characteristic distribution correction subunit is used for correcting the characteristic values of all positions in the first characteristic matrix based on the second characteristic matrix to obtain a corrected first characteristic matrix;
the second characteristic distribution correction subunit is used for correcting the characteristic values of all positions in the second characteristic matrix based on the first characteristic matrix to obtain a corrected second characteristic matrix;
and the position-based fusion subunit is used for calculating a position-based weighted sum between the corrected first feature matrix and the corrected second feature matrix to obtain the multi-scale index associated feature matrix.
5. The multi-wind turbine power control optimization system of claim 4, wherein the first profile corrector subunit is further configured to: correcting the characteristic values of each position in the first characteristic matrix according to the following formula based on the second characteristic matrix to obtain a corrected first characteristic matrix;
wherein, the formula is:
wherein M is 1 And M 2 The first feature matrix and the second feature matrix,
and->
The feature values of the (i, j) th positions of the first feature matrix and the second feature matrix, respectively, and +.>
And->
The mean value of all eigenvalues of the first eigenvalue matrix and the second eigenvalue matrix, log representing a logarithmic function value based on 2,
and (c) a eigenvalue representing the (i, j) th position of the corrected first eigenvector.
6. The multi-wind turbine power control optimization system of claim 5, wherein the second characteristic distribution syndrome unit is further configured to: correcting the characteristic values of each position in the second characteristic matrix according to the following formula based on the first characteristic matrix to obtain a corrected second characteristic matrix;
Wherein, the formula is:
wherein M is 1 And M 2 The first feature matrix and the second feature matrix,
and->
The feature values of the (i, j) th positions of the first feature matrix and the second feature matrix, respectively, and +.>
And->
The mean value of all eigenvalues of the first eigenvalue matrix and the second eigenvalue matrix, log representing a logarithmic function value based on 2,
and (c) a eigenvalue representing the (i, j) th position of the corrected second eigenvector.
7. The multi-wind turbine power control optimization system of claim 6, wherein the decoding module is further configured to: performing decoding regression on the plurality of multi-scale index associated feature matrices using a decoder to obtain the plurality of decoded values according to a formula:
wherein X is each of the plurality of multi-scale index associated feature matrices, Y is each of the plurality of decoded values, and W is a weight matrix. />
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211543203.6A CN116151545A (en) | 2022-12-02 | 2022-12-02 | Multi-wind motor group power control optimization system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211543203.6A CN116151545A (en) | 2022-12-02 | 2022-12-02 | Multi-wind motor group power control optimization system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116151545A true CN116151545A (en) | 2023-05-23 |
Family
ID=86337996
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211543203.6A Pending CN116151545A (en) | 2022-12-02 | 2022-12-02 | Multi-wind motor group power control optimization system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116151545A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116578008A (en) * | 2023-06-06 | 2023-08-11 | 浙江汇驰厨房设备工程有限公司 | Control system and method for kitchen range hood |
| CN117913840A (en) * | 2024-03-19 | 2024-04-19 | 赛尔通信服务技术股份有限公司 | Energy-saving controller with peak and valley eliminating function and method thereof |
-
2022
- 2022-12-02 CN CN202211543203.6A patent/CN116151545A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116578008A (en) * | 2023-06-06 | 2023-08-11 | 浙江汇驰厨房设备工程有限公司 | Control system and method for kitchen range hood |
| CN117913840A (en) * | 2024-03-19 | 2024-04-19 | 赛尔通信服务技术股份有限公司 | Energy-saving controller with peak and valley eliminating function and method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116151545A (en) | Multi-wind motor group power control optimization system | |
| CN115759658B (en) | Enterprise energy consumption data management system suitable for smart city | |
| CN113157919B (en) | Sentence text aspect-level emotion classification method and sentence text aspect-level emotion classification system | |
| CN110852565B (en) | Grid frame planning method considering different functional attributes | |
| Yong | An improved artificial fish swarm algorithm for optimal operation of cascade reservoirs | |
| CN112560456A (en) | Generation type abstract generation method and system based on improved neural network | |
| CN116307624A (en) | Resource scheduling method and system of ERP system | |
| CN112819279A (en) | Planning evaluation method and system for expansion adaptability of distributed energy and power distribution network | |
| CN115456046A (en) | Wind driven generator speed and pitch changing control system and method based on active disturbance rejection control | |
| CN116624977B (en) | Building automatic control system and method based on artificial intelligence | |
| CN116404212B (en) | Capacity equalization control method and system for zinc-iron flow battery system | |
| CN115357065B (en) | Remote intelligent dehumidification control system and method for offshore wind turbine | |
| CN116151604A (en) | Office system flow analysis system and method under web environment | |
| CN115456256A (en) | Wind power plant reactive voltage control system and method based on self-adaptive control | |
| CN116483132A (en) | Coal flow control system and method based on drive motor current control coordination | |
| Lan et al. | Efficient converted spiking neural network for 3d and 2d classification | |
| Zhang et al. | Stochastic distributionally robust unit commitment with deep scenario clustering | |
| Wu et al. | t-SNE based on Sobol sequence initialized exchange market algorithm | |
| CN114021437A (en) | Wind power photovoltaic active scene generation method and device, electronic equipment and storage medium | |
| CN116561696B (en) | Multi-dimensional user adjustable load rapid aggregation method and system thereof | |
| Wu et al. | Hierarchical memory decoding for video captioning | |
| CN116794975B (en) | Intelligent control method and system for electric butterfly valve | |
| CN115912375B (en) | Low-voltage power supply compensation method and device and electronic equipment | |
| CN117293916B (en) | User-oriented power grid dispatching method and device and computing equipment | |
| CN117040030B (en) | New energy consumption capacity risk management and control method and system |
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 |